99448 Haiti Toward a New Narrative Systematic Country Diagnostic Raju Jan Singh Mary Barton-Dock Haiti Toward a New Narrative Raju Jan Singh Mary Barton-Dock © 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 findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 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: Singh, Raju Jan and Mary Barton-Dock. 2015. Haiti: Toward a New Narrative. Systematic Country Diagnostic. Washington, DC: World Bank. 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 official content or error in this translation. World Bank translation. The World Bank shall not be liable for any ­ 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 contained 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. Cover photo: © World Bank. Further permission required for reuse. Cover design: Florencia Micheltorena. Contents Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii About the Authors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Map. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Country Profile: What Makes Haiti Haiti ?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Trends and Profile in Poverty and Shared Prosperity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Trends and Drivers of Growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Sustainability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Priorities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1. Country Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Opportunities and a Vision. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 A Broken Social Contract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Vested Interests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Political Instability and Violence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Natural Disasters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Limited Data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 2. Trends and Profile in Poverty and Shared Prosperity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Trends. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Drivers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3. Trends and Drivers of Growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Trends. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Drivers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 4. Sustainability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Social Tensions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Macroeconomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Recent Progress in Poverty Reduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 5. Priorities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Prioritization Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Quantitative Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Qualitative Assessment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 iii Five Priority Areas for Policy Action. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Data Gaps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Appendix A: Price Comparison Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Appendix B: Bottlenecks and Correlates of Firm Productivity . . . . . . . . . . . . . . . . . . . . . . . . 95 Appendix C: Most Significant Data Gaps in Haiti . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 Boxes 1.1 Common Features of PIM in Donor-Dependent Countries. . . . . . . . . . . . . . . . . . . . . . . . 13 1.2 Product Market Concentration Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.1 Gender Inequalities in Haiti . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.1 Petrocaribe and Haiti. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 4.1 Drivers of Conflict in Haiti—An Empirical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 4.2 Electricité d’Haïti (EDH). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 5.1 Benchmarking the Drivers of Shared Prosperity: An Application to Haiti . . . . . . . . . . . 73 5.2 Conflict and Welfare Spending in Haiti: What Could We Learn from Cross-Country Evidence?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 5.3 Bank-Sponsored Competition of Academic Papers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 Figures 1.1 Inclusiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1.2 Competition Intensity and Extent of Market Dominance, 2014–15. . . . . . . . . . . . . . . . . 17 1.3 Business Risks Related to Weak Competition Policies (by Components). . . . . . . . . . . . . 18 1.4 Concentration Levels Based on HHI in the 18 Most Important Haitian Product Markets, 2012. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 1.5 Vulnerability Index, 2013 (Average Score of Susceptibility, Coping and Adaptive Capacity). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1.6 Migrants International Comparison, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1.7 Migrants by Destination Country, 2010. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1.8 Remittances, 2012. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 1.9 Foreign Flows, 1998–2012. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 1.10 Effects of Aid and Transfers on the Trade Balance, 1980–2013. . . . . . . . . . . . . . . . . . . . . 25 1.11 Imports of Goods and Services—LAC Region, 2011–13 . . . . . . . . . . . . . . . . . . . . . . . . . . 26 1.12 Merchandise Imports, Aid and Remittances, 2005–14. . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 1.13 Statistical Capacity Indicator, 2014 (0=Lowest, 100=Highest). . . . . . . . . . . . . . . . . . . . . . 27 2.1 Trends in Poverty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 iv Contents 2.2 Food Insecurity, 2012. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2.3 Change in Composition of Labor Market, Workforce Ages 15+, 2007–12. . . . . . . . . . . . 36 2.4 Composition of the Labor Market, Workforce Ages 15+, 2012. . . . . . . . . . . . . . . . . . . . . 36 2.5 Breakdown of Haiti’s Population by Labor Status (2012) . . . . . . . . . . . . . . . . . . . . . . . . . . 37 2.6 Born Elsewhere, 2011 (Total Population, Area of Living). . . . . . . . . . . . . . . . . . . . . . . . . . 38 2.7 Schooling of Adults Living Outside Department of Birth (15+), 2012. . . . . . . . . . . . . . . 38 2.8 Contribution to Extreme Poverty Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.1 GDP, 1970–2013 (1970 = 100) (Constant 2005). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.2 GDP per Capita, 1970–2013 (1970 = 100) (Constant 2005). . . . . . . . . . . . . . . . . . . . . . . . 44 3.3 Annual GDP Growth vs. Occurrence of Natural Disasters, 1971–2013. . . . . . . . . . . . . . 46 3.4 Annual GDP Growth vs. People Affected by Natural Disasters, 1971–2013 . . . . . . . . . . 46 3.5 Annual GDP Growth vs. Changes in Government, 1971–2013. . . . . . . . . . . . . . . . . . . . . 47 3.6 Economic Structure, 1970–2012. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.7 Decomposition of Value Added Growth by Sector, 1971–2013. . . . . . . . . . . . . . . . . . . . . 49 3.8 Urban Population, 1971–2013. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.9 Obstacles to Growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 3.10 Change in Governance Indicators, 2004–13. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.11 Credit by Sector, as of Q2 of 2014. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.12 Logistic Performance Index, 2014 (1=Lowest, 5=Highest). . . . . . . . . . . . . . . . . . . . . . . . . 54 3.13 Port Tariffs Estimated Cost per TEU, 2009. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 3.14 Electric Power Consumption, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 3.15 Informal Employment, 2012–22 (Working Age Population 15+). . . . . . . . . . . . . . . . . . . 56 4.1 Population Living in Flood Prone Areas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 4.2 Population Exposed to Hurricane Damage—High and Medium Intensity. . . . . . . . . . . 62 4.3 Political Violence, 2003–06. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 4.4 Criminal Activity, 2010–14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 4.5 Macroeconomic Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.6 Effective Exchange Rate, 2006–14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.7 Central Government Fiscal Balance, 2004–14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 4.8 Current Account Balance, 2004–14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 4.9 International Aid, 2008–25. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 4.10 Government Deposits, 2009–14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 4.11 Petrocaribe Financing, 2008–17. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 4.12 Project Activities Financed by Petrocaribe, 2008–13. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 4.13 Histogram of Annual Per Capita Consumption, 2012. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 5.1 Prioritization Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 5.2 Extreme Poverty Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 5.3 Economic Magnitude of Estimated Parameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 5.4 Life Expectancy at Birth, 2010. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 5.5 Cabinet Changes, 2003. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 5.6 Income Effects of Closing the Gap. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Contents v Maps 2.1 Extreme Poverty Rates by Department, 2012. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.1 Violence and Criminal Activity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Tables 1.1 Import Quotas for 19 Major Families, 1984–85. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 1.2 Frequency and Impact of Natural Disasters, 1971–2014. . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.1 Access to Basic Services—Coverage Rates (2001–12). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.2 Basic Sociodemographic and Socioeconomic Characteristics of Poor, Extreme Poor and Nonpoor Households, 2012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.1 Contributions to Growth (α = 40%) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.2 Haiti’s Governments, 1971–2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 vi Contents Acknowledgments We would like to thank the members of events across Haiti. In this regard, we wish to the Haiti Country Team from all Global express our gratitude to the ONPES, IHSI, Practices, CCSAs, IFC, and MIGA, as well as and MINUSTAH for making this possible. all the partners and stakeholders in Haiti This report also draws heavily on the recent- who have contributed to the preparation of ly-completed Poverty Assessment and the this document in a strong collaborative pro- ongoing Public Expenditure Review. In cess. We are very grateful for the generosity ­ addition, a number of background papers exhibited in providing us with substantive have been written by the Country Team on inputs, knowledge and advice, particularly specific themes. In this respect, the Team given the time limitations. The table below is appreciative to Prof. James Robinson identifies the full list of team members who (Harvard University) and Prof. Suresh Naidu have contributed their time, effort and ex- (Columbia University) for discussions on the pertise, and their affiliations. role of Haitian business elites in Haiti’s de- The team was led by Raju Jan Singh velopment, and particularly to Lauren Young (Program Leader, LCC8C). The work was (Columbia University) for presenting pre- carried out jointly with the IFC (Sylvain liminary results of their research at the Kakou) and MIGA (Petal Hacket) under World Bank. We wish also to thank Prof. the overall guidance of Mary Barton-Dock Cristina Bodea and Masaaki Higashijima (Special Envoy for Haiti, LCC8C) and Jun (both from Michigan State University) for Zhang (Senior Regional Manager for the their work on public spending and conflict. Caribbean, IFC). The analytical work was validated by We wish to thank for their helpful sugges- wide consultations. In this regard, we wish to tions and insights the Systematic Country thank Bernard Craan, Executive Director of Diagnostic (SCD) peer reviewers, Nancy the Private Sector Economic Forum for orga- Benjamin, Senior Country Economist nizing a meeting with the members of his as- (GMFDR), Oscar Calvo-Gonzalez, Program sociation; Delphine Colbeau, UNDP, who Leader (LCC2C), and Philip Keefer, coordinated a workshop on violence with the Principal Advisor (IDB), as well as Rolf Parta heads of all UN agencies present in Port-au- for moderating our two-day Country Team Prince; Gilles Damais, IDB, for inviting us to retreat, and Augusto de la Torre (LCR Chief present our work at one of its Wednesdays Economist) and Daniel Lederman (LCR of Reflection with members of the academia, Deputy Chief Economist) for their advice civil society, private sector, and Haitian ad- throughout the stages of the SCD process. ministration; Kesner Pharel to provide us The SCD not only draws on existing liter- with the opportunity to share our ideas on ature from within and outside the Bank, but his TV show; and Mariam Yazdani from benefitted from the results of the recent Vivario for interesting discussions on gang household survey (ECVMAS 2012), as well dynamics. We are also very grateful to as access to the MINUSTAH data on crime ­Hans-Muller Thomas, Kore Fanmi National vii Coordinator, Germanite Phanord, Kore (President, Haiti Conciliation and Fanmi Regional Coordinator, and Jean Arbitration Chamber), Charles Cadet Raynold Saint Hilaire, Chief Social Worker (Ministry of Economy and Finance, Haiti), for organizing our field trip to the com- Kathleen Dorsainvil (American munes of Boucan Carré and Thomassique, University), Fritz Jean (President, North Central Plateau Department, as well as to the East Chamber of Commerce, Haiti), Eddy National Association of Haitian Professionals Labossière (President, Association of for having invited us at Harvard University Haitian Economists) and Guy Pierre for their Third Annual Conference with the (Autonomous University of Mexico), and Haitian diaspora. from the World Bank: Dorsati Madani Finally, we wish to thank Ricardo (Senior Economist, GMFDR), Gael Augustin (Dean, School of Economics— Raballand (Senior Public Sector Specialist, Notre Dame University of Haiti), Raulin GGODR), and Erik von Uexkull (Country Cadet (Dean, School of Economics Dean— Economist, GMFDR). A special thanks Quisqueya University), Fritz Deshommes should be given to all our participants from (Deputy Dean of the State University of Haiti, Canada and the United States, but Haiti), Amos Durosier (Dean, Advanced particularly to our laureates: Jose Minerve Commercial and Economic Studies Cayo (State University of Haiti), Jean Institute), and Lionel Metellus (Dean, Ribert Francois (State University of Haiti), Quisqueya American University Institute) Carl-Henri Prophète (Centre d’Études for setting up our competition of academic Diplomatiques et Internationales, CEDI), papers (“The Twin Goals Awards”), as well Alendy Saint-Fort (FDSE), Jean Carrington as all the members of our Selection Saintima (IHECE), and Guimard Syvrain Committee from Haiti: Henri Bazin (CTPEA). viii Acknowledgments Haiti SCD team Global practice/cross-cutting area Team members Agriculture Pierre Olivier Colleye, Katie Freeman, Christophe Grosjean, Eli Weiss Communication Christelle Chapoy, Berdine Edmond Education Melissa Adelman, Juan Baron, Axelle Latortue Energy and extractives Susana Moreira, Remi Pelon, Frederic Verdol Environment Nyaneba Nkrumah Finance and markets Juan Buchenau, Caroline Cerruti Governance Alexandre Berg, Mamadou Deme, Onur Erdem, Sheila Grandio, Fabienne Mroczka Haiti CMU Mary Barton-Dock, Pierre Bonneau, Gabrielle Dujour, Nellie Sew Kwan Kan, Michelle Keane, David Lighton, Deo Ndikumana, Raju Singh, Kanae Watanabe, Paula White Health, nutrition and population Eleonora Cavagnero, Sunil Rajkumar IFC Ary Naim, Sylvain Kakou, Lina Sun Kee, Jean Francois Pean, Frank Sader, Jun Zhang Macroeconomics and fiscal Kassia Antoine, Calvin Djiofack, Evans Jadotte, Julie Lohi, Sandra Milord, management Konstantin Wacker MIGA Petal Hacket Poverty Facundo Cuevas, Federica Marzo, Aude-Sophie Rodella, Thiago Scot Social protection Lucy Bassett, Carine Clert, Maki Noda Social, urban, rural and resilience Ali Alwahti, Paul Blanchard, Sylvie Debomy, Sergio Dell’Anna, Joan Fomi, Van Anh Vu Hong, Oscar Ishizawa, Peter Lafere, Michel Matera, Bernhard Metz, Claudia Soto Orozco, Rafael Van der Borght, Gaetano Vivo, Javier Sanchez- Reaza, Alys Willman Trade and competitiveness Babatunde Abidoye, Massimiliano Cali, Emiliano Duch, Tanja Goodwin, Maria Kim, Martha Licetti, Siobhan Murray, Georgiana Pop, Lucia Jimena Villaran, Joaquin Zentner Transport and ICT Malaika Becoulet Water and sanitation Jean-Martin Brault Acknowledgments ix About the Authors Raju Jan Singh is the program leader for fiscal decentralization and public finance, Haiti, leading and overseeing the World Bank’s banking, trade, and remittances. Raju holds work on economic policy, private sector de- a master’s degree and a doctorate in eco- velopment, and education and social pro- nomics from the Graduate Institute of Inter- tection, and was previously sector leader national Studies in Geneva. and lead economist on Central ­ African states, stationed several years in Yaoundé, Mary Barton-Dock is the special envoy and Cameroon. Prior to joining the World Bank, director for Haiti. Prior to taking this posi- Raju was working as a senior economist and tion, she was the director of climate change mission chief at the International Monetary and environment for the World Bank. She Fund, where he held positions in the Fiscal has also served as the country director Affairs, Asian and Pacific, and African De- for Cameroon, Chad, the Central African partments, working on a wide range of Republic, Equatorial Guinea, Gabon, and countries and leading missions to China, São Tomé and Principe. Prior to becoming Cyprus, and Tonga. He has been an advisor a country director, she was the manager of in the Swiss Executive Director Office, and the World Bank’s Agriculture, Environment, worked at the Swiss ­ Finance Administration and Social Development programs in West in Bern, as well as at Lombard Odier & Cie Africa. In addition, she has been the World (private banking) in Geneva. He has also Bank’s resident representative in Chad, and been a consultant for the Swiss Agency for the team leader from programs in Southern Development and Cooperation, working Africa. Prior to joining the Africa region, with the central banks of Rwanda and Tan- she also worked in South East Asia, and she zania, and has taught at the Graduate Institute started her career with the World Bank of International Studies in Geneva. He has working on Bolivia. Mary holds a master’s published on a wide set of issues, including degree in public policy from Harvard. x Abbreviations ACD Armed Conflict Dataset ACLED Armed Conflict Location & Event Data ASCUYDA Automated System for Customs Data BMPAD  Bureau de Monétisation du Programme d’Aide au Développement (Bureau of Monetization of Development Aid Programs) CCRIF Caribbean Catastrophe Risk Insurance Facility CPIA Country Policy and Institutional Assessment CSCCA  Cour Supérieure des Comptes et du Contentieux Administratif (Supreme Court of Accounts and Contentious Administrative Proceedings) DHS Demographic Health Survey DINEPA  Direction de l’Eau Potable et de l’Assainissement (Water and Sanitation Authority) DR Dominican Republic ECVMAS Enquête des Conditions de Vie des Ménages (Household survey) EDE PÈP Social Assistance Program “Help the People” EDH Electricité d’Haiti (Public Electricity Company) EM-DAT Emergency Events Database FDI foreign direct investment FSAP Financial Sector Assessment Program GCI Global Competitiveness Index GDP gross domestic product HELP Haiti Economic Lift Program HHI Herfindahl-Hirschman Index HIPC highly indebted poor country HNP Haiti National Police HS Harmonized Coding System HTG Haitian gourde IFC International Finance Corporation IHSI Institut Haïtien de Statistique et d’Informatique (Haiti’s Statistical Institute) IICA Inter-American Institute for Cooperation on Agriculture IMF International Monetary Fund IPPs independent power producers LAC Latin America and Caribbean LIC low income countries LPI Logistics Performance Index LSCI Liner Shipping Connectivity Index MDG Millennium Development Goals MDRI Multilateral Debt Relief Initiative xi MEF Ministry of Economy and Finance MENFP Ministère de l’Education Nationale (Ministry of Education) MINUSTAH United Nations Stabilization Mission in Haiti MSMEs micro, small, and medium enterprises MSPP  Ministère de la Santé Publique and de la Population (Ministry of Public Health and Population) NEER nominal effective exchange rate NGOs non-governmental organizations ODA official development assistance OECD Organization for Economic Cooperation and Development ONPES Observation Nationale de la Pauvreté et l’Exclusion Sociale PARDH  Plan d’Action pour le Relèvement et le Développement d’Haiti (Action Plan for the Recovery and Development of Haiti) PDNAs Post-Disaster Needs Assessments PIM public investment management PIP Public Investment Program PIU Project Implementation Units POVCALNET Online Poverty Analysis Tool—World Bank PPP public-private partnership PSDH  Plan Stratégique de Développement d’Haiti (Strategic Plan for Development of Haiti) REER real effective exchange rate PRSP Poverty Reduction Strategy Paper SAM Social Accounting Matrix SCD Systematic Country Diagnostic SCI Statistical Capacity Indicator TEU twenty foot equivalent units TFP total factor productivity UN United Nations WDI World Development Indicators WEF World Economic Forum WHO World Health Organization xii Abbreviations Map xiii 7 3° W 7 2° W HAITI AT LA N T I C OCEAN This map was produced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of The World Bank SELECTED CITIES AND TOWNS Île de la Tortue Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. Palmiste DEPARTMENT CAPITALS 20°N 20°N Port-de-Paix ge NATIONAL CAPITAL a ss RIVERS Pa To d MAIN ROADS Môle St.-Nicolas NORD - OUE ST Monte Les ar Christi dw Cap-Haitien Tro RAILROADS is in W Limbé DEPARTMENT BOUNDARIES Gros-Morne Fort-Liberte Baie de NORD To Santiago INTERNATIONAL BOUNDARIES Trou- Ferrier Henne Grande Rivière du-Nord du Nord 74°W Ennery Saint- N O R D - ES T Gonaives Raphaël Saint Michel C n de l'Attalaye e tr Gu Golfe de a ay l am la Gonâve Pl pu at o ARTIBONITE ea Maïssade HAITI u Hinche DO MI N I C A N Ar Verrettes on CENTRE tib ite 19°N Lac de Péligre REPUB LI C La Cayenne Île de la Gonâve Anse-à-Galets Mirebalais To San Juan Pointe-à-Raquette Grande Jeremie Cayemite Roseaux Croix des PORT-AU-PRINCE Bouquets GRANDE- Anse d'Hainault ANSE Léogâne Étang Lago To if de la H o tt e NIPPES OUEST Saumâtre Enriquillo Barahona M a ss Miragoâne Miragoâne Petit- Goâve Chaine de la Selle (2680 m ) Les Anglais Camp-Perrin SUD Vieux Bourg d'Aquin SUD - EST Belle- Anse Thiotte Jacmel Côtes-de-fer Marigot Les Cayes Port-Salut Île à Vache 18°N 18°N 0 10 20 30 40 Kilometers To C a r i b b e an Sea Oviedo 0 10 20 30 Miles 74°W 7 3° W 7 2° W Executive Summary Haiti has a vision to become an emerg- This Systematic Country Diagnostic ing economy by 2030. Haiti’s geography, (SCD) seeks to identify the most important resources, and history provide it with oppor- ­ constraints to and opportunities for inclusive tunities. The country has comparative advan- and sustainable growth in Haiti. To identify tages, including its proximity and access to the key constraints to Haiti’s growth and major markets; a young labor force and a shared prosperity, an extensive review of the dynamic diaspora; and substantial geo- ­ literature (from both within and outside the graphic, historical, and cultural assets. Areas World Bank) was first carried out. Economic of economic opportunity for Haiti include and sector work on Haiti produced in 1980s agribusiness, light manufacturing and tour- and the early 1990s had already identified ism. Building on these opportunities, the most of the country’s challenges and demon- Government of Haiti issued in May 2012 a strated that better functioning institutions, Strategic Development Plan (PSDH), aiming stronger human capital, and improvements at building a new modern, diversified, resil- in infrastructure were all needed for Haiti’s ient, competitive and inclusive economy, re- economic growth and shared prosperity. spectful of its environment and in which Rather than listing these again, this report people’s basic needs are met. This objective attempts to provide some prioritization would require ambitious double digit growth and identify the most binding constraints, rates, a significant break from the past, based both quantitatively and through a series on an expansion of agriculture, construction, of consultations with stakeholders and the manufacturing, and tourism. country team. Overall, Haiti’s growth performance in the last four decades has been disappointing, however, and poverty remains endemic. A Country Profile: What history of vested interests, political instabil- ity, and natural disasters has prevented the Makes Haiti Haiti ? country from realizing its aspirations, trap- A social contract is missing between the ping the country in a low equilibrium and State and its citizens. While overall income keeping it as one of the poorest and least growth is a necessary condition for increas- equal countries in the world. GDP per capita ing shared prosperity, it is not sufficient. fell by 0.7 percent per year on average be- Growth that is inclusive of the poor requires tween 1971 and 2013. As a result, in 2012 additional mechanisms such as a pro-poor 59 percent of Haitians remained poor and fiscal regime, as well as targeted social pro- 24 percent suffered from extreme poverty, grams and expenditures, not only to redis- indicating that almost 6.3 million Haitians tribute resources towards the poor but also could not meet their basic needs and 2.5 mil- more importantly to ensure that the less lion could not even cover their food needs. well-off are an integral part of the process Executive Summary 1 and that opportunities improve for all. The structure of the private sector shows Previous reports have noted, however, that signs of high degrees of concentration, Haiti has never had a tradition of providing hampering the entry of new actors and re- services to the population or creating an en- sulting in high prices for consumers. From vironment conducive to sustainable growth. the beginning of the twentieth century, au- Haiti’s tax system generates limited resources tocratic leaders in Haiti have traded politi- for the government and tends to be regres- cal support from the elite for economic sive. Furthermore, public spending in health, advantages to this elite. Though publicly education, and social protection remains available information on privately held limited, constraining the government’s abil- businesses is limited, many of the same ity to provide services and offer equal oppor- families who dominated the Haitian econ- tunities to its citizens. In the absence of omy during the era of Duvalier in the 1970s government, basic services such as health and the 1980s seem to remain in control of and education are mainly provided by non-­ large segments of the economy today, re- government actors, placing a substantial fi- sulting in high concentration in a number nancial burden on households and delivering of key industries, distorted competition, achievements closely linked with household and non-transparent business practices in income. many instances. Several of the most impor- The reliance on non-government actors tant food products in the Haitian consump- has also weakened public investment tion basket are sold in concentrated management. Haiti’s public investment man- ­ markets, and a preliminary analysis indi- agement exhibits a number of distinctive fea- cates that the prices of these products are tures and practices common to countries on average about 30–60 percent higher in that are aid-dependent, including weak Haiti than in other countries from the re- appraisal capacity and reliance on donors ­ gion. This translates into limited opport­ to design good projects, hampering the unities for a substantial expansion of the ­ effective use of public resources. Sectoral formal private sector across most sectors. strategies to guide the prioritization of Few of Haiti’s established private firms have projects are lacking. This leads to a Public ­ modern capital and governance structures Investment Program composed of projects with professional management, limiting that are neither fully assessed nor prioritized. their access to long-term financing. Furthermore, there is no effective ex-ante Political violence has occurred regularly control on disbursements based on the phys- throughout Haiti’s history, leading to insta- ical progress of projects against plans. While bility. At Independence in 1804, Haiti was progress is being made in fiscal reporting by at the forefront of history, being the first na- rolling out the use of a single treasury ac- tion to abolish slavery. Since then, however, count, domestically-funded capital expendi- with some exceptions such as the 30-year tures are not yet properly accounted for, period of autocratic rule under Francois ­ tracked and reported, creating an environ- Duvalier (Papa Doc) and his son Jean- ment conducive to a lack of transparency Claude Duvalier (Baby Doc) (1957–86), and accountability. Haiti has known a succession of short-lived 2 Executive Summary governments. Lacking sufficiently long peri- now seem to do so because they cannot find ods of stability, the country has struggled work opportunities in Haiti. This large dias- to develop the institutional mechanisms pora is a significant source of remittances: and policy fundamentals essential to eco- remittances received by Haiti are the highest nomic development and the rule of law. among Latin America and Caribbean (LAC) Disenfranchised and without effective chan- countries in terms of GDP and the fourth nels to voice needs and demands, citizens highest in the world in terms of export have taken to the streets in protest, some- earnings. times violently. While violent airing of griev- ances in Haiti’s early history generally took place in the rural areas, contemporary unrest Trends and Profile in tends to break out in the cities, reflecting the Poverty and Shared country’s demographic evolution and urban- ization over the years. Against this backdrop, Prosperity the post-earthquake period has been com- While remaining high, poverty has de- paratively stable. clined in Haiti. Recent findings indicate that Furthermore, the Haitian population is extreme poverty has declined in Haiti from one of the most exposed in the world to 31 percent of the population in 2000 to 24 ­ natural disasters—hurricanes, floods and percent in 2012. Progress was mainly con- earthquakes. Between 1971 and 2013, Haiti’s centrated in urban areas, however. This trend economy has been subjected to natural di- is confirmed by both monetary and non- sasters almost every year with adverse effects monetary poverty indicators, with the big- on growth. The country has a higher number gest non-monetary progress recorded in of disasters per km2 than the average of education. All school-age children go to the Caribbean countries. In 2008, tropical school in about 90 percent of the households storms and hurricanes caused losses esti- compared to about 80 percent in 2001. mated at 15 percent of GDP. The earthquake Immunization rates are also up. on January 12, 2010 killed 220,000 people, Recent evidence suggests that this decline displaced 1.5 million people, and destroyed in extreme poverty was driven by labor the equivalent of 120 percent of GDP. income, private transfers, and aid. Non- ­ In this unfavorable environment, migra- agricultural labor income increased by about tion has become a key avenue for Haitians 3½ percent on average per year, especially seeking a better life. Substantial internal mi- among men, with expansions in construc- gration is taking place, particularly from tion, telecommunication and transport, all rural to urban areas, as people seek better concentrated in urban areas. Formal employ- economic opportunities and better services. ment remains small (13 percent of the labor In addition, for both political and economic force) with agriculture and urban informal reasons, large numbers of Haitians have emi- sectors still providing most of the employ- grated throughout the twentieth century ment with about 40 percent and 47 percent building an important diaspora. A vast ma- of the labor market, respectively. Workers’ jority of Haitians who continue to emigrate transfers from abroad have represented more Executive Summary 3 than a fifth of Haiti’s GDP in recent years, a period of intense political instability in and the percentage of households receiving Haiti. Between 1986 and 2014 the country private transfers (domestic transfers or re- had 18 changes of president and important mittances from abroad) in Haiti increased changes in regime. Such political instability from 42 percent to 69 percent between 2000 has often been accompanied by violence and and 2012. Furthermore, the 2010 earthquake a continuous weakening of state institutions, resulted in unprecedented aid flows in the the rule of law, and the investment climate, form of money, goods and services. These undermining investor confidence. external flows have also contributed in re- Uncertainty as to whether investors can ob- ducing poverty over the period, especially tain returns from their investments rep- in the metropolitan area which attracted resents one of the main constraints to growth most of the assistance (in large part because in Haiti. Political instability has also resulted Port-au-Prince was hit hardest by the in a trade embargo in the first half of the earthquake). 1990s that crippled private sector activities. Haiti’s business environment is hampered by institutional weaknesses. Although gover- Trends and Drivers of nance indicators have improved, Haiti still ranks lowest in the region in control of Growth corruption or government effectiveness. ­ Overall, Haiti’s growth performance in Efficient mechanisms for international the last four decades has been disappointing. ­ arbitration and mediation are lacking. From 1971 to 2013, GDP growth averaged Guarantees for the protection of investors’ 1.2 percent a year, much lower than the aver- private property rights are insufficient. Legal age of the LAC region (3.5 percent) and the and regulatory frameworks are fragmented average of economies at the same level of de- and dysfunctional. In particular, a real prop- velopment (3.3 percent). The few periods of erty cadaster and land registry system is positive growth were short-lived, often fol- needed. Furthermore, access to finance is lowed by a contraction in economic activity. challenging for both, households and me- Furthermore, in light of the country’s impor- dium- and small-sized enterprises. tant demographic growth, the level of GDP Haiti’s infrastructure also falls short. per capita even fell by 0.7 percent per year Island economies are extremely dependent on average between 1971 and 2013. Whereas on the quality, frequency and cost of the low income countries (LICs) have on average means of transport that link them to markets seen their GDP per capita taking off since which represent both outlets for their prod- the mid-1990s, Haiti was left behind. ucts and supply sources for the needed im- Political instability and natural disasters ported goods. The efficiency and effectiveness have taken a toll on growth. Despite invest- of transport, whether by road, by sea, or by ment and increases in the labor force, Haiti’s air, therefore strongly affects their competi- growth performance has remained weak, re- tiveness. The quality of transport and logistics flecting the natural disasters and political in- services in Haiti is low, however, with large stability the country has experienced. The parts of the territory still poorly connected. departure of Jean-Claude Duvalier initiated Recent evidence indicates, for instance, that 4 Executive Summary only 40 percent of people living in rural areas labor market and, more importantly, for have access to all-weather roads. Haiti is also their ability to absorb post-school training less integrated into the global shipping line either on or off the job, and to adapt to network than many developing countries, changing job requirements. and the costs of loading and unloading a The constraints discussed above seem standard container at Port-au-Prince are by to be particularly present in the agriculture far the highest of the Caribbean ports. and the non-agriculture informal sector, In particular, the provision of electricity where most Haitians are likely to continue is problematic. The supply of electricity cov- working for the foreseeable future. The for- ers only a small proportion of the country, mal manufacturing and service sectors have with rural areas being particularly neglected. the potential to be an important source of Only 35 percent of Haitians have access to employment, but because they hire such a electricity (11 percent in rural areas). The small share of the labor force, even with very provision of electricity in Haiti is also unsta- high growth rates, they will not be able to ble and there are frequent power cuts and absorb more than a fraction of the new en- surges, which can result in serious damage to trants. The agriculture sector in Haiti has industrial equipment. Business owners cite been declining for many years, the result of the lack of reliable electricity supply as the neglected rural infrastructure, weak research most binding constraint to private sector de- and extension, poorly defined land tenure, velopment. Despite this poor service, the limited access to credit, and under-investment cost of electricity is among the highest in the in human capital. With regard to the infor- region. Furthermore, per capita consump- mal non-agriculture sector, empirical evi- tion of electricity in Haiti is substantially dence suggests that a little investment in lower than in other Caribbean countries, for years of education or remedial training for example, it is only two percent of the level skills acquisition among people who have observed in the Dominican Republic. been out of the school system for a long Surveys also point to the lack of qualified time, as well as improved access to inputs, labor. One major constraint faced by enter- such as electricity and water, are associated prises in Haiti is the difficulty to find techni- with substantial increases in income. cians that are well qualified, particularly in new technologies and trained managers for middle management. This forces the country Sustainability to position itself as a low cost producer for Urbanization is increasing Haiti’s vul- goods and services requiring limited skills. nerability to natural disasters. As a safety The level of education among the adult pop- valve, migration probably contains poverty ulation remains low: 45.7 percent of the and provides relief to many families through adult population (60.5 percent of house- private transfers. However, moving to urban holds’ heads) have never attended school or areas is putting pressure on the environment, have not completed primary education. Not and leading to the settlement of large popu- mastering basic skills such as literacy and lations in vulnerable places. Urban areas numeracy when starting work represents a have experienced an explosive expansion major impediment for their insertion in the in the past two decades with a near total Executive Summary 5 absence of urban planning and growth regu- aftermath of the 2010 earthquake. The real lation. This unplanned urban growth has effective exchange rate remained aligned created vulnerabilities for both the physi- with fundamentals. cal environment and those who live in it. Budget constraints are becoming tighter, Furthermore, settlers tend to occupy areas however, and could put at risk some of the that are not only prone to risk from disasters, recent progress achieved in human develop- but are also sometimes nature’s protection ment. The country has taken the opportunity from risks such as coastal ecosystems. Slow of its greater borrowing space to contract degradation of the environment in such loans on concessional terms with Republica areas in turn increases the country’s vulnera- Bolivariana de Venezuela through the bility as a whole to natural disasters. Petrocaribe agreement. While the availability Rapid urbanization could also be feeding of this concessional financing has allowed greater violence. Youth violence is often ex- higher levels of public investments and pressed as a reaction to perceived social and spending on social programs, it has also al- economic exclusion—grievances that can lowed a widening of Haiti’s twin fiscal and easily be mobilized by political interests. current account deficits, increasing the coun- Rapid migration to urban neighborhoods, try’s level of indebtedness, although keeping associated with poverty, unemployment, po- it at a sustainable level. The decline in aid litical and social marginalization, limited and in international oil prices (which affect access to services, and weak governance, ­ the availability of Haiti’s concessional re- could fuel the formation of gangs that engage sources) is reducing the country’s fiscal at-risk youth and draw them in. Evidence space. These tighter constraints could put at suggests that migration is associated with risk some of the recent progress achieved in greater violence, as social ties erode. New poverty reduction and human development, migrants are likely to have less access to local making the country’s balancing act between social networks which provide economic and developmental needs and fiscal sustainability social support. These grievances have also even more challenging. contributed to intergenerational conflict, un- dermining social cohesion and hindering the capacity of many communities for produc- Priorities tive collective action. A quantitative approach would suggest Macroeconomic stability has been main- a policy mix including higher investment in tained in the aftermath of the earthquake. human and physical capital, which would Haiti’s macroeconomic environment has sig- also support greater political stability. Faster nificantly improved over the past decade. economic growth alone will not be enough Inflation was brought down to single digits to bring significant improvements in the liv- and international reserves built up to ade- ing standards of most Haitians. Without any quate levels. Between 2009 and 2011, Haiti’s change in the income distribution, reducing stock of external debt drastically shrank fol- extreme poverty to 3 percent by 2030 would lowing debt relief. In addition to benefiting require a two- to three-fold acceleration in from the HIPC and the MDRIs, Haiti re- Haiti’s growth rate with respect to its best ceived additional debt cancellation in the performing years: a very ambitious outcome. 6 Executive Summary Policies to ensure more inclusiveness are • State capacity needs to be strengthened. thus needed to allow the incomes of the bot- Poor governance and the need to tom 40 percent to grow faster than the aver- strengthen the State were generally viewed age. In this regard, estimations suggest that as the most important constraints. The the bottom 40 percent of the income distri- lengthiness of the process should not bution would see their incomes grow faster deter immediate action, but call for more than the average if Haiti were to invest in realistic expectations. In this regard, many health, education, and infrastructure, as well stakeholders perceived that donors by- as achieve greater political stability. In turn, passing the Haitian government in favor simulations for Haiti indicate that political of non-State actors, while providing per- stability could be increased—or more haps a quick and short-term response, broadly conflict risk reduced—through continued to undermine the public sector higher welfare spending (education, health, and could not bring a long-term solution and social protection). to the lack of government capacity. Qualitative research based on consulta- • Restoring the social contract was flagged tions brought additional important insights. as particularly important. Most modern Broad consultations were carried out with democracies use fiscal revenue to finance stakeholders and the country team. In this the State, regulate private activity, and re- regard, the general story line was validated, distribute wealth. In return for these reve- but in addition the consultations highlighted nues, the State is expected to provide key that: services and a level playing field for the private sector. In Haiti, both sides of this • Maintaining the stability of the macroeco- contract are very weak—few people or nomic environment called for immediate businesses pay taxes, and the State strug- attention. The decline in donor assistance—­ gles to provide services and appropriate returning to pre-earthquake levels—and regulation. lower availability of concessional financ- • Barriers to market entry need to be ing following the drop in international oil removed, competition enhanced, and ­ prices raised concerns about the sustain- small private sector actors supported. ability of many social programs and of the Representatives of the established private recent progress in improving access to sector recognized the lack of competition some basic services. In this regard, greater existing in Haiti’s economy, a reflection mobilization of own fiscal revenue and in their view of the country’s unfavorable improved efficiency in public expenditure business environment. In this regard, management were seen as critical. civil society expressed concern about the • Turning to more medium-term chal- higher prices and the possible lack of lenges, progress needs to be made simul- ­ innovation that limited competition taneously on several fronts. For Haiti to implied. break from its past and achieve meaning- • There was not only a need for more jobs, ful progress towards poverty reduction, a but also for better jobs. It was pointed piecemeal approach to reform focusing on out that the private sector in Haiti one sector only would not work. was not only limited to a couple of large Executive Summary 7 enterprises. Most people in Haiti were oc- Policy packages—or buckets—rather cupied in a less visible private sector: in than independent activities in sectors were the agriculture and informal sectors, two thus recommended. The SCD points out sectors that had been neglected so far. five broad themes around which activities • The importance of reducing vulnerabilities need to be organized in order to ignite a was also stressed. In the face of recurring process whereby Haiti could set itself on a shocks and vulnerability, better risk man- new development path: (a) balancing mac- agement strategies and better targeting in roeconomic stability with developmental social protection should be a priority to needs; (b) improving statistics and analyt- protect households and individual liveli- ics; (c) creating greater economic opportu- hoods. Disaster risk management should nities and better jobs, including through be strengthened. Risks need to be identified infrastructure and human capital; (d) (re) and understood better, as well as reduced building the social contract; and (e) re- in the future through greater awareness ducing vulnerabilities and building resil- when investing. Financial protection needs ience. Progress on all these themes is to be designed in case risks materialize. needed simultaneously. In light of the • Finally, the consultations made it clear tighter budget constraints facing the gov- that there was limited evidence to inform ernment, maintaining the stability of the planning and targeting of programs and macroeconomic environment, and im- policies. Discussions with stakeholders, as proving knowledge and statistics to in- well as discussions with the country team crease the effectiveness of public policy confirmed that it is difficult to carry out (including more transparent fiscal report- effective development policies without ap- ing) call more particularly for immediate propriate statistics or analytics. attention. 8 Executive Summary 1. Country Profile Opportunities and a areas of work for the recovery and develop- ment of Haiti: territorial reform, economic Vision reform, social reform and institutional re- Haiti’s geography, resources, and his- form. The Plan aims at building a new mod- tory provide it with opportunities. Haiti oc- ern, diversified, resilient, competitive and cupies the western, smaller portion of the inclusive economy, respectful of its environ- island of Hispaniola, which it shares with the ment and in which people’s basic needs are Dominican Republic. Both by area and pop- met. This objective would require ambitious ulation, Haiti is the third largest Caribbean na- double digit growth rates, a significant break tion (after Cuba and the Dominican Re­ public), from the past, based on an expansion of agri- with 27,750 square kilometers and an esti- culture, construction, manufacturing, and mated population of 10.4 million people. The tourism. country has comparative advantages, includ- Overall, Haiti’s growth performance in ing its proximity and access to major mar- the last four decades has been disappoint- kets; a young labor force and a dynamic ing, however, and poverty remains endemic. diaspora; and substantial geographic, histori- A history of vested interests, political insta- cal, and cultural assets.1 There is a pent-up bility, and natural disasters has prevented demand and many untapped markets that the country from realizing up to now its as- the private sector can explore. Areas of eco- pirations, trapping it in a low equilibrium nomic opportunity for Haiti include in agri- and keeping it as one of the poorest and business, light manufacturing and tourism. most unequal countries in the world. GDP Furthermore, Haiti possesses mineral poten- per capita fell by 0.7 percent per year on av- tial. According to the World Economic erage between 1971 and 2013. As a result, Forum, Haiti’s economic fundamentals could the overall poverty headcount amounts to allow the country to become a vibrant econ- about 59 percent and extreme poverty to omy and grow by 6–8 percent a year if ade- 24 percent in 2012, indicating that almost quate policies were in place World Economic 6.3 million Haitians cannot meet their basic Forum (2011). needs and 2.5 million cannot even cover Building on these opportunities, the their food needs. country has a vision to become an emerging economy by 2030. In May 2012, the Govern- ment of Haiti issued a Strategic Development A Broken Social Plan (PSDH), building on the Action Plan for the Recovery and Development of Haiti Contract (PARDH) prepared following the 2010 While overall income growth is a neces- earthquake. The PSDH details the PARDH’s sary condition for increasing shared prosperity, vision and strategic plan for the country’s it is not sufficient. Growth is not a cure-all, long-term development, and four major but it is essential to allow Haitians to escape Country Profile 9 poverty. Unfortunately, Haiti has yet to achieve and economics which creates little incentive a sufficiently strong growth momentum. How­ for broad-based coalition building. “Haitian ever, even the modest resources the country institutions have never provided justice, edu- has generated in the past have not been dis- cation or healthcare to the majority of the tributed in an inclusive manner. Growth that population” (Lockhart and Forman 2013). is inclusive of the poor requires additional Instead, a small economic elite has supported mech­ anisms such as a pro-poor fiscal regime, a state that makes only negligible invest- as well as targeted social programs and expen- ments in human resources and basic infra- ditures, not only to redistribute resources to- structure (World Bank 1998). “Governance wards the poor but also more importantly to and state capacity to effectively formulate ensure that the less well-off are an integral part and implement sound policies, and to deliver of the process and that opportunities im- core public services to the population, are prove for all.2 weak. The state is present largely in the major A social contract is missing between the urban centers and has been unable to pro- State and its citizens. Haiti is not very inclu- vide basic services or infrastructure to large sive. The Bank’s 1998 Poverty Report noted portions of the population” (Buss 2013). that “Haiti has never had a tradition of gov- Haiti’s tax system generates limited re- ernance aimed at providing services to the sources for the government and tends to be population or creating an environment con- regressive (figure 1.1). Despite recent im- ducive to sustainable growth.” A number of provements in tax collection, Haiti col- Haitian and international observers broadly lects fewer domestic revenues in terms of agree that the Haitian State remains ineffec- GDP than comparable countries in the tive and delivers little to its population. To a Latin America and Caribbean region (LAC). large extent, this results from the particular Furthermore, Haiti’s tax revenues rely heav- winner-takes-it-all nature of Haitian politics ily on indirect taxes which affect consumers Figure 1.1  Inclusiveness a. Fiscal revenues are low … b. ...and taxation could be made more progressive Fiscal revenue, 2011 (percentage of GDP) Ratio of direct to indirect taxation, 2009 or 201 1 35 Panama 30 Low income 25 countries Jamaica 20 Costa Rica 15 Honduras Dominican 10 Republic 5 Nicaragua Haiti 0 0 20 40 60 80 a ica ala a ago nad e os ia ze a aica s ama blic aiti Bar es dura Gre and th ragu inic nad Luc Beli bad ta R tem in H epu Tob Jam Pan Dom Gre Source: IMF. Nica Hon St. Gua Cos an R and ent Vinc inic idad Dom St. Trin Sources: World Bank and IMF. 10 Country Profile Figure 1.1  continued c. Social spending remains limited… d. …and some major expenses favor the Social spending 2013 or latest (health, education, richer households and social protection in percentage of GDP) Cumulative distribution of fuel subsidies, 2013 (per decile) 100.0 25 100 20 80 15 60 10 40 16.8 5 20 1.1 1.6 3.7 6.0 0.1 0.1 0.3 0.5 0 0 1 2 3 4 5 6 7 8 9 10 (20 a (20 s (20 a (20 r (20 a (20 a 13) do 11) 12) 11) 12) 11) 11) dura ic am ragu al ta R tem i (20 alva Pan Source: World Bank staff calculations. Hon Nica Gua Cos El S Hait Source: World Bank and Ministry of Economy and Finance (MEF). e. Money remains the main obstacle to access f. …leading income groups to experience differences to many basic services such as health care... in health and other human development indicators Obstacles to access to healthcare services, 2012 (per quintile) Stunting rate under 5, 2012 (percentage per quintile) Q5 35 30 Q4 25 Q3 20 15 Q2 10 Q1 5 0 0 50 100 Q1 Q2 Q3 Q4 Q5 Not willing to go alone Distance to health provider Source: MSPP, IHE, and ICF International 2013. Not having money for Not having permission to go treatment Source: MSPP, IHE, and ICF International 2013. independently of their income level. The differently on economic agents of varying in- ratio of direct to indirect taxes stood at about come levels. 30 percent in 2011, a level inferior to that of Furthermore, social spending remains most countries in LAC and to the average of limited and the delivery of basic services low income countries (LIC): a fact largely strongly inequitable. Public spending in explained by the sizable share of Haiti’s health, education, and social protection revenues coming from international trade. ­ amounts to 5 percent of GDP, below com- Tax systems relying relatively more on direct parator countries, limiting the government’s taxes tend to be more progressive because ability to offer equal opportunities to its citi- in such systems the burden of taxes weighs zens. At the same time, many large spending Country Profile 11 items such as fuel subsidies clearly favor the and fragmented among themselves, leading rich. In this regard, the rise in fuel prices in to highly uneven service provision. October 2014, as well as the recent decline in The reliance on non-State actors, par- international oil prices, could be conducive ticularly donors, has also weakened ­ public to more inclusiveness in public spending. In investment management. Haiti’s public in- the absence of government, basic services vestment management exhibits a number of such as health and education are mainly pro- distinctive features and practices common to vided by non-government actors. Eighty to countries that are aid-dependent, including ninety percent of primary school students weak ­ appraisal capacity and reliance on do- attend non-public schools, placing a sub- nors to design good projects. These hamper stantial financial burden on households and the effectiveness of public investments delivering achievements closely linked with (box 1.1). While the government’s Strategic household income. Outcomes are equally Development Plan (PSDH) provides broad unfavorable to the poor in the health sector: guidance on government priorities, sectoral in 2012, infant mortality was 62 per 1,000 strategies to guide the prioritization of proj- live births for the richest quintile compared ects are lacking. This leads to a Public Invest- to 104 for the poorest quintile.3 ment Program (PIP) composed of projects In the absence of public services, non-­ that are neither fully assessed nor prioritized. State service providers have stepped in. In Furthermore, there is no effective ex-ante response to limited government services, ­ control on disbursements based on projects’ foreign assistance—as well as a growing physical progress against plans. More impor- number of international non-governmental tantly, although progress in fiscal reporting organizations (NGOs) since the 1950s—has is being made by rolling out the use of a sin- provided a broad range of dispersed public gle treasury account, domestically-funded services. In many remote rural areas, the capital expenditures are not yet properly ac- State has extremely limited delivery capac- counted for, tracked and reported, creating ity, and NGOs provide services to the popu- an environment conducive to a lack of trans- lation. Their services are also important parency and accountability, as well as to mis- among the urban poor. In health, about 50 management of scarce public resources. percent of total health expenditure is pro- Fi­nally, even though the existing legal frame- vided by NGOs, which for the most part work is acceptable for the management of deliver primary health care services and ­ public investments, its requirements are NGOs or private, for-profit institutions run rarely respected, with numerous processes over 80 percent of all primary and second- and ­ procedures that, when not redundant, ary schools. In water and sanitation, while excessively elaborate. are ­ the National Water and Sanitation Director- ate (DINEPA) has greatly improved its pres- ence in remote rural areas by creating, Vested Interests training and supporting local water com- Haiti is reported  to have been one of the mittees, interventions are often funded richest colonies. At the end of the eighteenth through NGOs. These non-state actors are ­ century, on the eve of its independence, largely disconnected from the state system ­ Saint-Domingue—as the territory was then 12 Country Profile BOX 1.1  Common Features of PIM in Donor-Dependent Countries Public investment management (PIM) systems in donor-dependent settings tend to exhibit the following distinctive features: Investment guidance, project development and preliminary screening: government strategy docu- ments, such as a Poverty Reduction Strategy Paper (PRSP), tend to be directed towards the donors, rather than covering both external and domestic investment in an integrated and coherent manner. They are at a level of generality that limits the extent to which they can provide a basis for prelimi- nary screening of projects, and are often not supported by effective sector strategies. Formal project appraisal: there is a reliance on donors to conduct appraisal, with a serious lack of appraisal capacity within government; and a lack of guidance on defining the project preparation process and on how to appraise domestically-financed projects and PPPs. Donor capacity-building on appraisal tends to be agency-specific, with little or no domestic training capacity. Independent review of appraisal: reflecting reliance on donors, there is a lack of capacity for inde- pendent review, either of donor projects or domestically-financed projects. Project selection and budgeting: the budget is divided into a recurrent and a development budget, with weak integration between the two and substantial off-budget aid. The use of Public Investment Programs (PIPs) remains quite common, but these can be poorly connected to fiscal policy and the budget. In practice, a PIP tends to be more a coordination tool than a tool to manage the project port- folio strategically or to help enforce review of individual project proposals before they can be consid- ered for budget funding. Agreement by a donor to finance a project is tantamount to the project being included in the budget—subject to basic screening for consistency with a PRSP (which is not difficult given their generality) and any required counterpart financing being affordable. Project implementation: unpredictability of donor funding (especially budget support) interrupts project implementation due to lack of alternative financing. Weak project management capacity induces donors to set up multiple Project Implementation Units (PIUs) within implementing agencies that initially help to speed implementation and compliance with fiduciary standards, but which cut across and negatively impact on in-line capacities and accounting and reporting systems. Procurement is undertaken by PIUs or donors to donor standards rather than national procurement standards. Project adjustment: reliance on donors to trigger review of any projects that are off-track. No simi- lar mechanism for domestically-financed projects. Facility operation: formal hand-over procedures on completion of donor projects, but inadequate asset registration systems; and inadequate funding for operations and maintenance, in part due to weak integration of recurrent costs of donor projects in fiscal policy and budgets. Basic completion review and evaluation: reliance on donors to undertake reviews and evaluations of their projects. Otherwise, little or no systematic basic post-project review, let alone evaluation, and little systematic use made of findings from donor evaluations to ­ improve future project design and implementation. Source: Rajaram et al. 2010. Country Profile 13 known—was considered the world’s richest goods. The migrants typically stayed in Haiti and most productive colony. It was also re- as resident aliens to maintain the protection puted to be the most violent, with the ma- of foreign governments, but often married jority of its population working as slaves into elite Haitian society to circumvent re- in plantations (Acemoglu and Robinson strictions on foreign ownership of Haitian 2012; Dubois 2013). Brutal punishments property (Plummer 1988). These families were common, and slaves died at stagger- of mixed Haitian and European descent ing rates. In 1804, Haiti emerged from were followed beginning in the 1890s by 13 years of revolution as the second nation Syrian and Lebanese immigrants. By the in the Western Hemisphere to attain inde- early turn of last century, these immigrants pendence from a European colonial power owned the major trading houses, as well as (after the United States). transportation and communication systems Independence did not lead, however, to (Plummer 1988). the development of inclusive institutions. From the beginning of the twentieth The country’s successful revolt left the new century, autocratic leaders in Haiti have republic isolated on the international stage, ­ traded political support from the elite for as evidenced by the early lack of diplomatic economic advantages to this elite. During recognition from Europe and the United the 1910s, presidents began floating do- States. This had a negative impact on the mestic loans that were largely subscribed willingness of foreigners to invest in Haiti. by the foreign commercial community It is also argued that there were insufficient (Rotberg 1971). Plummer (1988) writes checks and balances on the power of post- that “merchant-bankers had to cooperate independence leaders who set themselves with official requests for funds if they up as a new elite, exploiting Haiti’s people wished to continue conducting business through the very means as the French had unhindered, but they were frequently re- earlier. Furthermore, to gain recognition, warded for their pains by favors at the cus- Haiti had to agree to compensate France for tomhouses.” François Duvalier (‘Papa Doc’) the property losses faced by colonial plant- continued this tradition of cutting deals ers during Haiti’s revolution (1791–1804). with the economic elite. In 1958, for in- Payments began in 1825 and only stopped stance, following a first abortive coup at- 122 years later, in 1947, draining the econ- tempt, Duvalier demanded “displays of omy of much needed capital. loyalty—especially from wealthy business- An elite, often of foreign descent, build men, who made the first of many special important economic interests. Beginning in contributions to the defense of the realm” the mid-nineteenth century, Haiti’s economy (Rotberg 1971). The practice was contin- had grown increasingly dependent on im- ued in the early 1960s, when Haiti was si- ports, and wealth was accumulating in the multaneously hit by an economic slump families of importers. With looser restric- and cut off from U.S. aid. “Foreign-owned tions on foreign merchants in the mid-1840s, companies and men of wealth with impor- migrants primarily from Britain, France and burden: tant stakes in the country shared the ­ Germany began to establish themselves on they were encouraged to contribute to the the island to meet the demand for European stability of the regime by subscribing to 14 Country Profile national bonds, government lotteries” professional management, limiting their (Rotberg 1971). access to long-term financing. In return, statutory monopolies and ex- The open institutions needed to create a clusivities were usually granted for coopera- level playing field for economic actors and tive economic elites. A long list of industries underpin a rule of law based on impersonal became monopolies by presidential decree ­ exchanges have not emerged. To prosper, during the 1950s and 1960s: “mineral and economies need institutional arrangements petroleum exploration and exploitation, the to resolve disputes among firms, citizens, construction and operation of television and governments, to clarify ambiguities in stations, the planting and processing of ses- laws and regulations, and to enforce compli- ame, the processing of guano, the manufac- ance (World Bank 1998). A State needs to ture of ­chocolate, a fertilizer industry, the use its coercive power to enforce property development of casinos and hotels, the con- rights at low cost without abusing it and struction of a sugar factory, the improve- exploiting its citizens in order to develop ­ ment of the telephone system…” (Rotberg mechanisms to enforce effectively imper- 1971). By 1985, some 19 families held al- sonal agreements (North 2010).4 In such most exclusive rights to import many of agreements, an economic operator’s decision the most commonly consumed products to transact is independent of his partner’s in Haiti, as detailed in table 1.1. Another personal past history. It is made without twenty to thirty families held import ­ licenses knowledge of that partner’s past conduct, for another 92 items (Fass 1990). Though or the expectation of future transaction with publicly available information on privately him or her, or the ability to report miscon- held businesses is limited and import quotas duct to future trading partners (Greif 2006). has been dismantled, many of the same elite In particular, mechanisms of arbitration are families who dominated the Haitian econ- required to settle possible disputes between omy during the era of Duvalier in the 1970s traders and ensure contract enforcement. remain in control of large segments of the The inability of the Haitian State to manage economy today. its power appropriately has not allowed the Against this backdrop, the structure of emergence of these mechanisms and the es- the private sector shows signs of high de- tablished private sector had little incentive to grees of concentration, hampering the create a level playing field that would have entry of new actors. The established private primarily favored new entrants. sector is dominated by a number of fami- In the absence of effective contract en- lies, resulting in high concentration in a forcement, economic operators tend to re- number of key industries, distorted compe- main within their family or social networks. tition, and non-transparent business prac- Without effective legal systems making low tices in many instances. In practice, this cost enforcement of contracts possible, translates into limited opportunities for a ­ operators have to rely on personal knowl- substantial expansion of the ­ formal private edge, reputation, or the possibility of some sector across most sectors. Few of Haiti’s form of retaliation in selecting their busi- established private firms have modern ness ­partners. Family, ethnic or social net- ­ capital and governance structures with works have been shown to be able to impose Country Profile 15 16 Table 1.1  Import Quotas for 19 Major Families, 1984–85 Percent Total Total A B C D E F G H I J K L M N O P Q R S share (%) importers Household utensils, metal 7 92 48 99 2 Household utensils, 2 2 15 26 49 94 5 plastic Shoes 70 70 1 Slippers 92 92 1 Nails 65 65 1 V-8 juice 29 71 100 2 Vegetable and fruit juices 48 48 96 2 Toothpaste 3 62 65 2 Liquid disinfectant 11 14 14 39 78 4 Wrapping paper, 66 16 82 2 cardboard Paper and plastic bags 94 94 1 Irons for pressing clothes 43 28 28 99 3 Paint 44 22 22 88 3 Hand soap 10 18 45 73 3 Candies 65 32 97 2 Textiles 31 12 37 80 3 Milk 63 28 91 2 Spaghetti, macaroni, etc. 44 56 100 2 Country Profile Source: Reveco 1984. effective informal sanctions and enforce con- indicates few firms dominate key markets) tracts in weak institutional environments, (figure 1.2). This suggests that in Haiti, prices, and thereby foster economic transactions availability and quality of goods and services that would ­ otherwise not take place (Greif are perceived to be determined by a small 1989, 1993; Rauch 2001). Empirically, the number of firms rather than by market mech­ presence of migrant networks, for instance, a­nisms and that these incumbent firms ap- has been shown to compensate for the lack pear to face little competitive pressure from of good ­ institutions and is associated with new entrants or smaller domestic firms. This greater ­international trade (Ehrhart et al. situation could be the result of high legal or 2014). While these relationships are useful, behavioral entry barriers in the market that they are not conducive to greater inclusion. prevent new competitors from participating social organization tends to undermine This ­ in key markets. These market structures collective action by providing alternative, could facilitate tacit agreements among fami- non-­ collective avenues for ensuring individ- lies/groups to allocate markets among them- ual and family survival. selves, which may harm productivity and As a result, competition is perceived as incentive to innovate. low in Haiti. According to the Global Com- Weak competition contributes to a high petitiveness Report of 2014–15, Haiti ranks degree of operational business risk. Accord- low in terms of competition: Haiti ranks 140th ing to the Economist Intelligence Unit’s op- out of 148 countries in terms of intensity of erational risk model (2013), investors in local competition and ranks 140th in terms Haiti perceive that the risks associated with of extent of market dominance (a high rank discriminatory policies favoring Figure 1.2  Competition Intensity and Extent of Market Dominance, 2014–15 a. Intensity of local competition, 2014–15 b. Extent of market dominance, 2014–15 (1=not intense at all, 7=extremely intense) (1=dominated by few business groups, 7=spread among many firms) 6 4.5 4.0 5 3.5 4 3.0 2.5 3 2.0 2 1.5 1.0 1 0.5 0 0 at o To a ca Ja go a Ho gua Ja la a at a Ho ica as Pa lic Pa as ad m ic Re ity ad Pa y R ay To y ca iti ca iti Gu ag an am al gu Gu am ua t a om bl b ai Do n C dur ur an uni ba Ha Ha gu a em em n pu b Tr n C epu ra ra ag m m ica mu n n nd a n r Ca min Par m om d Ni Ni d n be n rib ca i in a a id id be in in Do rib Tr Ca Source: World Economic Forum 2014. Country Profile 17 incumbents and unfair business practices Trade data seem to confirm these per- are higher than in other countries in the ceptions about limited competition. Ac- Caribbean region and only comparable to ­ cording to a product market concentration those in Guatemala, Honduras, and Nicara- analysis based on available import data gua (figure 1.3). The sources of these risks (see box 1.2), Haiti imports most of its are mainly related to: a) price control and final products and most of the imported the control of other market variables that goods sold in Haiti are provided by few both limit the benefits that firms could ob- large firms. About 70 percent of total im- tain by competing effectively and harm the ports in value are in goods whose markets incentives to invest; and b) vested interest are either highly or moderately concen- and discrimination against foreign firms trated according to common thresholds that appear to play the most important role used by the U.S. Department of Justice and in generating distortions of the level-playing Federal Trade Commission (­ figure 1.4). field, hinder innovation, and preclude more While market concentration is not unusual efficient firms from competitively gaining given the size of the economy, high and market shares. stable market shares (above 40–50 percent of the market) may increase the potential risks of firm abusive behavior with negative impact on consumers. Highly concentrated Figure 1.3  Business Risks Related product markets (representing around 40 to Weak Competition Policies percent of imports with a Herfindahl- (by Components) Hirschman Index [HHI] exceeding 2,500) include petroleum, food and consumer 12 goods markets (telephone sets, motorcy- cles). Imported goods in moderately con- 10 centrated markets (representing another 30 8 percent of imports with an HHI between 1,500 and 2,500) include food, consumer 6 goods (electric generating sets, motor vehicles) and cement. ­ 4 The concentrated nature of the Haitian 2 market may be leading to higher consumer prices. The market analysis aims at detect- 0 ing the market characteristics that favor m a ub an a Pa go ay Ho ala Ni ras a at i Do aic an am gu Gu ait the existence of market power and its po- gu ba ic c em u ra H i m Re in ad Pan nd l ra To ca Ja tential risks in terms of abusive practices. p d For instance, the risk of potential anticom- id in Tr Unfair competitive practices petitive behavior given high market con- Price controls centration may translate into high prices Discrimination against foreign companies for consumers, which may be particularly Vested interests/cronyism damaging in Haiti where more than half of Source: The Economist Intelligence Unit. the population lives on less than US$2 per 18 Country Profile BOX 1.2  Product Market Concentration Analysisa An analysis of product market concentration was conducted jointly between the World Bank and the Ministry of Economy and Finance of the Republic of Haiti with the collaboration of a team of researchers from Harvard and Columbia Universities. The analysis was based on Herfindahl-Hirschman indices (HHI), calculated using customs im- port data based on the Harmonized Coding System (HS) disaggregated at a four-digit level. The HS is a multipurpose international product nomenclature developed by the World Customs Organization. The HHI is a commonly accepted measure of market concentration based on market share esti- mates for each firm in the market. It is calculated by squaring the market share of each firm com- peting in a market (showing the relative space they occupy in the specific market being analyzed), and then summing up the resulting numbers for the fifty largest firms (or overall firms if the indus- try counts fewer than fifty companies). The HHI is preferred to other concentration measures (such as concentration ratio) because it accounts for the heterogeneity in the size of the firms in the mar- ket, giving more weight to larger firms. The HHI can range from close to zero to 10,000. The closer a market is to being a monopoly the higher the HHI. If, for example, there was only one firm in an industry (a monopolist), that firm would have 100 percent market share, and the HHI would be equal to 10,000 (100^2). Conversely, if there were thousands of firms competing in an industry, each firm would have nearly 0 percent market share, and the HH index would be close to zero. For the purpose of this analysis, the concentration levels based on the Horizontal Merger Guidelines (U.S. Department of Justice and Federal Trade Commission) were used. As such, three HHI levels help identify the characteristics of product market concentration: • Unconcentrated Markets: HHI below 1,500. • Moderately Concentrated Markets: HHI between 1,500 and 2,500. • Highly Concentrated Markets: HHI above 2,500. a. Based on a background paper prepared for the SCD jointly by the World Bank (Georgiana Pop, Tanja Goodwin, Martha Martinez Licetti, Lucia Villaran, and Joaquin Zentner) and the Ministry of Economy and Finance (Jimy Dorsainvil, Clifford Reginald Nau, Edward Nsimba, and Reginald Surin). day. Several of the most important food as well as the cost to import containers and products in the Haitian consumption bas- import tariff levels (see appendix A for ket are sold in concentrated markets. A further details).5 preliminary analysis over time of prices for such products seems to suggest that on av- erage, prices in Haiti are approximately Political Instability and 30–60 percent higher than in other coun- tries from the LAC region, even taking Violence into account differences in income levels Political violence has occurred regu- and purchasing power of the population, larly throughout Haiti’s history, leading to Country Profile 19 Figure 1.4  Concentration Levels Based on HHI in the 18 Most Important Haitian Product Markets, 2012 Unconcentrated HHI<1500 Articles for the conveyance or packing of goods Cane or beet sugar and chemically pure sucrose Moderately concentrated New pneumatic tyres, of rubber 2500>HHI>1500 Food preparations not elsewhere specified Motor vehicles for the transport of goods Motor cars Meat and edible offal Rice Electric generating sets and rotary converters Pasta, couscous, whether or not prepared Cement Highly concentrated Bread, pastry, cakes and biscuits Telephone set HHI>2500 Petroleum oils and oils from bituminous minerals Soups and broths and preparations Animal or vegetable fats and oils Palm oil and its fractions Milk and cream, concentrated 0 1,000 2,000 3,000 4,000 5,000 HHI Source: Ministry of Economy and Finance, 2014, Haiti’s Import Database. Note: Each product is identified by a four-digit Harmonized Coding System (HS). government instability. At Independence in The past decade has been nevertheless 1804, Haiti was at the forefront of history, comparatively stable. Over the past ten years, being the first nation to abolish slavery. with security provided by a large United Since then, however, with some exceptions Nations Stabilization Mission in Haiti (MI- such as of the 30-year period of autocratic NUSTAH), two presidents have been chosen rule under Francois Duvalier (Papa Doc) by election and the current President, Michel and his son Jean-Claude Duvalier (Baby Martelly, was the first to accede to the Presi- Doc) (1957–86), Haiti has known a suc- dency by election from a party in the opposi- cession of short-lived governments. tion. However, stability remains fragile, Lacking sufficiently long periods of stabil- with frequent changes in government and ity, the country has struggled to develop ­ repeated delays in electoral calendars. Such the institutional mechanisms and policy delays led to the lapsing of Parliament in fundamentals essential to economic January 2015, leaving the executive to try to ­ development. broker a deal with the opposition and agree 20 Country Profile on a transitional government who would or- country caught in a vicious circle in which ganize elections. unemployment, inequality, and poor educa- The distribution of resources seems to tion feed into lawlessness and violence, feed grievances. From the early days of inde- making it difficult for the economy to grow pendence until the end of the nineteenth and create jobs, thus perpetuating unem- century, almost all of Haiti’s heads of state ployment and inequality.” For instance, the granted land to the military and to high pro- political turmoil of the late 1980s and 1990s, file civil servants, stripping peasants of their coupled with a US embargo, severely de- land rights (Dubois 2013). As a result of this pressed Haiti’s exports. Exports of assem- history, calls for and conflicts over land re- bled electronic goods virtually disappeared distribution repeatedly emerge as a source of and never recovered. External aid flows instability and uncertainty, with serious con- have also been highly volatile, responding sequences for growth. Many observers agree to political events. All of these elements that Haitians perceive political and economic have negatively affected employment of a disputes as a zero sum game with only win- very young, and increasingly urbanized and ners and losers, each with very long disenfranchised population, the level of memories. public investments, and the overall sustain- Grievances lead to conflicts. Distribu- ability of the social and political situation in tional conflicts arguably have as a corollary the country. inhibited the development of social capital and hampered the building of trust within Haitian society further preventing agreement Natural Disasters across groups and the emergence of social The Haitian population is one of the most cohesion at the national level. Disenfran- exposed in the world to natural disasters— chised and without effective channels to hurricanes, floods and earthquakes.6 Haiti voice needs and demands, citizens have has a higher number of disasters per km2 taken to the streets in protest, sometimes than the average of the Caribbean countries violently. The Bank’s 1998 Poverty Report ­ (table 1.2). In 2008, tropical storms and hur- noted “the absence of a culture of democratic ricanes caused losses estimated at 15 percent decision making and peaceful consensus of GDP. The earthquake on January 12, 2010 building … has generated tensions…and killed 220,000 people, displaced 1.5 million hampered Haiti’s rehabilitation effort.” While people, and destroyed the equivalent of 120 violent airing of grievances in Haiti’s early percent of GDP. These disasters tend to affect history generally took place in the rural disproportionally the poorest and marginal areas, contemporary unrest tends to break pop­ lations settling in flood zones and u­ out in the cities, reflecting the country’s de- coastal areas, in particular in the case of mographic evolution and urbanization over tropical storms where almost 50 percent of the years. damages and losses to the productive sectors Violence in turn hampers economic op- have been concentrated in the agricultural portunities. A report on governance and so- sector.7 Overall, based on available historic cial justice in the Caribbean (Dollar 2000) data, weather-related disasters are estimated argued that Haiti is an “extreme case of a to have caused damages and losses Country Profile 21 Table 1.2  Frequency and Impact of Natural Disasters, 1971–2014 Number of Disasters/ Disasters/ Deaths/ Total Disasters/ Country/group natural land surface population population damage/ year disasters (`000 sq. km) (millions) (millions) GDP Haiti 137 3.1 5.0 13 23,427 1.776   Excluding earthquake (2010) 136 3.1 4.9 13 1,855 0.22 Dominican Republic 60 1.4 1.2 6 311 0.05 Jamaica 34 0.8 3.1 13 102 n.a. Nicaragua 66 1.5 0.5 11 2,363 0.33 Honduras 70 1.6 0.6 9 3,298 0.40 El Salvador 51 1.2 2.5 8 687 0.34 Guatemala 82 1.9 0.8 5 1,754 0.12 Costa Rica 58 1.3 1.1 12 72 0.04 Panama 46 1.0 0.6 12 80 0.01 Other Caribbean Statesa 129 2.9 0.3 30 86 0.19 Source: EM-DAT: The OFDA/CRED International Disaster Database. a. Antigua and Barbuda, The Bahamas, Barbados, Belize, Guyana, Suriname, Dominica, Grenada, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, and Trinidad and Tobago. Simple average. Note: n.a. = not applicable. amounting to about 2 percent of GDP on av- because Haiti experienced more than twice erage per year during 1975–2012.8 as many floods, as a consequence of a Moreover, the effects of these natural storm, and three times the number of disasters are much larger in Haiti, indicat- drought periods than the Dominican Re- ing higher vulnerability and institutional public. Floods are one of the most com- weaknesses. While Haiti appears to be mon weather-related events that affect somewhat more exposed to natural haz- Haiti and are partly related to the country’s ards than some countries in the Carib- severe deforestation. Haiti’s greater vulner- bean, it is also more vulnerable. Even ability is reflected in the consequences of when the earthquake of 2010 is excluded these events in terms of losses, both from the statistics, the number of casual- human and economic, reflecting the inade- ties to the population is very large, simi- quacy of drainage, large populations set- lar only to Honduras and Nicaragua tling in flood zones, and the lack of sound (table 1.2). In Haiti, population density and buildings or building codes. lack of coping capacity have translated into large numbers of people affected by natural disasters ­(figure 1.5). In this regard, the Migration comparison between Haiti and the Domin- In this unfavorable environment, mi- ican Republic, both sharing the same is- gration has become a key avenue for Haitians land, is instructive. The number of seeking a better life. Substantial internal mi- weather-related disasters from 1971 to gration is taking place, particularly from 2014 has been more than double in Haiti rural to urban areas, as people seek better than in the Dominican Republic, mostly economic opportunities and better services. 22 Country Profile Figure 1.5  Vulnerability Index, Figure 1.6  Migrants International 2013 (Average Score of Comparison, 2010 Susceptibility, Coping and Percent of total population Adaptive Capacity) 70 100 60 90 80 2/171 50 70 40 60 50 30 40 20 30 20 10 10 0 0 B r r iti B s do m S ic t i Gu d n u l v P Ha c li ur Ho u El m Gu oli m n nd rin lv lit l r bi r l p tiv ilit cit n Gr J Su pi Ni ib c dp cit Ov co pt of p sc r of ck Source: World Bank. c ln Su ck vu L L Haiti LIC LAC Source: World risk report. Figure 1.7  Migrants by Destination Country, 2010 Thousands In addition, for both political and economic 600 reasons, large numbers of Haitians have emi- 500 grated throughout the twentieth century building an important diaspora (figure 1.6). 400 A vast majority of Haitians who continue to 300 emigrate now seem to do so because they 200 cannot find work opportunities in Haiti Orozco (2006). They move mostly to the 100 United States, the Dominican Republic, 0 Canada and France ­ ­ (figure 1.7).9 According s ic th nc Th d bl t ou pu n to the Global Bilateral Migration Matrix, one s, St rs Fr C R m d h n Ot it million Haitians were estimated to be living h Un ic B in m abroad in 2010. About 540,000 of those Do were in the United States, 280,000 in the Source: World Bank. Country Profile 23 Dominican Republic,10 74,000 in Canada and of GDP for ODA, 13.2 percent of GDP for ex- 42,000 in France.11 ports and 2 percent of GDP for FDI. This large diaspora is a source of signifi- Remittances are, however, predominantly cant remittances. Remittances received by an urban phenomenon and migrants tend to Haiti represent a large share of GDP, the be better educated. Emigration is relatively largest share among LAC countries, and the ­ costly and accessible mainly to urban and fourth highest in the world in terms of export wealthier families. While for the country as earnings (figure 1.8). For every year in the a whole, about a third of all households re- period 2000–09, remittances were larger ceive remittances, this share reaches over 35 than official development assistance (ODA), percent for urban households and drops to foreign direct investment (FDI) and the total 20 percent for households in rural areas. exports of goods and services (­ figure 1.9).12 Migrants tend also to be better educated, ­ ­ Despite the sharp increase in development representing an important “brain drain.”13 assistance in the year after the earthquake, Overall, the profile of Haitian migrants in 2012, remittances once again surpassed de- is very similar to those of the average velopment assistance, representing the equiv- ­ Caribbean migrant. However, individuals alent of 20.4 percent of GDP vs. 16.1 percent with advanced education represent only 3 Figure 1.8  Remittances, 2012 Percentage of export earnings 300 250 200 150 100 50 0 n os vo M lic Po ho r G nk sh m aiti L ti l do st v si ic ri ni o ud b or so do t p ist m W on rib so pu d H nd t B b n m -L rm m m Ko lv T ol N jik Li Ki S l or Ar Co l s R J S n B T El h Ti B nc r K Fr Source: World Bank. 24 Country Profile percent of the resident population in Haiti would suggest that this ­ c onsumption falls compared for example to almost 20 percent mainly on imports. The continuous in- in the Dominican Republic and the Baha- crease in remittances—and aid—observed mas. These figures suggest that over 45 over the past has been associated with an percent of Haitians with a college degree expansion of imports and a widening of currently live outside of the country.14 the trade deficit, making Haiti one of the Remittances have tended to finance the countries that import the most in terms of consumption of foreign goods. Remit- GDP ­ Import in- (figures 1.10 and 1.11). ­ tances represent an important source of creased from about 30 percent of GDP in livelihood, amounting to about 20 percent the early 1980s to more than 50 percent in of household income. Controlling for a set the early 2010s, largely consumption goods of household characteristics, remittances financed by international aid and remit- have been shown to be positively associ- tances. Major items imported by Haiti are ated with higher expenditure on food, food (representing around 35 percent ­e ducation and health. 15 On the of total imports over the past decade), macro-economic level, however, data ­ manufactured goods (around 18 percent ­ Figure 1.9  Foreign Flows, 1998–2012 Figure 1.10  Effects of Aid and Percentage of GDP Transfers on the Trade Balance, 1980–2013 50 Percentage of GDP 80 40 60 30 40 20 20 10 0 0 –20 19 8 20 9 00 20 1 02 20 3 20 4 05 20 6 20 7 08 20 9 20 0 20 1 12 0 1 0 0 0 9 0 0 9 1 20 20 19 20 20 N t ODA –40 Exports of G&S 19 0 19 2 84 19 6 19 8 19 0 19 2 94 19 6 20 8 20 0 2002 2004 2006 20 8 20 0 12 8 9 8 9 8 9 0 8 9 0 1 19 19 19 P rson l r mitt nc s For i n dir ct inv stm nts Trade balance (as % GDP) Aid + remittances (% GDP) Source: World Bank. a. Includes in kind donations. Source: World Bank. Country Profile 25 Figure 1.11  Imports of Goods and Services—LAC Region, 2011–13 Percentage of GDP 90 80 70 60 50 40 30 20 10 0 Ec ico s s M ic l u st or u r lo B m i il R il t P n Ur do Gr ic i n d r in t i Ni n m Ho u P nd l Do ic c th B ci bl Ar mb Co , R v i n Ch St li n Co v d ur Ri Ha S u u Gr rbu x in nd nd . Lu Gu oli nt m P di pu Br B nd m Gu u r B l r c u J El n ic V in nt u m Do i in nt .V A c St Source: World Bank. of imports), machinery and transport that takes into account statistical methodol- equipment, fuel, and raw materials (figure ogy, source data and periodicity, Haiti 1.12).16 ranks below comparator countries (figure 1.13). As a result and despite the progress that has resulted from surveys such as the Demographic and Health Survey (MSPP, Limited Data IHE, and ICF International 1994, 2000, There is limited evidence to inform 2006, 2013), Haiti is still characterized by ­ planning and targeting of programs and a dearth of data and ­reliable statistics. The policies. According to the Statistical census is being ­continuously delayed, for Capacity Indicator (SCI), a composite index ­ instance. Notwithstanding improvement in 26 Country Profile Figure 1.12  Merchandise Imports, Aid and Remittances, 2005–14 Percentage of GDP 15 30 10 20 5 10 0 0 P 08 09 10 11 12 13 05 06 07 14 20 20 20 20 20 20 20 20 20 20 Official transfers credit (rhs) Private transfers credit (rhs) Food Other consumer goods POL and other energy Intermediate goods Capital goods Source: Central Bank of Haiti (BRH). Note: P = Preliminary. Figure 1.13  Statistical Capacity recent years data provision has serious Indicator, 2014 (0=Lowest, shortcomings that severely hamper IMF 100=Highest) surveillance (IMF 2014a). The most af- fected areas are the national accounts, fis- cal data, and external sector statistics. In Periodicity and particular, constant price GDP data are timeliness published using a base year of still being ­ Source data 1986–87, which does not adequately re- flect the structural economic changes re- sulting from the various civil conflicts Statistical and natural disasters Haiti has experienced methodology over the recent decades, and thus intro- Overall score duce major uncertainty in policy analysis. Equally problematic are estimates of the 0 20 40 60 80 100 informal sector and no labor statistics are ­ Haiti disseminated. Low income countries Latin America and Caribbean Source: World Bank. Country Profile 27 Notes 1. Haiti has traditionally enjoyed substan- Insurance Facility (CCRIF) is a joint reserve tial preferential market access to the United mechanism that allows member countries, States and Canada, but has not succeeded in including Haiti, to quickly access liquidity for fully exploiting this advantage. For exam- infrequent, high-impact disasters. The proba- ple, the country has not been able to fill its bilistic model used by the CCRIF to calculate quota for exports to the United States under insurance premiums offered to the countries the Hemispheric Opportunities through and reinsurance markets is the MPRES model Partnership Encouragement II (HOPE II) developed by the Kinanco firm: http://www.ki- initiative or under the Haiti Economic Lift nanco.com. Program (HELP) Act. These laws allow Haiti 9. For a more detailed discussion on these flows to assemble textiles, whatever the origin of the see Orozco (2006) and Jadotte (2008). imported fabrics, and to export them to the 10. Extrapolation from older data and migrant United States duty free and tax free but are due flows suggest that the actual number of to end in 2020, unless renewed. Haitians living in the Dominican Republic 2. For a discussion on the policies needed to should be between 380,000 and 500,000 greater inclusion, see for instance Narayan, (Wooding and Moseley-Williams 2004). Chanduvi, and Tiwari (2013). 11. For details on the methodology of the esti- 3. MSPP, IHE, and ICF International 2013. mates see Ratha and Shaw (2007). 4. For an example of how such institutions 12. All data for remittances and GDP aggregates are could  develop, see Greif (1989, 1993, 2006). from the World Development Indicators (WDI). The study compares the evolving structure of 13. For more evidence on the education attain- political and economic institutions of Genoese ment of Haitians migrants in the US (in the traders, which ultimately provided the essential city of NY, more specifically), see the survey institutions for impersonal exchange, and the on remittance senders discussed in Orazco practice of the Mahgribi (Jewish merchants but (2006). in a Muslim culture), who fail to make the nec- 14. This estimate considers the share of col- essary institutional adjustments and lose out in lege-educated residents from ECVMAS the competitive trade of the Mediterranean. 2012; OECD data for education attain- 5. The availability of data to carry out a simi- ment of ­ migrants as of 2010–2011 and es- lar exercise for LICs is limited, but including timates for education of Haitians living in Kenya and Mauritius in the sample suggests the Dominican Republic. Niimi and Ozden that prices in Haiti are 35–50 percent higher. (2006), based on data from Docquier and 6. From 1993 to 2012, Haiti has experienced Marfouk (2006) and Barro and Lee (2000), two  droughts, one earthquake, 31 floods and reported that 80 percent of Haiti born peo- 26 tropical storms/hurricanes. ple with college degrees lived abroad as of 7. Calculations based on existing Post-Disaster 2000. This result is driven by the estimates Needs Assessments (PDNAs). that only 0.7 percent of Haiti residents had 8. Estimations based on EM-DAT and existing college degree in that year. Post-Disaster Needs Assessments (PDNAs). 15. Econometric analyses carried out for the 2014 According to the probabilistic model used by Poverty Assessment tend to show that a 10 the CCRIF, tropical cyclones cause an esti- percent increase in remittances would be asso- mated US$118 million in damages each year, ciated with an 8–13 percent increase in spend- which is equivalent to 1.4 percent of GDP for ing on education and health, and a 5 percent 2013, while earthquake-related damages aver- increase in food expenditures. age US$26 million each year (0.31 percent of 16. An estimated 60 percent of food consumed in GDP in 2013). The Caribbean Catastrophe Risk Haiti is imported. 28 Country Profile 2. Trends and Profile in Poverty and Shared Prosperity1 Trends rural areas, where 38 percent of the popula- tion is not able to satisfy its nutritional needs, Recent findings would suggest that ex- compared to 12 percent in urban areas and 5 treme poverty has declined in Haiti over the percent in metropolitan areas (i.e., the capital past decade (figure 2.1). Extreme poverty city). The poorest regions are the furthest decreased in Haiti from 31 percent in 2000 from the capital city and the most isolated. to 24 percent in 2012 (World Bank and The poorest departments are geographically ONPES 2014).2 Progress was mainly concen- concentrated in the northern region of the trated in urban areas, however. The share of country where the extreme poverty rate ex- extreme poor has declined from 21 percent ceeds 40 percent in the North East and North to 12 percent and from 20 percent to 5 per- West, compared to 5 percent in metropolitan cent in other urban and metropolitan areas, area of Port-­au-Prince (map 2.1). respectively, while it has stagnated in rural Furthermore, income inequality is the areas.3 This trend is confirmed by both mon- highest in the region and has widened in etary and non-monetary poverty indicators, rural areas. The Gini coefficient stagnated at with the biggest non-monetary progress about 0.6 between 2001 and 2012. This re- recorded in education (table 2.1). All school- sult hides, however, important regional dis- age children go now to school in about 90 parities: while inequality widened in rural percent of the households compared to areas (from 0.49 to 0.56), it narrowed in about 80 percent in 2001. Immunization urban areas (from 0.64 to 0.59). Preliminary rates are also up. While data is not available evidence would suggest that increases in to assess trends in moderate consumption labor income and private transfers could poverty, it is also expected to have improved have contributed to the narrowing of in- in the last decade. equality in urban areas. On the other hand, Poverty remains, however, endemic in the contraction in agricultural production Haiti and varies widely. Haiti is the poorest caused by ­ repeated weather-related shocks country in Latin America and among the could explain the widening of inequality in poorest in the world in terms of per capita rural areas, reducing earnings for agricul- GDP and human development.4 The overall tural labor and increasing the prices of food, poverty headcount amounts to about 59 per- both factors affecting the poorest dispropor- cent and extreme poverty to 24 percent in tionately. As a result, in 2012 the top one 2012, indicating that almost 6.3 million percent of the population lived on almost Haitians cannot meet their basic needs and 50 times the resources of the bottom 10 per- 2.5 million cannot even cover food needs.5 cent.6 This result makes Haiti the most un- Geographically, poverty and extreme poverty equal country in Latin America and one of are considerably higher in rural areas. More the most unequal in the world. than 80 percent of the extreme poor live in Trends and Profile in Poverty and Shared Prosperity 29 Figure 2.1  Trends in Poverty a. Extreme poverty has declined, especially in urban areas… b. ...But strong regional disparities remain... Poverty incidence, 2000 and 2012 Incidence of extreme poverty, 2012 (percentage of population) (percentage of poor in the total population) 40 37.6 37.8 Nord-Est 44 31.4 Nord-Ouest 43 30 23.8 Nord 37 21.0 20.0 20 Grand´Anse 36 11.9 Centre 34 10 Nippes 30 4.6 Sud 26 0 National Rural Autre urbain Aire Sud-Est 25 métropolitaine Artibonite 23 2000 2012 Ouest 10 Sources: ECB 1999/2001 (Pedersen and Lockwood 2001) and official poverty rates (ECVMAS 2012). Source: ECVMAS 2012. c. Driven by private transfers and labor income... d. … Particularly in commerce and construction Income composition, bottom quintiles Sectoral contribution to growth 2001–12 100 (percentage of GDP growth) 11 18 24 17 1 32 1 26 Commerce, restaurants 80 8 18 1 21 4 and hotels 12 19 5 60 19 Construction 20 17 20 Transport and 25 17 40 16 communication 18 Manufacturing 15 37 24 27 20 19 21 Non retail services 24 13 9 15 13 15 Other retail services 0 2001 2012 2001 2012 2001 2012 Extractive industries Q1 Q2 Q3 Water and electricity Transfers Agriculture labor income Agriculture Capital + pensions Non-agriculture labor income –10 0 10 20 30 40 Autoconsumption Imputed rent Sources: IHSI and World Bank staff calculations. Sources: ECVMAS 2001, 2012. e. Inequality is nevertheless important... f. … And one of the highest in the world Gini coefficient, 2001 and 2012 Gini index, 2012 or latest (20 highest inequalities) (SEDLAC income methodology) 70 0.70 65 0.65 60 0.65 0.61 0.61 55 0.60 0.60 50 0.56 45 0.55 40 0.50 0.50 35 30 0.45 e ica s. bia ras blic Colo via mbia ala Par tho Swa ma Com es Nam s ibia ay Pan e e an wanda i Les zil d Hait incip oro Chil . St zilan National Rural Urban ll Bra agu rone uth Afr Boli tem Zam Afric Hondu a o epu che Fed d Pr R Gua Sey an R 2001 2012 sia, So Source: World Bank staff calculations. Tom Note: 0=equality, 1=inequality Mic tral Sao Cen Source: World Bank. 30 Trends and Profile in Poverty and Shared Prosperity Table 2.1  Access to Basic Services—Coverage Rates (2001–12) Percent National Urban Rural Indicator 2001 2012 2001 2012 2001 2012 School-age children in school 78 90 84 93 74 87 Under-5 mortality (per 1,000 live births) 137.7 92 111.7 88 149.4 99 Children (12–23 months) fully vaccinated 33.5 45.2 33.6 44.5 33.5 45.6 Access to improved drinking water sources   WHO definitiona — 53 — 55 — 52    Access to tap water (in house) 7 11 13 18 3 5   Expanded definition b — 73 — 91 — 56    Treated water (purchased) — 20 — 36 — 4 Access to energy c 32 36 62 63 11 11 Rate of open defecationd 63 33 44 11 76 53 Access to improved sanitatione — 31 — 48 — 16 Habitat, nonhazardous building materials 48 60 71 81 33 41 Sources: ECVH 2001; ECVMAS 2012; World Bank and ONPES 2014. Note: — = not available; WHO = World Health Organization. a. According to the international definition (WHO), access to improved drinking water is the proportion of people using improved drinking water sources: household connection, public standpipe, borehole, protected dug well, protected spring, rainwater. b. The expanded definition includes the international definition (WHO), plus treated water (purchased). c. Includes electricity, solar, and generators. d. Rate of open defecation refers to the proportion of individuals who do not have access to improved or unimproved sanitation. This indicator is part of the Millennium Development Goals (MDG) and is a key element of discussion for the post-2015 agenda. e. Improved sanitation is access to a flush toilet or an improved public or private latrine. Despite improvements in access to basic with 34 percent among non-poor house- services since 2001, the poor still face sig- holds (table 2.2). The poorest also have lim- nificantly larger barriers. Poverty is an im- ited access to maternal health services and portant barrier to both school enrollment are more likely to die during delivery, with and health service utilization, for instance. fewer than one woman in ten in the lowest According to the most recent household quintile benefitting from assisted delivery survey, cost is the main reason for keeping (compared to seven out of ten among the children out of school or not consulting a better off).8 These obstacles are even greater doctor if they are sick in 83 percent and 49 in rural areas, where the supply of services respectively.7 As a result, in percent of cases, ­ is more limited. 2012, while 87 percent of 6- to 14-year-olds Most of the poor work, but their earn- in poor households were in school, this ratio ings are insufficient to lift them out of pov- reached 96 percent for non-poor house- erty. Almost 70 percent of poor households holds. In turn, this more limited access to have heads holding jobs (against 73 percent education perpetuates poverty: the poor live among the non-poor). However, among in households in which the heads have an the former, 61 percent work in agriculture, average of three times fewer years of educa- where average earnings are less than 20 tion, and as many as 61 percent of the heads percent of the earnings in the formal sector. of poor households are illiterate, compared Among the remainder, 35 percent work in Trends and Profile in Poverty and Shared Prosperity 31 Map 2.1  Extreme Poverty Rates by Department, 2012 NORD-OUEST NORD NORD- EST BELOW 25% ARTIBONITE BETWEEN 26% AND 32% BETWEEN 33% AND 37% MORE THAN 37% CENTRE PORT-AU-PRINCE GRANDE- ANSE OUEST NIPPES SUD SUD-EST IBRD 41835 | AUGUST 2015 Sources: ECVMAS 2012; World Bank and ONPES 2014. the informal sector, where earnings are less experience repeated food shortages World than half those in the formal sector. As a Bank and ONPES (2014). As a result, one- result, more than half of poor households ­ fifth of under-5-year-olds are chronically mal- undertake two or more income generating nourished (MSPP, IHE, and ICF International activities. This underscores the urgent need 2013). This is a particular cause of concern to create higher labor productivity jobs. because proper nutrition in early life is crucial The poor suffer from poor nutrition early for brain development and subsequent life in life and from food insecurity. Food insecu- outcomes (Alderman and King 2006). rity is significant in Haiti, at 28 percent na- The poor in Haiti are particularly vul- tionwide and 34 percent in rural areas.9 Poor nerable and access to social safety nets is household members are much more likely to limited. A typical Haitian household faces report frequent hunger or lack of food at bed multiple shocks annually: nearly 75 per- time relative to members of nonpoor house- cent of households suffer economic con- holds (­ figure 2.2). Households with children sequences following a shock (e.g., natural under the age of 5 are much more likely to disaster or health-related). Households in 32 Trends and Profile in Poverty and Shared Prosperity Table 2.2  Basic Sociodemographic and Socioeconomic Characteristics of Poor, Extreme Poor and Trends and Profile in Poverty and Shared Prosperity 33 Nonpoor Households, 2012 National Urban Rural Variable Extreme Extreme Extreme Non poor Poor Non poor Poor Non poor Poor poor poor poor Household size, numbera 3.9 5.3 6.2 4.1 5.7 6.4 3.6 5.1 6.2 Age of the head, years 44.7 46.7 48.9 43.1 44.1 43.8 48.2 48.9 50.0 Under-5-year-olds, %a 0.3 0.7 1.0 0.4 0.7 1.0 0.3 0.7 1.0 Dependency ratio a 53.8 88.4 114.0 53.6 79.0 115.2 54.2 95.2 113.7 Head, man, % 56.8 56.0 59.6 53.5 49.6 48.1 63.9 61.1 62.0 Education, head, years a 6.3 3.5 1.7 7.5 5.0 3.2 4.1 2.5 1.4 Literacy, heads w/>5 years of education, % 65.8 38.5 19.6 74.2 54.1 34.5 47.6 26.1 16.5 Employed head, % 73.2 67.8 72.9 68.7 59.8 53.2 82.8 74.2 77.1 Unemployed head, % 15.8 18.3 11.9 20.1 28.7 32.1 6.4 10.1 7.7 Inactive head, % 11.0 13.9 15.2 11.2 11.5 14.7 10.7 15.7 15.3 Employed household members, numbera 1.4 1.5 1.6 1.3 1.3 1.2 1.6 1.6 1.7 Head employed in agriculture, % 25.5 49.1 77.5 6.1 16.3 41.1 60.7 70.2 82.7 Head employed in the formal sector, % 17.5 6.1 1.6 24.6 9.5 2.2 4.7 4.0 1.6 Head employed in the informal sector, % 57.0 44.7 20.9 69.3 74.3 56.7 34.5 25.8 15.7 Households receiving private transfers 58.58 60.56 58.23 57.56 64.14 64.81 60.5 58.84 56.91 (excluding remittances), % Households receiving remittances, % 37.76 18.21 13.83 40.72 25.72 14.87 32.13 14.59 13.62 Average per capita consumption, G 58,372 22,335 10,300 60,989 23,360 11,322 52,657 21,520 10,086 Average food share in total consumption, % 46.7 57.5 62.4 42.4 48.9 47.2 56.0 64.3 65.5 Access to improved sanitation, % 49.6 23.2 11.1 57.9 35.4 24.9 31.3 13.4 8.2 Access to tap water, % 15.3 10.6 5.4 18.4 18.1 17.8 8.6 4.6 2.8 Access to a sustainable source of energy, % 58.3 28.2 7.9 73.0 51.3 32.4 26.1 9.8 2.8 Dwellings made w/nonhazardous materials, % 80.7 57.1 28.6 88.2 75.8 53.6 64.3 42.2 23.4 Food security rate, % 88.2 72.2 43.4 88.0 71.9 33.7 88.6 72.5 45.5 Sources: ECVMAS 2012; World Bank and ONPES 2014. Note: The estimates for the poor exclude the extreme poor. Variables reflect the percent share of individuals. a. Share of households relative to the average household. Figure 2.2  Food Insecurity, 2012 unable to cope with shocks and adopt coping Percentage of population strategies that damage human capital. Overall, 23 percent of households in extreme 100 poverty changed their nutritional profile in 80 response to a major shock. In particular, if the shock hits the entire community, a stag- 60 gering 56 percent of households in extreme 40 poverty change their nutritional behavior, as opposed to 37 percent of resilient households. 20 The extreme poor are also more likely to re- 0 move their children from school, particularly Non Poor Non Poor Non Poor if the household is experiencing a change in poor poor poor No food Went to Entire day composition (such as the birth or death of a bed hungry with no food household member) or a decline in the mon- Never (0 days) etary support from outside the household, Rarely (3–10 times/mo) which is often used to pay school fees. Often (more than 10 times/mo) Women and girls are particularly vulner- Sources: ECVMAS (2012) and World Bank staff calculations. able. Although poverty does not differ by gender, despite sizable progress in both edu- cation and health outcomes, adult women poverty are even more vulnerable, particu- are still less well educated than adult men larly those in extreme poverty: among poor and are more likely to be illiterate. Their use households, 95 percent experience at least of health services is also still very limited. one economically damaging shock per year. Women in Haiti are also significantly less The poor and vulnerable have limited access likely to be employed and earn significantly to public support, given the low capacity of less than men (box 2.1). Finally, ­­gender-based the State.10 Most assistance arrives to them in violence and low participation in the public the form of remittances or support from sphere are widespread in Haiti. churches, other nongovernmental institu- tions, and donors. Despite recent efforts to expand social assistance provision under the Drivers EDE PÈP framework since 2013, the pro- Recent findings suggest that the decline grams remain too fragmented, too small, and in extreme poverty was partly driven by not targeted enough to make a significant ­ rising non-agricultural labor income. Labor difference. The majority of the poor continue income increased by about 3½ percent on to lack access to formal safety nets that could average per year, especially among men. allow them to smooth their consumption GDP data suggest that the sectors that grew over time, prevent irreversible loss of human the most over the recent past were construc- capital, and avoid destitution. tion, telecommunication and transport, all Coping strategies perpetuate poverty. In concentrated in urban areas, partly explain- the absence of formal instruments to manage ing the improvements of urban living con- risk effectively, the poorest households are ditions. As a result, employment probably 34 Trends and Profile in Poverty and Shared Prosperity BOX 2.1  Gender Inequalities in Haiti Women and girls in Haiti still face greater obstacles to attend school. Despite progress in school en- rollment among younger cohorts, adult women are still worse educated than adult men and are more likely to be illiterate. Adult men have, on average, two more years of education than women and are over 10 percentage points more likely to be literate. Early withdrawal from school can have lifelong consequences. Underage marriage and teen pregnancy, for instance, represent additional risks for girls who do not attend school: 17 percent of Haitian women are married in adolescence, compared with 2 percent of men, this number dropping among girls with higher education (Cicmil 2013). They deal with greater health challenges. Maternal mortality, at 380 deaths per 100,000 live births, is still five times higher than the regional average (WHO 2014)a, partly because a high percentage of birth de- liveries take place outside a health care facility (65 percent). Poor nutrition is a threat for both children and mothers: according to the MSPP, IHE, and ICF International (2013), 22 percent of children are stunted or too short for their age, while nearly half of women aged 15–49 have anemia. The prevalence of HIV/AIDS is higher among women (2.7 percent) than men (1.7 percent), reflecting both knowledge differentials among other factors: only 15 percent of young women have correct information on how to prevent sexual HIV transmission compared to 28 percent of young men (Boesten and Poku 2009). Women are significantly disadvantaged in the labor market. Holding constant several social and de- mographic characteristics, one finds that women are 20 percentage points more likely than men to be unemployed and, if working, 6 percentage points more likely to be in the informal sector. Wages among women are also 32 percent lower than wages among men. Over two-thirds of this difference is unexplained by observable characteristics, suggesting that discrimination could play a role. Gender-based violence is widespread. Gender-based violence is a chronic problem: according to the MSPP, IHE, and ICF International (2013), 13 percent of Haitian women have experienced sexual vio- lence, and 29 percent of women who have ever been married have experienced spousal violence, whether emotional, physical, or sexual. Vulnerability is particularly high among internally displaced people in camps and areas affected by the 2010 earthquake: a survey in 2011 indicated that 64 percent of interviewed pregnant adolescent girls had become pregnant after being raped (PotoFi 2012). Female participation in the public sphere is low. Only 4 percent of all parliamentary seats were occupied by women in the most recent legislature, placing Haiti well below the regional average of 26 percent and among the countries with the lowest rate (136th out of 142 countries). At the national level, as of April 2014, 8 of 23 ministers and 3 of 20 secretaries of state were women.b At the local level, women account for only 12 percent of all mayors. The government has taken steps to expand women’s representation, including by creating the Gender Equality Office in Parliament and amending the Constitution to stip- ulate a quota of at least 30 percent women in all public offices, but enforcement is problematic. a. This number is not recognized by the Ministry of Public Health and Population (MSPP). b. CEPALSTAT Database (http://estadisticas.cepal.org/cepalstat/WEB_CEPALSTAT/Portada.asp?idioma=i), Statistics Division, United Nations Economic Commission for Latin America and the Caribbean, Santiago, Chile. experienced a structural shift over this pe- higher than in the agricultural sector riod towards non-agriculture activities, con- (ECVMAS, 2012).11 tributing to better living standards: the Most of the labor force remains em- average monthly salaries in the formal or ployed in the agriculture and urban infor- informal sectors are two to four times mal sectors. Reliable data on employment Trends and Profile in Poverty and Shared Prosperity 35 in Haiti are scarce. Available employment highly dependent on agriculture: almost and household surveys would suggest, nev- 80 percent of households engage in farming. ertheless, that since 2007 the share of for- Moreover, for half the households, farming mal employment in total employment has is the sole economic activity. Unfortunately, increased by about 3 percentage points, the returns to agriculture are low and unreli- driven mainly by jobs in the administra- able, and the activity resembles a subsistence tion and NGOs (figure 2.3).12 Despite this ­ strategy rather than reliance on a produc- expansion, formal employment remains tive economic sector.14 Labor productivity small with agriculture and urban ­ informal is particularly low among the poor with sectors still providing most of the employ- only 20 percent of them using fertilizers ment with about 40 percent and 47 per- and pesticides. Moreover, even though the cent of the labor market, respectively area of cultivated land is only slightly f igures 2.4 and 2.5). 13 While wages have (­ smaller among the poor than among the tended to increase in the small formal sec- non-poor (1.2 hectares versus 1.6 hectares, tor over that period, they have stagnated in respectively), the poor spend two to four agriculture and in the informal sector, and times less on fertilizers, pesticides, seeds, overall 70 percent of workers still earn less and labor. Increased productivity is cor- than the minimum wage. related to greater access and use of inputs, The persistence of rural poverty reflects as well as crop diversification for both poor households’ reliance on a low-performing and non-poor agricultural households agricultural sector. Rural livelihoods are (World Bank and ONPES 2014). Figure 2.3  Change in Composition Figure 2.4  Composition of the of Labor Market, Workforce Ages 15+, Labor Market, Workforce Ages 15+, 2007–12 2012 Percentage points Percent 6 7.5 4 2 1.4 1.1 1.0 0.7 0.4 0.1 –4.7 0 –2 –4 39.0 –6 47.0 re Os n al Es d al tio ol 3.0 m m tu SO NG eh r r 1.4 ra ul fo fo us 1.2 ric ist in e Ho 1.0 at Ag in e at m iv A ricultur Pr Admin SOEs iv Ad NGOs Pr Hous hold Priv t inform l Priv t form l Sources: Enquête sur l’Emploi et l’Economie Informelle (EEEI) 2007 and ECVMAS 2012. Source: ECVMAS 2012. 36 Trends and Profile in Poverty and Shared Prosperity Figure 2.5  Breakdown of Haiti’s Population by Labor Status (2012) H iti n popul tion (2012) 10.8 m Workin Youn r th n popul tion 15 rs-old 6.96 m (64%) 3.84 m (36%) Unp id work rs P id work rs Un mplo d In ctiv ( id –f mili l ) 3.1 m (44%) 1.4 m (20%) 1.8 m (26%) 0.7 m (10%) Non-f rm A ricultur Non-f rm s lf- Non-f rm Un mplo d busin ss Discour d s ctor mplo m nt w work (ILO) own rs 0.9 m (65%) 1.2 m (39%) 0.8 m (28%) 0.8 m (26%) 0.5 m (35%) 0.2 m (7%) Inform l Form l 0.4 m (52%) 0.4 m (48%) NGO, f mil , Priv t Public nd oth r 0.22 m (56%) 0.14 m (34%) 0.04 m (10%) Sources: World Bank calculation; ECVMAS 2012. Participation in the nonfarm sector would private transfers (domestic transfers or remit- seem critical to emerge from poverty in rural tances from abroad) in Haiti increased from Haiti. Engaging in the nonfarm ­ sector in 42 percent to 69 percent between 2000 and rural areas reduces the probability of being 2012. Workers’ transfers from abroad have poor by 10 percentage points. The typical represented more than a fifth of Haiti’s GDP nonfarm job in rural areas is a one- or in recent years, and originate mainly from the two-person shop engaged in small retail. United States and the Dominican Republic Although still low paying, this type of activity (DR). While remittance flows from the U.S. yields better returns than those accruing to are larger, transfers from the DR are more farming. About 40 percent of non-poor likely to reduce poverty since they tend to households participate in the nonfarm sector, benefit poorer households located in rural a participation rate that is 1.5 times higher areas. Furthermore, the 2010 earthquake than the participation rate among the poor. resulted in unprecedented aid flows in the ­ Private transfers and international aid, form of money, goods and services. These have also contributed to the decrease in pov- ­ external flows have also contributed in reduc- erty. The percentage of households receiving ing poverty over the period, especially in the Trends and Profile in Poverty and Shared Prosperity 37 Metropolitan area which attracted most of Migrating, both domestically and abroad, the assistance (in large part because Port-au- seems to be a profitable income generation Prince was hit hardest by the earthquake). solution for many households. In many devel- In addition to migration abroad, domes- oping countries, households turn to migration tic migration is highly prevalent, with a clear as a way to increase available income, in a bias towards urban areas. While one million context of limited economic opportunities or Haitians are estimated to live abroad (see returns to education. In particular, the deci- Chapter 1), almost two million Haitians have sion of a household to educate or send one of migrated internally. Indeed, 24 percent of the its members abroad can be seen as an invest- population has moved from their place of ment: families incur upfront costs (such as ed- birth (compared to 20 percent in 2001).15 In ucation fees, transport tickets, seed money, terms of areas of residence, it is clear that the network mobilization, visa) in order to reap flow of migrants has an urban bias, and espe- future income gains from better labor oppor- cially the Aire Metropolitaine. Over half of tunities.16 Data suggests a strong link between the current population of the Metropolitan schooling and mobility in Haiti (­ figure 2.7). If area is composed of migrants, against less the initial costs of international migration than 30 percent in other urban areas and less are too high for poorer households, moving than 10 percent in rural settings (figure 2.6). within the country represents a second-best In terms of departments, 65 percent of the option: when controlling for individual and total migrant population now resides in the households characteristics, educated migrants Ouest, followed by 9 percent in Artibonite and 8 percent in the Nord. This pattern sug- gests the attraction of migrants to urban cen- Figure 2.7  Schooling of Adults ters: Gonaïves in Artibonite and Cap Haitian Living Outside Department of in the North are the two largest centers after Birth (15+), 2012 Port-au-Prince. Percentage of total population 35 Figure 2.6  Born Elsewhere, 2011 (Total Population, Area of Living) 30 Percent 25 20 60 52.4 15 50 10 40 28.5 5 30 0 20 g y y ar r ov r nd e ab pe lin ar ar co low y e 8.4 im im d p o an te u ho 10 pr pr se e sc et ar ple e e pl et et No m nd m pl pl 0 Co co Co m m y co Co Metropolitan Other urban Rural In area se Sources: Haitian authorities and World Bank staff calculations. Source: ECVMAS 2012. 38 Trends and Profile in Poverty and Shared Prosperity in urban areas earn on average between 20 40 (from 10 to 15 percent in the first quintile and 30 percent more than their counterparts and from 7 to 12 percent in the second). in rural areas.17 Private transfers are associated with lower People have been moving to urban areas poverty levels. Since over 60 percent of poor also to seek better public services. The State and extreme poor households rely on some being absent in many places outside the main kind of transfers, excluding those has a size- urban centers (as discussed in Chapter 1), peo- able effect on poverty headcounts.19 If private ple have been migrating to the Metropolitan transfers were to stop flowing, extreme pov- areas and to other urban areas to seek better erty would increase 5 points (from 23.8 per- access to services. Although, the quality of cent to 28.9 percent), and moderate poverty sanitation remains low (only 31 percent of the ­ would rise 4 percentage points (from 58.5 population had access in 2012 to improved percent to 63 percent). As poor households sanitation), the ratio for rural areas is even have less access to remittances, excluding much lower (16 percent).18 Access to im- them from the total budget has a smaller im- proved sources of drinking water is similar in pact on poverty rates: extreme poverty rises urban and rural areas, at 55 and 52 percent, to 25.4 percent and moderate poverty to 60.3 ­ respectively. However, most of the remainder percent.20 Some regions like the South East of the urban population (36 percent) pur- or the South benefit particularly from private chases safe water directly from vendors, while domestic transfers, while in the North, the most of the remainder of the rural population North East, and the Nippes remittances from (44 percent) does not have this option and abroad play a greater role in reducing pov- uses unimproved water sources (river water erty (figure 2.8). or unprotected wells) with a high probability of contamination. Access to energy (electric- ity, solar, or generators) expanded only Figure 2.8  Contribution to slightly because of gains in urban areas, Extreme Poverty Reduction accompanied by stagnating levels in rural ­ Percentage points areas, which remained at 11 percent. 0 Private transfers have played an impor- –1 tant role in supporting households’ income in Haiti. Larger private transfers coming –2 from domestic and international migrants –3 have played a significant role in raising in- –4 come in urban areas, together with more –5 nonagricultural employment opportunities. –6 The share of private transfers in total house- hold income has increased from 20 to 42 –7 ite h es percent in the first and from 15 to 20 percent st h st Ce t er an st Ar nse rt es No ut pp nt Ea No Ea Gr We on No W So A Ni tib d´ h h in the second quintile of the income distri- h ut rt rt So bution. In rural areas, where poverty has Domestic transfers Remittances stagnated and inequality increased, the share Sources: ECVMAS 2012; Haitian authorities and World Bank staff of private transfers also rose for the bottom calculations. Trends and Profile in Poverty and Shared Prosperity 39 Notes 1. All statistics in this section come from 9. According to the National Food Security ECVMAS 2012 and ECVH 2001, unless oth- Coordination Unit, the food insecurity rate erwise indicated. was 28 percent nationwide and 48 percent in 2. Pedersen and Lockwood (2001) based on the rural areas in 2011. To measure food insecurity, 1999/2000 Budget and Consumption Survey. the unit uses a composite indicator composed As a point of comparison, Azevedo et al. of both quantitative and qualitative measures. (2013) studies 16 countries having substan- The numbers contained in this chapter, on the tially reduced poverty over broadly a decade, other hand, refer exclusively to the food se- ranging from a 10 percentage points decline curity indicator of the Food and Agriculture (Costa Rica) to more than 40 percentage Organization of the United Nations, which is points (Romania). based on food intake. 3. Although not completely comparable, the 10. Only about 8 percent of the Haitian popula- monetary poverty indicator from 2000 was tion received non-contributory social assis- calculated using a methodology similar to the tance benefits (such as scholarship, food aid, new official one, including the use of a national or other transfer) in 2012 (World Bank and food poverty line against per capita household ONPES 2014). consumption. The reduction in consumption 11. The informal sector is defined by the poverty is also confirmed by trends in non- International Labor Organization as unincor- monetary well-being indicators. Income based porated enterprises (household businesses) measures of moderate poverty suggest that that are not registered, do not keep formal it declined from 77 percent in 2001 (ECVH accounts, and are not in the primary sector 2001) to 72 percent in 2012 (ECVMAS 2012). (agriculture). Consumption based poverty measures are 12. The increase in the share of employment in considered the most accurate in capturing agriculture stems from a rise in family aid. welfare levels, especially in countries with high The intuition would be that in dire times, levels of rural poverty and significant income young people and women who would not have volatility. The new official Haitian poverty worked were called into the labor force. measure is consumption based. 13. The private informal sector in figure 2.5 in- 4. Haiti’s GDP per capita amounted to US$1,575 cludes non-farm self-employment, non-farm (PPP) in 2013 and ranked 161 out of 186 coun- business owners, and the informal non-farm tries in the UNDP Human Development Index. wage work in figure 2.6. 5. These rates are based on per capita consump- 14. Since 2000, the sector has performed poorly, tion and were calculated using the 2012 official contracting by 0.6 percent annually as a con- Haitian moderate and extreme poverty lines of sequence of repeated adverse climatic shocks. 82.2 HTG (US$1.98) and 41.7 HTG (US$1), In 2012, agricultural production narrowed respectively. by 1.3 percent following a series of droughts, 6. Based on per capita income data. heavy rains, and hurricanes, which generated 7. According to the 2012 demographic and crop and seasonal income losses of 40–80 health survey ([DHS] MSPP, IHE, and ICF percent. See “Haiti Food Security Outlook” International 2013), 7 in 10 women aged (October 2012–March 2013), Famine Early 15–49 years do not seek medical support for Warning System Network, Washington, DC, lack of money, while 43 percent do not do so http://www.fews.net/central-america-and- for lack of transport (see chapter 3). caribbean/haiti/food-security-outlook​ 8. In 2012, the coverage of deliveries within insti- /­october-2012. tutions was 8.4 times greater among the high- 15. Enquête des Conditions de Vie en Haiti (ECVH est income quintile (76 percent) than among 2001) and Enquête des Conditions de Vie des the lowest quintile (9 percent). Ménages Apres le Séisme ECVMAS 2012). 40 Trends and Profile in Poverty and Shared Prosperity 16. Clemens (2011) estimates that unskilled 19. Official poverty rates are based on consump- Haitian farmers migrating to the US could tion, not on income. The exercise above increase their annual incomes by a factor consists on subtracting transfers from total of 20. consumption and recalculating poverty rates, 17. Clemens (2014) gathers evidence that mi- thus relying on the assumption that house- gration tends to increase with income until holds consume all of income received from a certain threshold, suggesting that poorer transfers (no savings). households would like to migrate, but do not 20. Acosta et al. (2006) use the 2011 ECVH to have the means to afford it. estimate the effect of remittances on pov- Improved sanitation includes flush toi- 18. erty. Using an income-based welfare measure lets and  improved latrines. According to and  the international poverty lines of U$1 the United Nations Children’s Fund and and U$2 a day for extreme and moderate the World Health Organization, an im- poverty, respectively, they find that excluding proved sanitation latrine is one that hygien- ­ remittances increased extreme poverty from ically separates human excreta from human 53 percent to 60 percent and moderate poverty contact. from 71 percent to 76 percent. Trends and Profile in Poverty and Shared Prosperity 41 3. Trends and Drivers of Growth Trends lost infrastructure.1 Access to concessional borrowing has also ­ financed a higher level of Overall, Haiti’s growth performance in public investment (box 3.1). the last four decades has been disappointing. The economy’s performance has, ­ however, From 1971 to 2013, GDP growth averaged 1.2 been held back by negative productivity percent a year, much lower than the average of shocks. Real GDP growth in Haiti during the the LAC region (3.5 percent) and the average period was disappointing, despite ­ increases in of economies at the same level of develop- the factors of production, implying negative ment (3.3 percent) (­ figure 3.1). The few peri- total factor productivity (TFP). This negative ods of positive growth were short-lived, often con­tribution could ­reflect the natural disasters followed by a contraction in economic activity. and political instability Haiti has experienced.2 Furthermore, in light of the country’s impor- Only during a limited period of time after the tant demographic growth, the level of GDP per ­ return to democracy (1995–99), growth in TFP capita even fell by 0.7 percent per year on was positive. During these episodes, improve- average between 1971 and 2013. While Haiti’s ­ ments in productivity could have stemmed GDP per capita has expanded in line with the from the reallocation of factors towards eco- average of LAC during the 1970’s, it has lost nomic sectors with higher productivity, higher ground ever since (figure 3.2). Even low income “quality” of inputs (labor and capital) or the countries (LIC) have on average seen their adoption of new technologies (IMF 2004). GDP per capita taking off since the mid-1990s, Natural disasters have indeed hampered leaving Haiti behind. Haiti’s economic performance over the last Haiti’s growth has been largely driven by forty years. Between 1971 and 2013, Haiti’s an expanding labor force. Looking at the fac- economy has been subjected to numerous tor decomposition of Haiti’s growth, one no- shocks with adverse effects on growth. While tices that labor has been the main driver of the natural disasters have occurred almost every country’s economic expansion (table 3.1). year during this period (­ figure 3.3), the ex- Increases in life expectancy and reductions of tent of the economic consequences depends fertility rates (from 5.7 births/woman in 1971 on the nature of the disasters: the effects of to 3.2 in 2012) have contributed to the increase floods tend to be limited to certain areas, with in the share of the working age population: fewer people affected, while those of hurri- population ages 15–64 represented in 2012 canes tend to be more widespread (­ figure 3.4). 59 percent of the total population, up from In 2008, for instance, Haiti was hit by four 54 percent in 1971. Capital accumulation, al- hurricanes, causing a contraction in agri- though timid for most of the period, has been cultural production by more than 7 percent. picking up since 2000. Even if overestimated, The 2010 earthquake was also destructive capital accumulation is likely to have been lifted and led to a significant loss of human life by the re-engagement of ­ donors and the aid and displacements, as well as damage to flows following the 2010 earthquake to r ­ ebuild Trends and Drivers of Growth 43 Figure 3.1  GDP, 1970–2013 (1970 Figure 3.2  GDP per Capita, 1970– = 100) (Constant 2005) 2013 (1970 = 100) (Constant 2005) US dollars US dollars 450 220 400 200 180 350 160 300 140 250 120 200 100 150 80 100 60 19 0 19 3 76 19 9 82 19 5 88 19 1 94 20 7 00 20 3 06 20 9 12 9 9 8 7 0 7 0 7 19 0 19 3 76 19 9 82 19 5 88 19 1 94 20 7 00 20 3 06 20 9 12 19 19 19 19 9 19 20 20 9 8 7 0 7 0 7 19 19 19 19 19 20 20 LAC Haiti LIC LAC Haiti LIC Source: UN and World Bank. Source: UN and World Bank. Table 3.1  Contributions to Growth (` = 40%) Percentage points 1970–2012 1970–1981 1981–1995 1995–1999 1999–2012 Capital stock 0.3 –0.2 0.1 0.3 1.0 Labor 1.3 1.2 1.2 1.6 1.3 Human capital per labor 0.5 0.4 0.6 0.4 0.3 Total factor productivity –1.0 2.5 –3.2 0.3 –1.8 Real GDP 1.1 4.0 –1.3 2.6 0.8 Source: World Bank staff calculations. infrastructure, dwellings, and, to a lesser extent, 1986 and 2014 the country had 18 changes of jobs. In 2012, the country was hit by two hurri- president and important changes in regime canes (Isaac and Sandy) and one drought, dition, according to the Cross- (table 3.2). In ad­ leading to negative growth of 1.3 percent in National Time Series data archive, there were the agricultural sector. 20 major cabinet changes in Haiti from 1986 to Political instability has also taken a toll 2006 (meaning a change of Prime minister or on growth. The departure of Jean Claude 50 percent of the cabinet posts are assumed ­ Du­valier initiated a period of intense politi- by new ministers). Empirical investigations cal instability in Haiti (figure 3.5). Between have shown that such cabinet changes are 44 Trends and Drivers of Growth BOX 3.1  Petrocaribe and Haiti The Petrocaribe initiative, created in 2005, has 18 members.a The agreement aims to promote cooperation among state energy operators, in terms of technology, energy policy, joint exploration, refining, sales and investment in the energy sector. The most relevant aspect of the Petrocaribe agreement is the stable supply of oil from Republica Bolivariana de Venezuela to other members, at favorable financing conditions. Through the agreement, Republica Bolivariana de Venezuela commits to providing oil to the members based on quotas established bilaterally at international market prices. Haiti joined the initiative in March 2007 and started benefiting from the agreement in October 2007, with a quota of 14 thousand barrels per day (b/d). However, only a fraction of the bill needs to be paid in cash, the remaining is financed on concessional terms. Petrocaribe Financing Conditions Price per barrel (dollars) Share of financing (% of total) Repayment period (years) Grace period (years) 150 70 23 2 100 60 23 2 80 50 23 2 50 40 23 2 40 30 23 2 30 25 15 2 Source: BMPAD. The sale of these oil products in the domestic market creates considerable resources for the government. Since 2008, the Petrocaribe flows have amounted cumulatively to 25 percent of GDP. An autonomous agency of the Ministry of Finance (the Bureau de Monétisation du Programme d’Aide au Développement, BMPAD) manages these resources on behalf of the Haitian government. BMPAD plays an intermediary role between Haitian fuel purchasers and the Venezuelan supplier. Petrocaribe resources have been used to finance investment and social projects, as well as support the electricity sector. Whenever the government decides that new projects will be financed using these resources, it publishes a resolution listing these projects and the amounts to be financed in the official gazette. Disbursements into the projects are re- ported in BMPAD’s webpage on a regular basis. These projects suffer nevertheless from the same shortcomings than the rest of the Public Investment Program, notably a lack of proper assessment, prioritization and monitoring (see box 1.1). A recent audit of these in- vestments by the CSCCA highlights serious shortcomings in the procurement and super- vision of contracts, including over-priced infrastructure and abandoned projects. a. Antigua and Barbuda, the Bahamas, Belize, Cuba, Dominica, Grenada, Guatemala, Guyana, Haiti, Honduras, Jamaica, Nicaragua, the Dominican Republic, St. Kitts and Nevis, Saint Lucia, Saint Vincent and the Grenadines, Suriname and Republica Bolivariana de Venezuela. Trends and Drivers of Growth 45 Figure 3.3  Annual GDP Growth vs. Occurrence of Natural Disasters, 1971–2013 10 5 0 –5 –10 –15 74 80 83 86 89 95 98 04 10 13 92 71 77 01 07 20 20 19 19 19 20 19 19 20 19 19 19 19 19 20 Ocurrence GDP growth Sources: UN, World Bank and EM-DAT: The OFDA/CRED International Disaster Database. Figure 3.4  Annual GDP Growth vs. People Affected by Natural Disasters, 1971–2013 10 4,500 4,000 5 3,500 3,000 0 2,500 2,000 –5 1,500 –10 1,000 500 –15 0 71 74 77 80 83 86 89 92 95 98 01 04 07 10 13 19 20 19 20 20 19 20 19 19 19 20 19 19 19 19 People affected (thousands) (rhs) GDP growth Sources: UN, World Bank and EM-DAT: The OFDA/CRED International Disaster Database. detrimental to growth (e.g., Aisen and Persistent and growing urban crime has Veiga 2013). Their results imply that Haiti also hampered investment and growth. Al­ would have grown by 1.2 percentage points though the Haitian countryside has histori- faster if it had achieved an average level of cally been relatively safe, the country’s homi­ political stability.3 cide rate doubled between 2007 and 2012 46 Trends and Drivers of Growth Figure 3.5  Annual GDP Growth vs. Changes in Government, 1971–2013 10 5 0 –5 –10 –15 74 80 86 10 13 77 83 89 92 98 01 04 71 95 07 20 20 19 19 19 20 19 19 19 19 20 19 19 20 19 Occurrence GDP growth Sources: UN, World Bank, and International Media. Table 3.2  Haiti’s Governments, 1971–2014 Adm. Position and name Periods served Time served 1 President Jean Claude Duvalier 4/1971 to 2/1986 15 years 2 President Henri Namphy (civilian-military junta) 2/1986 to 8/1988 2 years 3 President Leslie F. Manigat 2/1988 to 6/1988 4 months 4 President Henri Namphy (military junta) 6/1988 to 9/1988 3 months 5 President Prosper Avril (military junta) 9/1988 to 4/1990 20 months 6 President Hérard Abraham (military junta) 3 days in 4/1991 3 days 7 President Ertha Pascal Trouillo 4/1990 to 2/1991 10 months 8 President Jean-Bertrand Aristide 2/1991 to 9/1991 7 months in Haiti 9 President Joseph C. Nérette 10/1991 to 5/1992 7 months 10 No de-facto president n.a. n.a. 12 President Emile Jonassaint 6/1994 to 9/1994 3 months 13 President Jean-Bertrand Aristide 10/1994 to 2/1996 16 months 14 President René Préval 2/1996 to 2/2001 5 years 15 President Jean-Bertrand Aristide 2/2001 to 2/2004 3 years 16 President Boniface Alexandre 2/2004 to 5/2006 2 years 17 President René Préval 5/2006 to 5/2011 5 years 18 President Michel Martelly 5/2011 to present 4 years (ongoing) Sources: World Bank and International Media. Note: n.a. = not applicable. Trends and Drivers of Growth 47 from 5.1 to 10.2 homicides per 100,000 in- when the embargo ended, and never returned habitants.4,5 Crim­ inal activity has been espe- to pre-embargo levels afterwards.6 In the be- cially widespread near the industrial areas of ginning of the 1970, agriculture was the most Port-au-Prince and in urban slums, forcing important sector, representing almost 40 per­ businesses to absorb high security costs and cent of the value added, but fell to less than periodically shut down or move their opera- 20 percent in 2012. Authors have argued that tions. The effects of violence are felt most the embargo, because it hampered access to deeply on the most vulnerable members of important inputs such as seeds, contributed the population. Residents of poor, marginal- to the decline of productivity in the sector ized neighborhoods are 40 times more likely (IICA 1997). High fragmentation of land, low to be murdered than other urban dwellers. levels of technology applied in the sector, and These events have significantly reshaped soil deterioration have also constrained agri- the structure of the Haitian economy over the cultural produc­ tivity. As seen above, the sec- last four decades. The US embargo (1991–94) tor is also very vulnerable to natural disasters. triggered a strong decline in manufacturing Some of those factors have also contributed (figure 3.6). This sector represented 18 percent to the great volatility in the contribution of of GDP in 1990, shrunk to 10 percent in 1994 the agricultural sector to GDP growth while Figure 3.6  Economic Structure, 1970–2012 Percentage of value added 40 30 20 10 0 70 73 76 79 82 85 88 91 94 97 00 03 06 12 09 20 19 19 19 19 19 19 19 19 19 20 19 20 20 20 Agriculture Manufacturing and mining Wholesale and retail Other services Transport and communications Construction Source: World Bank staff calculations. 48 Trends and Drivers of Growth Figure 3.7  Decomposition of Value Added Growth by Sector, 1971–2013 Percentage points 10 5 0 –5 –10 –15 –20 91 93 95 97 99 01 03 05 07 09 11 13 71 73 75 77 79 81 83 85 87 89 20 19 20 19 19 19 20 19 19 19 19 19 20 19 19 19 19 19 20 20 19 20 Primary Secondary Tertiary Not defined Value added growth Source: UN data and World Bank staff calculations. Figure 3.8  Urban Population, construction and services have been more 1971–2013 steady drivers of growth, especially since the Percent total population earthquake. (figure 3.7). The decline of agriculture has been ac- 60 companied by an acceleration of urbanization (figure 3.8). An increase in population con­ 55 centration in the cities has also pushed the 50 demand for services. Together, the activities of transport and communication, commerce 45 and other services currently represent almost 40 half of the value added in Haiti. Another sec- tor that has risen in importance is construc- 35 tion, also driven by urbanization. Growth in 30 construction was the main driver of the rise in the share of secondary activities in GDP. 25 20 Drivers 15 Weak structural policies and institu- 19 1 74 19 7 80 19 3 86 19 9 19 2 95 20 8 20 1 04 20 7 20 0 13 7 0 tions seem to continue to be at fault (figure 3.9). 7 0 9 8 8 9 1 19 19 20 19 19 19 Source: World Bank. Poor governance and political instability Trends and Drivers of Growth 49 have been identified in previous assessments 143 out of 148 countries. According to these as the major impediments to sustainable de- indicators and the investors’ survey of the ­ velopment, along with weak public sector GCI, the major constraints in Haiti’s business capacity and accountability, followed by low ­ environment include poor infrastructure and levels of education and badly deteriorated limited access to finance, as well as cumber- infrastructure World Bank (2002). Doing ­ some administrative procedures to start a Business Indicators rank Haiti at 174 out of business and comply with tax obligations. 185 countries, while the 2013–14 Global Although governance indicators have Competitiveness Index (GCI) ranks it improved, they remain low. The increase in ­ Figure 3.9  Obstacles to Growth a. Governance is a concern... b. … Infrastructure inappropriate ... Governance indicators, 2013 (percentile rank) Quality of port infrastructure, 2013 Control of Low income corruption countries Rule of law Latin America and Caribbean Government Small Caribbean effectiveness countries Voice and Haiti accountability 0 50 100 0 1 2 3 4 5 6 7 Haiti LIC LAC Source: World Economic Forum. Note: 1=extremely underdeveloped, 7=well developed and efficient Source: World Bank. by international standards. c. ... As well as access to electricity ... d. … Human development is low… Access to electricity, 2011 Human Development Index, 2013 (overall score) (percentage of population) Least developed Low income countries countries Caribbean Latin America and Caribbean Latin America and Caribbean Haiti Haiti 0 50 100 0 0.2 0.4 0.6 0.8 Source: World Bank. Source: UNDP. figure continues next page 50 Trends and Drivers of Growth Figure 3.9  continued e. … And the business climate unfavorable… f. … Restricting competition and Business indicators, 2015 or latest (doing business) opportunities… Intensity of local competition, 2015 49 Private credit (in % 31 of GDP, 2013) Latin America 19 and Caribbean Strengh of minority 5 inverstor protection 5 Low income (0–10) 2 countries 31 Starting a business 29 Haiti 140/144 (days) 97 0 50 100 0 3 5 7 LAC LIC Haiti Source: World Economic Forum. Source: World Bank. Note: 1=low, 7=high (scores and position of Haiti). g. ...And encouraging the population h. ... Or in low paid jobs to seek its future abroad… Underemployment, 2012 Stock of migrants of Haitian origin, (percentage of employed population) 1960–2010 (in thousands) Sud-Est 79.7 1,200 Nippes 79.5 Grand’Anse 79.3 1,000 Nord-Ouest 78.2 Sud 77.7 800 Centre 77.1 Nord 76.0 600 Artibonite 75.5 Nord-Est 74.6 400 Ouest 56.4 200 National 70.0 Urban 57.3 Rural 80.3 0 1960 1970 1980 1990 2000 2010 0 20 40 60 80 100 Source: World Bank. Source: ECVMAS. political stability after the low-point of 2004 for control of corruption, rule of law, govern- came with an improvement in a number of ment effectiveness, as well as voice and ac- governance indicators (figure 3.10). The im- countability have all registered improvements plementation of the Automated System for in Haiti over that period, although remain- Customs Data (ASCUYDA) was stepped up, ing low compared to the average of the LAC for instance, after the earthquake.7 Indicators region and to that of low-income economies Trends and Drivers of Growth 51 Figure 3.10  Change in Governance inhabitants) results in a low level of usage of Indicators, 2004–13 financial services. Today, only 27 percent of Percentile rank all adults in Haiti have an account at a formal financial institution (compared with 45 percent in LAC). Coupled with limited credit Control of corruption information availability, unclear property rights and weak contract enforcement, ac- Rule of law cess to credit is problematic, particularly for small- and medium-sized enterprises, lim- Government effectiveness iting their capacity of taking advantage of op- portunities, growing their operations and generating employment. Furthermore, Voice and accountability competition to provide financial services is 0 5 10 15 20 limited. The sector is highly concentrated, with close to 80 percent of bank assets Source: World Bank. being held by the three largest banks, and— according to the 2008 FSAP—10 percent of (figure 3.9). Haiti still ranks lowest in the re- individual borrowers receiving approximately ­ gion in control of corruption or government 80 per­cent of total loans. Roughly 40 percent effectiveness. of total credit is extended to commerce and In particular, a real property cadaster and services activities, whereas agriculture re- land registry system is needed. Less than five ceives less than 0.2 percent (figure 3.11). percent of Haiti’s land has been surveyed, Haiti’s infrastructure also falls short. Island and more than 75 percent of rural land con- economies are extremely dependent on the tracts are drawn up according to traditional quality, frequency and cost of the means of procedures and are not officially registered. transport that link them to markets which The situation is just as complicated in urban represent both market outlets for their prod- areas. The courts are clogged by numerous ucts and supply sources for the needed im- land disputes. Increased legal security of ported goods. The efficiency and effectiveness land rights is essential to protect both indi- of transport, whether by road, by sea, or by vidual households and investors (who hesi- air, therefore contribute to the competitive- tate to engage in investment projects that ness of these countries. The quality of trans- might end up being challenged in court), port and logistics services in Haiti is poor, and to develop and maintain a real estate however, the country ranking 144th out of market. Several attempts have been made at 160 countries on the World Bank’s logistics the national level to develop a registry sys- performance index (LPI) in 2013 and trail- tem, without success. ing behind its competitors (figure 3.12).8 Access to finance is challenging for both, The road network in Haiti is in poor condi- households and medium- and small-sized en- tion and many parts of the territory are not terprises. The very small number of well connected: 60 percent of people living branches and service points of formal in rural areas do not have access to main ­ institutions in the country (6.35/100,000 roads during the rainy season.9 Haiti is also 52 Trends and Drivers of Growth Figure 3.11  Credit by Sector, as of Q2 of 2014 a. Bank assets, June 2014 b. Exposure of the banking sector portfolio by sector, (percent of total bank assets) June 2014 (percent of the portfolio) 6.1 3.7 3.4 6.0 23.3 17.0 3.1 3.1 9.2 2.5 0.2 1.8 11.3 24.2 29.7 27.0 15.4 13.0 Capital Bank Citibank Scotiabank Transport and Construction Electricity, gas, Sogebel BUH BPH communication and water Unibank Sogebank BNC Agriculture Commerce Services Manufacturing Personnal Real estate Source: Central Bank of Haiti (BRH). less integrated into the global shipping line Haiti’s competitors in light manufacturing network than many developing countries.10 and apparel industry. Since reliability is low, Furthermore, the costs of loading and unload- Haitian industries must have backup power ing a standard container at Port-au-Prince sources, generally diesel generators. Fur­ ther­ are by far the highest of the Caribbean ports more, the supply of electricity covers only a (figure 3.13).11 small proportion of the country, with rural In particular, the provision of electricity is areas being particularly neglected. Per capita problematic. The electricity current is unsta- consumption of electricity in Haiti is substan- ble and there are frequent power cuts and tially lower than in other Caribbean countries, surges, which can result in serious damage to for example, it is only two percent of the level industrial equipment. Despite this poor service, in the Dominican Republic (figure 3.14). the cost of electricity is among the highest in Access to clean water and improved sani- the region. In 2011, industry was charged tation in Haiti also remains limited. Access to $0.32/Kwh, compared to $0.18 in Nicaragua, quality water is a significant challenge in $0.17 in the Dominican Republic and $0.06 in rural areas and small towns of Haiti (table 2.1). Bangladesh IFC (2011). These countries are Rural areas are generally served through water Trends and Drivers of Growth 53 Figure 3.12  Logistic Performance Index, 2014 (1=Lowest, 5=Highest) 3.5 144 of 166 3.0 2.5 2.0 1.5 1.0 0.5 0 s s ss om nt r ur lit cin co ct lin st ll s pm qu ru tr Cu m st ics hi nd r Ti ls ist Ov fr In n in Lo tio ck rn Tr t In B h m s, Th Dominic n R public J m ic Gu n Haiti Source: World Bank. Figure 3.13  Port Tariffs Estimated Figure 3.14  Electric Power Cost per TEUa, 2009 Consumption, 2011 US dollars kWh per capita 900 Port Ev r l d s 800 Mi mi 700 Port of Sp in, T&T 600 Port of Piont Lis s, T&T 500 Dominic n Ports Authorit 400 Pur to Rico Ports Authorit Haiti 300 T rmin l V rr ux 200 Priv t Op r tors (public b rths) 100 0 200 400 600 0 s bl n Ho or iti St v dorin Wh rf Ship ch r sb i l u liv ur Ha d pu c m ic Bo R ini nd lv r t m c Source: TranSystems. S Ni Do Gu El Based on a 700 TEU ship, discharging and loading a mixture of a.  20-ft and 40-ft containers. b. Pilotage, Port Dues, Light Dues, and Dockage. Source: World Bank. 54 Trends and Drivers of Growth points equipped with hand pumps, while small off the job, and to adapt to changing job towns are served with gravity-fed piped sys- requirements. tems from which water is delivered through Against this backdrop, light manufactur- stand posts, kiosks and household connec- ing is a success story, but with challenges. The tions. A substantial portion of the system is apparel industry has expanded rapidly since not operational, however, for lack of suffi- 2009 with exports especially to the US mar- cient funds for operation and maintenance, ket helped by preferential access agreements. and less than ten percent are equipped with These exports have been growing at 18 percent chlorination devices.12 Sani­ tation is equally a per year. The industry is estimated to employ challenge with only 16 per­ cent of households about 30’000 workers at present and is ex- having access to improved sanitation in rural pected to expand further in the future. It is by areas. Both urban and rural populations far the largest formal employer. Wages remain, mostly rely on individual on-site sanitation nevertheless, low, while rapid migration and solutions, as sustainable collection and treat- informal housing settlements around these ment of sewage are practically non-​­ existent. firms can rapidly create developmental chal- Due to low access to improved sanitation fa- lenges due to limited infrastructure and gov- cilities, open defecation is frequent. In this ernment services to meet workers’ needs. fragile environment, facilitated by flooding Nonetheless, apparel wage-earners have a and heavy rain, cholera broke out and rapidly vested interest in stability, and could be a spread through Haiti in 2010. source of poverty reduction and shared pros- Surveys also point to the lack of quali- perity. However, as long as port and electricity fied labor ILO (2010). One major con- costs remain high compared to its com­ pe­ straint faced by enterprises in Haiti is the titors, Haiti’s comparative advantage rests on difficulty to find technicians that are well cheap labor, which will force wages to re- qualified, particularly in new technologies. main low. This forces the country to position itself as a Tourism has a long way to go, but could low cost producer for goods and services contribute significantly to growth. Haiti’s s requiring little skills. An opaque labor potential in tourism is rich, not only because market may be at fault, with no institutional of beautiful nature and beaches, but also be- mechanism to enable the ex­ change of infor- cause of its world-class cultural and natural mation between labor demand and sup- heritage, such as the Citadelle, Palais Sans ply. More generally, the level of education Souci, and Ramiers, classified as World among the adult population remains low: Heritage sites by UNESCO. Demand for 45.7 percent of the adult population (60.5 tourism services is growing with close to percent of households’ heads) have never one million visitors registered in 2013 ac- attended school or have not completed cording to the World Tour­ ism Organization. primary education. Not mastering basic Most of these visitors are from the Haitian skills such as literacy and numeracy when diaspora in the US and cruise ship visitors, starting work represents a major impedi- with overnight visitors in some of the emerg- ment for their insertion in the labor mar- ing tourist resorts and high-end hotels repre- ket and, more importantly, for their ability senting a much smaller number, thus limiting to absorb post-school training either on or the economic impact of tourism so far. Trends and Drivers of Growth 55 24,000 people arrive by cruise ship every Figure 3.15  Informal Employment, week, but there are no excursions within Haiti 2012–22 (Working Age organized. Haiti’s performance also pales in Population 15+) comparison to its neighbor, the Dominican Percent Republic, where 90 percent of its more than five million visitors annually stayed overnight 90 and expenditure amounted to about 8 percent 88 of GDP in 2013. Overall, while the potential 86 is significant, Haiti’s tourism industry has a 84 long way to go, including overcoming the 82 image of a dangerous country, marred by vio- lence and cholera, and, once again, its weak 80 transport and electricity infrastructure. 78 Mining holds potential, but comes with 76 social and environmental challenges. Haiti 74 shows significant potential in mineral and 72 metal mining. Based on preliminary explor- atory activities, Haiti is expected to have sub- 70 stantial deposits, especially in gold, silver, 20 22 18 16 19 13 15 14 21 12 17 20 20 20 20 20 20 20 20 20 20 20 copper, and aluminum bauxite. Given the Growth of priv t form l mplo m nt 3% limited information available, determining the Growth of priv t form l mplo m nt 10% economic potential of mining in the country Sources: ECVMAS 2012 and World Bank staff calculations. has to be approached with caution. Estimates suggest a reserve value of about US$5–6 billion at current market prices. Successful ­ How could the earnings of the under- exploitation of this wealth will, however, employed and those employed in the in- require a much more concerted effort by formal sector be improved? The formal the government to put in place the appro- manufacturing and service sectors have the priate policy, regulatory and institutional potential to be an important source of em- framework. At present, the framework for ployment, but because they hire such a awarding exploration and exploitation li- small share of the labor force, even with censes suffers from a lack of transparency, very high growth rates, they will not be able leading some larger mining companies to to absorb more than a fraction of the new halt their exploration work. Most impor- entrants. Figure 3.15 simulates the share of tantly, managing the environmental and the labor force in the informal and agricul- social impact of mining will need to be ad- ture sectors. Even with a ten percent annual dressed, including through an open and growth rate—in line with the strong job cre- transparent community and stakeholder ation observed over the past five years, im- engagement mechanism to ensure that plying about 30,000 new formal jobs a local communities affected by these opera- year—formal employment would only ac- tions can help shape these operations and count for about 20 percent of the labor benefit from them. force in a decade. Most Haitians are thus 56 Trends and Drivers of Growth likely to continue working in low-productiv- range of 40 percent. Transport services are ity agriculture and non-agriculture informal often unavailable, and Haiti’s two rainy sea- sector activities over the foreseeable future. sons often make roads impassable. When The challenge is thus to enhance the pro- factoring in mangoes that spoil in the field ductivity—hence the earnings—of those al- before they can be transported, one ex- ready employed, while at the same time porter estimated that losses could reach up creating more formal jobs.13 to 50 percent. In production zones with poor The constraints discussed above seem to accessibility, producers and traders are forced be particularly present in agriculture. Haiti’s to transport goods on horse­ back or by foot. agricultural potential has yet to translate into The first segment of transport, from the farm commercial agribusiness. Notwithstanding to the first point of collection, is the most the fact that 40 percent of jobs in Haiti are in costly in terms of actual and implicit trans- agriculture, the country is far from develop- port costs and losses. Lastly, poor storage ing a commercial agribusiness sector. The ag­ conditions reduce the shelf-life of pro­ duce, culture sector in Haiti has been declining ri­ making producers anxious to sell at any for many years, the result of neglected rural price and cutting the season down to its infrastructure, weak research and extension, min­ imum time range. poorly defined land tenure, limited access to Regarding the informal sector, improve- credit, and under-investment in human capi- ments in skills and access to inputs could en- tal. Even in potentially successful niches such hance income prospects. Most of the nonfarm as coffee, cacao or mangoes, the vast majority enterprises in Haiti operate on a small scale of agricultural production continues to be re- and are in the informal sector. Nonfarm busi- lated to low-productivity subsistence farm- nesses are micro in nature and have an average ing. There has been a general shift out of long of 1.5 workers, including the owner. An analy- cycle towards short cycle production (annual sis of correlates of successful self-employed cereal cultures and vegetables). These factors, performance was carried out, using three mea- along with high population growth, have con­ sures: revenues per worker (in log), revenues tri­buted to shrinking parcel sizes; complex, per worker in the top 30 of the distribution, informal tenure systems that discourage long- profits per worker in the top 30 of the distri- term investment; and cultivation of marginal bution (results are presented in appendix B). lands with steep slopes, encroaching on for- With the caveat that this exercise indicates a ests, destroying watersheds, and aggravating conditional correlation rather than a causal the tenuous environmental situation. relation, the analysis helps nevertheless iden- Limitations in logistics are particularly tify the characteristics and inputs that could constraining for agriculture supply chains. be significantly more relevant to improve Firms in the agriculture sector do not get to ­ understanding of what will it take to elevate choose where to set up shop and need to productivity and generate jobs in the self-­ transport their products to the markets. employed and small business sector. The re- Supply chains analyses carried out on six sults suggest that human capital (experience, products showed that transport-associated skills), and access to physical capital such as losses could be extremely high.14 For infor- electricity and water, are associated with mal products, loss levels could be in the higher returns. Trends and Drivers of Growth 57 Notes 1. There are reasons to believe that investment infrastructure, international shipments, lo- figures in Haiti are over estimated. Both trea- gistics competence, tracking and tracing, and sury financed and donor’s financed projects timeliness. have a significant share of current expendi- 9. The sharp deterioration of Haiti’s road net- tures recorded as investment. work, the inadequate packaging of fruit 2. A simple regression would suggest that natural and  vegetables and the use of vehicles un- disasters and Cabinet changes could account suited to the transport of agricultural goods for 60 percent of the variance in TFP. are responsible for considerable commer- 3. Calculations based on a background paper cial  losses. In the North West, for instance, prepared for the SCD by Kassia Antoine, Raju transport costs for a stem of bananas from Singh, and Konstantin M. Wacker. the farm to the primary market are esti- 4. UNODC Homicide data. mated  to represent 25  percent of the sale 5. The main state penitentiary collapsed follow- price or 45 percent of the profit margin ing the earthquake, allowing most of the crim- (World Bank 2013). inals to escape. Combined with a still weak 10. According to UNCTAD’s liner shipping law enforcement and justice system, this event connectivity index (LSCI), Haiti ranks lower ­ enabled criminal gangs to grow, as well as po- than the main players in the Caribbean: litical assassinations, carjackings, and kidnap- Haiti’s score of 5 is much lower than scores pings to continue. by Jamaica (21) and the Bahamas (27), and 6. The Inter-American Development Bank the Dominican Republic (23) that host (2009) mentions for instance that in 1984 the ­ transshipments ports. The liner shipping vast majority of U.S. baseballs were assembled ­ connectivity index (LSCI) is made up of five in Haiti. The same seems to have been true for components: 1) the number of companies that the Rubik’s Cube. 125,000 workers were em- provide services from/to a country’s ports; 2) ployed in the sector at the time and the em- the size of the largest ship providing services bargo brought this number close to zero ten from/to a country’s port (measured in Twenty years later. foot Equivalent Units–TEU); 3) the number 7. Customs provides about 30 percent of bud- of services that connect the country’s port(s) get resources. In addition to fraud preven- to other countries’ ports; 4) the total number tion, the customs department also has the of ships operating from/to the country’s port; responsibility of stopping unfair competition and 5) the total container carrying capacity of and unwarranted earnings, while at the same those ships. time facilitating fluid and transparent trade 11. According to a 2009 study mandated by the transactions. The main entry point for indi- Haitian Chamber of Commerce, costs in viduals and merchandise is Port-au-Prince. Port-au-Prince are five times higher than The customs services situated at the port and in  the ports of the Dominican Republic the airport of Port au Prince cover respec- (World Bank 2013). tively more than 90  percent of imports and 12. DINEPA’s performance monitoring system, 70 exports of the country. They also generate which monitors roughly half of the water sup- 84 percent of customs receipts. In addition, ply systems, estimates that 41 percent of the eight customs posts are situated around the stand posts and 45 percent of the kiosks do not perimeter of the country and there are also deliver water. mobile brigades. 13. For an early discussion of these themes, 8. The LPI is based on six core dimensions of see  for instance Delatour and Dnggleby trade-related services, including customs, (1993). 58 Trends and Drivers of Growth 14. Supply chain analysis shows how the effects of Gros-Morne), rice (Estère, Pont Sondé, constraints (costs, times and losses) are distrib- Desdunes, Verrettes, Artibonite) and vetiver uted throughout the logistics chain. In case of (Les Cayes, Sud). In addition, three informal Haiti, surveys and semi-structured interviews supply chains have been examined: mangoes have been conducted with government offi- (Carice and Capotille, Nord-Est), and coffee and cials, shipping companies, exporters for three avocadoes (Central Plateau Region—Savanette, formal supply chains: mangoes (Artibonite, Baptiste—Centre). Trends and Drivers of Growth 59 4. Sustainability Environment increases the country’s vulnerability as a whole to natural disasters. The metropolitan Urbanization is increasing Haiti’s vul- area of Port-au-Prince, for instance, is a nerability to natural disasters. As a safety coastal plain surrounded by the mountain valve, migration probably contains poverty chain Morne l’Hôpital, which severely limits and provides relief to many families through the availability of land for growth. However, private transfers. However, moving to urban the urban area has experienced an explosive areas is putting pressure on the environment, expansion in the past two decades with a and leading to the settlement of large popu- near total absence of urban planning and lations in vulnerable places (figures 4.1 and growth regulation, much of it on precarious 4.2). Their settlement in urban areas takes terrain. The trees and other vegetation that place usually in the cheapest, less accessible, used to cover the Morne have now almost and least serviced land, more often than not, completely disappeared, giving way to a in disaster-prone areas. Floods, hurricanes spontaneous and anarchic urbanization oc- and earthquakes have considerable inten- curring on both state-owned and private sity in urban areas that make the precarious land. and informal urban dwellers particularly Furthermore, natural disasters imply sig- vulnerable. In terms of the number of com- nificant fiscal costs. Work was carried out munes, urban areas are the most prone to jointly with the Ministry of Economy and suffer floods. Twenty-seven communes lay Finance (MEF) to develop several hypothe- within the highest risk of flood-prone area. ses regarding contingent liabilities for the The severity of hurricanes seems also to be government in the event of disasters. The more acute in urban areas. Two-thirds of results suggest that the annual average fiscal urban communes and over 70 percent of cost for the government can be estimated at urban population are affected by hurricanes around one percent of GDP. This estimated with medium to high impact. cost was based on modeling used by the This unplanned urban growth has created CCRIF, and is linked to damages caused by vulnerabilities for both the physical environ- tropical hurricanes (0.8 percent of GDP) and ment and those who live in it. Settlers tend by earth­quakes (0.2 percent of GDP). In the to occupy areas that are not only prone to absence of a financial strategy to manage risk from disasters, but are also sometimes this risk and provide appropriate fiscal buf- nature’s protection from risk. Coastal cities fers, natural disasters will require reallocat- are throughout the world seeing an increase ing public funds and involve high in shoreline retreat, which exposes them to opportunity costs. a greater risk of flooding partly due to de- Less frequent, higher-impact events can graded coastal ecosystems by human activity pose a real threat to the sustainability of public (Sanchez-Reaza and Carletto 2013). This slow finances. While frequent, low-impact events degradation of the environment in turn Sustainability 61 Figure 4.1  Population Living in Figure 4.2  Population Exposed Flood Prone Areas to Hurricane Damage—High and Percent of population Medium Intensity Percent of population 18 80 16 70 14 60 12 50 10 40 8 6 30 4 20 2 10 0 0 Urban Intermediate Rural Urban Intermediate Rural Source: World Bank staff calculations. Source: World Bank staff calculations. er- constrain fiscal space, less frequent, high­ Rapid migration to urban neighborhoods, impact events can imply significant fiscal associated with poverty, unemployment, po- shocks and pose a real threat to the sustain- litical and social marginalization, limited ability of public finances. Although with large access to services, and weak governance, ­ opportunity costs as discussed above, the gov- could fuel the formation of gangs that en- ernment can mobilize a portion of its financ- gage at-risk youth and draw them in. These ing with respect to low-impact events through grievances have also contributed to inter- budget reallocations or the use of its deposits. generational conflict, undermining social However, these funds will not be sufficient to cohesion and hindering the capacity of many address major events that give rise to costs communities for productive collective action that far exceed the government’s financing ca- Marc et al. (2012). pacity. The fiscal shock associated with a hur- Violence in Haiti tends to be predomi- ricane that occurs once every 50 years, for nantly an urban phenomenon. Looking at instance, is estimated at about 10 percent of the number of politically violent events GDP, representing a much greater risk for fis- across arrondissements between 2000 and cal sustainability and macroeconomic stability. 2010, one observes that urban centers are by far the arrondissements with the largest number of conflict events per capita (map 4.1, Social Tensions figure 4.3). This observation is confirmed Rapid urbanization could also be using the MINUSTAH data for the most feeding greater violence. Youth violence ­ recent period covering criminal events is often expressed as a reaction to perceived (map 4.1, figure 4.4). Even when controlling social and economic exclusion—grievances that for relevant variables, our empirical work can easily be mobilized by political interests.1 suggests that the share of urban population 62 Sustainability 0 5 10 15 20 25 30 Por Les 0 50 100 150 200 250 300 t-au - Por Gonaïv t- e Fort Prince Le C au-Prin s -Lib ap-H ce Mira erté goâ a Sain ïtien n t-M Cro Hinc e arc ix-d he Oua Hinch es- Jacm nam e Le C Bouque el in ap-H ts Le T Mireb the rou- a Les aïtien du- lais Cay La G Nord Jéré es on mi Léo âve Oua Léogâne g Per 100,000 inhabitants Per 100,000 inhabitants nam e Sain Mira âne goâ L´Ar inthe t- ne Por c a Les Raphaë t- haie Côt l Les de-Pai Las eaux Gon x cah oba Mire aïves s L bal Gro Jacme s-M l Le T ascaho ais rou- bas Fort orne -Lib Les du-Nor Côt d J erté eau Cro Port érémie P Source: Armed Conflict Location & Event Data Project (ACLED). Sain ix-d -de- t-Lo ort-Sa x es-B P Source: United Nations Stabilization Mission in Haiti (MINUSTAH). uis d lut ouq aix u Des uets Bell Nord e salin M -An Figure 4.4  Criminal Activity, 2010–14 Cer Plais es Figure 4.3  Political Violence, 2003–06 L´Ac armel se ca la anc ul-d ad L´Ac So e u-N e ul-d urce ord u-N Sain Corai ord Gro t-M l B s-M arc Les ainet orne Mar Cayes Aqu m L´Ar elade Cer Le Li in cah ca m Môl ai Ans la Sou bé e Sa A e e d´ rce int- quin Nico Sain Hainau t-R lt Sain la t-Lo Valliè s Des aphaël uis d res salin u No Môl e Sa Baradè es Les P r int- r Cha ort-Sa d rdon lut Gra nde Nico es ni -Riv Va las Gra L Le ères ière llières nde ´Anse- Limbé du N -Riv à o ière -Veau Bor rd du N gne ord B Cor La G ainet onâ B ail Les P v Bell orgne e-A Cha laisan e rdon ce n L´An n Ans Baradè se e d´ se-à ières Hain res -Ve ault au Sustainability 63 Map 4.1  Violence and Criminal Activity a. Political violence, 2003–06 b. Criminal activities, 2010–14 Source: Armed Conflict Location & Event Data (ACLED). Source: United Nations Stabilization Mission in Haiti (MINUSTAH) in the arrondissement is positively and sig- Macroeconomy nificantly correlated with conflict intensity (box 4.1). Macroeconomic stability has been Migration is associated with greater vio- maintained. Haiti’s macroeconomic envi- lence, as social ties erode. The share of im- ronment has significantly improved over the migrants in the arrondissement’s population past decade, as was noted in the most recent is also positively correlated with conflict in- IMF ex post assessment (IMF 2014c). tensity. New migrants are likely to have less Inflation was brought back to single digits access to local social networks which pro- and has remained contained at these levels vide economic and social support in case of since 2009, reaching 4.5 percent at end-FY14 negative shocks. Such buffer mechanisms (figure 4.5). International reserves have been are an effective way to reduce tensions in built up and maintained at an adequate level, Haiti. Controlling for the presence of ex- covering about 5 months of imports. tended families confirms this interpretation. Domestic revenue mobilization has also im- Given the dearth of data at the arrondisse- proved, although it is still low by regional ment level on such measures, we resort to standards (13 percent of GDP in 2012). the 2001 household survey which enables us Between 2009 and 2011, Haiti’s stock of to construct department-wise share of ‘ex- ­ external debt drastically shrank following tended families’ and ‘extended kinship’ debt relief. In addition to benefiting from the combined. The former is a household com- HIPC and the MDRI, Haiti received addi- posed of more nuclear fam­ i­ lies and the latter tional debt cancellation in the aftermath of is a network of more families not necessarily the 2010 earthquake. As a result, total exter- linked by blood re­ lations. Adding this nal debt fell from 29.4 percent of GDP to variable to the regression explains the mi- 8.9 percent of GDP in 2011. gration variable away: a denser social net- The IMF in its most recent assessment es- work can help prevent or resolve disputes timated that the exchange rate remains through non-violent means. broadly in line with its medium-term 64 Sustainability BOX 4.1  Drivers of Conflict in Haiti—An Empirical Analysisa In order to understand better the role of various socio-economic factors in explaining the ebbs and flows of conflict across Haiti in the 2000s, a cross-sectional analysis across the 42 Haitian arron- dissements was run. This strategy has the advantage of allowing us to control for national level fac- tors that may not be properly observable in cross-country analyses and may be endogenous to conflict. In addition the within-country analysis relies on a more homogenous definition of con- flict events than a cross-country analysis, although coming at a cost of not allowing for the investi- gation of the role of factors that do not vary across spatial units or over the period of analysis, such as the political regime and the oligopolistic nature of the economy. We used data on conflict events from the Armed Conflict Location & Event Data (ACLED) Project, which collects and codes media reports on civil and communal conflicts, militia interactions, vio- lence against civilians, rioting, and protests across various countries since 1997. In the case of Haiti the data is available between 1997 and July 2010. We aggregated the individual conflict events at the department and sub-department (“Arrondissement”) level to construct time varying conflict inten- sity measures at our unit of analysis. We complement these data with data on crimes collected by MINUSTAH mainly on the basis of Haiti National Police (HNP) reports. These data are more recent, covering the period 2010 to September 2014, although not strictly comparable with the conflict events data as they include a broad variety of criminal activities, such as robberies, thefts, homicides, rapes, illegal possession of firearms. Most of these events are not related to political violence or political actions as it is the case for the events recorded by ACLED. Nevertheless, they provide a useful additional dimension and allow us to gauge the latest developments in violence in Haiti. The results of our analysis suggest that violence in Haiti in the last decade tends to be predomi- nantly an urban phenomenon. Both political violence and criminal activities are concentrated in urban centers. A higher share of immigrants from other areas of the country is also associated with higher conflict intensity. This result is mainly driven by the fact that a higher migrants’ share is as- sociated with a lower presence of social networks (measured as extended families and kinship), which reduce conflict. The analysis would also indicate that past conflict and land ownership are all important correlates of violence. In addition, the results suggest that economic shocks matter a great deal in explaining the pattern of conflict in Haiti. In particular, increases in remittances are associated with a reduction in conflict intensity, consistent with the idea that by raising incomes they raise the opportunity cost of partici- pating into violent activities. Finally, the evidence suggests that election periods amplify the effect of economic shocks on conflict. a. Based on a background paper prepared for the SCD by Babatunde Abidoye, Massimiliano Calì, and Evans Jadotte. fundamentals. The real effective exchange 2007, 2010, and 2013 Article IV rate appreciated by about 23 percent during Consultations concluded that, on balance, 2006–14, against the background of large re- the Haitian gourde was not misaligned with mittances and external aid inflows (figure fundamentals, and that the economy experi- 4.6). IMF Staff analysis accompanying the enced an equilibrium response to large Sustainability 65 Figure 4.5  Macroeconomic Environment a. Inflation, 2004–14 (percentage) b. Gross international reserves, 2004–14 (millions of US$) 30 3,000 25 2,500 20 2,000 15 1,500 10 1,000 5 500 0 0 14 14 10 11 12 13 10 11 12 13 04 04 09 09 08 08 06 06 05 05 07 07 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Sources: IMF and Haiti’s Statistical Institute (IHSI). Source: Central Bank of Haiti (BRH). Figure 4.6  Effective Exchange Rate, 2006–14 Index, 2010 = 100 120 110 100 90 80 70 4 9 8 08 10 06 10 06 11 12 13 13 11 07 7 -1 l-0 -0 -0 r- v- b- n- p- g- n- c- n- g- ar t- ay Ap ar Ju No Au Fe Ju Se Ja De Oc M Au Ja M M NEER REER Source: IFM. external aid and private remittance inflows. Haitian currency. Enhancing competitive- Since 2010, a moderate depreciation of the ness in Haiti is predicated on structural nominal effective exchange rate coupled with ­ reforms to boost productivity through im- inflation above that of key trade partners has proving infrastructure and the business envi- led to a gradual real appreciation of the ronment, as discussed later on.2 66 Sustainability Haiti’s twin fiscal and current account Bank-Fund Debt Sustainability Analysis deficits have, nevertheless, widened. The concluded that Haiti’s debt profile faced fiscal deficit averaged 2.2 percent of GDP only a moderate risk of debt distress during 2006–09, but widened to 7.1 per­ (IMF 2015). cent and 6.3 percent of GDP in 2013 and 2014, respectively (figures 4.7 and 4.8). This widening fiscal deficit mainly reflected Recent Progress in transfers to the public electricity company (EDH), which amounted to about 1.5 per­ Poverty Reduction cent of GDP in 2013, and the fuel retail The exceptional level of aid following price freeze that had a fiscal cost of almost the 2010 earthquake is rapidly declining 2 percent point of GDP (box 4.2). The coun­ eroding fiscal buffers. Donor assistance has terpart of the rising fiscal deficit was a ris- declined from its peak after the earthquake ing external current account deficit. Both to 7 percent of GDP in 2013 (figure 4.9). deficits are largely financed by concessional Additional fiscal revenue could not be mobi- external borrowing. Despite this steady lized or public spending cut to offset fully accumulation of debt, the latest joint this substantial decline in resources and Figure 4.7  Central Government Figure 4.8  Current Account Fiscal Balance, 2004–14 Balance, 2004–14 Percentage of GDP Percentage of GDP 3 2 2 1 1 0 0 –1 –1 –2 –2 –3 –3 –4 –4 –5 –5 –6 –6 –7 –7 –8 –8 04 05 06 07 08 09 10 11 12 13 14 20 20 20 20 20 20 20 20 20 20 20 04 05 06 07 08 09 10 11 12 13 14 20 20 20 20 20 20 20 20 20 20 20 Sources: Ministry of Finance (MEF), IMF, and World Bank staff calculations. Sources: IMF and Central Bank of Haiti (BRH). Sustainability 67 BOX 4.2  Electricité d’Haïti (EDH) Due to poor management, the performance of the national, vertically integrated electricity utility Electricité d’Haïti (EDH) has deteriorated over time. Worsening commercial performance has led to a lack of infrastructure maintenance and has rapidly deteriorated the quality of electricity ser- vice, including frequent service interruptions and large voltage fluctuations. The sector’s institu- tional framework is obsolete (e.g., not allowing EDH to tap into important renewable energy potential), sector policies are out of date, planning and monitoring of sector activities is inade- quate, and vested interests have hampered reform efforts. The effect of the earthquake on power generation, transmission, and distribution further aggravated EDH’s weak performance by delay- ing key modernization activities (e.g., new billing system and the rehabilitation of the Peligre hy- dropower plant). In 2015, EDH’s losses stood at 57 percent of the electricity generated (of which commercial losses represented nearly 35 percent). EDH’s weak grid infrastructure, poor commercial performance, and inadequate controls over sub- contracted electricity generation by independent power producers (IPPs) have led to a financial drain on government resources. Due to its inability to meet electricity demand and in an attempt to expand electricity availability, EDH has subcontracted part of the production of electricity to IPPs.a Unable to cover its operating expenses, including fuel costs and power purchases in part be- cause of low bill collection rates, EDH has relied on fiscal transfers from the Treasury averaging US$200 million annually in recent years (equivalent to 10 percent of the national budget and 1–2 percent of GDP).b Past governance reforms to preserve the integrity of public resource management in the electric- ity sector have produced modest results. Achievements to-date include: (a) EDH Commercial Recovery action plans (2012–2013; 2013–2014; and 2014–2015); (b) a National Directions Paper (endorsed by Cabinet on September 15, 2013); and (c) a government Energy White Paper (in draft). Since 2014, an active dialogue on the management and reform of the energy sector has been initiated within the government’s Energy Commission and with the strategic partners in the sector. a. IPPs provided 60 percent of total electricity generation in 2011. b. Electricity transfers are unbounded due to their dependence on international oil prices, electricity supply, unsanctioned theft and bill non-payment, inter alia. government deposits needed to be drawn lower international oil prices imply a higher down (figure 4.10). Wider fiscal and current payment rate and less concessional financ- account deficits are also raising concerns ing. With lower international oil prices, Haiti about a crowding out of private sector credit has experienced a substantial decline in the with a monetary stance becoming tighter to availability of concessional financing. contain the mounting pressures on Haiti’s International oil prices are expected to re- currency. main low in the near future, making it even The recent decline in international oil more challenging for the country to balance prices is further tightening fiscal resources. the need to ensure macroeconomic stability According to the Petrocaribe agreement, while protecting the social and investment 68 Sustainability Figure 4.9  International Aid, Figure 4.10  Government 2008–25 Deposits, 2009–14 Percentage of GDP Percentage of GDP 20 8 18 7 16 6 14 5 12 4 10 3 8 2 6 1 4 0 2008 20 9 20 0 20 11 2012 2013 2014 2015 2016 20 17 2018 20 19 20 0 20 21 20 22 2023 20 24 25 0 1 2 2010 2011 2012 2013 2014 20 Source: IMF. Sources: IMF and World Bank staff calculations. Figure 4.11  Petrocaribe Financing, Figure 4.12  Project Activities 2008–17 Financed by Petrocaribe, 2008–13 Millions of US dollars Percentage of GDP 450 400 9% 350 6% 300 250 9% 200 65% 150 4% 2% 100 5% 50 0 08 09 10 11 12 13 14 P P P 15 16 17 20 20 20 20 20 20 20 20 20 20 Social Agriculture Others Water and Actual World Bank projections sanitation Crude at US$100 Crude at US$50 Economy Energy Infrastruture Crude below US$40 Source: BMPAD. Sources: Government of Haiti and World Bank staff calculations. Sustainability 69 Figure 4.13  Histogram of Annual Per Capita Consumption, 2012 Gourdes Extreme Moderate line Vulnerability line 200 line (G$30,080) (G$36,096) (G$15,262) 180 Number of individuals in thousands 160 140 120 100 80 60 40 20 0 0 12 0 18 0 24 0 30 0 36 0 42 0 48 0 54 0 60 0 66 0 72 0 78 0 84 0 90 0 1,0 00 1,0 00 1,1 00 1,2 00 1,2 00 1,3 00 1,3 00 1,4 00 1,5 00 0 50 50 0 0 0 0 0 0 0 0 0 0 0 0 0 50 ,5 ,5 ,5 ,5 ,5 ,5 ,5 ,5 ,5 ,5 ,5 ,5 ,5 ,5 5 5 5 5 5 5 5 5 6, 2, 8, 4, 0, 6, 2, 8, 4, 0, Annual per capita consumption in gourdes Sources: ECVMAS 2012 and official poverty lines; World Bank and ONPES calculations. spending needed for social cohesion and that most households have a consump- growth (figures 4.11 and 4.12). tion budget close to the poverty line (figure Against this backdrop, poverty could rise 4.13). In this context, a decline in activities again. As mentioned in Chapter 2, increases related to aid and reconstruction could ad- in non-agriculture income linked to sectors versely affect the living standards of many related to aid or reconstruction have driven vulnerable households, potentially reversing part of the poverty reduction observed in part of the recent gains in poverty reduction. Haiti. In 2012, more than 80 per­cent of the Some progress in service delivery could also population lived with less than four dollars a be jeopardized, such as in health where al- day (PPP) and only two percent lived with most 80 percent of spending in 2012 was ten dollars or more. These numbers indicate donor-financed. Notes 1. See World Bank (2010a) and for a more detailed exchange rate is market determined, with the analysis of the data Willman and Marcelin (2010). central bank intervening to smooth out excess 2. Haiti’s exchange rate system is classified as volatility. a crawl-like arrangement, under which the 70 Sustainability 5. Priorities Prioritization Process the average growth registered in post-earth­ quake/reconstruction years (2011–14). As­ The systematic country diagnostic suming unchanged income distribution, per (SCD) seeks to identify the most important capita GDP would need to grow by about 7 constraints to and opportunities for inclusive percent per year for extreme poverty to fall and sustainable growth in Haiti. The previous to 3 percent by 2030. This would require a chapters on Haiti’s growth performance and two- to three-fold acceleration in Haiti’s growth lack of inclusiveness, as well as the diagnostic rate with respect to its best performing years: of poverty and growth, are the first steps in a very ambitious outcome. identifying priority areas and opportunities to Policies to ensure more inclusiveness are increase growth and welfare. The report now needed. Increasing growth of the bottom 40 attempts to provide a sense of prioritization percent by around one percentage point more and identify the most binding constraints to than the best observed average growth rate faster economic growth and greater shared (2005–09) would add considerable impetus prosperity, both quantitatively and through a to poverty reduction. In this scenario, per series of consultations with stakeholders and capita real GDP of the bottom 40 percent ­ the Country Team (figure 5.1). would grow twice as fast as the mean. Such a performance would cause poverty to d ­ ecrease Quantitative Assessment by 5 percentage points more than in the dis­ While overall income growth is a neces­ tribution neutral growth scenario, coming sary condition for reducing extreme poverty, it closer but still falling short from reaching the 3 is not sufficient. The obstacles to faster eco­ percent target for extreme poverty by 2030. To nomic growth and greater poverty reduction achieve the 3 percent target, a combination of have been described above, but faster economic faster and more inclusive growth would be growth alone will not be enough to bring sig­ needed: a growth of about 4 percent per year nificant improvements in the living standards with the income of the bottom 40 growing at of most Haitians. Simulations show that if twice that speed. But what would be the mea­ growth in Haiti up to 2030 were to follow its sures needed to deliver this faster growth in historical performance, poverty would hardly lower income groups? Once identified, could decline. Under the more sustained perfor­ they be ranked to assist the setting of priorities? mance of one percent real per capita growth Human capital and political stability mat­ rate observed in Haiti over the 2005–09, pov­ ter more for lower income groups. First, the erty reduction would still fall significantly correlation between potential drivers of income short of reaching the goal of extreme poverty and the income growth of the poorest 20 and of 3 percent or less by 2030 (figure 5.2). Re­ poorest 40 percent of households were esti­ ducing extreme poverty by almost half in 15 mated, using household data for a sample of years, to about 14 percent, would require an 100 countries (box 5.1). This measure gives us overall GDP growth of 3.3 percent per year, the expected impact of changing one of these Priorities 71 Figure 5.1  Prioritization Process Benchmarking Identified Country Identified Diagnostic country priority areas knowledge opportunities knowledge Figure 5.2  Extreme Poverty Simulations Percentage of population a. Poverty rate projections, 2013–30 b. Poverty rate projections, 2013–30 30 30 23.8 23.8 25 20 18.0 20 15 13.8 13.8 13.0 10 10 5 3.0 3.0 0 0 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 GDP growth 2.5% (best spell 2005/09) GDP growth best spell–bottom 40% grows double GDP growth 3.3% (average 2011/14) GDP growth 3.3% (average 2011/14) GDP growth 7.2% (goal of 3% ext. pov.) GDP growth 3.6% (bottom 40% grows double the average) Source: World Bank staff calculations. drivers on the income growth of the poorest. matter slightly more to the lower income Second, following Araujo et al. (2014), the stan­ groups. Infrastructure (measured by mobile dard deviation of each variable is taken to il­ phone usage, a common measure in the lit­ lustrate the magnitude of possible or realistic erature) and transparency, although not dis­ change one could expect in this variable. A vari­ proportionally favoring the lowest income able with a higher standard deviation would group, are nevertheless ­important for overall be expected to be “easy to change.” Multiplying income growth. the two gives a rough estimate of the expected As an alternative, Haiti’s performance in key effect of a regular change in a variable for in­ variables was benchmarked to a reference group. come developments. The analysis finds that As a reference group, the average of an aspi­ health (proxied by life expectancy), as well as rational group of countries was taken (Belize, education (measured by school enrollment) the Dominican Republic, Ecuador, El Salvador, are important for increasing income genera­ Jamaica, Paraguay, and Peru). In terms of in­ tion at the bottom of the income distribution come levels, these LAC countries broadly (figure 5.3). Reasonable changes in political stand today where Haiti aspires to be after stability (measured by the number of cabinet 2030, i.e., beyond a threshold of US$2,310, changes and within-regime instability) taken to illustrate economic emergence. 72 Priorities BOX 5.1  Benchmarking the Drivers of Shared Prosperity: An Application to Haitia We estimate economic magnitudes for key correlates of income growth at the mean, the bottom 40 and 20 percent of the income distribution across a set of about 100 countries. While econometric exercises investigating determinants of GDP growth have been numerous, less is known about fac­ tors influencing household incomes at different segments of the income distribution. Empirical work by Dollar and Kraay (2002) and Dollar, Kleineberg, and Kraay (2013) broadly rejects the idea that other factors than mean income growth would influence incomes of the poor, thus sug­ gesting that growth would be mostly distribution neutral, whereas other studies have attempted to outline key drivers of pro-poor growth (e.g., Bourguignon 2003; Ravallion and Chen 2007; Christiaensen, Punam, and Sanoh 2013). Our results add to this discussion and suggest that some factors indeed allow growth to be more beneficial for the bottom 40 or 20 percent of the income distribution. Our data set is mainly based on Dollar, Kleineberg, and Kraay (2013). The data covers household data from the World Bank’s POVCALNET and the Luxembourg Income Study databases. Household income data is organized in “spells,” i.e., income changes between two survey years, calculated as average annual log differences.b These are calculated for average income, income at the bottom 20 percent or the bottom 40 percent. We focus on those non-overlapping spells that are at least five years long, providing 299 spells for 117 countries with a median spell length of 6 years, which is also the preferred sample of Dollar, Kleineberg, and Kraay (2013). To this data, we add macroeco­ nomic variables, especially measures of political stability or institutions that are of particular inter­ est for Haiti. Based on a fixed effect estimation, the results confirm that the income at the bottom 20 and 40 percent grow with mean income. The estimated effect is a little smaller than previously thought, al­ though the estimated elasticity around 0.85 is statistically not significantly different from one. We also find that better health and education outcomes, as well as lower political instability, have an economically relevant effect for income growth that goes beyond the effect via mean income (i.e., it is stronger at the bottom of the distribution). a. Based on a background paper prepared for the SCD by Kassia Antoine, Raju Singh, and Konstantin M. Wacker. b. POVCALNET data is either income or consumption, LIS data is disposable income. We still refer to “income” in our paper. Figure 5.4 illustrates Haiti’s position relative to Again, closing the gaps in human capital its aspirational peers with respect to life ex­ and achieving greater political stability seem pectancy. It shows that the country is in the to have the highest potential for lower income lowest third percentile of the health distribu­ groups. Multiplying the gap in the indicators tion and lags behind the group average by between Haiti and the reference aspirational approximately ten years. Figure 5.5 does the group by the unconditional effects estimated same for cabinet changes, highlighting that above captures the potential income gains Haiti other peers suffered from comparable political may experience from closing the gap in these instability (especially Peru, Ecuador, and key variables. This idea, based on Araujo et al. Paraguay). (2014), not only highlights the areas where the Priorities 73 Figure 5.3  Economic Magnitude Figure 5.4  Life Expectancy at of Estimated Parameters Birth, 2010 Percentage points Years 80 80 Cabinet changes 70 69 70 Inflation 60 60 Inflation variation 50 50 50 School enrollment 42 40 40 Mobile phone use 33 30 31 30 22 Transparency 20 19 20 Life expectancy 10 10 3 0 0 –0.03 0.00 0.03 0.06 0.09 0.12 lv i in Par or Re ay Ja lic ca e Ec ru Gr r p C Sa ait do liz ou LA ad Pe b ai u Bottom 20 Bottom 40 Mean H Be ua pu ica ag m Source: World Bank. n El m Do 2010 value p-rank (rhs) Source: World Bank. Figure 5.5  Cabinet Changes, 2003 10 years average country lags behind most but implicitly weights this gap by the economic relevance of closing 1 .2 120 it.1 Figure 5.6 shows the income effect from 1.0 100 closing the gap relative to the reference aspi­ 97 rational group for different parts of the income 0.8 82 82 80 distribution. Variables relating to health, edu­ 71 cation, political stability clearly stand out. Again, 0.6 60 closing Haiti’s gap in infrastructure, although not disproportionally favoring the lowest in­ 0.4 40 32 come groups, is important for overall growth. 0.2 20 In turn, political instability—and conflict 9 9 risk more generally—could be reduced through 6 0 0 higher spending on social services and social protection. Security and political instability iti Sa lize Ja dor Re ica Pa blic Ec ay r ru p C do ou LA Ha Pe gu a e a ua pu Gr ica m B lv seems to be taken very often as exogenous fac­ ra n El tors, but recent international evidence suggests in m that these factors may at least partly be under Do 2003 value p-rank (rhs) the control of policy makers. Common sources Source: World Bank. of conflict include grievance, injustice, and 74 Priorities Figure 5.6  Income Effects of in agricultural and textile exports appear to Closing the Gap reduce violence in departments which employ Percentage of the gap many people in these sectors. At the same time, incidents of violence are also clustered around Life expectancy at birth political events such as elections and transi­ Mobile phone use tions, suggesting the combination of economic Cabinet changes and political factors contributing to violence. Transparency School enrollment Qualitative Assessment Inflation variation The next step in the prioritization pro­ Inflation cess was a series of consultations with stake­ 0 5 10 15 holders and the Country Team. Recognizing Mean Bottom 40 Bottom 20 the data limitations and the difficulty to un­ Source: World Bank. tangle association and causality, any quanti­ tative assessment needed to be validated and lack of economic opportunities Cederman, comple­ mented by a qualitative assessment. The Weidmann, and Gleditsch (2011) or Wimmer, results of cross-country regressions are not to be Cederman, and Min (2009). A growing liter­ interpreted mechanically, but serve as a starting ature on public spending and conflict show point for a discussion about policy priorities. For evidence that spending on education or health, this purpose, a two-day Country Team retreat for instance, could contribute reducing these was set up. In addition, continued consulta­ sources of conflict Singh, Bodea, and tions with the authorities and other stakeholders Higashijima (2014). Based on an empirical (including representatives of the established analysis that covers 148 countries from 1960 private sector and civil society) were carried out.2 to 2009, simulations for Haiti suggest that higher The Team also reached out to the Haitian dias­ welfare spending (education, health, and so­ pora and to beneficiaries of the Kore Fanmi cial protection) is associated with lower risk program.3 Furthermore, a Bank-sponsored of small-scale conflict onset (box 5.2). competition of academic papers was launched Lack of economic opportunities seems to (“The Twin Goals Awards”) to foster research be driving violence and instability within Haiti. and debate, as well as identify any issue that Building on the work presented in box 4.1, the may have been over-looked (box 5.3). These spatial differences in conflict intensity across discussions led to the identification of broad Haiti’s territory were examined to shed empir­ themes or opportunities going forward. ical light on drivers of conflict. The results sug­ Consultations validated the overall story gest that economic shocks matter a great deal line. Discussions confirmed that institutions, in explaining the pattern of conflict in Haiti. human capital, and infrastructure all mat­ In particular, increases in remittances appear tered for Haiti’s economic growth and shared to reduce conflict intensity, consistent with the prosperity. It was pointed out that economic idea that by raising incomes they raise the op­ and sector work on Haiti produced in 1980s portunity cost of participating into violent ac­ and the early 1990s had already identified tivities. Similarly, the study finds that an increase most of the challenges presented in the Priorities 75 BOX 5.2  Conflict and Welfare Spending in Haiti: What Could We Learn from Cross-Country Evidence?a Azam (2001) observed that “the occurrence of civil conflict is intimately related to the failure of gov­ ernment to deliver the type of public expenditure that people want; i.e., with a strong redistributive component such as health and education.” Governments could thus foster their legitimacy, and thereby reduce the risk of conflict, by signaling that they care about their population. This could be done in a variety of ways, such as increased spending for water sanitation, securing basic health needs, or provid­ ing a strong system of education (Stasavage 2005). Through welfare policies that influence positively the 0.08 living standards of citizens, governments can out­ Probability of small conflict spend the opposition, help­ ing gain support from a 0.06 broad segment of the population, co-opt political opposition, and decrease the incentives for organiz­ ing a rebellion. Social safety nets, transfers and in­ 0.04 vestment in public goods show people that the state cares, and prevents disadvantaged members of soci­ 0.02 ety from falling below a certain level of poverty and experiencing absolute desperation. Positive encoun­ 0 ters with frontline service officials is argued to pro­ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 vide a source of legitimacy for the state, particularly Welfare spending (% of GDP) in fragile and conflict-affected situations where the Confidence intervals (90%) state was previously mistrusted, or outright feared Predicted probabilities (Brinkerhoff, Wetterberg, and Stephen 2012). Thyne (2006) suggests that education promotes social cohesion by encouraging students to cultivate interper­ sonal skills and thus learn how to resolve disputes peacefully. Our empirical analysis covers 148 countries over 1960–2009 and uses a random effects logit model with regional dummies and a set of standard control variables to predict the probability of conflict. To measure the onset of violent conflict, we used the Armed Conflict Dataset (ACD) from the Uppsala Conflict Data Program PRIO for years 1960–2009 (Gleditsch et al. 2002, Version 4.1.) to generate a dichotomous variable that takes the value of one in years with a new conflict onset and zero otherwise. This dataset has become the standard reference for cross-country analyses of con­ flict determinants. The onset of conflict was defined with those above a threshold of 25 battle deaths per year. The results suggest that higher welfare expenditure (education, health and social) is associated with lower risk of conflict. In order to simulate the effects for Haiti, the coefficients of the cross-country regression were taken and all variables in the model set at their average for Haiti with the exception for welfare spending that we let vary. The Figure above illustrates the reduction in the risk on con­ flict onset associated with higher welfare spending. a. Based on a background paper for the SCD, prepared by Cristina Bodea, Masaaki Higashijima, and Raju Singh. 76 Priorities previous sections. It was thought already at the gangs recognized that greater economic op­ time unlikely that much economic progress portunities were in turn needed to reduce vi­ could be achieved without an improvement in olence. Improving education without better educational quality and access. Infrastructure economic opportunities fed greater emigra­ had deteriorated badly. Migration to urban tion. Discussions in villages highlighted that areas had strained water and sanitation facil­ improving seeds or production techniques ities, and increased urban vulnerabilities and without better transport or access to markets people exposures. It was argued that it would just depressed prices and could make farmers be difficult to tackle these problems effectively worse off. without first addressing the issue of gover­ Controlling cholera illustrates the need for nance. In addition, the consultations high­ a multi-sectoral approach. Cholera control will lighted a number of other dimensions. contribute to improving the health of Haitians, Maintaining the stability of the macroeco­ particularly of the poor, and fostering the long- nomic environment called for immediate at­ term development of the country. The most tention. There was a broad recognition that effective way to achieve this involves multi-sec­ progress had been made in restoring macro­ toral interventions aimed at improving health economic stability and this did not seem to be care, hygiene, an access to clean water and im­ at the moment a constraint to faster growth or proved sanitation. Cholera cannot be con­ greater shared prosperity. This assessment was, trolled without addressing primary vectors of however, backward looking and concerns were transmission of the disease such as the lack of expressed going forward. The decline in donor safe water supply, inappropriate excreta man­ assistance and lower availability of concessional agement and the absence of sanitation. Health financing following the drop in international interventions are critical to eliminate deaths oil prices raised questions about the sustain­ from cholera, but to sustainably prevent new ability of many social programs and of the re­ cases of cholera from emerging, these alone cent progress in improving access to many basic will not be enough. Concurrent and substan­ services. Even the established private sector tial improvements in water supply and sani­ expressed concern about the tighter mone­ tation coverage, including medium-term tary stance being followed. invest­ ments in this sector targeted at high Turning to more medium-term challenges, incidence and high risk zones are essential. it was stressed that progress needed to be made Building the social contract was flagged as simultaneously on several fronts. For Haiti to particularly important. Most modern democ­ break from its past and achieve meaningful racies use fiscal measures to finance the State progress towards poverty reduction, a piece­ and redistribute wealth. In return, the State is meal approach to reform focusing on a spe­ expected to provide key services. In Haiti, both cific sector would not work. Actions or reforms sides of this contract are very weak—few peo­ should not be seen as independent from each ple or businesses pay taxes and the State other, but needed to fit within a package. Vio­ struggles to provide services. State institutions lence and political instability, for instance, are are needed to provide security and public ser­ seen as a binding constraint to faster growth vices, and to write the regulations that will allow and greater economic opportunities. However, a market economy to operate. The established both the UN forces and NGOs dealing with private sector complained about the power Priorities 77 BOX 5.3  Bank-Sponsored Competition of Academic Papers In order to foster research and debate, a call was issued for research papers identifying the most critical constraints and opportunities facing Haiti in accelerating progress toward the goals of end­ ing extreme poverty and promoting shared prosperity in a sustainable manner. The competition was open to all Haitian scholars and was intended to provide a Haitian perspective to the debate, as well as identify any issue that the existing literature may have over-looked. This proposal and the rules were prepared with the Deans of the universities present in Port-au- Prince. The competition was widely advertised in academic circles and in regular newspapers, and submissions were received from Haiti (Port-au-Prince and Cap Haïtien), Canada and the United States. The papers were reviewed by a panel of World Bank researchers and leading Haitian figures, including the Presidents of the Association of Haitian Economists, the Haitian Arbitration and Mediation Division, and the North-East Chamber of Commerce, as well as professors from the American University and the Autonomous University of Mexico. The three winners were awarded cash prizes for their contributions and presented their work at a conference at the State University of Haiti in November, 2014. The event was attended by represen­ tatives from academia, civil society, government, and the international community. This was an opportunity to discuss issues related to inequality, inadequate competition and economic opportu­ nities, and the absence of a social contract between the State and its citizens which, among other is­ sues, were identified by the participants as impediments to growth. More particularly, while the papers submitted varied in approaches and emphases, they converged on a common set of themes underscoring Haiti’s scarcity of human and institutional capital. Most authors agreed that Haiti’s history of violence, exclusion, and foreign interference explained the country’s difficulty in developing a shared national vision for its future, as well as the needed insti­ tutions and social capital to implement it. All papers argued that aid dependence and the interfer­ ence of external governmental and non-governmental actors continued to prevent Haiti from developing home-grown solutions and institutional capacity. In this regard, authors stressed the State’s ineffectiveness in fulfilling basic functions, such as enforcing the rule of law, collecting taxes, and delivering services. Poor outcomes in education were seen by most as a principal cause of Haiti’s poor governance, lack of competitiveness, low productivity, and high unemployment. Urbanization and population growth were identified as major contributors to vulnerability, and the need for a more balanced spatial organization of the country was underscored. Source: World Bank 2014e. surges and outage, the poor road infrastruc­ accountable State. There was a very strong ture, or health conditions that forced it to pro­ social demand for a balanced and coherent de­ vide the missing services that it would expect velopment of the country, easier access to from the State. In line with the results of the public services as well as more job and edu­ less,” there 2010 report entitled “Voice of the Voice­ cational opportunities outside what is often was a feeling that there was a need to foster a referred to as the “republic of Port-au- greater sense of civic responsibility, overcome Prince.” A very large majority wanted to be social divisions, and establish a more able to live near places of origin without 78 Priorities having to relocate to the capital to study, to be reluctant to fund the government. make a living or access public services. For Concerns about the lack of capacity within this to happen, civil society and representa­ the Government of Haiti, as well as en­ tives of the private sector stressed the need to trenched systems of patronage, corruption, hold regular elections. There also needs to be and inefficiency could be valid. However, more productive citizen engagement, includ­ government capacity will never be built or ing mechanisms that effectively channel citi­ improved if donors continue to bypass local zens’ voice and demands to the State so that institutions in favor of non-State actors. By their concerns can be addressed. circumventing the Haitian government, do­ State capacity needs to be strengthened. The nors were perceived by many as prolonging most binding constraints are probably also this process and continuing to undermine those that would take most time to fix. Poor the public sector. governance and the need to strengthen the state To foster greater competition, barriers to were generally viewed as the most important market entry needed to be removed and constraints. Without improvements in these small private sector actors supported. areas, it was thought that little else could be Representatives of the established private achieved. Even for the fastest transforming sector recognized the lack of competition ex­ countries, improving institutional quality from isting in Haiti’s economy, a reflection for the level of a country like Haiti to that of Ghana them of the country’s unfavorable business takes in the range of 15–30 years (World Bank environment. Without proper contract en­ 2011). Relatedly, Pritchett, Woolcock, and forcement, business had to be kept between Andrews (2013) find that rates of improvement trusted partners, from credit to the whole lo­ in measures such as quality of bureaucracy gistics chain. Civil society expressed concern and administrative capacity are typically very about the higher prices and the possible lack low in developing countries, even with sub­ of innovation that limited competition im­ stantial donor effort. They conclude that “coun­ plied. Consultations flagged thus the impor­ tries like Haiti or Liberia will take many decades tance of supporting domestic informal and to reach even a moderate capability country smaller private sector actors in strengthening like India.” The lengthiness of the process their capacity to add value to the Haitian should, however, not deter immediate action, economy. This included facilitating access to but only call for more realistic expectations. available capital, stronger business manage­ In this regard, there was a lot of criticism ment capacity, better linkage of producers to about the limited aid effectiveness and ca­ markets and improved infrastructure to sup­ pacity building. The international community port such links. Direct connections with eco­ was criticized for not contributing enough to nomic partners outside of Haiti were also build the needed capacity. There was a rec­ suggested as an avenue to explore. The need ognition that in the immediate aftermath of for developing such access for agribusiness, the earthquake, the Haitian government had tourism, and manufacturing, among others, very limited capacity as almost all government outside of Port-au-Prince was underlined as buildings were destroyed and agencies were particularly important to encourage growth operating in a state of emergency. Yet years that would be geographically better balanced later, it was observed that donors continued and more pro-poor. Such growth could also Priorities 79 contribute to reducing migration to overly untargeted. Furthermore, it was argued that crowded and vulnerable urban environ­ it would be important to continue strength­ ments, particularly in Port-au-Prince. ening and mainstreaming disaster risk man­ Furthermore, there was not only a need agement activities and make sure disaster risk for more jobs, but also for better jobs. There management becomes a core component of a was a call to recognize where people actually sustainable poverty reduction and economic worked. It was pointed out that the private sec­ growth strategy. Financial protection strate­ tor in Haiti was not only limited to a couple gies, particularly if they are designed to meet of large enterprises. Most people in Haiti the needs of the population in extreme poverty, were occupied in a less visible private sec­ can help protect the government and house­ tor: in the agriculture and informal sectors, holds from the economic burden of shocks and two sectors that had been neglected so far. disasters. The potential contribution of the formal sec­ Finally, the consultations made it clear that tor in creating jobs was recognized, but it there was limited evidence to inform planning was not seen as becoming the main employer and targeting of programs and policies. The re­ in Haiti any time soon. Incomes generated in cently finalized Poverty Assessment conducted agriculture or in the informal sector are, jointly by the Bank and the Haitian authorities, however, not sufficient to pull people out of as well as the ongoing Public Expen­ diture poverty. The challenge is thus to enhance Review, are filling some of the knowledge gaps, the productivity—hence the earnings—of but many more remain, making the conclu­ those already employed in agriculture and sions of this SCD tentative. Dis­­cussions with the informal sector, while at the same time the Country Team and with stakeholders con­ creating more formal jobs. Providing greater firmed that it was difficult to carry out effec­ economic opportunities was also flagged as tive development policies without appropriate a critical factor to reduce criminal or political statistics or analytics (see appendix C for a violence, especially in urban areas, as this list of gaps). For instance, the knowledge and would decrease grievances and frustrations monitoring of the actual situation of WSS while increasing the opportunity costs of services—critical to contain water-borne dis­ joining gangs. eases such as cholera—in Haiti is limited and The importance of reducing vulnerabilities fragmented, hampering the efficient prioriti­ was also stressed. In the face of recurring shocks zation and programming of investments. and vulnerability, better risk management strat­ Policy packages—or buckets—rather than egies and targeting in social protection should independent activities in sectors were thus rec­ be a priority to protect households and indi­ ommended. Consultations identified five broad vidual livelihoods. Disaster risks need to be themes around which activities needed to be identified and understood better. Risks need organized in order to ignite a process whereby to be reduced through better awareness when Haiti could set itself on a new development investing. One million people are vulnerable path: (a) maintaining macroeconomic stabil­ to shocks that could push them into poverty. ity, while meeting developmental needs; (b) im­ Despite the significant expansion of social as­ proving statistics and analytics; (c) creating sistance provision within the EDE PÈP frame­ greater economic opportunities and better jobs, work, the system remains very fragmented and including through infrastructure and human 80 Priorities capital; (d) (re)building the social contract; and More particularly, there is an urgent need (e) reducing vulnerabilities and building resil­ to increase tax revenue and to diversify the rev­ ience. Progress on all these themes is needed enue base away from trade taxes. Despite a rise simultaneously. In light of the tighter budget in revenue of the past decade, Haiti has one of constraints facing the government, maintain­ the lowest revenue mobilization ratios in the ing the stability of the macroeconomic envi­ region. Furthermore, Haiti’s tax system also ronment, and improving knowledge and statis­ tics tends to be regressive, relying heavily on in­ to increase the effectiveness of public policy direct taxes. Corporate and personal income (including more transparent fiscal reporting) tax rates are largely comparable to the regional call particularly for immediate attention. Further average. A large share of corporate income work will be needed to flesh out in more de­ tax revenue is, however, lost because of exemp­ tail what would need to be included in the pack­ tions. Anecdotal evidence suggests a number age of reforms and actions to achieve these of taxes that generate little revenue but impose objectives, but the possible content of these high compliance costs on businesses (“nui­ buckets are spelled out in a bit more detail in sance” taxes) and reducing them would be a the next section. better alternative to attract investors. The per­ sonal income tax brackets are also not well aligned with the country’s socio-economic struc­ Five Priority Areas for ture with top income tax brackets only apply­ Policy Action ing to a very small share of taxpayers. Ongoing work to move from the current turn-over tax Short-Term to a simple VAT is welcome. Maintaining macroeconomic stability, Fuel price subsidies have been eliminated, while meeting develop­ ment needs. but could return. The rise in retail fuel prices Fiscal policy needs to find a balance between in October 2014, combined with a decline contributing to higher and more inclusive in international oil prices has allowed the growth while maintaining sustainability. government to eliminate the fuel price subsi­ Haiti’s dependence on external assistance dies that were burdening the budget (about at a time of declining aid is a challenge, but 2 percent of GDP in 2013). Such subsidies favor could also be seen as an opportunity for the the rich, more so than subsidies on food, health country to reduce its reliance on the donor or education. International oil prices are, how­ community. The reduced availability of con­ ever, volatile and without the introduction of cessional financing following the decline in an automatic price adjustment mechanism for international oil prices has added urgency to petroleum products, fuel price subsidies could this need. In this regard, achieving a greater return to haunt the budget, this time under mobilization of own fiscal revenue will be im­ tighter financing constraints. Fuel price sub­ portant as already initiated, as well as identi­ sidies are, however, only one dimension of fying options to improve the composition of broader development challenges: more gener­ public spending (reduce fuel subsidies and ally, the overall policy framework to develop transfers to the state-own electricity company) urban transport, and promote more sustain­ as well as its efficiency (especially public able and cost effective energy needs to be investment).4 strengthened. Priorities 81 Despite rising public investment, eco­ Medium-Term nomic growth has not accelerated so far in (Re)building the social contract. Haiti. Haiti still performs poorly for selected Rebuilding or building the social contract infrastructure indicators, such as access to between Haiti’s state and its citizens will re­ electricity, roads or ports. Against this back­ quire greater transparency and accountabil­ drop, high public investment should have ity. More resources and time should be devoted been expected to contribute to reducing bot­ to build capacity within the Haitian adminis­ tlenecks to faster growth. This paradox is, tration rather than use non-State actors to de­ however, nothing new in Haiti. Several rea­ liver services. Developing an appropriate fiscal sons have been put forward, ranging from revenue base will require a renewed climate deficiencies in the country’s national ac­ of accountability and transparency from both counting system to chronic lack of mainte­ the government and the private sector. This mu­ nance or simply the unproductive nature of tual accountability should be centered around the investment itself. Poor past donor coor­ the private sector’s commitment to full tax com­ dination and the high volatility of external pliance, and the government’s commitment to aid have been argued to have affected the meeting minimum service levels and manag­ impact of investments in Haiti. This disap­ ing budgets, decision-making and funding in pointing outcome stems also in part from full transparency. More broadly, it will require very weak public investment management, a well-functioning civil society and set of in­ hampering the effectiveness of public stitutions, including regular elections and investments. greater autonomy at the community level al­ lowing for the emergence of local leadership. Improving statistics and analytics. Greater legitimacy will also be needed, Statistical and analytical capacities would through the provision of basic services across need to be strengthened. Access to timely the country. The provision of services for which and consolidated fiscal data would be people do care would enhance the legitimacy important to monitor appropriately public ­ of the State and provide more equal economic spending, especially with tighter budget opportunities. In particular, the public provi­ constraints. More regular statistical surveys ­ sion of services to increase the human capital on critical sectors for the Haitian economy accumulation capacity of the poor—such as will enable better analysis on poverty and health or education—will be essential in break­ growth, as well as allow the government to ing the vicious circle of intergenerational pov­ monitor progress and adopt appropriate pol­ erty. Expanding access and the quality of icies. In particular, the absence of a recent services, while reducing costs among house­ census hampers the design and costing of holds, will be critical to improving health and policies for basic services such as health or education outcomes, particularly among chil­ education. Regular monitoring of poverty dren and women. Achieving universal primary and living conditions is a necessary step to enrollment will also require a short- to medi­ promoting evidence-based and effective pol­ um-term financing plan and an improved co­ icy making. In this regard, strengthening the ordination with social protection programs. national statistical system through invest­ On the health care front, policies should aim ments in this sector will be critical. 82 Priorities at improving the accountability of service pro­ capacity in Haiti’s public sector, further com­ viders, increasing service utilization and qual­ pounding the overall difficult business ity, and expanding preventive health care environment. services to reduce costs. In both sectors, fur­ Policies to boost households’ income— thermore, the establishment of an informa­ wherever they work—are essential to sustain­ tion system allowing for better identification ing and accelerating welfare gains. In urban and targeting of vulnerable populations, as well areas, achieving this objective will have to in­ as for services quality control, will be critical volve the creation of economic opportuni­ in optimizing the use of available resources. ties and better jobs. A higher level of education and skills, for example, is cor­ Creating greater economic opportuni­ related with higher labor income. In rural ties and better jobs. Private economic areas, the stagnation of both extreme pov­ ­ activities are the engines for faster and more erty and income inequality observed be­ inclusive growth. More vibrant private eco­ tween 2000 and 2012 reflects the increasing nomic activities will be critical to create reliance on the low-productivity agricul­ greater opportunities to escape poverty. In tural sector. Because 80 percent of the ex­ this regard, macro-economic stability has to treme poor live in rural areas, it will be be maintained, as well as vulnerabilities con­ necessary to develop this sector by means of tained and resilience strengthened. More gen­ policies that support income diversification. erally, the business environment should be made Such a diversification could contribute at more favorable. All stakeholders, including the spreading risk, strengthening food security, government, the established private sector, and and preserving biodiversity. Furthermore, donors, should agree and adhere to a consis­ both in urban and rural areas, it is neces­ tent and comprehensive private sector devel­ sary to promote expanded access to inputs opment strategy to maximize the positive role (e.g., seeds, electricity, finance, water) and to of the private sector in channeling investment product markets (e.g., transport), as well as into viable and productive sectors to create jobs improve the business environment in order and generate income. to increase the profitability of employment. Foreign direct investment, encouraged by Greater economic opportunities and better a more favorable business environment, could jobs, especially in urban areas, would contrib­ provide new players and enhance competition. ute in reducing crime and violence, further Haiti has had a limited, but broadly positive improving the business climate. experience with FDI, especially in infrastruc­ To reduce poverty and increase productiv­ ture. The entry of Digicel in the telecommu­ ity in a more balanced manner across Haiti’s nication sector in 2006 has resulted in widening territory, a multipronged approach is critical. access to 2.4 million people. The E-Power proj­ Opportunities and advantages presented by ect introduced a lower cost producer while various regions outside of the ­ capital should increasing power generation capacity in Port- be developed. This approach would aim to au-Prince by 35 percent. These successes are broaden the pool of economic actors and few, unfortunately, hampered by a lack of trans­ support their contribution to ­ productive ac­ parency in existing sectoral policy and regu­ tivities and growth, in an environment en­ latory environments, and limited institutional compassing most of the poor and the poorest. Priorities 83 Multipronged investments in physical and ser­ Progress made thus far in understand­ vice connectivity between rural communities ing and analyzing urban resiliency needs to and producers, and small and regional town be supported. Following the work on urban markets and consumers, would add value to growth and resiliency recently carried out, the local economy, help limit post-harvest more research is needed to document (a) production losses, reduce the isolation of local urban demographic trends, (b) rural to populations and their disconnection from the urban migration trends, (c) urban surface State, and stimulate ­economic activity. If con­ growth and possible sprawl including land ments in a given geo­ ceived as a set of invest­ tenure, (d) the determinants of urbaniza­ graphical space, together with the productivity tion and the role of secondary cities, (e) support discussed above and the expansion of urban-rural linkages and the limitations services discussed below, such an approach imposed on the rural economy by urban would provide an avenue for accelerating sprawl, (f) investments and public-service growth for the bottom 40 percent, for building delivery needs for urban areas to address greater confidence of the poor in a better fu­ the double-challenge of poverty and risk, (g) ture, and, in the medium term, possibly pro­ the role of cities in reducing poverty and vide alternatives to migration. through which channels it is occurring, and (h) the link between economic and disas­ Reducing vulnerabilities and building ter-risk vulnerabilities, i.e., whether the resilience. Institutional arrangements for group of people that just came out of pov­ emergency preparedness and disaster preven­ erty are still vulnerable to poverty reversals tion and mitigation would need to be strength­ as a result of future disasters. ened. A key initial step toward upgrading the Social protection would also need to be management of disaster risks involves improv­ less fragmented and better targeted. The ing the identification and understanding of foundational blocks of social protection disaster risks in Haiti by quantifying and antic­ ­ should be built, starting with a targeting sys­ ipating the potential impacts of natural haz­ tem. Deve­ loping a targeting system is an es­ ards on Haitian society and the economy. It sential step to enhancing human capital would also be important to reduce existing accumulation among the poor and vulnerable risks and avoid the creation of new risks by in­ using, among others, a national poverty map tegrating risk awareness in public policies and allowing for the identification of pockets of investments. In light of the vulnerability of ag­ poverty. This would include moving forward riculture and its volatile contribution to eco­ with the development of a unique registry nomic growth, efforts to make production of social protection beneficiaries in priority more resilient to adverse weather conditions areas. In this regard, the Haitian authorities are called for. Given the potential threat natu­ have developed a national targeting tool and ral hazards pose to the sustainability of public are in the process of establishing a national finances, developing an integrated disaster risk registry of beneficiaries. Both of these mea­ financing and insurance strategy would also sures will improve the efficiency and effec­ help ensure greater fiscal capacity in the after­ tiveness of social protection programs. In a math of a disaster, while managing the volatil­ second stage, the coverage of social safety ity on fiscal accounts. nets could be increased, especially among 84 Priorities households with children, while ensuring • Employment and labor productivity: a bet­ sound targeting and improving the quality ter understanding of the dynamics of job of relevant programs, particularly those able creation and the factors behind low labor to enhance human capital. productivity is needed, especially in the in­ formal sector. • Agriculture census: the last census dates back Data Gaps from 2008. A better idea of what is pro­ The SCD has highlighted a number of data duced where and factors preventing peo­ gaps that would call for attention: ple moving out of subsistence farming is needed. • Consolidation and access to timely fiscal • Single registry of beneficiaries: the social data: tighter financing constraints make it protection system in Haiti is fragmented even more urgent to track public spend­ and lacks transparency. A single registry ing appropriately. Differences in budget of beneficiaries is needed to improve the classifications and the absence of an effec­ efficiency of the system through a better tive single treasury account lead to a frag­ targeting of its beneficiaries. mentation of fiscal data and prevents a • Survey on governance and corruption: the comprehensive monitoring of public survey measures the perception of corrup­ spending from budget appropriation to tion and governance from households, pub­ payment. lic servants, the private sector, and NGOs, • A new census: the last census dates back to but the last one dates back to 2007. 2003. Many events have taken place since, and the dynamics and location of people Rebasing the National Accounts and revis­ may have substantially changed. Without ing the Social Accounting Matrix (SAM): the knowing the size of the country’s popula­ most recent available SAM dates back to 1986. tion and its distribution across the national Haiti’s economic structure has evolved and territory, it is very difficult to design and a better understanding of its current inter- cost any public policy. sectoral linkages is needed. Notes 1. While it is implicitly assumed that a one-unit 37th Anniversary Conference on June 9 and at improvement in closing a large gap demands the launch of the Competitive Industries and less effort than an equivalent one-unit im­ Innovation Program Training Program at the provement in a small gap, this might not nec­ Limonade Campus on June 17. These facts were essarily be the case. Our exercise hence does also presented to representatives of the private not explicitly take into account the costs or sector on May 19, at the occasion of the 35th efforts of closing this gap. Haitian Business Delegation visit to Washington 2. These stylized facts, or parts of them, have DC. They have also served to frame the discus­ been discussed with the authorities and vari­ sions with the government, donors, and the rep­ ous stakeholders, particularly at the Conference resentatives of the private sector during the MD on Growth and Employment organized by and the RVP’s recent visit to Haiti, July 7–9, 2014. the Ministry of Economy and Finance in Port- 3. The Community Social Worker Initiative au-Prince on March 17, at the Central Bank’s (KORE FANMI) is a pilot program of the Priorities 85 Haitian Government, established with World vaccines, and latrines, with a view to moving Bank financing in partnership with a number away from a national fragmented and unequal of United Nations agencies such as UNICEF. system to one of systematic coverage focusing This program seeks to harmonize and improve on the rights and needs of families. the provision of basic services to poor and vul­ 4. This work is being done as part of the ongoing nerable families, such as access to education, Public Expenditure Review. 86 Priorities References Acemoglu, A., and J. Robinson. 2012. Why Banks, A. S., and K. A. Wilson. 2014. Cross- Nations Fail: Origin of Power, Prosperity, and National Time-Series Data Archive. Databanks Poverty. Random House. International, Jerusalem, Israel. Acosta, P., C. Calderón, P. Fajnzylber, and Barro, R., and J. Lee. 2000. “International Data H. López. 2006. “Do Remittances Lower on Educational Attainment Updates and Poverty Levels in Latin America?” In Implications.” NBER Working Paper 7911, Remittances and Development: Lessons from National Bureau of Economic Research. Latin America, edited by P. Fajnzylber, and Berg, L. A. 2010. Crime, Politics and Violence J. Humberto López. World Bank Latin in Post-Earthquake Haiti. Institute of Peace, America Development Forum Series, Washington, DC. Washington, DC: World Bank. Bodea, C. B., M. Higashijima, and R. J. Singh. Aisen, A., and F. J. Veiga. 2013. “How Does 2014. “Conflict and Public Spending in Haiti: Political Instability Affect Economic What Could We Learn from Cross-Country Growth?” European Journal of Political Evidence?” SCD Background Paper, World Economy 29: 151–67. Bank, Washington, DC. Alderman, H., and E. M. King. 2006. “Investing Boesten, J., and N. K. Poku. 2009. Gender and in Early Childhood Development.” Research HIV/AIDS: Critical Perspectives from the Brief, World Bank, Washington, DC. Developing World. Farnham, Ashgate. Alys, W., L. A. Berg, and P. Lafare. 2014. Bourguignon, F. 2003. “The Growth Elasticity of “Violence, Fragility and Resilience in Haiti.” Poverty Reduction: Explaining Heterogeneity SCD Background Paper, World Bank, across Countries and Time Periods.” In Washington, DC. Inequality and Growth: Theory and Policy Antoine, K. B. da S. 2014. “Haiti: Growth Implications, edited by T. S. Eicher and Dynamics.” SCD Background Paper, World S. J. Turnovsky, 3–26. Cambridge, MA: Bank, Washington, DC. MIT Press. Antoine, K. B. da S., R. Singh, and K. M. Wacker. Brinkerhoff, D., A. Wetterberg, and D. Stephen. 2014. “Benchmarking the Drivers of Income 2012. “Service Delivery and Legitimacy in Growth at the Bottom 40 Percent: An Fragile and Conflict-Affected States.” Public Application to Haiti.” SCD Background Paper, Management Review 14: 273–93. World Bank, Washington, DC. Bureau de Monétisation du Programme d’Aide Araujo, J., M. Brueckner, M. Clavijo, au Développement (BMPAD). 2014. http:// E. Vostroknutova, and K. Wacker. 2014. www.bmpad.gouv.ht/petrocaribe/les-fonds​ Benchmarking the Determinants of Economic -petrocaribe.html. Growth in Latin America and the Caribbean. Buss, T. F. 2013. “Foreign Aid and the Failure of World Bank Report 91015-LAC. Washington, State Building in Haiti under the Duvaliers, DC: World Bank. Aristide, Préval, and Martelly.” Working Paper Azam, J.-P. 2001. “The Redistributive State and 2013/104, WIDER. Conflicts in Africa.” Journal of Peace Research Cali, M., B. Abidoye, and E. Jadotte. 2014. 38: 429–44. “What Drives Conflict in Haiti? An Empirical Azevedo, J. P., G. Inchauste, S. Olivieri, J. Saavedra, Analysis in the 2000s.” SCD Background and H. Winkler. 2013. “Is Labor Income Paper, World Bank, Washington, DC. Responsible for Poverty Reduction? A Calvin, D., J. Lohi, and E. R. Bulmer. 2014. Decomposition Approach.” Policy Research “Haiti Systematic Country Diagnostic. Trade Working Paper 6414, World Bank, Background Note.” SCD Background Paper, Washington, DC. World Bank, Washington, DC. References 87 Cederman, L. E., N. Weidmann, and Dollar, D., and A. Kraay. 2002. “Growth Is Good K. S. Gleditsch. 2011. “Horizontal Inequalities for the Poor.” Journal of Economic Growth 7: and Ethnonationalist Civil War: A Global 195–225. Comparison.” American Political Science Dorte, V. 2008. “Labor Markets in Urban and Review 105: 478–95. Rural Haiti: Based on the First Household Christiaensen, L., C. P. Punam, and A. Sanoh. Survey for Haiti.” Policy Research Working 2013. “Africa’s Growth, Poverty and Inequality Paper 4574, World Bank, Washington, DC. Nexus—Fostering Shared Prosperity.” Mimeo, Dubois, L. 2013. Haiti: The Aftershocks of History. World Bank. https://editorialexpress.com​ Reprint edition. Picador. /­cgi-bin/conference/download. Echevin, D. 2011. “Vulnerability and Livelihoods cgi?db_name=CSAE2014&paper_id=381. before and after the Haiti Earthquake.” Policy Cicmil, H. 2013. “Whose Education? Haiti’s Girls Research Working Paper 5850, World Bank, and Haiti’s Recovery—Thinking Development Washington, DC. Organization.” http://www.­thinkingdevelopment​ Economic Commission for Latin America and .org/news/whose-education-haitis​-girls​-and​ the Caribbean (database), Statistics Division, -haitis​-­recovery. United Nations Economic Commission for Clemens, M. A. 2011. “Economic Impacts of H-2 Latin America and the Caribbean, Santiago, Nonimmigrant Visa Eligibility for Haiti.” Center Chile. http://estadisticas.cepal.org/cepalstat​ for Global Development, Washington, DC. /­WEB_CEPALSTAT/Portada.asp?idioma=i. ———. 2014. “Does Development Reduce Economist Intelligence Unit. 2013. Operational Migration?” Working Paper 359, Center for Risk Model. Global Development. ECVH (Enquête sur les conditions de Vie des Collier, P. 2009. Haiti: From Natural Catastrophe Ménages en Haïti). 2001. Institut Haïtien de to Economic Security. A Report for the Statistique et d’Informatique (IHSI). Secretary-General of the United Nations, ECVMAS (Enquête sur les Conditions de Department of Economics, Oxford University. Vie des Ménages après le Séisme). 2012. Crespo, C. J., S. Klasen, and K. M. Wacker. 2013. Institut Haïtien de Statistique et “Why We Don’t See Poverty Convergence: ­d’Informatique (IHSI). The Role of Macroeconomic Volatility.” Ehrhart, H., M. Le Goff, E. Rocher, and R. J. Singh. Courant Research Centre Discussion Paper 2014. “Does Migration Foster Exports? 153, University of Gottingen. Evidence from Africa.” Policy Research Delatour, L., and T. J. Dnggleby. 1993. Working Paper 6739, World Bank, Strengthening the Informal Sector in Haiti: Washington, DC. Constraints, Possible Interventions and Factors EM-DAT. The OFDA/CRED International for Success. USAID Haiti and PADO. Disaster Database. Université catholique de Docquier, F., and A. Marfouk. 2006. “Inter­ na­ Louvain—Brussels—Belgium. http://www​ tional Migration by Education Attainment, .emdat.be. 1990–2000.” In International Migration, Facundo, C., F. Marzo, and T. Scott. 2014. “Haiti Remittances and the Brain Drain, edited by Poverty Assessment Background Note: Ç. Özden and M. Schiff. Washington, DC: Migration and Remittances in Haiti.” SCD World Bank and Palgrave Macmillan. Background Paper, World Bank, Dollar, D. 2000. Governance and Social Justice in Washington, DC. Caribbean States. Report 20449, World Bank Facundo, C., and A. S. Rodella. 2014. “Making Development Research Group. Self-Employment Work for Haiti: The Scope Dollar, D., T. Kleineberg, and A. Kraay. 2013. of the Challenge.” SCD Background Paper, “Growth Still Is Good for the Poor.” Policy World Bank, Washington, DC. Research Working Paper 6568, World Bank, Famine Early Warning System Network. 2012. Washington, DC. “Haiti Food Security Outlook.” Washington, DC, 88 References http://www.fews.net/central-­america-and​ ———. 2013. Haiti: Selected Issues. Country -­caribbean/haiti/food-security-outlook​ Report 13/91, Washington, DC. /­october-2012. ———. 2014a. Haiti: Seventh Review Under the Fass, S. M. 1990. Political Economy in Haiti: The Extended Credit Facility, Requests for Waiver Drama of Survival. Transaction Publishers. of Nonobservance of Performance Criterion, Fries, G., and R. Fernandez. 2012. “Agro- and Modification of Performance Criteria- Logistics in Central America—A Supply Staff Report; Press Release; and Statement by Chain Approach.” Background paper, the Executive Director for Haiti. Country World Bank, Washington, DC. Report 14/105, Washington, DC. Gleditsch, N. P., P. Wallensteen, M. Eriksson, ———. 2014b. Haiti: Eighth Review under the M. Sollenberg, and H. Strand. 2002. “Armed Extended Credit Facility and Request for Conflict 1946–2001: A New Dataset.” Journal Waiver of Nonobservance of Performance of Peace Research 39: 615–37. Criterion-Staff Report; Press Release; and Government of the Republic of Haiti. 2014. Plan Statement by the Executive Director for Haiti. Stratégique de Développement d’Haïti (PSDH). Country Report 15/3, Washington, DC. Greif, A. 1989. “Reputation and Coalition in ———. 2014c. Haiti: Ex Post Assessment of Medieval Trade: Evidence on the Maghribi Longer-Term Program Engagement. Country Traders.” Journal of Economic History 49: 857–82. Report 15/4, Washington, DC. ———. 1993. “Contract Enforceability and ———. 2015. Haiti: Staff Report for the Article IV Economic Institutions in Early Trade: The Consultation and Request for a Three-Year Maghribi Traders’ Coalition.” American Arrangement under the Extended Credit Economic Review 83: 525–48. Facility. Washington, DC. ———. 2006. “The Birth of Impersonal Exchange: Jadotte, E. 2008. “Labor Supply Response to The Community Responsibility System and International Migration and Remittances in Impartial Justice.” Journal of Economic the Republic of Haiti.” Document de treball Perspective 20: 221–36. 0808, Departament d’Economia Aplicada, Institut Haitien de Statistique et d’Informatique Universitat Autònoma de Barcelona, (ISHI). 2007. “Enquête sur l’Emploi et l’Econ- Barcelona. omie Informelle (EEEI).” Jean, F. 2014. La Fin d’une Histoire Economique Inter-American Development Bank. 2009. en Haïti. Les Editions de l’Université d’Haïti. “Haiti: A New Paradigm.” Conférence sur le Loayza, N., C. Calderón, and P. Fajnzylber. 2005. Développement Economique et Social d’Haiti III: “Economic Growth in Latin America and the Vers Un Nouveau Paradigme de Coopération. Caribbean: Stylized Facts, Explanations, and Inter-American Institute for Cooperation on Forecasts.” World Bank, 7315. Agriculture. 1997. Agriculture in Haiti, Lockhart, C., and J. M. Forman. 2013. “Escaping ­ 1991–1995 and Beyond. Port of Spain. the Crisis Trap: New Options for Haiti.” International Finance Corporation. 2011. Transitions Forum Case Study, Legatum “Economic Integrated Zone, Market Institute and Institute for State Effectiveness, Analysis.” Washington, DC. London and Washington, DC. ———. 2013. “IFC Jobs Study.” Washington, DC. Marc, A., A. Willman, G. Aslam, M. Rebosio, International Labor Organisation. 2010. and K. Balasuryia. 2012. Societal Dynamics “La Promotion du Travail Décent dans la and Fragility. World Bank, Washington, DC. Reconstruction et le Développement d’Haïti Matera, M., and R. Van der Borght. 2014. après le Tremblement de Terre de 2010.” “Economic and Fiscal Impact of Disasters in Geneva, Switzerland. Haiti.” SCD Background Paper, World Bank, International Monetary Fund. 2008. Haiti: Washington, DC. Financial System Stability Assessment. Ministère de la Santé Publique and de la Washington, DC. Population (MSPP), Institut Haïtien de References 89 l’Enfance (IHE), and ICF International. 1994. Rajaram, A., T. Minh Le, N. Biletska, and Demographic and Health Survey. J. Brumby. 2010. “A Diagnostic Framework for ———. 2000. Demographic and Health Survey. Assessing Public Investment Management.” ———. 2006. Demographic and Health Survey. WPS5397, World Bank, Washington, DC. ———. 2013. 2012 Haiti Mortality, Morbidity, and Ratha, D., and W. Shaw. 2007. “South-South Service Utilization Survey: Key Findings. Migration and Remittances.” Working Paper Calverton, MD: MSPP, IHE, and ICF 102, World Bank, Washington, DC. International. Rauch, J. E. 2001. “Business and Social Networks Narayan, A., J. S. Chanduvi, and S. Tiwari. 2013. in International Trade.” Journal of Economic “Shared Prosperity—Links to Growth, Literature 39: 1177–203. Inequality and Inequality of Opportunity.” Ravallion, M., and S. Chen. 2007. “China’s Policy Research Working Paper 6649, World (Uneven) Progress against Poverty.” Journal Bank, Washington, DC. of Development Economics 82: 1–42. Niimi, Y., and C. Ozden. 2006. “Migration and Reveco, L. 1984. “Bases et Critères Pour Une Remittances in Latin America: Patterns and Politique Des Prix.” United Nations Determinants.” In Remittances and Development Program, Port-au-Prince. Development: Lessons from Latin America, Rotberg, R. I. 1971. Haiti: The Politics of Squalor. edited by P. Fajnzylber and J. Humberto ­ Boston, MA: Houghton Mifflin Company. López. Latin America Development Forum Sanchez-Reaza, J., and A. Carletto. 2013. Series. Washington, DC: World Bank. “Resilient Coastal Cities: The Economic, North, D. 2010. Understanding the Process of Social and Environmental Dimensions of Economic Change. Princeton, NJ: Princeton Risk.” LAC Opportunities for All, no. 14, University Press. World Bank, Washington, DC. Orozco, M. 2006. “Understanding the Singh, R., C. Bodea, and M. Higashijima. 2014. Remittance Economy in Haiti.” Paper “Oil and Conflict: Can Public Spending commissioned by the World Bank, mimeo. ­ Have a Mitigation Effect?” Policy Research Pedersen, J., and K. Lockwood. 2001. Working Paper 7100, World Bank, “Determination of a Poverty Line in Haiti.” Washington, DC. FAFO Institute of Applied International Stasavage, D. 2005. “Democracy and Education Studies Working Paper, Oslo. Spending in Africa.” American Journal of Plummer, B. G. 1988. Haiti and the Great Powers, Political Science 49: 343–58. 1902–1915. Lousiana State University Press, Thyne, C. 2006. “ABC’s, 123’s and the Golden Baton Rouge, LA. Rule: The Pacifying Effect of Education on Pop, G., J. Dorsainvil, T. Goodwin, M. M. Licetti, Civil War, 1980–1999.” International Studies C. R. Nau, E. Nsimba, R. Surin, L. Villaran, Quarterly 50: 733–54. and J. Zentner. 2014. “Brief Analysis of UNDP (United Nations Development Market Functioning in Haiti.” SCD Programme). 2014. Human Development Background Paper, World Bank. Index. http://hdr.undp.org/en/data. PotoFi Haiti Girls Initiative (“PotoFi”). 2012. UNODC (United Nations Office for Drugs and Gender Aftershocks: Teen Pregnancy and Crimes). 2013. Homicide Statistics. http:// Sexual Violence in Haitian Girls. http://potofi​ www.unodc.org/gsh/en/data.html. .files.wordpress.com/2012/12/summary-report​ United Nations Statistics Division. 2014. National -poto-fi-girls-gbv-field-survey.pdf. Accounts (database). https://data.un.org/. Pritchett, L., M. Woolcock, and M. Andrews. US Agency for International Development. 2013. “Looking Like a State: Techniques of 2006. Haiti Conflict Assessment. Persistent Failure in State Capability for Washington, DC. Implementation.” The Journal of Development Willman, A., and L. H. Marcelin. 2010. “If They Studies 49: 1–18. Could Make Us Disappear, They Would! 90 References Youth and Violence in Cité Soleil, Haiti.” ———. 2013. “Strengthening Competitiveness Journal of Community Psychology 38: 515–31. and Promoting Diversification in Haiti— Wimmer, A., L. E. Cederman, and B. Min. Diagnostic Trade Integration Study.” 2009. “Ethnic Politics and Armed Conflict: Washington, DC. A Configurational Analysis.” American ———. 2014a. “Analytical Approaches for a Sociological Review 74: 316–37. Systematic Country Diagnostic (SCD)— Wooding, B., and R. Moseley-Williams. 2004. A Resource Document.” Washington, DC. “Needed but Unwanted: Haitian Immigrants ———. 2014b. World Development Indicators. and Their Descendants in the Dominican http://data.worldbank.org/data-catalog​ Republic.” Catholic Institute International /­world-development-indicators/wdi-2014. Relations, CIIR Briefing. ———. 2014c. Doing Business Indicators. http:// World Bank. 1998. Haiti. The Challenges of www.doingbusiness.org/. Poverty Reduction. Caribbean Country ———. 2014d. Statistical Capacity Index. http:// Management Unit Report, 17242-HA. datatopics.worldbank.org/statisticalcapacity/ Washington, DC: World Bank. ———. 2014e. “Prizes Awarded to Young ———. 2002. Haiti: Country Assistance Haitians Researchers for Their Work on the Evaluation. Report 23637, Washington, Future of the Haitian Economy.” http://www​ DC: Operations Evaluation Department. .worldbank.org/en/news/feature/2014/11/11​ ———. 2006. Haiti: Options and Opportunities for /­young-haitians-rewarded-for-their-work-on Inclusive Growth. Washington, DC: Country -haiti-economy. Economic Memorandum. World Bank and ONPES. 2014. Investing in ———. 2007. Social Resilience and State Fragility People to Fight Poverty in Haiti, Reflections for in Haiti. Washington, DC. Evidence-Based Policy Making. Washington, ———. 2010a. Violence in the City: Understanding DC: World Bank. and Supporting Community Responses to World Economic Forum. 2011. Private Sector Urban Violence. Washington, DC. Development in Haiti: Opportunities for ———. 2010b. Haiti Earthquake PDNA Investment, Job Creation and Growth. World (­ Post-Disaster Needs Assessment): Assessment Economic Forum in Partnership with the of Damage, Losses, General and Sectoral World Bank, the Inter-American Needs. World Bank: Washington, DC. Development Bank and the International ———. 2010c. Global Bilateral Migration Matrix. Finance Corporation, Geneva, Switzerland. http://data.worldbank.org/data​-­catalog​/­global ———. 2014. Global Competitiveness Report. -bilateral-migration-database. Geneva, Switzerland. ———. 2011. World Development Report 2011 on World Health Organization. 2014. Haiti: Conflict, Security and Development. National Health Expenditures. Geneva, Washington, DC. Switzerland: WHO. References 91 Appendix A: Price Comparison Analysis The price comparison analysis uses three The baseline empirical specification for data sources to explore whether food prices in the price comparison analysis follows the Haiti are higher than other countries from equation: Ln(Priceijt) = β1GDPit + β2Ln(Xit) + Latin America and the Caribbean region (as β3Haiti + ηj + δt + εijt well as North America): (a) “Numbeo,” an on­ where: i = country; j = product; t = year; line database of user contributed data on cost Xit = GDP per capita, cost of imports, and of living, (b) Ministry of Economy and Finance duties imports (as control variables); and ­ of Haiti and (c) FAO GIEWS Food Price Data. ηj = product fixed effects. The Haiti dummy The first database applies a common variable captures the relative price levels in methodology in gathering price data across Haiti compared to the average across other countries, thus strengthening the comparabil­ countries after adjusting for the differences ity of price information used in this analysis. in per capita GDP PPP, import costs, cus­ However, since price data for Haiti is only avail­ toms duties, and product type as well as able for 2014 at “Numbeo,” this only ­ allows for time-specific effects. cross-section analysis. Data provided by the The food products were selected based Haitian Ministry of Economy and Fi­ nance on on availability across databases, the rele­ Haitian prices for 2010–2013 is com­ bined with vance in the Haitians consumption basket “Numbeo” data for the panel data analysis. and product characteristics. For example, Finally, for additional robustness, the analysis products were selected that are relatively is replicated with prices from the FAO similar (or homogeneous) across countries GIEWS database for the available food prod­ in order to minimize the differences associ­ uct (rice). ated with product differentiation. Appendix A: Price Comparison Analysis 93 Appendix B: Bottlenecks and Correlates of Firm Productivity Log. revenue per Successful firm Successful firm Variables worker (revenues in top 30) (profits in top 30) Sell transformed products 0.106 0.00585 0.0295 (0.105) (0.0351) (0.0360) Sell non transformed products 0.357*** 0.0619* 0.0297 (0.120) (0.0368) (0.0381) Age 0.128*** 0.0263*** 0.0241*** (0.0127) (0.00350) (0.00338) Age square –0.00144*** –0.000302*** –0.000280*** (0.000136) (3.64e–05) (3.53e–05) Man 0.402*** 0.135*** 0.137*** (0.0949) (0.0307) (0.0312) Head of household 0.138** 0.0430* 0.0379 (0.0659) (0.0242) (0.0243) Complete elementary 0.256*** 0.0640** 0.0689** (0.0890) (0.0317) (0.0311) Complete secondary 0.321*** 0.0602* 0.0528 (0.0898) (0.0333) (0.0331) Complete tertiary 0.675*** 0.168** 0.125* (0.189) (0.0653) (0.0680) Access to electricity 0.274*** 0.0745*** 0.0510* (0.0763) (0.0277) (0.0273) Access to water 0.163* 0.0898** 0.0424 (0.0893) (0.0355) (0.0349) HH received remittances 0.00656 –0.00287 –0.00852 (0.102) (0.0354) (0.0356) HH received domestic private transfers 0.0116 0.0224 0.0299 (0.0653) (0.0224) (0.0222) Urban area 0.387*** 0.0771*** 0.0662** (0.0771) (0.0271) (0.0271) Constant 5.428*** –0.424*** –0.369*** (0.319) (0.0824) (0.0818) Observations 1,609 1,642 1,631 R2 0.217 0.105 0.096 Source: Module L-ECVMAS I. Note: Robust standard errors in parentheses. Successful firm: Revenues/Profits per worker in top 30 percent of revenues/profits per worker distribution. Estimates control for firm type (own account, household, employer), sector, activity. *p < 0.1, **p < 0.05, ***p < 0.01. Appendix B: Bottlenecks and Correlates of Firm Productivity 95 Appendix C: Most Significant Data Gaps in Haiti Sector/ Data sets/survey Frequency Comments theme descriptions Cross sectoral data GAP Population census Every 10 years Last available 2003—Planned but not budgeted for in 2015. GAP Continuous employment/ Yearly Last available 2003—Urban only if nationwide labor survey not achievable (should include informal sector) GAP Yearly vital statistics report Yearly Births, deaths, causes of death and basic demographics—Requires institutional strengthening on collection of data at various institutions Poverty data Periodic GAP Poverty assessment Every 5 years Last available 2014/2015. GAP Poverty headcount by TBD Data on poverty at national level exists, but not section communal for each communal section. Important to make spatial link between risk and poverty. Public financial and governance data Ongoing but Regular reporting on Monthly Supported by a variety of donors, but broadly many GAPs government financial data, unavailable. especially; public investment expenditures commitments, and final payments; local government financial data consolidated financial data for State owned enterprises GAP Data on special public Yearly PSUGO, FER, FNE, etc. programs/funds GAP Public Expenditure and Periodicity TBC Latest available 2011 (EU) Financial Accountability Survey (PEFA) GAP Survey on governance and Periodicity TBD Latest available 2007, 2011 (Government corruption Anti-Corruption Unit) Economic data GAP Rebase GDP Periodic (long term) GAP Social Accounting Matrix Periodicity TBD Last available 1986 update table continues next page Appendix C: Most Significant Data Gaps in Haiti 97 Sector/ Data sets/survey Frequency Comments theme descriptions Education data GAP School census Yearly Needs to be of good quality. Current surveys don’t meet quality standards needed for adequate use. Must include infrastructure questions for Water and Sanitation and schools characteristics. Planned Cartography of all schools Baseline 2015 Baseline will be started by IDB January 2015. Yearly update Health data Periodic GAP Demographic and Health Every 5 years Last available DHS 2012 Survey (DHS) Every 5 years Mini-DHS (in between) Periodic GAP Health facility census Every 3 years Last available 2013. Quality needs improvement GAP Detailed data set on TBD Includes (a) geographical commune level government and donor information on programs, (b) commodities programs provided, (c) personnel hired (including type), (d) spending (per category of spending). Baseline to be developed. Periodic GAP Health system indicators TBD Collected from National Health Information collection System, SNIS, include inputs (Human resources, equipment, drugs, etc.), outputs (number of key procedures, consultations, etc.) and intermediate outcome indicators. GAP Improved surveillance TBD Building on cholera surveillance systems to systems disease control track signs of epidemic outbreaks and enable rapid response. Data for disaster risk analysis and prevention Ongoing High resolution digital Ongoing—WB Light detection and ranging, using an airborne elevation model financing laser techniques to densely sample the surface of the earth and produce highly accurate topographic models Ongoing National Emergency Shelter Ongoing—WB Information needed: functional survey and Survey financing ad-hoc structural assessment of all the shelters used in case of emergency; scope countrywide Ongoing Natural Hazards Risk Atlas Ongoing—WB First Atlas, will summarize all the major risks financing to non-practitioners with a resolution of 1:100,000 scale Ongoing Baseline assessment of Planned—WB Estimated available in 2017–2018 hydro-meteorological Financing (including marine) and climate data Ongoing Systematic and consolidated System and baseline Estimated available in 2017–2018. Will require meteorological and planned—WB long term engagement of Government and hydrological/hydrogeological financing partners. database Ongoing Updated return period for Planned—WB Periodicity of storm; wind; storm surge. select hazards financing 98 Appendix C: Most Significant Data Gaps in Haiti Sector/ Data sets/survey Frequency Comments theme descriptions GAP National public and private TBD Assessment of resilience of schools, hospitals, critical facilities survey and critical administrative offices such as police assessment station, etc., to disaster risk (Earthquakes, Floods, Hurricanes). Financial sector data Ongoing FINDEX survey on financial Baseline ongoing— baseline inclusion WB financing Periodic GAP Ongoing Financial capabilities survey Planned—WB financing GAP Enterprise survey TBD Survey should provide data about firm financing, covering also the informal sector. GAP FinScope surveys TBD In depth surveys about demand and use of financial services (e.g., carried out among households and MSMEs) GAP MixMarket datasets or TBD Provide financial data about the structure and similar surveys performance of all financial cooperatives and microfinance entities and data on the composition and performance of their loan portfolios. Agriculture data GAP Agricultural census Every 5 years Last available 2008 raw data made available recently (FAO). But design and methodology was poor, data is out of date and may not be usable. Survey needs to be consistent across years and digitally collected. Periodic GAP Update/validation of Yearly Must use representative samples. IHSI agricultural census/ tasked with annual updates but lacks ability agricultural survey to carry out. GAP Time-series satellite imagery TBD Remote sensing to collect data on crop cover, and crop rotations by seasons and years GAP Market information survey TBD Stock and flows of agricultural goods by agricultural goods department, including consumption, imports, exports and goods sold into local markets. Ongoing Updated and scaled-up the Ongoing—WB Completion date baseline TBD. National Farmer Registry financing Water data Planned Comprehensive country Planned—WB Information: boreholes, water networks, baseline for water supply financing household connections, kiosks, etc., and how many people are served by these; Level of detail: community level for the whole territory. Planned National Sanitation Baseline Planned—WB Information required: open defecation, improved Survey financing vs. non improved toilets, collective vs. individual septic tanks, etc., Level of detail: community level for the whole territory. table continues next page Appendix C: Most Significant Data Gaps in Haiti 99 Sector/ Data sets/survey Frequency Comments theme descriptions GAP Survey for water availability Periodic TBD Latest available 2009 from Ministry of in schools and health Education survey of public and private schools. facilities Confirm whether imminent IDB survey of schools includes the water and sanitation. Transport and infrastructure data Ongoing Rural access index Ongoing—WB Includes condition of the road networks, road financing density, and provide reliable geo-referenced and structured road network data. 100 Appendix C: Most Significant Data Gaps in Haiti SKU K8422