The State of Social Safety Nets 2018 The State of Social Safety Nets 2018 The State of Social Safety Nets 2018 © 2018 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 21 20 19 18 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: World Bank. 2018. The State of Social Safety Nets 2018. Washington, DC: World Bank. doi:10.1596/978-1-4648-1254-5. 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 World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content 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 World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. ISBN (print): 978-1-4648-1254-5 ISBN (electronic): 978-1-4648-1255-2 DOI: 10.1596/978-1-4648-1254-5 Cover photo: Third grader from Niger, Mariama. © Stephan Gladieu/World Bank. Further permission required for reuse. Part 1 opener photo: Children in the village of Ambohimahatsinjo, Madagaskar. © Mohammad Al-Arief/World Bank. Further permission required for reuse. Part 2 opener photo: A young girl in an evacuation center, Philippines. © Jerome Ascano/World Bank. Further permission required for reuse. Cover design: Bill Pragluski, Critical Stages, LLC. Library of Congress Cataloging-in-Publication Data has been requested. Contents Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii About the Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv Structure of the Book. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix Executive Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 PART I  ANALYTICS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Chapter 1  Explaining the Social Safety Net’s Data Landscape. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 What Are Social Protection and Social Safety Net Interventions? What Is the ASPIRE Classification of Social Safety Net Programs?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 What Is the ASPIRE Database? How Does the ASPIRE Team Collect and Ensure Quality of the Data? What Are the Limitations of the Data? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 How Is the Performance of Social Safety Net Programs Measured? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Chapter 2  Spending on Social Safety Nets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 How Much Do Regions and Countries Spend on Social Safety Nets?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Do Higher-Income Countries Spend More on Social Safety Nets? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 How Has Spending Changed over Time?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Which Social Safety Net Instruments Do Countries Fund? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Chapter 3  Analyzing the Performance of Social Safety Net Programs. . . . . . . . . . . . . . . . . . . . . 31 Who Is Covered by Social Protection and Labor Programs? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Which Types of Social Safety Net Programs Cover the Poor? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 What Is the Beneficiary Incidence of Various Social Safety Net Instruments?. . . . . . . . . . . . . . . . . . . . . . . . 41 What Are the Benefit Levels of Social Safety Net Programs? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 What Are the Poverty and Inequality Impacts of Social Safety Net Programs?. . . . . . . . . . . . . . . . . . . . . . . 59 What Factors Affect the Impact of Social Safety Net Transfers on Poverty and Inequality? . . . . . . . . . . . . 61 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Highlight 1: Productive Outcomes of Social Safety Net Programs: Evidence from Impact Evaluations in Sub-Saharan Africa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 v PART II SPECIAL TOPICS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 Chapter 4  Social Assistance and Aging. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 What Are Old-Age Social Pensions, and Why Are They on the Rise?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Why Do Countries Introduce Old-Age Social Pensions? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 What Have Old-Age Social Pensions Accomplished? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Highlight 2.  Policy Considerations for Introducing Old-Age Social Pensions. . . . . . . . . . . . . . . . . . . . . . . 82 Chapter 5  The Emergence of Adaptive Social Protection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Why Does the World Need Adaptive Social Protection?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Focus Area 1: Building Household Resilience Before Shocks Occur. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Focus Area 2: Increasing the Capability of Safety Nets to Respond to Shocks after They Occur . . . . . . . . 89 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Appendix A  Methodological Framework, Definitions, and Data Sources. . . . . . . . . . . . . . . . . . 94 Appendix B  Household Surveys Used in the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Appendix C  Global Program Inventory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Appendix D  Spending on Social Safety Net Programs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 Appendix E  Monthly Benefit Level Per Household. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 Appendix F  Performance Indicators. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Appendix G  Old-Age Social Pensions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 Appendix H  Basic Characteristics of Countries Included in the Book . . . . . . . . . . . . . . . . . . . 161 Glossary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 Boxes 1.1 Leveraging Household Survey Data to Monitor and Measure Social Protection and Labor Program Performance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3.1 Universal Social Protection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.2 Measuring the Impact of Social Protection and Labor Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 5.1 Horizontal and Vertical Expansions through Ethiopia’s Productive Safety Net Program . . . . . . . . . 90 5.2 Responding Rapidly to Disasters through Vertical Expansions in Fiji and the Philippines. . . . . . . . 91 5.3 Investing in Risk and Vulnerability Information and Tying It to Safety Net Programming in the Dominican Republic, Kenya, and the Republic of Yemen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 vi Contents Figures 2.1 Average Global and Regional Spending on Social Safety Nets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.2 Share of Donor-Funded Safety Nets in Sub-Saharan African Countries . . . . . . . . . . . . . . . . . . . . . . . 18 2.3 Social Safety Net Spending Variations across Countries and Regions: East Asia and Pacific, Latin America and the Caribbean, and Europe and Central Asia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.4 Social Safety Net Spending Variations across Countries and Regions: Africa, Middle East and North Africa, and South Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2.5 Social Safety Net Spending across Country Income Groups versus the OECD. . . . . . . . . . . . . . . . . . 21 2.6 Total Social Safety Net Spending and Income Levels across Regions . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.7 Absolute Annual Spending on Social Safety Nets per Capita across Countries and Regions: East Asia and Pacific, Europe and Central Asia, and Latin America and the Caribbean. . . . . . . . . . 23 2.8 Absolute Annual Spending on Social Safety Nets per Capita across Countries, Economies, and Regions: Middle East and North Africa, Sub-Saharan Africa, and South Asia . . . . . . . . . . . . . . 24 2.9 Regional Median Annual Social Safety Net Spending per Capita . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.10 Transfer Amount for Cash Transfer Programs, by Income Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.11 Trends in Social Safety Net Spending in Latin America and the Caribbean . . . . . . . . . . . . . . . . . . . . 26 2.12 Trends in Social Safety Net Spending in Europe and Central Asia, 2003–14. . . . . . . . . . . . . . . . . . . . 27 2.13 Expansion of Flagship Cash Transfer Programs in Tanzania, Senegal, the Philippines, and Indonesia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 2.14 Social Safety Net Spending across Regions, by Instrument . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 3.1 Share of Total Population and the Poorest Quintile That Receives Any Social Protection and Labor Programs, as Captured in Household Surveys, by Region. . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.2 Share of Total Population and the Poorest Quintile That Receives Any Social Protection and Labor Programs, as Captured in Household Surveys, by Country Income Group . . . . . . . . . . . 33 3.3 Share of Poorest Quintile That Receives Any Social Protection and Labor Program, as Captured in Household Surveys, by Type of Social Protection and Labor Area and Country Income Group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.4 Share of the Poorest Quintile That Receives Unconditional Cash Transfer Programs, as Captured in Household Surveys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 3.5 Share of the Poorest Quintile That Receives Conditional Cash Transfer Programs, as Captured in Household Surveys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 3.6 Share of the Poorest Quintile That Receives Social Pensions, as Captured in Household Surveys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.7 Share of the Poorest Quintile That Receives Public Works, as Captured in Household Surveys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 3.8 Share of the Poorest Quintile That Receives Fee Waivers and Targeted Subsidies, as Captured in Household Surveys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 3.9 Share of the Poorest Quintile That Receives School Feeding Programs, as Captured in Household Surveys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.10 Share of the Poorest Quintile That Receives In-Kind Transfers, as Captured in Household Surveys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.11 Global Distribution of Beneficiaries by Type of Social Safety Net Instrument, as Captured in Household Surveys, by Quintile of Pretransfer Welfare. . . . . . . . . . . . . . . . . . . . . . . . 44 3.12 Distribution of Unconditional Cash Transfer Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Contents vii 3.13 Distribution of Conditional Cash Transfer Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3.14 Distribution of Social Pensions Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.15 Distribution of Public Works Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.16 Distribution of Fee Waivers and Targeted Subsidies Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 3.17 Distribution of School Feeding Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 3.18 Distribution of In-Kind Transfer Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.19 Social Protection and Labor Transfer Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Total Population. . . . . . . . . . . . . . . . . . . 53 3.20 Social Protection and Labor Transfer Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile. . . . . . . . . . . . . . . . . . . 54 3.21 Unconditional Cash Transfer Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 3.22 Conditional Cash Transfer Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 3.23 Social Pensions’ Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 3.24 Public Works’ Value Captured in Household Surveys as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 3.25 Fee Waivers and Targeted Subsidies Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 3.26 World Reductions in Poverty from Social Safety Net Transfers, as Captured in Household Surveys, as a Share of Pretransfer Indicator Levels, by Relative and Absolute Poverty Lines . . . . . . 60 3.27 Reductions in Poverty and Inequality from Social Safety Net Transfers, as Captured in Household Surveys, as a Share of Pretransfer Indicator Levels, by Country Income Group Using Relative Poverty Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 3.28 Poverty Headcount Reduction from Coverage and Level of Social Assistance Benefits for the Poorest Quintile, as Captured in Household Surveys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 3.29 Poverty Gap Reduction from Coverage and Level of Social Assistance Benefits for the Poorest Quintile, as Captured in Household Surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 4.1 Population, Age 64 Years and Older, as a Percentage of Total Population, by Region . . . . . . . . . . . . 72 4.2 Number of Countries with Old-Age Social Pensions, 1898–2012. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 4.3 Introduction of Old-Age Social Pensions, 2001–13. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 4.4 Distribution of Old-Age Pension Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 4.5 Age of Eligibility for Pension Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4.6 Old-Age Pension Coverage of Population Age 60 Years and Older, by Region. . . . . . . . . . . . . . . . . . 77 4.7 Distribution of Old-Age Social Pension Beneficiaries, by Income Quintile. . . . . . . . . . . . . . . . . . . . . 78 4.8 Old-Age Social Pensions as a Share of Beneficiaries’ Welfare, Poorest and Second-Poorest Quintiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 4.9 Impact of Old-Age Social Pensions on Poverty Headcount, Poverty Gap, and Gini Inequality Index Reduction, as a Share of Pretransfer Indicator Levels, Using Relative Poverty Line (Poorest 20 Percent). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 viii Contents 4.10 Benefit–Cost Ratio of Old-Age Social Pensions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 H2.1 Elderly and Labor Force Coverage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 5.1 Total Number of Disasters and Affected People, 1980–2012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 5.2 Total Number of Displaced People, 1951–2015 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 5.3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Ranking of Natural Disasters and Safety Net Coverage 5.4 Social Safety Net Coverage of the Poor and Humanitarian Spending, 2010–15 . . . . . . . . . . . . . . . . . 88 5.5 Program Scalability to Enable Responsiveness to Shocks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Map 4.1 Countries with Old-Age Social Pensions and Their Main Purpose. . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Tables 1.1 Social Protection and Labor Market Intervention Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.2 ASPIRE Classification of Social Safety Net Programs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.3 Countries with the Household and Administrative Data Used in This Book, by Country Income Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.4 Matching of Administrative and Household Survey Data for Social Safety Net Programs for Selected Countries/Economies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.1 Social Safety Net Spending across and within Regions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.2 Variations in Social Safety Net Spending across Country Income Groups . . . . . . . . . . . . . . . . . . . . . . 22 A.1 ASPIRE Social Protection and Labor Program Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 B.1 Household Surveys Used in the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 C.1 Conditional Cash Transfers and Unconditional Cash Transfers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 C.2 Food and In Kind and School Feeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 C.3 Public Works and Fee Waivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 D.1 Spending on Social Safety Net Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 E.1 Monthly Benefit Level Per Household for Selected Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 F.1 Key Performance Indicators of Social Protection and Labor Programs. . . . . . . . . . . . . . . . . . . . . . . 145 F.2 Key Performance Indicators of Social Safety Nets Programs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 F.3 Poverty and Inequality Reduction as a Result of Social Safety Nets Programs . . . . . . . . . . . . . . . . . 150 G.1 Old-Age Social Pensions around the World. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 G.2 Old-Age Social Pensions Captured in ASPIRE Household Surveys . . . . . . . . . . . . . . . . . . . . . . . . . . 160 H.1 Basic Characteristics of Countries Included in the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Contents ix Foreword T he need for social safety net/social assis- major data collection and processing effort tance (SSN/SA) is a critical concern for is  required. This has been the goal of a governments across the globe. Which World  Bank initiative called Atlas of Social SSN/SA programs to choose, how to best struc- Protection: Indicators of Resilience and ture and deliver them, and how to make them Equity  (ASPIRE), a compilation of compre- fiscally sustainable over the long term are hensive social protection indicators derived important questions because the answers to from administrative and household survey these questions affect the well-being of millions data (http://datatopics.worldbank.org/aspire/). of poor and vulnerable people around the The empirical analysis presented in this edition world. As the interest in and the use of SSN/SA of the book uses administrative (program-level) programs continue to grow, countries are also data for 142 countries and household ­ survey exploring how to better integrate SSN/SA pro- data for 96 countries. grams into their overall social protection and The evidence presented unequivocally indi- jobs agenda. cates that SSN/SA programs matter. The book The global focus on social protection and shows that SSN investments in coverage and jobs in general and on the role of SSN in partic- adequacy reduce the poverty gap/headcount ular has intensified. For the first time, social and lower income inequality, and coverage of protection is part of a comprehensive agenda of the poor tends to be larger in those places the Sustainable Development Goals (SDGs). where coverage of the general population is also SDG 1 calls to end (extreme) poverty in all its ­ substantial. It is not surprising that coverage manifestations by 2030, ensure social protec- and adequacy of SSN/SA programs come at a tion for the poor and vulnerable, increase access fiscal cost; globally, developing and transition to basic services, and support people harmed by economies spend an average of 1.5 percent climate-related extreme events and other eco- of  gross domestic product (GDP) on these nomic, social, and environmental shocks and programs. Whereas many countries still do not ­ disasters. Target 1.3 (Goal 1) seeks to imple- spend enough on SSN/SA programs to affect ment nationally appropriate social protection poverty, others have dedicated spending that systems and measures for all, including floors, has helped millions escape extreme poverty and and by 2030 achieve substantial coverage of the millions more to become less poor. poor and the vulnerable. Naturally, many ques- For the poor and vulnerable around the world, tions arise in implementing this agenda; for much more needs to be done, and much more example, what is deemed “nationally appropri- can be done regarding SSN/SA programs. ate” in a given country or context? What is a Significant gaps in coverage and benefit levels mix of SSN/SA programs and interventions remain. Even more disconcerting is that the gaps that makes sense in a specific context or a given are more pronounced in low-income countries. set of policy objectives? How much of the SSN The data suggest that in low-income countries, spending is too little versus too much? SSN/SA programs cover only 18 percent of A robust evidence base is needed to answer the  poorest quintile, and the average transfer these questions. The main objective of this accounts for only 13 percent of the lowest quin- book is to benchmark where individual coun- tile’s consumption. The international develop- tries, regions, and the world stand in terms of ment community needs to stand ready to work SSN/SA spending and key performance indi- further with countries in addressing the gaps. cators, such as program coverage, beneficiary Beyond presenting the key numbers on incidence, benefit level, and impacts on spending and performance around the world, reducing poverty and inequality. To evaluate this book also dives deeper into two thematic and benchmark these indicators consistently areas pertinent to managing risk and vulnera- across space (countries/programs) and time, a bility. The first is social assistance and aging, xi which looks specifically into the role of old-age hope that you will keep coming back to this book social pensions. The s ­econd is adaptive social as a reference guide and a compass to chart your protection, which discusses shocks and how thinking on the issues presented here. In the SSN/SA programs can be adapted to better meantime, we look forward to producing, shar- respond to them. It is clear that the risk of ing, and disseminating the latest global, regional, old  age is more predictable, but the risk of and country-level data and developments in ­ natural disasters is much less so; hence, differ- this crucial field of social safety nets, through this ent approaches and instruments are needed to 2018 edition and the ones to come. The reader is help people manage those risks. encouraged to further explore the rich dataset We are excited to offer you the full range of that the ASPIRE online platform offers. data and analysis that inform this book, and we I hope you enjoy reading this book. Michal Rutkowski Senior Director Social Protection and Jobs Global Practice World Bank Group xii FOREWORD Acknowledgments T his book was prepared by the Atlas who provided invaluable guidance on chapter 4 of  Social Protection: Indicators of of the book. The authors also acknowledge valu- Resilience and Equity (ASPIRE) team led able support from Jewel McFadden (­acquisitions by Alex (Oleksiy) Ivaschenko and composed of editor), Rumit  Pancholi (­ production editor), Marina Novikova (lead  author for chapter 2), and Deb Appel-Barker (print coordinator) on Claudia P. Rodríguez Alas (lead author for design, layout, management, and printing of chapter 3), Carolina Romero (lead author for this book. chapter 4), Thomas Bowen (lead author for The authors give special thanks to the country chapter 5), and Linghui (Jude) Zhu (lead author teams for collecting, sharing, and validating for the cross-chapter data analysis). detailed program-level data on SSN programs. Overall guidance was provided by Michal The team members include Pablo Acosta, Ihsan Rutkowski (senior director), Steen Jorgensen Ajwad, Mahamane Amadou, Diego Angel- (director), Margaret Grosh (senior advisor), Urdinola, Ignacio R. Apella, Philippe Auffret, Anush Bezhanyan (practice manager), and Clemente Avila Parra, Joao Pedro de Azevedo, Ruslan Yemtsov (lead economist and social Juan M. Berridi, Shrayana Bhattacharya, Gaston safety nets [SSNs] global lead), of the Social M. Blanco, John D. Blomquist, Gbetoho J. Boko, Protection and Jobs Global Practice of the Bénédicte de la Brière, Stefanie Brodmann, World Bank. Hugo Brousset, Tomas Damerau, Christabel E. Many thanks go to the reviewers for the Dadzie, Ivan Drabek, Puja Dutta, John van Dyck, ASPIRE database and the State of Social Safety Heba Elgazzar, Adrian Nicholas Gachet Racines, Nets report, whose views helped shape the Jordi Jose Gallego-Ayala, Sara Giannozzi, direction of this work, including Francesca Endashaw T. Gossa, Rebekka E. Grun, Nelson Bastagli, Margaret Grosh, Aline Coudouel, Gutierrez, Carlos S. Iguarán, Fatima El-Kadiri Philip O’Keefe, Phillippe Leite, Cem Mete, El-Yamani, Alex Kamurase, Toni Koleva, Emma Monsalve, Carlo Del Ninno, Aleksandra Matthieu Lefebvre, Raquel T. Lehmann, Victoria Posarac, and Ramya Sundaram. Levin, Ana Veronica Lopez, Zaineb Majoka, The authors thank the regional focal points Dimitris Mavridis, Emma S. Mistiaen, Muderis and teams for ongoing efforts with data sharing Mohammed, Vanessa Moreira, Matteo and verification: Aline Coudoue and Emma Morgandi, Ingrid Mujica, Iene  Muliati, Rose Monsalve of the Africa region; Pablo Acosta, Mungai, Michael Mutemi Munavu, Edmundo Jesse Doyle, and Puja Dutta of the East Asia and Murrugarra, Bojana Naceva, Suleiman Namara, Pacific region; Renata Gukovas, Aylin Isik- Minh Cong Nguyen, Ana Ocampo, Foluso Dikmelik, Mattia Makovec, and Frieda Okunmadewa, Katerina Petrina, Marina Vandeninden of the Europe and Central Asia Petrovic, Juul Pinxten, Serene Praveena Philip, region; Ursula Milagros Martinez Angulo, Lucian Pop, Avantika Prabhakar, Ali N. Qureshi, Lucia Solbes Castro, and Junko Onishi of Aneeka Rahman, Laura Ralston, Randa the  Latin America and Caribbean region; el-Rashidi, Laura B. Rawlings, Gonzalo Reyesy, Amr Moubarak and Wouter Takkenberg of the Nina Rosas Raffo, Solène Rougeaux, Manuel Middle East and North Africa region; and Cem Salazar, Nadia Selim, Veronica Silva Villalobos, Mete of the South Asia/Europe and Central Julia Smolyar, Victor Sulla, Hadyiat El-Tayeb Asia regions. Alyn, Cornelia M. Tesliuc, Fanta Toure, Maurizia Special thanks go to Maddalena Honorati Tovo,  Andrea Vermehren, Asha M. Williams, (former task team leader for the ASPIRE work) Sulaiman A. Yusuf, Giuseppe Zampaglione, and for the generous advice and guidance provided Eric Zapatero. to the team in the early stages of this work. The authors apologize to anyone who has The  team is also  grateful to Robert Palacios, been unintentionally not mentioned here. Acknowledgments xiii About the Authors Oleksiy Ivaschenko is a senior economist in the National Research University Higher School the World Bank’s Social Protection and Jobs of Economics in Moscow, Russia. Global Practice. He is a versatile empirical economist with extensive experience in opera- Carolina Romero is a research analyst in the tions and analytical work in social protection World Bank’s Social Protection and Jobs Global and labor, poverty analysis, migration, and Practice. She has more than a decade of experi- impact evaluations. His  work has been pub- ence reforming publicly and privately managed lished in many development journals including ­ pension systems worldwide. She is also a coau- the Journal of Comparative Economics, Journal thor of multiple books and articles on pension of Development and Migration, Journal of Policy ­ systems, labor markets, and youth and female Modeling, Migration Letters, and Economic empowerment. She holds an M.B.A. from Development and Cultural Change. Oleksiy The  Wharton School of the University of holds a Ph.D. in development economics Pennsylvania and a master’s degree in econom- from  the Gothenburg School of Economics ics from the Universidad de Los Andes in in Sweden. Colombia. Claudia P. Rodríguez Alas is a social protec- Thomas Bowen is a social protection specialist tion specialist in the World Bank’s Social in the World Bank’s Social Protection and Jobs Protection and Jobs Global Practice. Her work Global Practice. In this capacity, Thomas has at the World Bank focuses on generating global worked extensively on issues related to social knowledge products on social protection, safety nets, cash transfer programs and their role including development of the Atlas of Social in building household resilience to disasters, and Protection: Indicators of Resilience and Equity climate change, with a particular focus on the (ASPIRE) database. She has also worked with East Asia Pacific region. He holds an MA in eco- nonprofit organizations on community out- nomics and international relations from the reach and immigrants’ rights. She received her School of Advanced International Studies at The bachelor’s degree in economics from Montana Johns Hopkins University in Washington, DC. State University, where she was a Fulbright Scholar. Claudia also holds a master’s degree in Linghui (Jude) Zhu is a consultant in the World international development from American Bank’s Social Protection and Jobs Global University in Washington, DC. Practice with extensive experience in research, analysis, and technical assistance. He special- Marina Novikova is a consultant in the World izes in applied labor economics and has worked Bank’s Social Protection and Jobs Global widely on social protection, labor market, edu- Practice. She is a global focal point for adminis- cation, and migration. His current research is trative data for the ASPIRE database; her other on the labor mobility in China, exploring the projects comprise operational and analytical link between ­ migration and welfare distribu- work in social protection and labor, public tion. Jude is a Ph.D. candidate in economics at expenditure reviews, and labor market analysis. Kobe University in Japan and holds a master’s Marina holds a bachelor’s degree in economics degree from the University of Pittsburgh in and a master’s degree in labor economics from Pennsylvania. xv Structure of the Book T his book is the third edition of a publica- the social safety net instruments, and key defi- tions that monitors the state of social nitions used throughout the book. safety nets around the world by present- • Chapter 2 presents levels, trends, and patterns ing key global social safety net statistics on in countries’ social safety nets spending. spending, coverage, benefit level, and poverty/ • Chapter 3 discusses what happens with social inequality impact. safety nets around the world through the lens This 2018 edition of The State of Social of performance indicators. Safety Nets presents a richer and more com- • Chapter 4 looks at social assistance and aging. prehensive set of SSN/SA data compared with • Chapter 5 explores adaptive social safety nets. the 2015 edition. The main empirical analysis is presented in Part I, “Analytics.” This book The book also includes two highlights: also  takes a deeper conceptual dive into selected topics, which are discussed in Part II, • Highlight 1 looks at the evidence from impact “Special Topics.” The book consists of five evaluations in Sub-Saharan Africa on the role chapters: of social safety nets in enhancing productive inclusion. • Chapter 1 sets the stage for the discussion • Highlight 2 discusses policy considerations for of the social safety nets/social assistance by introducing old-age social pensions and the presenting the context of risks, rationale for special considerations that inform their design. xvii Abbreviations AOV Algemene Oudedags Voorzieningsfonds ASP adaptive social protection ASPIRE Atlas of Social Protection: Indicators of Resilience and Equity BEAM Basic Education Assistance Module BRP basic retirement pension CCT conditional cash transfer CGH Coady-Grosh-Hoddinott CGP Child Grant Program CPI consumer price index EECF enhanced elemental chlorine-free ESSPROS European System of Integrated Social Protection Statistics GDP gross domestic product GNI gross national income HH household HSCT harmonized social cash transfer HSNP Hunger Safety Net Program JSLU Jaminan Sosial Lanjut Usia LEAP Livelihood Empowerment against Poverty LEWIE Local Economy-Wide Impact Evaluation LIPW Labour-Intensive Public Works MASAF Malawi Social Action Fund MGNREG Mahatma Gandhi National Rural Employment Guarantee MOP Materijalno Obezbedenje Porodice (Family Material Support) MOPMED Ministry of Primary and Mass Education Department NCIP National Commission of Indigenous Peoples NHIS National Health Insurance Scheme OECD Organisation for Economic Co-operation and Development PASD Program Apoio Social Directo (Direct Social Support Program) PBS Pensión Básica Solidaria PCF processed chlorine-free PET Programa de Empleo Temporal PPP purchasing power parity PRAF Programa de Asignación Familiar (Family Allowance Program) PRODEP Projet National de Développement Communautaire Participatif (National Project of Community Participation Development) PRRO Protracted Relief and Recovery Operations PSNP Productive Safety Net Program PSSB Programa de Subsido Social Basico PSSN Productive Social Safety Net PtoP Protection to Production PW public works RCIW Rural Community Infrastructure Works S.V. Securite Vieillesse SA social assistance SCT social cash transfer SCTPP Social Cash Transfer Pilot Program xix SDG Sustainable Development Goal SIUBEN Sistema Único de Beneficiarios (Unique Registry of Beneficiaries) SOCX Social Expenditure Database (OECD) SPL social protection and labor SSN social safety net SUF Subsidio Familiar TCF totally chlorine-free TSA Targeted Social Assistance UCT unconditional cash transfer UNICEF United Nations Children’s Fund USP Universal social protection VUP Vision 2020 Umurenge Programme WEF World Economic Forum WFP World Food Progamme xx Abbreviations Executive Summary T he State of Social Safety Nets 2018 aims to achieve substantial coverage of the poor and the compile, analyze, and disseminate data vulnerable. Target 1.5 (Goal 1), which relates to and developments at the forefront of the adaptive social protection, aims to build the social safety net  (SSN)/social assistance (SA) resilience of the poor and those in vulnerable agenda.1 This series of periodic reports is part of situations and to reduce their exposure and vul- broader efforts to monitor implementation nerability to climate-related extreme events progress of the World Bank 2012–2022 Social and other economic, social, and environmental Protection and Labor Strategy against the stra- shocks and disasters. Measuring performance tegic goals of  increasing coverage—especially on those targets requires reliable data. among the poor—and enhancing the poverty As chapter 1 shows, most countries have impact of the programs.2 a  diverse set of SSN/SA instruments. Of This third edition of The State of Social Safety 142  countries in the ASPIRE administrative Nets examines trends in coverage, spending, database, 70  percent have unconditional cash and program performance using the World transfers, and 43 percent have conditional Bank Atlas of Social Protection Indicators of cash  transfers. More than 80 percent of coun- Resilience and Equity (ASPIRE) updated tries provide school feeding programs. Also, database.3 The book documents the main safety 67 percent of countries have public works, and net programs that exist around the world and 56 percent have various fee waivers. The num- their use to alleviate poverty and build shared ber of countries with old-age social pensions prosperity. The 2018 edition expands on the has also grown rapidly in the past two decades. 2015 version in its coverage of administrative A growing commitment to SSN/SA is also and household survey data. This edition is dis- evident; many countries tend to spend more on tinctive, in that for the first time it describes these programs over time. From the analysis of what happens with SSN/SA program spending the subset of countries with comparable data and coverage over time, when the data allow over time, chapter 2 shows that in the Latin such analysis. America and the Caribbean region, for exam- The State of Social Safety Nets 2018 also fea- ple, average spending on SSN/SA programs as tures two special themes—social assistance and a percentage of gross domestic product (GDP) aging, focusing on the role of old-age social increased from 0.4 percent of GDP in 2000 to pensions; and adaptive social protection, focus- 1.26 percent of GDP in 2015. This happened ing on what makes SSN systems and programs while regional GDP grew, which means that adaptive to various shocks. SSN spending has increased in relative and This book provides much-needed empirical absolute terms. Many countries in other evidence in the context of an increasing global regions, including Europe and Central Asia focus on social protection, as is evident in the and Sub-Saharan Africa, have also substan- Sustainable Development Goals (SDGs).4 For tially increased their spending on flagship the first time, social protection is part of a com- SSN programs. prehensive SDG agenda. SDG 1 calls to end Globally, developing and transition countries (extreme) poverty in all its manifestations by spend an average of 1.5 percent of GDP on SSN 2030, ensure social protection for the poor and programs. However, as chapter 2 highlights, vulnerable, increase access to basic services, spending varies across countries and regions. and support people harmed by climate-related The Europe and Central Asia region currently extreme events and other shocks and disasters. spends the most on SSN programs, with Target 1.3 (Goal 1) seeks to implement nation- average  spending of 2.2 percent of GDP; the ally appropriate social protection systems and Sub-Saharan Africa and Latin America and measures for all, including floors, and by 2030 the Caribbean regions are in the middle of the 1 spending range; and the Middle East and North situation is not much better in lower-middle-­ Africa and South Asia regions spend the least, income countries, where the ratio stands at at 1.0 percent and 0.9 percent, respectively. 18 percent. The book also shows that countries The increase in spending has translated into differ substantially in absolute average per cap- a  substantial increase in program coverage ita SSN spending (in terms of U.S. dollars, in around the world. For example, several coun- purchasing power parity terms). For example, tries are introducing flagship SSN programs Sub-Saharan African countries spend an aver- and  are rapidly expanding their coverage. In age of US$16 per citizen annually on SSN pro- Tanzania, the Productive Safety Net Program grams, whereas countries in the Latin America expanded from covering 2 percent to 10 percent and the Caribbean region spend an average of of the population between 2014 and 2016. In US$158 per citizen annually. Senegal, the National Cash Transfer Program It is important to close these gaps because expanded from 3 percent to 16 percent of the countries with low coverage and benefit levels population in four years. In the Philippines, the achieve only a very small reduction in poverty. Pantawid conditional cash transfer program has Analysis of the ASPIRE database indicates that expanded from 5 percent to 20 percent of the only countries with substantial coverage and population since 2010. These examples are only benefit levels make important gains in poverty a few of the rapidly expanding programs. reduction. Countries with the highest levels of Chapter 3 shows that SSN programs are coverage combined with high benefit levels making a substantial contribution to the fight achieve up to a 43 percent reduction in the against poverty. From the available household poverty headcount (the share of the population survey data, it is estimated that 36 percent of in the poorest quintile). Similar strong effects people escape absolute poverty because of are found with respect to reduction in the pov- receiving SSN transfers.5 In other words, in the erty gap and decline in income/consumption absence of transfers, many more people would inequality. be living in absolute poverty. Even if the SSN This book also goes beyond data analytics transfers do not lift the beneficiaries above the and considers two specific areas of social pro- poverty line, they reduce the poverty gap by tection policy that require further understand- about 45 percent.6 SSN programs also reduce ing and exploration: social assistance and aging consumption/income inequality by 2 percent, and adaptive social protection. Under the first on average. These positive effects of SSN trans- special topic, chapter 4 looks through the fers on the poverty headcount, poverty gap, numerical lens on the growing role of old-age and  inequality are observed for all country social pensions around the world. This is a income groups. global trend largely reflecting the limited cov- Despite the progress that has been made, the erage and adequacy of contributory pension social protection community needs to do more. schemes. The important contribution of the Significant gaps in program coverage persist chapter on old-age social pensions is its around the globe. These gaps are especially pro- attempt to quantify the poverty impact of this nounced in low-income countries, where only policy instrument using household surveys 18 percent of the poorest quintile are covered with reliable data. by SSN programs. Even in lower-middle-­ Chapter 5 discusses the key features that make income countries, less than 50 percent of the SSNs adaptive to various types of shocks, both poor have access to SSN programs. Moreover, natural (such as cyclones and droughts) and very few of the poor are included in social human-made (such as conflicts and forced dis- insurance programs. As the book suggests (see placement). Adaptive social protection instru- chapter 3), coverage is much better in upper- ments are important for people, irrespective of middle-­ income countries and high-income where they are in the life cycle. The chapter on countries, but even there the gaps remain.7 adaptive social protection aims to shed light Benefit levels also need to be increased. on  what adaptability is about and how to As chapter 3 shows, SSN benefits as a share of achieve  it. It also highlights examples of what the poor’s income/consumption are lowest in countries are already doing to make their social low-income countries, at only 13 percent. The protection schemes more flexible and efficient. 2 The State of Social Safety Nets 2018 NOTES 4. The ASPIRE database and the analysis for this book 1. The terms “social safety nets” and “social assistance” consider social protection to consist of social safety are used interchangeably in this book. They are non- nets social assistance, social insurance, and labor contributory measures designed to provide regular market programs. and predictable support to poor and vulnerable 5. Extreme poverty is measured with a poverty line of people. They are also referred to as “safety nets,” US$1.90 per day in purchasing power parity terms. “social assistance,” or “social transfers” and are a component of larger social protection systems. 6. The poverty gap is the distance between the pov- erty line and the average income of the poor. It is 2. The World Bank 2012–2022 Social Protection and typically expressed as the percentage shortfall Labor Strategy (www.worldbank.org/spstrategy) states in  income of the poor with respect to the that the “overarching goals of the strategy are to help poverty line. improve resilience, equity, and opportunity for people in both low- and middle-income countries through 7. In this book, the high-income countries category integrated social protection and labor systems, increas- includes only a few high-income countries that are ing coverage of social safety nets programs, especially members of the Word Bank Group and for which in lower-income countries, and improved evidence.” household survey data are available. For a list of these countries, see table 1.3 in chapter 1. 3. The ASPIRE database can be found at www.worldbank​ .org/aspire. EXECUTIVE SUMMARY 3 FPO PART I Analytics CHAPTER 1 Explaining the Social Safety Net’s Data Landscape INTRODUCTION policies and instruments that contribute to What are social protection and social safety net building human capital and facilitate access to (SSN) interventions? How does this book clas- jobs and investments in livelihoods. sify SSN programs? What is the Atlas of Social SPL instruments generally fall into the fol- Protection: Indicators of Resilience and Equity lowing three categories: (ASPIRE) database? How does the ASPIRE team collect data and ensure data quality? What 1. Social safety net (SSN)/social assistance (SA) are the limitations of the administrative and programs are noncontributory interventions household survey data used in this book? How designed to help individuals and house- is the performance of SSN programs measured? holds cope with chronic poverty, destitu- This chapter aims to answer these questions, tion, and vulnerability. SSN/SA programs and, by doing so, lays out the landscape for target the poor and vulnerable. Examples understanding the book in its entirety. include unconditional and conditional cash transfers, noncontributory social pensions, WHAT ARE SOCIAL PROTECTION AND food and in-kind transfers, school feeding SOCIAL SAFETY NET INTERVENTIONS? programs, public works, and fee waivers WHAT IS THE ASPIRE CLASSIFICATION (see table 1.1). OF SOCIAL SAFETY NET PROGRAMS? 2. Social insurance programs are contributory Social protection and labor (SPL) interventions interventions that are designed to help indi- are well recognized for promoting resilience, viduals manage sudden changes in income equity, and opportunity. The World Bank 2012– because of old age, sickness, disability, or 2022 Social Protection and Labor Strategy: natural disaster. Individuals pay insurance Resilience, Equity, and Opportunity argues that premiums to be eligible for coverage or con- SPL systems, policies, and instruments help tribute a percentage of their earnings to a individuals and societies manage risk and vola- mandatory insurance scheme. Examples tility and protect them from poverty and desti- include contributory old-age, survivor, and tution (World Bank 2012). Equity is enhanced disability pensions; sick leave and maternity/ through instruments that help protect against paternity benefits; and health insurance destitution and promote equality of opportu- coverage. nity. Resilience is promoted through programs 3. Labor market programs can be contributory or that minimize the negative effect of economic noncontributory programs and are designed to shocks and natural disasters on individuals and help protect individuals against loss of income families. Opportunity is enhanced through from unemployment (passive labor market Explaining the Social Safety Net’s Data Landscape 5 TABLE 1.1  Social Protection and Labor Market Intervention Areas Social protection and labor programs Objectives Types of programs Social safety nets/social assistance Reduce poverty and inequality • Unconditional cash transfers (noncontributory) • Conditional cash transfers • Social pensions • Food and in-kind transfers • School feeding programs • Public works • Fee waivers and targeted subsidies • Other interventions (social services) Social insurance (contributory) Ensure adequate living standards in • Contributory old-age, survivor, and the face of shocks and life changes disability pensions • Sick leave • Maternity/paternity benefits • Health insurance coverage • Other types of insurance Labor market programs (contributory and Improve chances of employment and • Active labor market programs (training, noncontributory) earnings; smooth income during employment intermediation services, wage unemployment subsidies) • Passive labor market programs (unemployment insurance, early retirement incentives) Source: World Bank 2012. Note: ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. policies) or help individuals acquire skills and There is clear demand for a tool that helps connect them to labor  markets (active labor monitor the scope, performance, and effect of market policies). Unemployment insurance SPL programs in countries worldwide. The and early retirement incentives are examples of Social Protection and Jobs Global Practice of passive labor market policies, whereas training, the World Bank Group is committed to devel- employment intermediation services, and oping and continuously updating a comprehen- wage subsidies are examples of active policies. sive set of comparable and accessible indicators to help measure the performance of SSN/SA This book, in its empirical analysis, focuses (as well as broader SPL) programs. on the state of SSN/SA programs. This focus The Social Protection and Jobs Global reflects an increased use of social safety net Practice has created a user-friendly bench- instruments as well as the need to capture marking tool that continuously updates key up-to-date data and to assess the status of SSN SSN/SPL indicators: ASPIRE. This portal serves programs globally. The advance of available as a one-stop shop for SPL indicators for both data resources positions the ASPIRE team well World Bank staff and external practitioners. to analyze SSN/SA programs. In addition, by ASPIRE links directly to the World Bank Group focusing on the SSN/SA programs (as a subset Databank to provide users with tools to search of SPL programs), this book provides continu- the database and to generate customized tables ity with the two previous books on the state of and charts. In addition, the portal includes social safety nets (2014 and 2015). related survey information from the World This book also extends to the broader SPL Bank Microdata Library. agenda. The “Special Topics” section considers For cross-country comparability, this book old-age social pensions, which are linked to the follows the ASPIRE harmonized classifica- social insurance agenda, and adaptive social tion of SSN/SA programs. ASPIRE groups protection, which can be achieved through both SSN/SA programs into eight harmonized cat- safety nets and contributory/insurance pro- egories on the basis of program objectives grams. Ultimately, the policy issues related to (see table 1.2 and appendix A). This classifi- social safety nets, social insurance, and labor cation is applied to each country in the data- market agendas are closely connected. base to generate comparable program 6 The State of Social Safety Nets 2018 TABLE 1.2  ASPIRE Classification of Social Safety Net Programs Program category Program subcategory Unconditional cash transfers a Poverty-targeted cash transfers, last-resort programs Family, children, orphan allowance, including orphans and vulnerable children benefits Noncontributory funeral grants, burial allowances Emergency cash support, including support to refugees and returning migrants ¯t Public charity, including zaka Conditional cash transfersb Conditional cash transfers Social pensions (noncontributory) c Old-age social pensions Disability benefits War veteran benefits Survivorship benefits Food and in-kind transfers Food stamps, rations, vouchers Nutrition programs (therapeutic, supplementary feeding) School supplies (free textbooks, uniforms) In-kind/nonfood emergency support Other in-kind transfers School feeding School feeding programs Public works, workfare, and direct job Cash-for-work creation Food-for-work, including food-for-training, food-for-assets Fee waivers and targeted subsidies Health insurance exemptions, reduced medical fees Education fee waivers Food subsidies Housing subsidies and allowances Utility and electricity subsidies and allowances Agricultural-inputs subsidies Transportation benefits Other social assistance Scholarships, education benefits Social services, transfers for caregivers (care for children, youth, family, working-age, disabled, and older persons) Tax exemptions Source: ASPIRE database. Note: a. Conditional cash transfer programs aim to reduce poverty by making welfare programs conditional upon actions by the beneficiary. The government (or an implementing agency) transfers the money only to those households or persons (beneficiaries) that meet certain criteria in the form of actions, such as enrolling children in public schools, getting regular check-ups at the doctor’s office, receiving vaccinations, or the like. Conditional cash transfer programs seek to help the current generation in poverty and to break the cycle of poverty for the next generation by developing human capital. b. Unconditional cash transfer programs do not require beneficiaries to perform any specific actions to be eligible for the benefit. However, these programs may require benficiaries to meet certain criteria or have a certain status to be eligible; for example, for a poverty-targeted benefit, a household must be below a poverty threshold. c. Social pensions here encompass various types of social pensions, such as old-age pensions, disability benefits, and survivorship benefits, whereas chapter 4 focuses exclusively on old-age social pensions. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. expenditure and performance indicators. WHAT IS THE ASPIRE DATABASE? HOW Whereas table 1.2 reflects various types of DOES THE ASPIRE TEAM COLLECT AND social pensions as captured by the ASPIRE ENSURE QUALITY OF THE DATA? WHAT database, this book’s Special Topics section ARE THE LIMITATIONS OF THE DATA? focuses exclusively on old-age social pen- The ASPIRE database has achieved significant sions, which facilitates a clean comparison of scale and has become the World Bank’s premier this instrument across countries. compilation of performance indicators for Explaining the Social Safety Net’s Data Landscape 7 social protection and labor programs. ASPIRE well as existing analysis such as public expendi- has two main sources of data: administrative ture reviews, constitute the secondary source of data, from which program expenditures and administrative data. number of beneficiaries are derived; and house- Although the SSN classification of programs hold survey data, which are used to estimate the facilitates cross-country comparison, it does coverage, benefit incidence, benefit levels, and not necessarily imply an easy and clean-cut poverty/inequality impact of SPL programs. differentiation of programs. As mentioned, The two data sources complement each other, ASPIRE classifies and aggregates individual and thus provide a more comprehensive view of SSN/SA programs into eight categories, largely SPL program performance around the world. on the basis of the objective and nature of each The ASPIRE work program supports con- program. However, in practice, program tinuous improvement in the quality, compara- objectives often tend to overlap, blurring bility, and availability of SPL/SSN data to the  line between classifications. For example, facilitate SPL benchmarking and inform poli- although a cash transfer program may not cies. As of November of 2017, the ASPIRE data- have explicit eligibility conditions (making it base included administrative spending data for an unconditional cash transfer), it may have 124 developing and transition countries and strong uptake incentive mechanisms or soft economies (see appendix D) and administra- conditions that influence decisions on how tive data on the number of beneficiaries of households spend the transfer, making it in the  largest programs for 142 countries (see principle a conditional cash transfer (Daidone appendix C).1 This book uses only the most and others 2015). recent subset of the ASPIRE data—specifically, Available program-level administrative the most recent year of data available per spending data currently covers 124 countries country. These data constitute the basis for the representing 80 percent of the world’s popula- analysis in chapter 2. For the performance tion. Updates are available for 28 countries analysis presented in chapter 3, the book uses through 2016; for 42 countries through 2015; the most recent household survey data from and for 41 countries through either 2013 or 96 countries (see appendix B).2 The examples 2014. The year of reference for the remaining of monthly benefit levels per household for countries in the database ranges from 2010 to selected programs are presented in a ­ ppendix E. 2013, except for Bhutan, Jordan, Marshall Islands, Appendix F presents key performance indica- and Vanuatu. For these four countries, only total tors. Appendix G takes stock of old-age social SSN spending is available from secondary pensions around the world (which are dis- sources, and the reference year is 2009. Countries cussed in chapter 4). The full list of countries with data points before 2009 are considered out- found in the household and administrative dated and are not included in the analysis. data used in this book is presented, by country A complete summary of spending indicators income group, in table  1.3. Basic characteris- disaggregated by program categories can be tics of these countries, such as the total coun- found in appendix D.3 The program-level admin- try population and gross national income GDP istrative data facilitates a granular look at country-­ per capita, can be found in appendix H. level spending on social safety nets/social World Bank staff and in-country consultants assistance. Furthermore, by comparing global collect and harmonize the administrative data spending trends and patterns, the spending pro- using standardized terms of reference, data files and program portfolios of countries and templates, and classifications. Publicly available regions can be benchmarked. government statistics, such as annual program The presence on the ground of the larger budget expenditures, are the primary source of ASPIRE and Social Protection and Jobs Global administrative data. In the case of donor- Practice teams, including consultants, facili- funded programs, the program budget pro- tates information flows that help improve the vided by the donor is also considered a primary quality of administrative data. The engagement source of data. Other information and data in the country helps establish a dialogue received from program and sector officials, as with  government counterparts and assists in 8 The State of Social Safety Nets 2018 TABLE 1.3  Countries with the Household and Administrative Data Used in This Book, by Country Income Group No. of countries Country/economy/region name Administrative Household Income group data data Administrative data Household data Low-income 26 22 Benin; Burkina Faso; Burundi; Central African Afghanistan; Burkina Faso; Central African countries Republic; Chad; Comoros; Congo, Dem. Rep., Republic; Chad; Congo, Dem. Rep.; Ethiopia; Ethiopia; Guinea; Guinea-Bissau; Liberia; The Gambia; Guinea; Haiti; Liberia; Madagascar; Madagascar; Malawi; Mali; Mozambique; Nepal; Malawi; Mozambique; Nepal; Niger; Rwanda; Niger; Rwanda; Senegal; Sierra Leone; Somalia; Senegal; Sierra Leone; South Sudan; Tanzania; South Sudan; Tanzania; Togo; Uganda; Uganda; Zimbabwe Zimbabwe Lower-middle- 48 37 Angola; Armenia; Bangladesh; Bhutan; Bolivia; Armenia; Bangladesh; Bhutan; Bolivia; income countries Cabo Verde; Cambodia; Cameroon; Congo, Cameroon; Côte d’Ivoire; Djibouti; Egypt, Arab Rep.; Côte d’Ivoire; Djibouti; Egypt, Arab Rep.; Rep.; El Salvador; Georgia; Ghana; Guatemala; El Salvador; Georgia; Ghana; Guatemala; Honduras; India; Indonesia; Jordan; Kosovo; Honduras; India; Indonesia; Jordan; Kenya; Kyrgyz Republic; Mauritania; Moldova; Kiribati; Kosovo; Kyrgyz Republic; Lao PDR; Mongolia; Morocco; Nicaragua; Nigeria; Lesotho; Mauritania; Moldova; Mongolia; Pakistan; Papua New Guinea; Philippines; Sri Morocco; Myanmar; Nicaragua; Nigeria; Lanka; Sudan; Swaziland; Tajikistan; Timor- Pakistan; Papua New Guinea; Philippines; São Leste; Tunisia; Ukraine; Vietnam; West Bank Tomé and Príncipe; Sri Lanka; Sudan; and Gaza; Zambia Swaziland; Tajikistan; Timor-Leste; Tunisia; Ukraine; Vanuatu; Vietnam; West Bank and Gaza; Zambia Upper-middle- 38 31 Albania; Azerbaijan; Belarus; Bosnia and Albania; Argentina; Belarus; Belize; Botswana; income countries Herzegovina; Botswana; Brazil; Bulgaria; Brazil; China; Colombia; Costa Rica; Croatia; Argentina; China; Colombia; Costa Rica; Dominican Republic; Ecuador; Fiji; Iraq; Croatia; Dominican Republic; Ecuador; Fiji; Jamaica; Kazakhstan; Malaysia; Maldives; Gabon; Grenada; Iraq; Kazakhstan; Lebanon; Mauritius; Mexico; Montenegro; Namibia; Macedonia, FYR; Malaysia; Maldives; Marshall Panama; Paraguay; Peru; Romania; Russian Islands; Mauritius; Mexico; Montenegro; Federation; Serbia; South Africa; Thailand; Namibia; Panama; Peru; Romania; Russian Turkey Federation; Samoa; Serbia; South Africa; St. Lucia; Thailand; Turkey High-income 12 6 Chile, Estonia, Hungary, Kuwait, Latvia, Chile; Latvia; Lithuania; Poland; Slovak countries Lithuania, Poland, Saudi Arabia, Seychelles, Republic; Uruguay Slovak Republic, Slovenia, Uruguay Total 124 96     Source: ASPIRE database. Note: Economies are divided among income groups according to 2016 gross national income per capita, calculated using the World Bank Atlas method. The groups are as follows: low-income, US$1,005 or less; lower-middle-income, US$1,006–3,955; upper-middle-income, US$3,956–12,235; and high-income, US$12,236 or more. See appendix H for gross national income per capita statistics for individual countries. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. gathering the required information/data, veri- updates are available (as mentioned) is consid- fying classification, and checking quality. ered the reference year for which expenditures Continuous improvement to the data is ensured are tallied. Unfortunately, for some countries, by close collaboration between the ASPIRE spending information for the latest year is not central team, the ASPIRE focal points in the available for all active programs, but in many regions, and the World Bank Social Protection cases prior-year information on spending is and Jobs Global Practice staff at the country available. In such cases, this prior-year spend- level, who have extensive program knowledge. ing (relative to the same-year gross domestic When estimating the amount a country product, GDP) is used. In sorting out the data, spends on SSN programs, the book uses the the focus is always on updating the largest pro- latest-available-year approach. For the list of all grams in terms of beneficiary numbers and active programs, the latest year for which spending amounts. The analysis of spending Explaining the Social Safety Net’s Data Landscape 9 (see chapter 2) also distinguishes the inclusion/ part of the International Comparison Program.7 exclusion of the health fee waivers in total SSN The consumption or income aggregates used to spending whenever possible.4 rank households by their welfare distribution Performance indicators are estimated using are validated by the World Bank’s regional pov- nationally representative household surveys. As erty teams. of November of 2017, the ASPIRE database There are important considerations to keep included 309 household surveys, correspond- in mind when going through the performance ing to 123 developing countries. The book uses analysis. First, this analysis is limited to the only the latest year for each country and only if programs captured in the household surveys. the data are from at least 2008; under this crite- Most household surveys capture only a frac- rion, 20 countries were excluded. In addition, tion of the programs administered in a given several countries whose surveys did not have country. Thus the data do not always include SSN information were excluded: Cambodia a comprehensive list of programs (which are (2013), Mali (2009), Myanmar (2009), Samoa likely to appear in administrative data) (2008), Togo (2011), and Tonga (2009).5 As a implemented in each country. Accordingly, result, the performance indicators are based on coverage indicators are underestimated with 96 countries (see appendix B for a full list of the respect to overall social spending. To illus- household surveys used). trate this point, the ASPIRE team conducted The ASPIRE team carefully reviews different a matching exercise, looking at program over- household surveys to identify relevant SPL pro- lap between the administrative data and gram information. Typically, the surveys household surveys for several countries (see include household income and expenditure table 1.4). A few key messages emerge from surveys; household budget surveys; living stan- this exercise: dards measurement surveys; integrated, multi- purpose, and socioeconomic surveys; or any 1. There is generally little overlap between other survey that is nationally representative administrative and household data. In the and captures information on social protection sample of counties (see table 1.4), on aver- and labor programs. In some cases, this work age only about 20 percent of programs can also leads to recommendations made to gov- be found in both administrative and house- ernment counterparts on how the design of the hold survey data; for some countries, the survey instrument/module can be changed to matching rate is less than 10 percent. better capture SPL programs. 2. Household surveys tend to capture larger pro- Individual variables are generated for each grams, although only in part. On average, the SSN program captured in the survey; they are matching programs capture about 50 percent then grouped into the eight harmonized pro- of the SSN budget, as seen by summing up the gram categories.6 The performance indicators budget for the matching programs on the are generated using these harmonized program basis of administrative budget data. categories. These indicators, in turn, can be dis- 3. Every country case is unique. There are aggregated by quintiles of welfare before and significant variations across countries in after transfers, extreme poverty status (defined terms of how many programs are captured as US$1.90/day in terms of purchasing power in the household survey (as a percentage parity, PPP), and rural/urban populations. of the total number of programs) and what Household weights are used to expand results percent of the total budget they account for to the total population of each country. in the administrative data. For example, in For cross-country comparability, all mone- Chile, 14 out of 135 programs (10 percent) tary variables are expressed in 2011 prices and match, accounting for 30 percent of SSN daily PPP in U.S. dollars. This also facilitates the programs’ total budget; in Romania, 10 out PPP US$1.90/day poverty-line metric to deter- of 65 (15 percent) match, accounting for mine the poverty status for each country/survey. 96 percent of the total budget; and in South Note that 2011 is used as a base year because Africa, 6  out of 16 (40 percent) match, this is the year when the most recent compre- accounting for 85 percent of the total budget hensive global price statistics were collected as a (see table 1.4). 10 The State of Social Safety Nets 2018 TABLE 1.4  Matching of Administrative and Household Survey Data for Social Safety Net Programs for Selected Countries/Economies Explaining the Social Safety Net’s Data Landscape Number of SSN Number of SSN Number of SSN Share of matched Share of all SSN budgets Reference year programs in programs/ programs/ programs in the total captured (in administrative Country/ (latest) for administrative Reference year for categories in HH categories matching number of programs data) by the matched Economy administrative data data HH survey data survey data in both sources (administrative data) programs (max = 1) Sub-Saharan Africa Ethiopia 2016 8 2010–11 3 1 0.13 0.68 Mauritania 2016 7 2014 12 1 0.14 0.23 Mozambique 2015 20 2014 3 2 0.10 0.41 Rwanda 2016 14 2014 10 4 0.29 0.55 South Africa 2015 16 2010 8 6 0.38 0.85 Tanzania 2016 14 2012–13 7 3 0.21 0.25 East Asia and Pacific Indonesia 2015 28 2014 8 3 0.11 0.46 Vietnam 2015 58 2014 16 7 0.12 0.20 Europe and Central Asia Armenia 2014 12 2014 8 3 0.25 0.62 Georgia 2013 18 2011 14 6 0.33 0.77 Lithuania 2016 15 2008 16 9 0.60 0.70 Poland 2013 45 2012 14 11 0.24 0.60 Romania 2014 65 2012 14 10 0.15 0.96 Ukraine 2014 52 2013 20 13 0.25 0.61 Latin America and the Caribbean Chile 2015 135 2013 23 14 0.10 0.32 Colombia 2015 37 2014 9 3 0.08 0.18 Guatemala 2013 10 2014 16 3 0.30 0.89 Middle East and North Africa Iraq 2013 5 2012 5 1 0.20 0.15 Morocco 2016 22 2009 12 5 0.23 0.10 West Bank & Gaza 2014 13 2009 3 1 0.08 0.10 South Asia India 2016 20 2010–12 2 1 0.05 0.53 Nepal 2014 57 2010–11 16 2 0.04 0.32 Pakistan 2016 30 2013–14 3 2 0.07 0.61 Sri Lanka 2015 40 2012 7 3 0.08 0.68 Source: ASPIRE team calculations, 2017. Note: For Vietnam, out of 58 programs, 21 are under Decree 136 (also called Program 136). Hence, Program 136 is a breakdown into 21 programs. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; HH = household; SSN = social safety net. 11 Also, some surveys collect information For  example, some surveys collect informa- only  on program participation without tion on social programs  mixed with private including transfer amounts. In such cases, transfers, making it difficult to isolate individ- only ­ coverage and beneficiary incidence ual SPL programs. Despite these  limitations, indicators  can be  estimated. Last, because ­ household surveys  have unique  advantages household surveys  differ in the method (see box 1.1) and are the sole source for calcu- for  collecting SPL  information across coun- ­ erformance indicators presented lating most p tries, the quality of the information varies. in this book. BOX 1.1  Leveraging Household Survey Data to Monitor and Measure Social Protection and Labor Program Performance Household surveys have great potential as the collection of SPL information in survey instruments to monitor and assess the perfor- instruments, including the following: mance of social protection and labor (SPL) pro- grams. However, not all countries use household 1. Review existing SPL programs in the country. surveys to estimate SPL program trends or gen- Obtaining a full list of programs and their erate basic performance indicators (such as cov- specifications (for example, target population, erage, benefit level, benefits and beneficiary benefit level, frequency of payments, and incidence, and effects on poverty). A main factor program size) will help make the list of SPL behind the low use of household data is the programs in the questionnaire more complete inadequate SPL information captured in most and better formulate the survey questions national household surveys. Thus there is a need to capture adequate information. In addition, to improve data collection and the quality of SPL information about the program size will help information in household surveys to better evaluate whether a program is large enough inform social policy. to be captured by the sample frame or if Why are household surveys necessary to oversampling is needed. measure SPL performance? Household  surveys 2. Identify and coordinate with key partners. are the only source of information regarding Coordination between policy makers, program potential beneficiaries and the basis for implementers, and National Statistical Office ex ante simulations for policy reform. By officials is crucial to design a good set of including information representative of the questions and sampling frame. The survey’s total population, household surveys allow the representation of programs can be imprecise identification of populations that, because of if the sample does not overlap with areas their characteristics, may be eligible for an where the programs are implemented. SPL program (for example, the poor, disabled, 3. Design the best format to collect SPL program and unemployed). Ex ante assessments can information. Whenever possible, survey be conducted for policy reforms by simulating questions should be specific for each program, the effect of a newly introduced program keeping answers at the individual level if the or parameter-adjusted existing programs. programs are provided to the individual (and In addition, ex post assessments facilitate not to the household). In addition, recording evaluating whether SPL programs are the value of the benefit (or an estimated reaching intended objectives. The availability value) makes performance analysis richer of total household income or consumption in because monetary-based indicators (such as household surveys also enables analysis of benefits incidence, benefit size, and effects the distributional effects of SPL programs and on poverty and inequality) can be estimated. their effects on poverty and inequality. Moreover, different collection formats can be How can household surveys be leveraged explored, such as designing modules specific to become key instruments for monitoring and to social safety net programs and/or placing evaluating social policy? Leite et al. (forthcoming) questions in sector-specific modules, given propose a series of recommendations to improve that SPL programs tend to be multisectorial. Source: Leite et al., forthcoming. 12 The State of Social Safety Nets 2018 Despite data limitations, the global SPL land- “Number of beneficiaries” is simply how scape today is much more accurate than even many people or households benefit from a  few years ago because of advances in the the program identification, capture, and harmonization of Usually, this information is available in the ASPIRE data. This global accumulation of administrative data. The intricacy here is related knowledge is reflected in this book, which builds to what the beneficiary unit is. Many programs on more extensive data (more countries/pro- are targeted at households as a beneficiary unit. grams captured) and adds sophistication of anal- In this case, it is often assumed that all house- ysis relative to the 2015 book on the state of social hold members benefit from the program/bene- safety nets (Honorati, Gentilini, and Yemtsov fit, and hence the number of individual 2015). Furthermore, for the first time, this book beneficiaries is simply the number of individu- attempts to look at SSN programs over time als living in the beneficiary households. For when the data allow intertemporal comparisons. some programs, such as conditional cash trans- fers, very often a subset of the household mem- HOW IS THE PERFORMANCE OF SOCIAL bers is assumed to benefit directly from the SAFETY NET PROGRAMS MEASURED? program (for example, children who get vacci- Performance measurement is the process of col- nations or go to school). In this case, the book lecting, analyzing, and/or reporting information simply uses the number of direct beneficiaries regarding the performance of an individual, provided (through primary and secondary data, group, organization, system, or program compo- as described earlier), without any further calcu- nent. The objective is to determine if the results/ lations. For individual-level benefits (for exam- outputs align with the intention or the intended ple, old-age social pensions), the number of achievement. Performance measurement esti- direct beneficiaries is reported (even though, mates the parameters under which programs are indirectly, all household members may benefit reaching the targeted results. By measuring per- from a household member receiving a benefit). formance, decisions can be made and interven- In any case, in the administrative data, the orig- tions carried out to improve programs. inal beneficiary units are always reported (see The analysis of SSN programs presented in this appendix C). For the household-level benefits, book relies on a number of key terms/parame- the number of recipient households and the ters. These parameters include spending/budget, number of individuals living in those house- number of beneficiaries, coverage, beneficiary/ holds are reported. benefit incidence, benefit size/adequacy, and poverty/inequality impact. This book focuses on “Coverage” indicates the absolute number of the core performance indicators found in the program beneficiaries or percentage of the ASPIRE database. Accordingly, the effects of SSN population or a given population group that programs in such areas as health or education benefits from a given SSN program outcomes, saving behavior, labor supply, fertility, Coverage is important because it indicates and migration are not considered; these effects the  size of the program “blanket” in both can be measured only through rigorous impact absolute and relative terms. In the ideal world, evaluations.8 The main contribution of this book the number of beneficiaries from the adminis- is to present a set of comparable core indicators trative data could be closely matched by cov- for many programs/countries, allowing a global erage (of the same program) from the picture of social safety nets to evolve. household survey data (using population weights). However, as table 1.4 demonstrates, “Spending” indicates the program budget this ideal is elusive. For the purposes of the In most cases, the costs of benefits provided performance analysis (chapter 3), this book account for most spending. While most pro- evaluates coverage relying on the household grams have administrative costs (which are survey data. This approach is taken because the costs of running/implementing the pro- it  would be helpful to know how various gram), those are rarely available and/or can- ­ population groups (for example, poor versus not be separated from the amount spent on ­ nonpoor) are covered by the same program benefits. (that can be found in the household survey). Explaining the Social Safety Net’s Data Landscape 13 This  level of analysis is simply not possible reduction in the measures of welfare (income/ using administrative data. Coverage, in com- consumption) inequality, such as the Gini bination with benefit size/adequacy, is very coefficient. often related to the program’s impact. “Beneficiary/benefit incidence” shows which NOTES segment of the population receives the 1. Appendix C presents information available in the ASPIRE database on the biggest programs (in terms program benefits of numbers of beneficiaries) in 142 countries by The beneficiary/benefit incidence can indicate aggregate program categories. Countries differ sig- what percentage of the total number of benefi- nificantly in the number of SSN programs operating ciaries/total amount of benefits go to the poor- in the country, ranging from fewer than 10–15 (such est quintile of the welfare distribution. The as in Bolivia, Croatia, or Timor-Leste), to more than calculation of this indicator requires the use of 50 programs (such as in Burkina Faso, Chile, or Vietnam). Thus, for some countries with a large household survey data that include the welfare number of programs, appendix C does not present indicator. Moreover, the household survey the full picture of coverage or versatility of programs needs to have the information about the SSN and should be treated with caution. programs for which the benefit incidence is 2. Data availability here refers to the most recent data being assessed. Thus, the data demands are available to the ASPIRE team. In some cases, more very high when it comes to estimating this recent household survey data may be available for a parameter. given country, but these data have not been properly processed yet, or the welfare aggregate has not yet been derived, rendering the data unusable for calcu- “Benefit level” indicates the amount of the lating the performance indicators. benefit, whereas “benefit adequacy” is a 3. To calculate total spending as a percentage of GDP, measure of the relative benefit level program-level spending is divided by GDP using the The main purpose of estimating benefit ade- GDP data from the corresponding year. In this chap- quacy is to get some idea of to what extent the ter and in appendix D, the World Development benefit size is small or large in comparison to a Indicators database (July 2017 version) is used for all benchmark (for example, average income/con- GDP data except for Timor-Leste, which uses the sumption in a country, poverty line, minimum World Economic Outlook database (April 2017 ver- sion). The World Bank income group classification as subsistence level, minimum wage, per capita of July of 2017 is used. GDP). The impact evaluation literature 4. To make clear which categories of spending are pre- (cited later in this book) often finds that frag- sented in chapter 2, the figures and notes to each rel- mented/small benefits fall short of achieving evant figure indicate whether the data cover total desired developmental effects. SSN spending (including health fee waivers) or “core” SSN spending (excluding health fee waivers). This “Poverty/inequality impact” reveals the technique could potentially be used with other cate- distributional effects of the benefit gories, such as educational fee waivers. Regarding poverty impact, two indicators are 5. For the household survey data to be included in the often looked at: percentage reduction in the analysis: (i) household surveys need to be nationally representative; (ii) they need to include information poverty headcount (prevalence) as a result of on social protection; (iii) there is a clearly defined the benefit; and percentage reduction in the welfare aggregate (either income or ­ consumption). poverty depth (distance to the poverty line). On the basis of these criteria, the household surveys The cost–benefit ratio can also be calculated. for Azerbaijan and Lesotho, for example, were not It indicates how much money, in U.S. dollars, used in the analysis. In the case of Azerbaijan, the it costs to reduce a poverty gap by US$1. As survey is a nonrandom sample of the applicants to Targeted Social Assistance; in the case of Lesotho, it empirical evidence around the world sug- is the only country in  the  sample where the asset gests, many SSN benefits help poor people index (rather than ­ consumption or income) is used become less poor (that is, reduce the poverty for welfare rankings. gap/depth) rather than graduate entirely 6. SSN/SA includes eight harmonized program catego- from poverty. Many SSN benefits also often ries, whereas a broader SPL includes 12 harmonized help make societies more equal. This is esti- program categories. See appendix A for further mated empirically by looking at the details. 14 The State of Social Safety Nets 2018 7. See http://siteresources.worldbank.org/ICPEXT​ Impacts of Cash Transfer Programs in Sub-Saharan /­Resources/ICP_2011.html/. Africa.” Food and Agriculture Organization of the Unitd Nations, Rome. 8. Chapter 3 reviews some of these studies on the role of SSN in enhancing productive inclusion. Honorati, M., U. Gentilini, and R. Yemtsov. 2015. The State of Social Safety Nets 2015. Washington, DC: World Bank. REFERENCES Leite, P., C. Rodríguez Alas, and V. Reboul. Forthcoming. ASPIRE (Atlas of Social Protection: Indicators of Resilience “Measuring Social Protection and Labor Programs and Equity). 2017. “Data Sources and Methodology.” through Household Surveys.” Policy Research Working Database, World Bank, Washington, DC. http://data​ Paper, World Bank, Washington, DC. topics.worldbank.org/aspire/~/documentation/. World Bank. 2012. “Resilience, Equity, and Opportunity: Daidone, S., S. Asfaw, B. Davis, S. Handa, and P. Winters. The World Bank Social Protection Strategy 2012–2022.” 2016. “The Household and Individual-Level Economic World Bank, Washington, DC. Explaining the Social Safety Net’s Data Landscape 15 CHAPTER 2 Spending on Social Safety Nets T his chapter aims to answer four main which is part of the SSN system, as well as the questions: How much do countries spend targeted social assistance program. on social safety net (SSN)/social assis- Countries in Sub-Saharan Africa spend tance (SA) programs in relative terms, as a per- around the global average on SSN. However, centage of gross domestic product (GDP), and many programs in the Africa region are donor- in absolute terms?1 Do higher-income coun- funded (see figure 2.2).2 About two-thirds of tries spend more, in relative and absolute terms, the United Nations High Commission on compared to lower-income countries? How has Refugees budget is allocated to programs in SSN spending changed over time? What is the Africa, and this humanitarian assistance is composition of SSN spending in terms of counted as SSN spending.3 The country with the main spending categories and instruments? the highest share of GDP spent on SSN is South Sudan (10 percent of GDP), which has only two HOW MUCH DO REGIONS AND COUNTRIES emergency assistance programs, both of which SPEND ON SOCIAL SAFETY NETS? are fully financed by donors, reflecting the frag- Developing countries spend, on average, ile environment in the country.4 1.5 percent of GDP on SSN programs. Aggregate The Africa region is very heterogenous in spending on SSNs, excluding general price subsi- its  SSN spending. Some of the world’s top dies, was examined for a sample of 124 develop- spenders, such as Lesotho (7 percent of ing countries for which data are available. SSN GDP)  and South Sudan (10 percent), are in spending is higher than the global average in Sub-Saharan Africa; but so are many countries Europe and Central Asia, at 2.2 percent of GDP, that spend very little on SSN as a percentage and about the global average in Sub-Saharan of  GDP. Those include Cameroon, Republic Africa, at 1.5 percent, and in Latin America of  Congo, Côte d’Ivoire, Guinea-Bissau, and the Caribbean, at 1.5 percent. East Asia and Madagascar, São Tomé and Príncipe, Somalia, Pacific, the Middle East and North Africa, and Togo, which spend less than 0.2 percent of and  South Asia spend 1.1 percent, 1.0 percent, GDP on SSN. and 0.9 percent of GDP, respectively (figure 2.1). In the Latin America and Caribbean region, Countries in the Europe and Central Asia the mean SSN spending is 1.5 percent of GDP, region spend on average the highest share of or 1.3 percent, excluding heath fee waivers. The GDP on SSN globally. Georgia, at 7 percent of highest spender is Chile (3.5 percent of GDP), GDP on SSN, spends the most in the region (see whereas the median country spends 1.5 percent appendix B). Spending in this country is driven of GDP (1.1 percent, excluding health fee waiv- by the universal old-age social pension scheme, ers). Guatemala (0.19 percent of GDP) and 16 The State of Social Safety Nets 2018 FIGURE 2.1  Average Global and Regional Spending on Social Safety Nets 2.5 2.2 2.1 2.0 Percentage of GDP 1.53 1.5 1.5 1.54 1.5 1.5 1.3 1.1 1.0 1.0 1.0 0.9 0.9 0.89 0.5 0 Europe and Sub-Saharan Latin America East Asia Middle East South Asia World Central Asia Africa and Caribbean and Pacific and North Africa (n = 7) (n = 124) (n = 27) (n = 45) (n = 18) (n = 17) (n = 10) Social safety net spending Social safety net spending without health fee waivers Source: ASPIRE database. Note: The number of countries in each region appears in parentheses. The difference in the regional average for Africa in this report as opposed to the Africa regional report (Beegle, Coudouel, and Monsalve, forthcoming) is that in the regional report, average social safety net spending (1.3 percent of GDP) does not include South Sudan as an outlier in terms of spending. The regional numbers presented in this figure are simple averages across countries. See appendix B for details. The conceptual treatment of health fee waivers is not straightforward because it depends on how countries arrange and report their provision of health care. Although in some cases the health fee waivers are reported under public health expenditures, in other cases they are counted under social protection expenditures. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. St. Lucia (0.48 percent of GDP) are the lowest North  Africa, South Asia, and Sub-Saharan SSN spenders (see appendix D). Africa, the median country spends almost 1 The East Asia and Pacific region percentage point less, around 0.7–0.8 percent spends  on  average 1 percent of GDP on SSN, of GDP (see table 2.1). but significant variation in spending exists Countries with very high SSN spending across countries. SSN spending ranges from ­ levels are often those that contend with fragil- 0.2 percent of GDP in Lao PDR and 0.3 percent ity, conflict, and violence. For example, Timor- in Myanmar to 2.0 percent in Mongolia and Leste introduced a universal social pension 6.5 percent in Timor-Leste (see appendix D). for war veterans in 2008 as a response to Timor-Leste spends the most on SSN in the violent conflicts in the mid-2000s. In South ­ region (figure 2.3). The median East Asia and Sudan, as mentioned, all SSN spending con- Pacific country spends 0.8 percent of GDP on sists of two large programs financed and SSNs, or 0.7 percent, excluding health fee waiv- implemented by the World Food Programme. ers (see table 2.1). These programs are in-kind and include The median SSN spending across the globe ­ multiple components, such as general food is 1.1 percent of GDP, or 1 percent, excluding distributions, blanket supplementary feeding ­ health fee waivers. The regions form two clus- programs, and targeted supplementary feed- ters in terms of median SSN spending. In Latin ing programs for internally displaced persons America and the Caribbean and in Europe and returnees. and Central Asia, the median country spends Another common explanation for the 1.5–1.9 percent of GDP on SSN, whereas observed high spending levels is the inclusion in  East Asia and Pacific, Middle East and of universal programs in the SSN portfolio in Spending on Social Safety Nets 17 FIGURE 2.2  Share of Donor-Funded Safety percentage of gross domestic product. as  a  ­ Nets in Sub-Saharan African Countries The  data suggest that high-income countries, Central African Republic at 1.9  percent of GDP, and upper-middle-­ Congo, Dem. Rep. income ­ countries, at 1.6 percent of GDP, tend Congo, Rep. to spend only somewhat more than lower- Ethiopia middle-­ income countries, at 1.4 percent of Malawi Somalia GDP, and low-­ income countries, at 1.5 percent South Sudan of GDP. Looking at spending levels excluding Guinea-Bissau health fee waivers, the patterns appear to be Liberia similar. Low-income, lower-middle-income, Uganda Sierra Leone and upper-middle-­ income countries spend on Cameroon average between 1.3 and 1.5 percent of GDP, Gambia, The whereas high-income countries spend on aver- Chad Benin age 1.9  percent of GDP (see figure 2.5 and Zimbabwe table 2.2). Mali The analysis using individual country Tanzania observations suggests that there is no global Mozambique Burkina Faso relationship between a country’s income level Mauritania and SSN spending as a percentage of GDP. In Kenya the Latin America and the Caribbean region, Sudan Ghana spending appears weakly, positively associ- Senegal ated with income levels, whereas in other Seychelles regions, spending is either negatively associ- Angola ated with income levels or has no correlation Botswana Gabon (­ figure 2.6). Globally, it appears that countries Mauritius with the same GDP per capita levels choose Namibia different levels of spending on SSNs reflecting 0 20 40 60 80 100 different policy preferences rather than eco- Percent nomic conditions. Donor-funded share Government share Globally, the median country spends around US$80 (US$66, excluding health fee Source: Beegle, Coudouel, and Monsalve, forthcoming. waivers) in purchasing parity power (PPP) terms annually per person (considering the the countries. For example, Georgia and total population, not just beneficiaries), while Lesotho are among the top spenders because the mean country spends around US$157 their SSN programs include a universal old- (US$150, excluding health fee waivers). As a age minimum social pension. In Georgia, complement to the relative spending analysis, spending of 4.6 percent of GDP on univer- absolute annual (PPP US$) spending per cap- sal  old-age pensions contributes more than ita can more accurately assess actual spending 60  ­percent of total SSN spending. Lesotho on SSNs in a country. For example, in abso- spends 2 percent of GDP on old-age social lute terms per person annually, the Latin ­ pensions (see appendix D). Mongolia also America and the Caribbean countries spend spends significantly more than the regional PPP US$158 (US$139, excluding health fee average because of its universal child benefit, waivers), whereas African countries spend called the  Child Money Program, which PPP US$16 (figures 2.7 and 2.8). Even though accounts for almost 80 percent of total SSN the Africa region is the second-largest spend- spending. ing region in the world in relative terms (per- centage of GDP), in absolute terms it is last DO HIGHER-INCOME COUNTRIES SPEND among the regions (figure 2.9). MORE ON SOCIAL SAFETY NETS? The absolute benefit level per household also Globally, country income levels appear to differs significantly across country income be  weakly associated with SSN spending groups. In a subsample of 36 countries that 18 The State of Social Safety Nets 2018 FIGURE 2.3  Social Safety Net Spending Variations across Countries and Regions: East Asia and Pacific, Latin America and the Caribbean, and Europe and Central Asia Papua New Guinea Lao PDR Myanmar Regional mean,1.1 East Asia and Pacific (n = 17) Vanuatu Thailand Global mean, 1.5 Philippines Kiribati Malaysia Samoa China Indonesia Cambodia Vietnam Marshall Islands Fiji Mongolia Timor-Leste Lithuania Tajikistan Latvia Azerbaijan Romania Macedonia, FYR Regional mean, 2.2 Moldova Turkey Europe and Central Asia (n = 27) Armenia Bulgaria Albania Kazakhstan Montenegro Russian Federation Serbia Poland Slovak Republic Estonia Slovenia Kosovo Belarus Hungary Kyrgyz Republic Croatia Bosnia and Herzegovina Ukraine Georgia Guatemala St. Lucia Latin America and Caribbean (n = 18) Costa Rica Honduras El Salvador Regional mean, 1.5 Uruguay Dominican Republic Brazil Peru Ecuador Panama Mexico Grenada Argentina Bolivia Nicaragua Colombia Chile 0 1 2 3 4 5 6 7 8 Spending on SSN, percentage of GDP Regional mean Global average Spending on SSN without health fee waivers, percentage of GDP Source: ASPIRE database. Note: Based on the most recent spending data available between 2010 and 2016 (except for the following four countries, for which only total spending data are available for years before 2010: Bhutan, Jordan, Marshall Islands, and Vanuatu). See appendix D for details. The number of countries in each region appears in parentheses. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. Spending on Social Safety Nets 19 FIGURE 2.4  Social Safety Net Spending Variations across Countries and Regions: Africa, Middle East and North Africa, and South Asia Egypt, Arab Rep. North Africa (n = 10) Djibouti Regional mean, 1 Middle East and Jordan Saudi Arabia Tunisia Kuwait Lebanon Morocco West Bank and Gaza Iraq Bhutan Pakistan South Asia Sri Lanka (n = 7) Bangladesh Maldives Regional mean, 0.9 Nepal India São Tomé and Príncipe Côte d'Ivoire Guinea-Bissau Cameroon Congo, Rep. Madagascar Somalia Regional mean, 1.5 Togo Gabon Zambia Nigeria Kenya Zimbabwe Tanzania Ghana Mali Niger Sub-Saharan Africa (n = 45) Comoros Chad Congo, Dem. Rep. Uganda Sierra Leone Ethiopia Senegal Sudan Mozambique Malawi Rwanda Guinea Botswana Swaziland Burkina Faso Burundi Angola Mauritania Cabo Verde Seychelles Liberia Central African Republic Benin Namibia South Africa Mauritius Lesotho South Sudan 0 1 2 3 4 5 6 7 8 Spending on social safety nets, percentage of GDP Regional mean Global average Spending on social safety nets without health fee waivers, percentage of GDP Source: ASPIRE database. Note: Based on the most recent spending data available between 2010 and 2016 (except for the following four countries, for which only total spending data are available for years before 2010: Bhutan, Jordan, Marshall Islands, and Vanuatu). See appendix D for details. The number of countries in each region appears in parentheses. The scale is restricted for convenience; the true value for South Sudan is 10.1 percent. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. 20 The State of Social Safety Nets 2018 TABLE 2.1  Social Safety Net Spending across and within Regions Percentage of GDP Region Spending on SSNs Mean Median Minimum Maximum Europe and Central Asia (n = 27) Total SSN 2.2 1.9 0.5 7.0 Excluding health fee waivers 2.1 1.9 0.5 5.7 Sub-Saharan Africa (n = 45) Total SSN 1.5 1.0 .. 10.1 Excluding health fee waivers 1.5 0.9 .. 10.1 Latin America and the Caribbean (n = 18) Total SSN 1.5 1.5 0.2 3.5 Excluding health fee waivers 1.3 1.1 0.2 3.5 East Asia and Pacific (n = 17) Total SSN 1.1 0.8 .. 6.5 Excluding health fee waivers 1.0 0.7 .. 6.5 Middle East and North Africa (n = 10) Total SSN 1.0 0.8 0.2 2.6 Excluding health fee waivers 0.9 0.7 0.2 2.6 South Asia (n = 7) Total SSN 0.9 0.7 0.3 1.5 Excluding health fee waivers 0.9 0.7 0.3 1.5 Source: ASPIRE database. Note: See appendix D for details. The number of countries in each region appears in parentheses. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; SSN = social safety net; .. = values below 0.01. FIGURE 2.5  Social Safety Net Spending across Country Income Groups versus the OECD 3.0 2.5 2.7 Average percentage of GDP 2.0 1.9 1.9 1.6 1.5 1.54 1.5 1.4 1.4 1.5 1.5 1.3 1.0 0.5 0 Low-income Lower-middle- Upper-middle- High-income World countries income countries income countries countries (n = 124) (n = 26) (n = 48) (n = 38) (n = 12) Social safety net spending Social safety net spending without health fee waivers OECD Source: ASPIRE database. Note: The number of countries in each country income group appears in parentheses. High-income countries included in the analysis are Chile, Estonia, Hungary, Kuwait, Latvia, Lithuania, Poland, Saudi Arabia, Seychelles, Slovak Republic, Slovenia, and Uruguay. Data for OECD countries refer to 2013 and are based on the Social Expenditure Database. Social safety net spending for OECD countries here is approximated by the sum of the “family” and “other social policy” social protection functions, as defined in the Social Expenditure Database. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; OECD = Organisation for Economic Co-operation and Development. have flagship (main) programs with the house- ­ountries—PPP US$106 versus PPP US$27, c hold as a beneficiary unit (see appendix E for respectively (figure 2.10). details), the benefit amount (in PPP US$) per Median values of the monthly transfer for household is four time greater in upper-­ these large programs illustrate similar disper- middle-income countries than in low-income sion across country income groups. Median Spending on Social Safety Nets 21 TABLE 2.2  Variations in Social Safety Net Spending across Country Income Groups Percentage of GDP Region Spending on SSNs Mean Minimum Maximum Median Low-income countries (n = 26) Total SSN 1.5 .. 10.1 1.0 Without health fee waivers 1.4 .. 10.1 0.9 Lower-middle-income countries (n = 48) Total SSN 1.4 .. 7.1 0.8 Without health fee waivers 1.3 .. 7.1 0.7 Upper-middle-income countries (n = 38) Total SSN 1.6 0.2 3.9 1.4 Without health fee waivers 1.5 0.1 3.9 1.2 High-income countries (n = 12) Total SSN 1.9 0.5 3.5 2.0 Without health fee waivers 1.9 0.5 3.5 2.0 OECD countries (n = 34) Family and other social protection areas 2.7 0.4 5.0 2.7 Source: ASPIRE database. Note: The number of countries in each country income group appears in parentheses. High-income countries included in the analysis are Chile, Estonia, Hungary, Kuwait, Latvia, Lithuania, Poland, Saudi Arabia, Seychelles, Slovak Republic, Slovenia, and Uruguay. Data for OECD countries refer to 2013 and are based on the Social Expenditure Database. SSN spending for OECD countries here is approximated by the sum of the “family” and “other social policy” social protection functions, as defined in the Social Expenditure Database. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; OECD = Organisation for Economic Co-operation and Development; SSN = social safety net; .. = values below 0.01. FIGURE 2.6  Total Social Safety Net Spending and Income Levels across Regions 12 10 Percentage of GDP 8 6 4 2 0 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 Log (GDP per capita) East Asia and Pacific Europe and Central Asia Latin America and Caribbean Middle East and North Africa South Asia Sub-Saharan Africa Linear (all countries) Sources: ASPIRE database; World Development Indicators for GDP per capita, PPP US$. Note: ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; PPP = purchasing power parity. 22 The State of Social Safety Nets 2018 FIGURE 2.7  Absolute Annual Spending on Social Safety Nets per Capita across Countries and Regions: East Asia and Pacific, Europe and Central Asia, and Latin America and the Caribbean Papua New Guinea Lao PDR Vanuatu East Asia and Pacific (n = 17) Kiribati Regional median, 43 Myanmar Cambodia Marshall Islands Samoa Philippines Vietnam Global median, 79 Thailand Indonesia China Fiji Timor-Leste Malaysia Mongolia Tajikistan Moldova Kyrgyz Republic Armenia Regional median, 253 Lithuania Azerbaijian Macedonia, Fyr Europe and Central Asia (n = 27) Albania Latvia Turkey Romania Blugaria Montenegro Serbia Kosovo Bosnia And Herzegovina Kazakhstan Ukraine Russian Federation Belarus Poland Georgia Slovak Republic Croatia Slovenia Hungary Estonia Guatemala Honduras Latin America and Caribbean (n = 18) St. Lucia El Salvador Nicaragua Costa Rica Regional median, 158 Bolivia Ecuador Dominican Republic Peru Brazil Uruguay Mexico Grenada Argentina Panama Colombia Chile 0 100 200 300 400 500 600 700 800 U.S. dollars Source: ASPIRE database. Note: Values are converted to constant 2011 prices using the PPP and CPI from the World Development Indicators. Also, 2011 is used as the base year value to calculate the CPI ratio, as deflator, between the observed year and 2011 for all sample countries. Then it is divided first by the CPI ratio and then by the 2011 PPP value to obtain the constant 2011 PPP US$. In cases where CPI series are not available from the World Development Indicators, the GDP deflator is used as a proxy for deflation, particularly for Argentina and Belarus. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; CPI = consumer price index; PPP = purchasing power parity. Spending on Social Safety Nets 23 FIGURE 2.8  Absolute Annual Spending on Social Safety Nets per Capita across Countries, Economies, and Regions: Middle East and North Africa, Sub-Saharan Africa, and South Asia Djibouti North Africa (n = 10) Egypt, Arab Rep. Middle East and Jordan Tunisia Morocco West Bank and Gaza Lebanon Saudi Arabia Iraq Regional median, 93 Kuwait Bhutan Bangladesh Regional median, 29 South Asia Pakistan (n = 7) Nepal Sri Lanka India Maldives São Tomé and Príncipe Guinea-Bissau Gobal median, 79 Cote d'Ivoire Cameroon Congo, Rep. Somalia Madagascar Togo Congo, Dem. Rep. Regional median, 16 Niger Zimbabwe Zambia Chad Tanzania Comoros Kenya Central African Republic Sub-Saharan Africa (n = 45) Mali Sierra Leone Nigeria Uganda Mozambique Malawi Ethiopia Guinea Burudi Liberia Rwanda Ghana Mauritania Gabon Burkina Faso Sudan Benin Regional median, 16 South Sudan Angola Swaziland Cabo Verde Lesotho Botswana Namibia South Africa Mauritius Seychelles 0 100 200 300 400 500 600 700 800 Source: ASPIRE database. Note: Values are converted to constant 2011 prices using the PPP and CPI from the World Development Indicators. Also, 2011 is used as the base year value to calculate the CPI ratio, as deflator, between the observed year and 2011 for all sample countries. Then it is divided first by the CPI ratio and then by the 2011 PPP value to obtain the constant 2011 PPP US$. In cases where CPI series are not available from the World Development Indicators, the GDP deflator is used as a proxy for deflation, particularly for Argentina and Belarus. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; CPI = consumer price index; PPP = purchasing power parity. 24 The State of Social Safety Nets 2018 FIGURE 2.9  Regional Median Annual Social Safety Net Spending per Capita 300 253 253 250 Annual in 2011 200 US$ PPP 158 150 139 93 80 66 100 69 43 40 50 16 15 29 29 0 Sub-Saharan South Asia East Asia Middle East Latin America Europe and World Africa (n = 45) (n = 7) and Pacific and North and Caribbean Central Asia (n = 124) (n = 17) Africa (n = 10) (n = 18) (n = 27) Median SSN spending (total) per capita Median SSN spending (without health fee waivers) per capita Source: ASPIRE database. Note: The number of countries in each region appears in parentheses. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; PPP = purchasing power parity; SSN = social safety net. FIGURE 2.10  Transfer Amount for Cash Transfer Programs, by Income Group 120 106 100 Monthly in 2011 79 80 US$ PPP 63 60 40 27 20 0 Low-income Lower-middle-income Upper-middle-income World countries (n = 8) countries (n = 15) countries (n = 13) (n = 36) Mean transfer amount Median transfer amount Source: ASPIRE database (see appendix E for details). Note: The number of countries (one program per country) appears in parentheses. The largest, or flagship, cash transfer program is selected per country. See the full list of selected programs in appendix E. Transfer amount values (as designed) are converted to constant 2011 prices using the PPP and CPI from the World Development Indicators. Also, 2011 is used as the base year value to calculate the CPI ratio, as deflator, between the observed year and 2011 for all sample countries. Then it is divided first by the CPI ratio and then by the 2011 PPP value to obtain constant 2011 PPP US$. In cases where CPI series are not available from the World Development Indicators, the GDP deflator is used as a proxy for deflation, particularly for Argentina and Belarus. High-income countries are excluded from this analysis because of a small sample. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; CPI = consumer price index; PPP = purchasing power parity. transfer amounts in low-income countries and HOW HAS SPENDING CHANGED lower-middle-income countries do not differ OVER TIME? significantly (averaging about PPP US$30). In general, SSN spending fluctuates a lot over However, the median upper-middle-income time in some countries, while it remains rela- country provides more than three times the tively stable in others. This section largely focuses median benefits of low-income countries and on time trends in SSN spending in the Latin lower-middle-­ income countries (a little less America and the Caribbean and the Europe and than PPP US$100), as figure 2.10 indicates. Central Asia regions because the other regions Spending on Social Safety Nets 25 lack consistent spending data for 10 years or Kazakhstan, Latvia, Macedonia, Montenegro, more. Hence, the findings reflect only these two Poland, Romania, Serbia, Turkey, and Ukraine) regions and do not represent global trends. with balanced panel time-series spending on However, the expansion in coverage and spend- SSN.5 Their total population represents about ing is also illustrated for many large (flagship) 60 percent of the Europe and Central Asia programs globally. countries.6 The analysis suggests that in this In Latin America and the Caribbean, social group of countries, average spending rose spending as a percentage of GDP increased sub- steadily, from 1.2 to 1.8 percent of GDP stantially over the past decade (2005–15). This from  2003 to 2009, and then fell slightly, to book analyzed a subsample of seven countries percent in 2014. Before the financial cri- 1.6  ­ in the region (Argentina, Brazil, Colombia, sis,  the region seems to have reached a Ecuador, Mexico, Peru, and Uruguay) with bal- steady  level of SSN spending; then spend- anced panel time-series spending on SSN. Their ing  grew in response to the financial crisis; total population represents about 75 percent of and now it is converging to the prior level (see the total Latin American and Caribbean popu- figure 2.12). lation. The analysis suggests that in this group of Many countries in Sub-Saharan Africa and Asia countries, average SSN spending increased from are introducing flagship SSN programs and are 0.43 to 1.26 percent of GDP from 2003 to 2015 rapidly expanding coverage. However, these initia- (see figure 2.11). The increase in SSN spending tives come at a fiscal cost. In Tanzania, accelerated around the time of the 2008 finan- the  Productive Safety Net Program expanded cial crisis, despite a reduction in the rate of eco- from 0.4 to 10 percent of the population from its nomic growth. Argentina and Peru show the launch in 2013 to 2016 (figure 2.13, panel a). This highest relative spending increases since 2009. coverage expansion was accompanied by a rapid In Europe and Central Asia, the increase increase in program spending, from 0.03 to almost in  social spending over a similar period 0.3 percent of GDP in two years. In Senegal, the was ­ moderate. This book analyzed a subsam- National Cash Transfer Program expanded from 3 ple of 15 countries in the region (Albania, to 16  percent of the population in four years Armenia, Azerbaijan, Belarus, Bulgaria, Estonia, ­(figure 2.13, panel b). The corresponding program FIGURE 2.11  Trends in Social Safety Net Spending in Latin America and the Caribbean 1.4 1.26 US$ trillion (constant 2010 US$) Spending, percentage of GDP 6 1.2 1.0 4 0.8 0.6 0.43 0.4 2 0.2 0 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 SSN spending, seven-country average (percentage of GDP) GDP (constant 2010 US$) Source: ASPIRE database. Note: GDP in Latin America and the Caribbean constitutes member countries of the International Development Association and International Bank for Reconstruction and Development. A balanced panel of seven countries (Argentina, Brazil, Colombia, Ecuador, Mexico, Peru, and Uruguay) is used. The average social safety net spending in Latin America and the Caribbean before 2010 should be interpreted with caution because data availability was more problematic, particularly for program-based disaggregated data up to 2009. Social safety net spending excludes health fee waivers. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; GDP = gross domestic product; SSN = social safety net. 26 The State of Social Safety Nets 2018 FIGURE 2.12  Trends in Social Safety Net Spending in Europe and Central Asia, 2003–14 2.0 1.80 US$ trillion (coonstant 2010 US$) 1.8 6 1.63 Spending, percentage of GDP 1.6 1.24 1.4 1.2 4 1.0 0.8 0.6 2 0.4 0.2 0 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 SSN spending, Europe and Central Asia average GDP (constant 2010 US$) Source: ASPIRE database. Note: GDP in Europe and Central Asia constitutes International Development Association and International Bank for Reconstruction and Development countries. Social safety net spending data do not include a data point for Poland in 2003 or for Montenegro, Poland, Serbia, and Turkey in 2014. The averages for these years should be interpreted with caution. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; SSN = social safety net. spending increased from 0.05 to 0.2 percent of and Central Asia has the largest cash transfer GDP during 2013–15. In Indonesia, the Program budget share among regions (with cash transfers Keluarga Harapan increased its coverage from 1 to consisting of unconditional and conditional cash 9  percent of the population between 2008 and transfers and social pensions). Cash transfers in 2016,  and the respective budget also increased Europe and Central Asia account for 76 percent ­(figure 2.13, panel c). In the Philippines, the flag- of the total SSN spending portfolio. At the same ship conditional cash transfer program called 4Ps time, the Middle East and North Africa coun- increased its coverage from 4 to 20 percent of tries, on average, allocate just over 40 percent of the  population between 2008 and 2015, and their budget to cash transfers (see figure 2.14). the  respective budget increased from 0.1 to The Latin America and Caribbean region 0.5  percent of GDP ­ (figure  2.13, panel d). The has the largest conditional cash transfer budget global inventory of the biggest SSN programs (by share. The region spends around 21 percent of its category) per country can be found in appendix C. total SSN budget on this instrument. However, the Latin America and Caribbean region is not WHICH SOCIAL SAFETY NET alone in its substantial reliance on conditional INSTRUMENTS DO COUNTRIES FUND? cash transfers. It is followed closely by Sub- Beyond the heterogeneity in total spending, Saharan Africa, where conditional cash transfers countries and regions also differ in their prefer- account for around 18 percent of the SSN budget. ences for various SSN instruments. The legacy East Asia and Pacific spends 12 percent of GDP of SSNs, cultural differences, demographic con- on conditional cash transfers (figure 2.14). ditions, the socioeconomic context, political The public works spending budget share is will, and other factors shape the structure of the highest in South Asia, where this type of countries’ SSN portfolios. Figure 2.14 maps the program is commonly implemented. South distribution of SSN budgets across different Asia spends 25 percent of its SSN budget on program types, by region. public works. In South Asia, Bangladesh and The analysis suggests that cash transfers take India spend the highest share (see up more than half of all SSN spending. Europe appendix  D). Sub-Saharan Africa spends on ­ Spending on Social Safety Nets 27 average 12 percent of the SSN budget on pub- Caribbean (9 percent). The spending on in-kind lic works. In Sub-Saharan Africa, Burundi, transfers in the Middle East and North Africa is Central African Republic, Ethiopia, and driven by such countries and economies as West Liberia spend the highest share of GDP on Bank and Gaza and Djibouti, where emergency public works (see appendix D). and fragile context leads to in-kind interventions In-kind transfers account for a significant (mostly donor funded). In Iraq, the spending on share of SSN spending in a number of regions. food rations accounts for more than 85 percent These include Middle East and North Africa of the total SSN spending.7 In South Asia, India’s (18  percent), Africa (11 percent), South Asia Public Distribution System program costs more (10  percent), and Latin America and the than 1 percent of GDP (see  appendix D) and FIGURE 2.13  Expansion of Flagship Cash Transfer Programs in Tanzania, Senegal, the Philippines, and Indonesia a. Tanzania, Productive Social Safety Net (CCT component) 1,200 12 1,000 10% 10 Percentage of beneficiaries Beneficiaries, thousands 10% in the total population 800 8 of households 600 6 Spending on PSSN = Spending 0.03% of GDP on PSSN = 400 4 2% 0.3% of GDP 200 2 0.4% 0 0 2013 2014 2015 2016 b. Senegal, National Cash Transfer Program (NCTP) 350 18 16% 16 300 14 Percentage of beneficiaries Beneficiaries, thousands 250 in the total population 10% 12 of households 200 10 Spending on 150 NCTP = 0.05% 5% 8 of GDP Spending on 6 100 NCTP = 0.02% 3% of GDP 4 50 2 0 0 2013 2014 2015 2016 (figure continues next page) 28 The State of Social Safety Nets 2018 FIGURE 2.13  Expansion of Flagship Cash Transfer Programs in Tanzania, Senegal, the Philippines, and Indonesia (Continued) c. Indonesia, Program Keluarga Harapan (PKH) 7,000 10 9% 9 6,000 8 Percentage of beneficiaries Beneficiarles, thousands 5,000 in the total population 7 5% of households 6 4,000 4% 5 4% 3,000 Spending on PKH 4 = 0.2% of GDP 2,000 Spending on 3 1% 2% PKH = 0.5% of GDP 2 1,000 1 0 0 11 12 13 14 15 16 10 08 09 20 20 20 20 20 20 20 20 20 d. Philippines, Pantawid Pamilyang Program (4Ps) 5,000 25 20% 4,500 19% 19% 4,000 20 Percentage of beneficiaries Beneficiaries, thousands 3,500 in the total population of households 3,000 15 2,500 Spending on 4Ps = 2,000 0.1% of GDP 10 Spending on 1,500 4Ps = 4% 0.5% of GDP 1,000 5 500 0 0 2009 2010 2011 2012 2013 2014 2015 Source: ASPIRE database. Note: Data for Tanzania include Zanzibar. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; CCT = conditional cash transfer. contributes almost 70  percent of the total SSN expand across space and time. The next chapter budget captured in administrative data. Regions explores the performance of SSN programs allocate between 4 and 9 percent to school feed- around the world. It looks at what countries ing programs (see figure 2.14). achieve in terms of coverage, benefit incidence, As this chapter has illustrated, SSN programs and poverty/inequality impact for the SSN bud- take many forms, and their budgets tend to get they spend. Spending on Social Safety Nets 29 FIGURE 2.14  Social Safety Net Spending across Regions, by Instrument Latin America and Caribbean 0.13 0.19 0.21 0.12 Sub-Saharan Africa 0.15 0.18 0.18 0.13 East Asia and Pacific 0.18 0.24 0.12 0.23 Middle East and North Africa 0.24 0.13 0.07 0.12 South Asia 0.26 0.19 0.05 0.04 Europe and Central Asia 0.36 0.30 0.10 0.06 0 10 20 30 40 50 60 70 80 90 100 Percent UCT Social pension CCT School feeding Public works In kind Fee waivers Other SA (excluding health) Source: ASPIRE database. Note: This figure shows estimates based on a sample of 112 countries with program-level data disaggregation available, as presented in appendix D. For comparability, health fee waivers are dropped from total spending and from the fee waivers category, which comprises educational fee waivers and utility fee waivers only. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; CCT = conditional cash transfer; SA = social assistance; UCT = unconditional cash transfer. NOTES 6. Those exclude high-income Organisation for 1. This chapter focuses on social safety nets only, as a sub- Economic Co-operation and Development (OECD) set of social protection and labor market programs. countries. 2. “Sub-Saharan Africa” and “Africa” are used inter- 7. The spending reference year for Iraq is 2012. changeably in this book. 3. See http://reporting.unhcr.org/sites/default/files/gr2016​ /pdf/02_Funding.pdf. REFERENCES 4. See http://pubdocs.worldbank.org/en/154851467143​ ASPIRE (Atlas of Social Protection: Indicators of Resilience 896227/FY17HLFS-Final-6272016.pdf. and Equity). 2017. Database, World Bank, Washington, DC. http://datatopics.worldbank.org/aspire/. 5. SSN spending data do not include a data point for Poland in 2003 or for Montenegro, Poland, Serbia, Beegle, K., A. Coudouel, and E. Monsalve,  eds. and Turkey in 2014. The averages for these years Forthcoming. Realizing the Full Potential of Social should be interpreted with caution. Safety Nets in Africa. Washington, DC World Bank. 30 The State of Social Safety Nets 2018 CHAPTER 3 Analyzing the Performance of Social Safety Net Programs T his chapter analyzes the performance of not replace results from impact evaluations. social safety net (SSN)/social assistance Performance indicators from household s ­ urvey (SA) programs using four indicators: data help monitor SSN programs’ performance coverage, beneficiary incidence, level of bene- over time as information from household sur- fits, and impact on poverty and inequality. To veys becomes available. Given that significant set the stage, the first section of this chapter resources are being invested by governments in presents an overview of the coverage of social the implementation of SSN programs, it is protection and labor (SPL) programs in important to continuously monitor the effec- general that encompasses SSN/SA programs, tiveness of these programs and inform social social insurance, and labor market programs. policy. Within this framework, this chapter then Performance indicators derived from analyzes the role of SSN programs as a main ­ household surveys assess the effect of the SPL instrument. Performance indicators pro- transfers on the welfare of beneficiaries (in vide answers to very important development terms of income or consumption) and their questions. Coverage indicates what percentage distributional effects. Impact evaluations, on of the total population or a specific population the other hand, are designed to measure a group benefits from SSN programs. Among all broader specific set of outcomes, such as the possible population groups, the poor have the effects on beneficiaries’ level of consumption, greatest need for social protection and are par- production, labor supply, human capital, and ticularly the focus of SSNs aimed at assisting risk management (see highlight 1 at the end the poor. Beneficiary incidence sheds light on of this chapter). The effects attributed to the how the total number of beneficiaries are dis- program are evaluated by using a counterfac- tributed along the welfare distribution of the tual to determine the potential outcomes for population. The level of benefits indicates the beneficiaries in the absence of the pro- the proportion of the benefits with respect to gram. However, impact evaluations are not the household’s total income or consumption. conducted very frequently and are not avail- Impact on poverty and inequality shows the able for all programs; accordingly, household reduction in the poverty headcount, poverty surveys play a crucial role in monitoring gap, and inequality (as measured by the Gini ­ programs, given that they are systematically index) because of the SPL transfers. conducted across years and may include a The performance indicators presented in larger set of programs for which impact eval- this chapter represent only a first step at uations are not available. Therefore, perfor- ­ monitoring SSN program performance and do mance indicators from household surveys Analyzing the Performance of Social Safety Net Programs 31 and impact evaluations provide a comple- Middle East North Africa (see figure 3.1).1 In mentary picture of what social protection terms of coverage of the poorest, Europe and programs are achieving. Central Asia has the highest coverage of indi- This chapter first presents evidence on cov- viduals in the poorest quintile (86 percent), fol- erage, then on beneficiary incidence, followed lowed by Latin American and the Caribbean by benefit size, and impacts on poverty and (76 percent). In all regions, a higher percentage inequality. In addition, a highlight focuses on of the poorest quintile (compared to the total productive outcomes of SSN programs in population) is covered. Sub-Saharan Africa based on impact evalua- The coverage rates presented in the first tions. The analysis uses a subset of the most three figures (see figures 3.1–3.3) are particularly recent household surveys (2008–16) from the high because they include all types of SPL pro- Atlas of Social Protection: Indicators of grams, not only SSN/SA programs. In other Resilience and Equity (ASPIRE) database, words, the figures include any type of social corresponding to 96 countries with informa- insurance (old-age pension and other social tion about SSN programs. The full sample of security), active and passive labor market pro- countries is used for coverage and beneficiary grams, and SSNs (unconditional cash transfers incidence indicators; however, for assessing [UCTs], conditional cash transfers [CCTs], social benefits level and impacts on poverty and pensions, public works, fee waivers and tar- inequality, only 79 countries are used, for geted  subsidies, school feeding, in-kind trans- which the monetary value of the transfers is fers, and other SA programs).2 In addition, other provided in the household surveys. This methodological factors drive this high coverage. chapter also provides country-level key per- ­ Specifically, the calculation includes direct and formance indicators of SPL programs in indirect beneficiaries, and the poorest quintile is appendix F.1 and key performance indicators estimated using pre-­ transfer welfare. solely for SSN/SA programs in appendix F.2. In a sample of countries capturing all coun- try income groups, the average SPL coverage WHO IS COVERED BY SOCIAL rate of the poorest quintile is 56 percent. PROTECTION AND LABOR PROGRAMS? The  coverage of SPL programs is highly Coverage is expressed as the percentage of the ­correlated with the countries’ level of income. population receiving a given type of SPL pro- Figure 3.2 shows that high- and upper-­middle- gram. In this analysis, coverage includes direct income countries cover 97 percent and 77 per- and indirect beneficiaries (all household mem- cent of the poorest quintile, respectively. In bers where at least one member receives a bene- contrast, lower-middle- and low-income coun- fit). The analysis first presents the global picture tries cover 54 and 19 percent of the poorest of SPL coverage by region and country income quintile, respectively. These coverage figures group, and then zooms in on country-level should be interpreted with caution because coverage rates by the type of SSN program. coverage rates derived from household surveys In discussing the coverage of the poor, the poor are likely to be underestimated.3 As a response are defined as individuals who belong to the to observed coverage gaps for the poor, such bottom 20 percent of the welfare distribution initiatives as universal social protection (USP) (in terms of household total income or con- have emerged (see box 3.1). sumption per capita). In all subsequent figures In terms of the coverage of the poor, low-­ and in appendix F, the pretransfer welfare indi- income countries lag in all three areas of social cator is used to rank households, except for the protection. Figure 3.3 shows that social indicator that expresses the social transfers as a insurance programs are more prevalent in share of total beneficiary welfare, which includes high-income countries, covering 60 percent of transfers. the poorest quintile; in contrast, in low-­income The analysis reveals that SPL programs cover countries only 2 percent of the poorest quintile on average 44 percent of the total population. is covered by this program type. SSN/SA pro- SPL programs cover more than half the popula- grams account for most SPL program cover- tion in East Asia and Pacific, Europe and Central age of the poor in all country income groups. Asia, Latin American and the Caribbean, and Yet, high-income countries report the highest 32 The State of Social Safety Nets 2018 FIGURE 3.1  Share of Total Population and the Poorest Quintile That Receives Any Social Protection and Labor Programs, as Captured in Household Surveys, by Region 90 86 80 76 70 64 65 60 60 56 55 51 Percent 50 44 43 40 30 28 29 24 21 20 10 0 World South Asia Sub-Saharan Middle East Asia Latin America Europe (n = 96) (n = 8) Africa East and and and the and (n = 31) North Africa Pacific Caribbean Central Asia (n = 7) (n = 10) (n = 20) (n = 20) Total population Poorest quintile Source: ASPIRE database. Note: The total number of countries per region included in the analysis appears in parentheses. Aggregated indicators are calculated using simple averages of country-level social protection and labor coverage rates across regions. Coverage is determined as follows: (number of individuals in the total population or poorest quintile who live in a household where at least one member receives the transfer)/(number of individuals in the total population). This figure underestimates total social protection and labor coverage because household surveys do not include all programs that exist in each country. The poorest quintile is calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. FIGURE 3.2  Share of Total Population and the Poorest Quintile That Receives Any Social Protection and Labor Programs, as Captured in Household Surveys, by Country Income Group 100 97 90 81 80 77 70 60 56 58 54 Percent 50 44 41 40 30 18 19 20 10 0 World Low-income Lower-middle- Upper-middle- High-income (n = 96) countries income countries income countries countries (n = 22) (n = 37) (n = 31) (n = 6) Total population Poorest quintile Source: ASPIRE database. Note: The total number of countries per country income group included in the analysis appears in parentheses. Aggregated indicators are calculated using simple averages of country-level social protection and labor coverage rates across country income groups. Coverage is determined as follows: (number of individuals in the total population or poorest quintile who live in a household where at least one member receives the transfer)/(number of individuals in the total population). This figure underestimates total social protection and labor coverage because household surveys do not include all programs that exist in each country. The poorest quintile is calculated using per capita pre- transfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. Analyzing the Performance of Social Safety Net Programs 33 BOX 3.1  Universal Social Protection The Universal Social Protection (USP) Initiative to make disability, maternity/paternity, and/ was launched by the World Bank Group, the or child benefits universal (Argentina, Nepal). International Labour Organisation, and other There are publicly financed child benefit social bilateral and multilateral partners in September pensions for all (Mongolia, Namibia) and min- 2016 to support the Sustainable  Development imum pensions for those who do not have  a Goal (SDG) agenda for social protection.a The contributory pension, ensuring universality USP Initiative aims to join the efforts of the (Azerbaijan, China). Some countries strategi- international agencies, donors, and govern- cally use transfers for the poor and vulnerable ments in providing social protection for all peo- who could fall further behind (Brazil, Chile, Fiji, ple in need. Access to adequate social protection and Georgia). is central to ending poverty and boosting shared The implementation of the USP Framework prosperity. The poorest countries continue to emphasizes both depth and breadth of cover- have enormous coverage/adequacy gaps, as age, or vertical and horizontal expansion. The the empirical evidence presented in this book depth of coverage is defined as areas of pro- clearly suggests. tection and can include income security, access Countries have many options and pathways to insurance and saving instruments, access to to achieve universal social protection. Some essential health care services, and other social countries have opted for an explicit univer- services or levels of support (or adequacy). sal coverage of specific population groups Expansion in the vertical sense means pro- (Botswana, Timor-Leste), whereas others have viding more protection to the same covered used a more gradual and progressive approach groups. In terms of horizontal expansion, there to building up coverage (Brazil, Thailand). Some are different population groups with respect countries have the principles of universalism to the stage in the life cycle (children, working (universal rights) embedded in their national age, and elderly) or level of income (poor, vul- constitutions (Bolivia, South Africa), whereas nerable, middle class, rich). The evidence sug- others have pursued those principles without gests that countries tend to gradually expand constitutional provisions (Swaziland, Uruguay). coverage both vertically and horizontally. The Universal social protection is most commonly degree of coverage of the poor is highly cor- started with (universal) old-age pensions (see related with the degree of coverage of the gen- chapter  4), but some countries have opted eral population. a. See http://www.ilo.org/global/topics/social-security/WCMS_378991/lang--en/index.htm. coverage of the poor by SSN programs (76 per- and target different population groups based cent), compared with only 18 percent in on needs and vulnerabilities. Countries gener- low-income countries. Labor market programs ally adopt a combination of SSN/SA programs cover the poor at a rate of 2 percent in low-­ based on their social policy objectives. This income countries and 8 percent in high-­ section analyzes the extent to which different income countries.4 SSN programs therefore SSN/SA programs cover individuals in the play a pivotal role in achieving social protec- poorest quintile. The analysis details the num- tion coverage of the poor.5 The rest of this ber of countries reporting each program typol- chapter focuses on analyzing the performance ogy in the 96 household surveys included in of SSN instruments in the countries included the ASPIRE database and presents the cover- in the ASPIRE database. age of the poor (direct and indirect beneficia- ries) by these instruments. WHICH TYPES OF SOCIAL SAFETY NET To facilitate analysis and cross-country PROGRAMS COVER THE POOR? ­ comparisons, the programs are grouped into Different countries focus on different SSN eight standard SSN categories. Therefore, instruments. There is no one-size-fits-all the  coverage indicator corresponds to the approach to SSN/SA programs. These noncon- aggregated program category and not neces- tributory programs address different issues sarily to an individual program. For example, 34 The State of Social Safety Nets 2018 FIGURE 3.3  Share of Poorest Quintile That Receives Any Social Protection and Labor Program, as Captured in Household Surveys, by Type of Social Protection and Labor Area and Country Income Group 90 80 76 70 62 60 60 50 Percent 45 43 40 30 28 24 23 20 18 10 6 7 6 8 2 2 0 World Low-income Lower-middle- Upper-middle- High-income (n = 96) countries income countries income countries countries (n = 22) (n = 37) (n = 31) (n = 6) All labor market All social insurance All social assistance/social safety nets Source: ASPIRE database. Note: The total number of countries per country income group included in the analysis appears in parentheses. Aggregated indicators are calculated using simple averages of country-level coverage rates for social insurance, social assistance, and labor market programs, across country income groups. Indicators do not count for overlap among programs types (people receiving more than one program); therefore, the sum of percentages by type of program may add up to more than 100 percent. Coverage is determined as follows: (number of individuals in the total population or poorest quintile who live in a household where at least one member receives the transfer)/(number of individuals in the total population). This figure underestimates total social protection and labor coverage because household surveys do not include all programs that exist in each country. The poorest quintile is calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. Latvia includes 10 programs under UCTs, household surveys in all regions. They cover 23 which embrace the means-tested Guaranteed percent of the poorest quintile, on average. In Minimal Income Allowance, together with the ASPIRE database, UCT programs are more universal child and family benefits. On reported in the household surveys of 63 coun- the other hand, Belize includes only one pro- tries, compared with 103 countries where the gram, the social welfare transfer, under UCTs. administrative data report having at least one Complete documentation of the programs program in this category.8 The most UCTs are that are included in each SSN category, per found in the household surveys for Europe and country, is available in the ASPIRE online Central Asia and Sub-Saharan Africa (20 and portal.6 In the discussion that follows, cover- 17 countries, respectively). Figure 3.4 shows age of the poor is presented by each SSN pro- the distribution of UCT  programs covering gram category, illustrating different patterns from 0.6 to almost 100 percent of the poorest in the use of specific SSN interventions, the quintile. Among the programs that achieve degree of variation in coverage rates across almost 100 percent coverage of the poor are countries, and the benchmarking of country the Child Money Program in Mongolia,9 and results against global program averages.7 the social transfers in Malaysia (94 percent UCTs constitute some of the most popular coverage of the poor), which may reflect the safety net tools and are included in most performance of the Bantuan Rakyat 1 Program, Analyzing the Performance of Social Safety Net Programs 35 FIGURE 3.4  Share of the Poorest Quintile That Receives Unconditional Cash Transfer Programs, as Captured in Household Surveys Mongolia 2012 Malaysia 2008 Russian Federation 2016 South Africa 2010 Romania 2012 Slovak Republic 2009 Indonesia 2015 Latvia 2009 Lithuania 2008 Georgia 2011 Chile 2013 Belarus 2013 Poland 2012 Croatia 2010 Vietnam 2014 Sri Lanka 2012 Botswana 2009 Armenia 2014 Mauritania 2014 Ukraine 2013 Kosovo 2013 Serbia 2013 Iraq 2012 Montenegro 2014 Bolivia 2012 Pakistan 2013 Average, 23.4 Jordan 2010 Moldova 2013 Turkey 2014 Albania 2012 China 2013 Kyrgyz Republic 2013 El Salvador 2014 Maldives 2009 West Bank and Gaza 2009 Tunisia 2010 Mauritius 2012 Egypt, Arab Rep. 2008 Jamaica 2010 Costa Rica 2014 Mozambique 2008 Djibouti 2012 Uruguay 2012 Niger 2014 Thailand 2013 Swaziland 2009 Belize 2009 Nepal 2010 Rwanda 2013 Namibia 2009 Sudan 2009 Liberia 2014 Mexico 2012 Zimbabwe 2011 Burkina Faso 2014 Kazakhstan 2010 Uganda 2012 Guinea 2012 Zambia 2010 South Sudan 2009 Colombia 2014 Morocco 2009 Tajikistan 2011 0 20 40 60 80 100 Percent Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 63); Europe and Central Asia (n = 20); Sub-Saharan Africa (n = 17); Latin America and the Caribbean (n = 9); Middle East and North Africa (n = 7); East Asia and Pacific (n = 6); and South Asia (n = 4). Unconditional cash transfers include any of the following: poverty alleviation and emergency programs, guaranteed minimum-income programs, and universal or poverty-targeted child and family allowances. They do not include social pensions or targeted subsidies in cash. The average coverage of unconditional cash transfers is estimated as the simple average of these programs’ coverage rates across countries. Coverage is determined as follows: (number of individuals in the total population or poorest quintile who live in a household where at least one member receives the transfer)/(number of individuals in the total population). This figure underestimates total coverage because household surveys do not include all programs that exist in each country. The poorest quintile is calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. 36 The State of Social Safety Nets 2018 the country’s flagship cash transfer program Latin America and the Caribbean c ­ apture CCT for the poor.10 In Europe and Central Asia, information. Only 19 countries in ASPIRE Russian Federation’s cash transfer pro- include information on CCT programs in their grams  show the  largest coverage of the poor household surveys (of which 16   are in Latin (79 ­percent). America and the Caribbean), compared with CCTs typically aim to reduce poverty and 64  programs observed in the administrative increase human capital by requiring beneficia- database. This instrument covers from 2.4 to 75 ries to comply with conditions such as school percent of the poor (see ­ figure 3.5). Asignaciones attendance and health checkups. The average Familiares in Uruguay has the largest coverage of coverage of the poorest quintile by CCTs in the the poor (75 percent), followed by Bonos Juancito sample of surveys is 40 percent. Pioneered Pinto and Juana Azurduy in Bolivia (73 percent), by  Brazil and Mexico in the late 1990s, CCTs and Prospera in Mexico (63 p ­ ercent). The CCT spread to other countries in the region and with large coverage outside Latin America and worldwide. Yet, few household surveys ­outside of the Caribbean is the Pantawid Pamilyang Pilipino FIGURE 3.5  Share of the Poorest Quintile That Receives Conditional Cash Transfer Programs, as Captured in Household Surveys Uruguay 2012 Bolivia 2012 Mexico 2012 Philippines 2015 Brazil 2015 Argentina 2013 Jamaica 2010 Colombia 2014 Dominican Republic 2014 Chile 2013 Ecuador 2016 Peru 2014 Average, 40.3 Panama 2014 Guatemala 2014 Costa Rica 2014 Honduras 2013 Bangladesh 2010 Paraguay 2011 Timor-Leste 2011 0 20 40 60 80 100 Percent Source: ASPIRE database. Note: The number of countries per region included in the analysis is as follows: total (n = 19); Latin America and the Caribbean (n = 16); East Asia and Pacific (n = 2); South Asia (n = 1); Europe and Central Asia (n = 0); Middle East and North Africa (n = 0); and Sub-Saharan Africa (n = 0). Conditional cash transfer programs include the following: Argentina 2013: Asignación Universal por Hijo. Bangladesh 2010: Maternity allowance; Program for the Poor Lactating; Stipend for Primary Students (MOPMED); Stipend for Dropout Students; Stipend for Secondary and Higher Secondary/Female Student. Bolivia 2012: Bono Juancito Pinto; and Bono Juana Azurduy. Brazil 2015: Bolsa Família. Chile 2013: Subsidio Familiar (SUF); Bono de Protección Familiar y de Egreso; Bono por control del niño sano; Bono por asistencia escolar; and Bono por logro escolar. Colombia 2014: Familias en Acción. Costa Rica 2014: Avancemos. Dominican Republic 2014: Solidaridad Program and other transfers. Ecuador 2016: Bono de Desarrollo Humano. Guatemala 2014: Programa Mi Bono Seguro. Honduras 2013: Asignaciones Familiares– Bonos PRAF and otro tipo de bonos. Jamaica 2010: Program of Advancement Through Health and Education (PATH)—child 0–71 months; 6–17 years; and pregnant and lactating women. Mexico 2012: Oportunidades. Panama 2014: Red de Oportunidades. Paraguay 2011: Tekopora. Peru 2014: Programa Juntos. Philippines 2015: Pantawid Pamilyang Pilipino Program (4Ps). Timor Leste 2011: Bolsa da Mae. Uruguay 2012: Asignaciones Familiares. The average coverage of conditional cash transfers is estimated as the simple average of these programs’ coverage rates across countries. Coverage is determined as follows: (number of individuals in the total population or poorest quintile who live in a household where at least one member receives the transfer)/(number of individuals in the total population). This figure underestimates total coverage because household surveys do not include all programs that exist in each country. The poorest quintile is calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. Analyzing the Performance of Social Safety Net Programs 37 Program (4Ps) in the Philippines, which ­ covers covariate and idiosyncratic shocks. Public 60 percent of the poor (see figure 3.5), demon- works programs include cash-, food-, and strating its focus on ensuring c ­overage of the inputs-for-work (Andrews et al. 2012). They are poorest. more often implemented in Sub-Saharan Africa For individuals who do not have access to and South Asia, although respective informa- social insurance benefits, social pensions aim to tion/data are not often captured in household overcome loss of income because of old age, dis- surveys. Only 9 countries in ASPIRE have spe- ability, or death of the bread winner. In the cific information about public works in the sample of countries, social pensions cover, on household surveys, as compared with 96 coun- average, 20  percent of the poorest quintile. tries in the administrative database. In these Social pensions presented here include noncon- countries, public works cover between 1 and 27 tributory disability and survivor pensions, and percent of the poorest quintile (figure 3.7); the thus represent a broader category than old-age largest coverage rates are observed for the social pensions (featured in chapter 4). Only MGNREG program in India (27 percent), a 36 countries in the ASPIRE household survey flagship national social safety net program with database capture any form of social pensions, a history going back decades; and MASAF in compared with 75 countries in the administra- Malawi (21 percent), the social fund program tive database. Most surveys with social pen- that has become the cornerstone of the national sion information are found in Europe and SSN system. Central Asia (n = 13) and Latin America and Fee waivers and targeted subsidies typi- the Caribbean (n = 10). In the sample, social cally subsidize services or provide access to pensions cover between 0.6 and 81 percent of low-priced food staples to the poor. They are individuals in the poorest quintile (see figure 3.6). common but generally provide limited cover- Georgia has the highest coverage of the poor- age of the poorest quintile—13 percent, on est quintile because of its universal old-age average, in the sample of countries. However, social pensions. A few countries in Africa have because this  category is not easily collected extensive coverage of the poorest quintile, by household surveys, this average is most such as Mauritius (79 percent) and South likely a considerable underestimate. Services Africa (62 percent). In all these countries, for- under this category usually relate to educa- mal social insurance has low coverage and tion, health, housing, transportation, or utili- social pensions constitute the main form of ties. When a beneficiary is exempt from social protection for the elderly. Thailand’s payment for such services and the cost is social pensions have high coverage of the low- borne by the government program, such fee est quintile (58 percent), also driven by provid- waivers provide conditional support for  the ing support to the elderly and disabled, taking targeted group using a specific service.12 into account their living arrangements within Out of 82 countries with information on fee extended families.11 On the opposite extreme, waivers and targeted subsidies in the admin- some countries in Europe and Central Asia istrative database, only 22 are observed in the have extended social insurance systems; there- household survey data. Among the regions, fore, social pensions cover only those who Europe and Central Asia (n = 10 countries) do  not benefit from social insurance, which and Latin America and the Caribbean coun- constitute rather narrow population groups. tries (n = 8 countries) capture this typology Hence the coverage is low (Latvia, Montenegro, of programs in the surveys most often. The Russian Federation, and Serbia). program covers between 0.4 and 56 percent Public works programs typically condition of the poor (see figure 3.8). Coverage rates for the transfer on participating in a community targeted subsidies and fee waivers in Europe project/activity. Very few public works pro- and Central Asia tend to be smaller because grams are captured in the sample of household these programs focus only on a subset of the surveys, and their coverage of the poorest quin- poor, but tend to include several different tile is limited, at 11 percent. Public works are forms of benefits (for example, subsidized implemented for many reasons, such as to pro- housing; and fee waivers for kindergartens, vide employment of last  resort or to ­ mitigate health care, public transportation). 38 The State of Social Safety Nets 2018 FIGURE 3.6  Share of the Poorest Quintile That Receives Social Pensions, as Captured in Household Surveys Georgia 2011 Mauritius 2012 South Africa 2010 Thailand 2013 Slovak Republic 2009 Timor-Leste 2011 Swaziland 2009 Kazakhstan 2010 Botswana 2009 Chile 2013 Namibia 2009 Nepal 2010 Mexico 2012 Costa Rica 2014 Poland 2012 Panama 2014 Bangladesh 2010 Lithuania 2008 Romania 2012 Moldova 2013 Brazil 2015 Average, 19.6 Maldives 2009 Turkey 2014 Russian Federation 2016 Colombia 2014 Sri Lanka 2012 Belize 2009 Sierra Leone 2011 Serbia 2013 Latvia 2009 Rwanda 2013 Honduras 2013 Paraguay 2011 Montenegro 2014 Guatemala 2014 Tajikistan 2011 0 20 40 60 80 100 Percent Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 36); Europe and Central Asia (n = 13); Latin America and the Caribbean (n = 10); Sub-Saharan Africa (n = 7); South Asia (n = 4); East Asia and Pacific (n = 2); and Middle East and North Africa (n = 0). Social pensions include any of the following: noncontributory old-age pensions; disability pensions; and survivor pensions. Social pensions average coverage is the simple average of social pensions coverage rates across countries. Coverage is determined as follows: (number of individuals in the total population or poorest quintile who live in a household where at least one member receives the transfer)/(number of individuals in the total population). This figure underestimates total coverage because household surveys do not include all programs that exist in each country. The poorest quintile is calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. School feeding programs provide meals to benefit a significant share of the poor—37 percent. students generally in poor and food-insecure Even though school feeding is a common safety areas, with the aim of improving nutrition, net program, it is not always captured in house- health, and educational outcomes. In the sam- hold surveys. Of the 117 countries reporting ple, these programs are found, on average, to school feeding in administrative data, only Analyzing the Performance of Social Safety Net Programs 39 FIGURE 3.7  Share of the Poorest Quintile That Receives Public Works, as Captured in Household Surveys India 2011 Malawi 2013 Ethiopia 2010 Afghanistan 2011 Nepal 2010 Niger 2014 Average, 10.8 Argentina 2013 Rwanda 2013 Mexico 2012 0 10 20 30 40 50 60 70 80 90 100 Percent Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 9); Sub-Saharan Africa (n = 4); Latin America and the Caribbean (n = 2); South Asia (n = 3); East Asia and Pacific (n = 0); Europe and Central Asia (n = 0); and Middle East and North Africa (n = 0). Public works programs included: Afghanistan 2011: Cash-for-work programs; food-for-work programs; or income-generating program/projects. Argentina 2013: Plan de Empleo. Ethiopia 2010: Productive Safety Net Program (PSNP). India 2011: MGNREG and other public works. Malawi 2013: MASAF; PWP; Inputs-for-work program. Mexico 2012: Programa de Empleo Temporal (PET). Nepal 2010: Rural Community Infrastructure Works Program (RCIW) and other food-for-work and cash for work programs. Niger 2014: Public works. Rwanda 2013: Public works from the Vision 2020 Umurenge Program. Public works average coverage is the simple average of public works coverage rates across countries. Coverage is determined as follows: (number of individuals in the total population or poorest quintile who live in a household where at least one member receives the transfer)/(number of individuals in the total population). This figure underestimates total coverage because household surveys do not include all programs that exist in each country. The poorest quintile is calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. 26 countries with this program are found in the information on in-kind transfers (see ­ figure household survey data. Of these 26 countries, 3.10), compared with 90 countries in the 15 are in Latin America and the Caribbean. administrative database. About one-third of Figure 3.9 shows that coverage varies across the surveys (14 out of  45) reporting in-kind countries from 2.21 to 86 percent of the poor. transfers are in countries in Sub-Saharan Such variation is more likely to reflect the dif- Africa, where they typically consist of pro- ferences in survey designs and efforts to capture grams to promote agricultural productiv- this form of benefit than real differences across ity  or  emergency relief. However, programs countries. The programs with the largest cover- report larger coverage of the poor in Latin age of the poor, among the countries with America and the Caribbean and Middle East ­ adequate data, are in Botswana (86 percent), and North Africa than in Sub-Saharan Africa. Bolivia (73 percent), El Salvador (69 percent), Supplemental food programs for children, preg- Nicaragua (67 percent), and Honduras and nant and nursing women, and the elderly are Panama (66 percent). common in Latin America and the Caribbean, In-kind transfers consist of food rations, as well as school supplies and uniforms. In a few clothes, school supplies, shelter, fertilizers, countries, these programs cover a high percent- seeds, agricultural tools or animals, and build- age of the poor, including Peru (84 percent), ing materials, among others. They are a very Ecuador (74 percent), and El Salvador and common SSN instrument, and in the sample Paraguay (70 percent). Coverage in Peru is par- cover, on average, 27 percent of the poorest ticularly large because the in-kind category quintile. Their objectives are usually to provide encompasses seven in-kind programs captured food security, improve nutrition, increase in the survey, including nutritional programs, ­ agricultural productivity, and deliver emer- school supplies, uniforms, and shoes and laptops gency relief. Forty-five countries in the to school children. In Middle East and North ASPIRE  ­ household survey database capture Africa, in-kind transfers take the form of food 40 The State of Social Safety Nets 2018 FIGURE 3.8  Share of the Poorest Quintile That Receives Fee Waivers and Targeted Subsidies, as Captured in Household Surveys Colombia 2014 China 2013 Zimbabwe 2011 Chile 2013 Albania 2012 Vietnam 2014 Russian Federation 2016 Costa Rica 2014 Panama 2014 Jamaica 2010 Poland 2012 Average, 12.8 Honduras 2013 Tajikistan 2011 Ukraine 2013 Lativa 2009 Romania 2012 Mauritius 2012 Slovak Republic 2009 Dominican Republic 2014 South Africa 2010 Belize 2009 Moldova 2013 0 10 20 30 40 50 60 70 80 90 100 Percent Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 22); Europe and Central Asia (n = 10); Latin America and the Caribbean (n = 8); Sub-Saharan Africa (n = 2); East Asia and Pacific (n = 2); Middle East and North Africa (n = 0); and South Asia (n = 0). Fee waivers and targeted subsidies include any of the following: energy products; education; utilities; housing or transportation fees waivers to specific households; or such fees discounted below the market cost. They do not include health fee waivers/subsidies, except for Zimbabwe. Fee waivers and targeted subsidies program coverage is the simple average of these programs’ coverage rates across countries. Coverage is determined as follows: (number of individuals in the total population or poorest quintile who live in a household where at least one member receives the transfer)/(number of individuals in the total population). This figure underestimates total coverage because household surveys do not include all programs that exist in each country. The poorest quintile is calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. and in-kind aid. They cover about 81 percent of WHAT IS THE BENEFICIARY INCIDENCE the poor in Iraq and 56 percent in the Arab OF VARIOUS SOCIAL SAFETY NET Republic of Egypt (see figure 3.10). In East Asia INSTRUMENTS? and Pacific, Indonesia’s Rastra Program (rice Beneficiary incidence indicates to what extent a subsidies) has the largest coverage of the poor given population group benefits from a pro- (71 percent).13 gram. For this analysis, individuals are ranked Analyzing the Performance of Social Safety Net Programs 41 FIGURE 3.9  Share of the Poorest Quintile That Receives School Feeding Programs, as Captured in Household Surveys Botswana 2009 Bolivia 2012 El Salvador 2014 Nicaragua 2014 Honduras 2013 Panama 2014 Jamaica 2010 Chile 2013 Costa Rica 2014 Thailand 2013 Paraguay 2011 Ecuador 2016 Uruguay 2012 Colombia 2014 Russian Federation 2016 Guatemala 2014 Malawi 2013 Average, 37.1 Morocco 2009 Sri Lanka 2012 Slovak Republic 2009 Latvia 2009 Liberia 2014 Peru 2014 Belize 2009 Nigeria 2015 Tanzania 2014 0 10 20 30 40 50 60 70 80 90 100 Percent Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 26); Latin America and the Caribbean (n = 15); Sub-Saharan Africa (n = 5); Europe and Central Asia (n = 3); Middle East and North Africa (n = 1); East Asia and Pacific (n = 1); and South Asia (n = 1). School feeding programs encompass any type of meals or food items provided at school. School feeding average coverage is the simple average of school feeding coverage rates across countries. Coverage is determined as follows: (number of individuals in the total population or poorest quintile who live in a household where at least one member receives the transfer)/(number of individuals in the total population). This figure underestimates total coverage because household surveys do not include all programs that exist in each country. The poorest quintile is calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. according to their position in the welfare 20 percent of the distribution (or if more than ­ distribution, based on quintiles of per capita 40 percent of its total beneficiaries belong to the pretransfer income or consumption; the pro- ­ bottom 40 percent of the distribution).15 portion of program beneficiaries belonging to Beneficiary incidence and benefits incidence each quintile is presented. Another correspond- help determine which population groups are ing indicator is benefits incidence, which shows benefiting from the program, and thus are use- the proportion of program benefits transferred ful indicators to analyze the performance of to individuals in each quintile.14 A program is SPL programs. Propoor beneficiary and benefits considered propoor if more than 20 percent incidence is the only way to ensure that a pro- of  its total beneficiaries belong to the bottom gram within a given budget achieves greater 42 The State of Social Safety Nets 2018 FIGURE 3.10  Share of the Poorest Quintile That Receives In-Kind Transfers, as Captured in Household Surveys Peru 2014 Iraq 2012 Ecuador 2016 Indonesia 2015 El Salvador 2014 Paraguay 2011 Chile 2013 Guatemala 2014 Egypt, Arab Rep. 2008 Morocco 2009 Nicaragua 2014 Romania 2012 Uruguay 2012 Côte d'Ivoire 2015 Vietnam 2014 Turkey 2014 Sierra Leone 2011 Djibouti 2012 Kosovo 2013 Mauritania 2014 Panama 2014 Armenia 2014 China 2013 West Bank and Gaza 2009 Costa Rica 2014 Average, 27.1 Haiti 2012 Malawi 2013 Sudan 2009 Niger2014 Georgia 2011 Russian Federation 2016 Nepal 2010 Tanzania 2014 Sri Lanka 2012 Tunisia 2010 Ethiopia 2010 Senegal 2011 South Sudan 2009 Kazakhstan 2010 Latvia 2009 Liberia 2014 Namibia 2009 Rwanda 2013 Croatia 2010 Nigeria 2015 0 10 20 30 40 50 60 70 80 90 100 Percent Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 45); Sub-Saharan Africa (n = 14); Latin America and the Caribbean (n = 11); Europe and Central Asia (n = 9); Middle East and North Africa (n = 6); East Asia and Pacific (n = 3); and South Asia (n = 2). In-kind transfers include any of the following: food aid; agricultural inputs; clothes; school supplies; and building materials. In-kind transfers average coverage is the simple average of in-kind transfer coverage rates across countries. Coverage is determined as follows: (number of individuals in the total population or poorest quintile who live in a household where at least one member receives the transfer)/(number of individuals in the total population). This figure underestimates total coverage because household surveys do not include all programs that exist in each country. The poorest quintile is calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. Analyzing the Performance of Social Safety Net Programs 43 impact in terms of poverty reduction (Yemtsov poorest quintile, which indicates that those et al., forthcoming). instruments are still propoor. The beneficiary incidence analysis con- The analysis of beneficiary incidence by SSN ducted by type of SSN instrument reveals that, instrument across countries shows the pres- on average, all types of SSN programs tend to ence of highly propoor programs in each pro- be propoor or favor the poor and near-poor. gram category. Figures 3.12 through 3.18 That is, a higher percentage of beneficiaries illustrate that despite wide variation in benefi- belong to the first and second poorest quin- ciary distribution across countries, most of tiles. This is illustrated in figure 3.11, where them favor the poor and near-poor, with more the lines representing each SSN instru- than 20 percent of the total beneficiaries ment  show a similar downward slope. CCTs belonging to the poorest quintile. However, for generally show a more propoor distribution every SSN category, there are also examples of compared with the other SSN instruments, programs that are proportionally distributed which is not surprising because these pro- or even favor the rich more than the poor and grams typically target poor households. the middle class. It is not always possible to Figure  3.11 shows that, among the observed draw general conclusions about the distribu- programs, 45 percent of CCT beneficiaries are tion of beneficiaries and benefits of a specific in the poorest quintile on average, while only program without knowing detailed informa- 4 ­percent are in the richest quintile. Between tion on the program’s design, eligibility criteria, 33 and 37 percent of beneficiaries of the other and implementation. Some programs may not SSN instruments, on average, belong to the be propoor  by design; for example, they may FIGURE 3.11  Global Distribution of Beneficiaries by Type of Social Safety Net Instrument, as Captured in Household Surveys, by Quintile of Pretransfer Welfare 50 45 40 35 30 Percent 25 20 15 10 5 0 Poorest Q2 Q3 Q4 Richest Unconditional cash transfers Conditional cash transfers Social pensions Public works Fee waivers School feeding In-kind transfers Source: ASPIRE database. Note: The total number of countries where the social safety net instrument is captured in household surveys is as follows: unconditional cash transfers (n = 63), conditional cash transfers (n = 19); social pensions (n = 36); public works (n = 9); fee waivers and targeted subsidies (n = 22); school feeding (n = 26); and in-kind transfers (n = 45). Beneficiaries’ incidence is calculated as follows: (number of direct and indirect beneficiaries [people who live in a household where at least one member receives the transfer] in a given quintile)/(total number of direct and indirect beneficiaries). The sum of percentages across quintiles per given instrument equals 100 percent. Aggregated indicators are calculated using simple averages of program instrument beneficiaries’ incidence rates across countries. Quintiles are calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; Q = quintile. 44 The State of Social Safety Nets 2018 not be addressed specifically to the poor but to second-­poorest quintile (figure 3.12). One rea- the general population (that is, in the case of son for this large variation of results across universal programs). Or their eligibility criteria countries is the fact that ASPIRE—in some may be categorical (for example, in terms of countries—aggregates many types of UCTs, disability, ethnicity, and war victims) and not and  the respective programs may have very means-tested. In those cases, beneficiaries different objectives and eligibility criteria. ­ meeting the categorical requirements may not However, for other countries, this category belong to the poor. captures only one program, making it easier to UCTs are characterized by a wide range of interpret the observed beneficiary incidence beneficiary incidence across countries, but results. Universal programs have an even on  average, 37 percent of UCT beneficiaries distribution. For instance, for the Child ­ belong to the poorest quintile and 23 to the Money Program in Mongolia, the participation FIGURE 3.12  Distribution of Unconditional Cash Transfer Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare Kosovo 2013 Montenegro 2014 Turkey 2014 Costa Rica 2014 Vietnam 2014 Mauritius 2012 Liberia 2014 Djibouti 2012 Kazakhstan 2010 Georgia 2011 West Bank and Gaza 2009 Jordan 2010 Indonesia 2015 El Salvador 2014 Armenia 2014 Rwanda 2013 Poland 2012 Serbia 2013 Kyrgyz Republic 2013 Jamaica 2010 Pakistan 2013 Sri Lanka 2012 Croatia 2010 Albania 2012 Egypt, Arab Rep. 2008 Sudan 2009 Morocco 2009 Belize 2009 Zambia 2010 Botswana 2009 Mexico 2012 0 10 20 30 40 50 60 70 80 90 100 Percent Poorest Q2 Q3 Q4 Richest (Figure continues next page) Analyzing the Performance of Social Safety Net Programs 45 FIGURE 3.12  Distribution of Unconditional Cash Transfer Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare (Continued) Namibia 2009 South Africa 2010 Moldova 2013 Zimbabwe 2011 Swaziland 2009 China 2013 Burkina Faso 2014 Ukraine 2013 Thailand 2013 Mozambique 2008 Belarus 2013 Tunisia 2010 Chile 2013 Iraq 2012 Russian Federation 2016 Nepal 2010 Romania 2012 Niger2014 Malaysia 2008 Bolivia 2012 Latvia 2009 Mauritania 2014 Mongolia 2012 Colombia 2014 Slovak Republic 2009 Maldives 2009 Lithuania 2008 Uganda 2012 Guinea 2012 South Sudan 2009 Uruguay 2012 Tajikistan 2011 0 10 20 30 40 50 60 70 80 90 100 Percent Poorest Q2 Q3 Q4 Richest Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 63); Sub-Saharan Africa (n = 17); Europe and Central Asia (n = 20); Latin America and the Caribbean (n = 9); Middle East and North Africa (n = 7); East Asia and Pacific (n = 6); and South Asia (n = 4). Unconditional cash transfers include any of the following: poverty alleviation and emergency programs; guaranteed minimum-income programs; and universal or poverty-targeted child and family allowances. They do not include social pensions or targeted subsidies in cash. Beneficiaries’ incidence is calculated as follows: (number of direct and indirect beneficiaries [people who live in a household where at least one member receives the transfer] in a given quintile)/(total number of direct and indirect beneficiaries). The sum of percentages across quintiles per given instrument equals 100 percent. Quintiles are calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; Q = quintile. of all  five quintiles is close to 20  percent (see Hogar Constituido (Transfer for Constituted ­ figure  3.12). In contrast, the programs in Household) belong to the poorest quintile. This Kosovo and Montenegro are much more program is provided only to public servants focused on the poorest quintile, which accounts who are married or have dependents whose for more than 70 percent of the total UCT ben- monthly gross salary is less than the sum of two eficiaries.16 At the same time, in Uruguay, only national minimum wages. This explains why 7 percent of the beneficiaries of Prima por the program is not propoor by design. 46 The State of Social Safety Nets 2018 In general, CCTS are, as has been mentioned quintile. The State Social Maintenance Benefit already, more propoor than other SSN program in Latvia has the highest proportion of benefi- types. Among all 19 CCT programs included in ciaries belonging to the poorest quintile (59 the ASPIRE database, the average beneficiary percent). In contrast, Programa Adulto Mayor incidence is 45 percent for the poorest quintile in Guatemala and Bono por Tercera Edad in and 26 percent for the second-poorest quintile. Honduras have only 8 and 9 percent of their Panama’s conditional cash transfer program, beneficiaries coming from the poorest quintile, Red de Oportunidades, has an incidence of par- respectively (see figure 3.14). Part of the ticipants from the poorest quintile of 75 ­percent. propoor performance shown by some coun- More than 65 percent of the beneficiaries of tries may be related to the way the p ­ retransfer Programa Juntos in Peru, belong to the poorest indicator is constructed. If social pensions 20 percent. Timor-Leste’s Bolsa da Mae has the cover a sizable part of the poor and their bene- lowest beneficiary incidence rate for the poorest fit level is high, most beneficiaries may tend to quintile, at 21 percent (see figure 3.13). depend on them and hence group in the lowest Social pensions also have a very propoor dis- quintile of the welfare distribution once such tribution of beneficiaries. An average of transfers are removed. On the other hand, 35 percent of beneficiaries belong to the p­ oorest social pensions in Honduras tend to cluster quintile and 22 percent to the second-­ poorest further up the income distribution, reflecting FIGURE 3.13  Distribution of Conditional Cash Transfer Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare Panama 2014 Peru 2014 Paraguay 2011 Brazil 2015 Argentina 2013 Mexico 2012 Ecuador 2016 Philippines 2015 Costa Rica 2014 Honduras 2013 Chile 2013 Uruguay 2012 Jamaica 2010 Colombia 2014 Bangladesh 2010 Dominican Republic 2014 Guatemala 2014 Bolivia 2012 Timor-Leste 2011 0 10 20 30 40 50 60 70 80 90 100 Percent Poorest Q2 Q3 Q4 Richest Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 19); Latin America and the Caribbean (n = 16); East Asia and Pacific (n = 2); South Asia (n = 1); Europe and Central Asia (n = 0); Middle East and North Africa (n = 0); and Sub-Saharan Africa (n = 0). Conditional cash transfer programs include the following: Argentina 2013: Asignación Universal por Hijo. Bangladesh 2010: Maternity allowance, Program for the Poor Lactating, Stipend for Primary Students (MOPMED), Stipend for Drop Out Students, Stipend for Secondary and Higher Secondary/ Female Student. Bolivia 2012: Bono Juancito Pinto and Bono Juana Azurduy. Brazil 2015: Bolsa Família. Chile 2013: Subsidio Familiar (SUF), Bono de Protección Familiar y de Egreso, Bono por control del niño sano, Bono por asistencia escolar and Bono por logro escolar. Colombia 2014: Familias en Acción. Costa Rica 2014: Avancemos. Dominican Republic 2014: Solidaridad Program and other transfers. Ecuador 2016: Bono de Desarrollo Humano. Guatemala 2014: Programa Mi Bono Seguro. Honduras 2013: Asignaciones Familiares–Bonos PRAF and otro tipo de bonos. Jamaica 2010: Program of Advancement Through Health and Education (PATH)—child (0–71 months); 6–17 years; and pregnant and lactating women. Mexico 2012: Oportunidades. Panama 2014: Red de Oportunidades. Paraguay 2011: Tekopora. Peru 2014: Programa Juntos. Philippines 2015: Pantawid Pamilyang Pilipino Program (4Ps). Timor-Leste 2011: Bolsa da Mae. Uruguay 2012: Asignaciones Familiares. Beneficiaries’ incidence is calculated as follows: (number of direct and indirect beneficiaries [people who live in a household where at least one member receives the transfer] in a given quintile)/(total number of direct and indirect beneficiaries). The sum of percentages across quintiles per given instrument equals 100 percent. Quintiles are calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; Q = quintile. Analyzing the Performance of Social Safety Net Programs 47 FIGURE 3.14  Distribution of Social Pensions Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare Latvia 2009 Russian Federation 2016 Serbia 2013 Brazil 2015 Rwanda 2013 Turkey 2014 Poland 2012 Costa Rica 2014 Romania 2012 Sri Lanka 2012 Lithuania 2008 South Africa 2010 Bangladesh 2010 Mexico 2012 Paraguay 2011 Mauritius 2012 Colombia 2014 Botswana 2009 Chile 2013 Moldova 2013 Panama 2014 Kazakhstan 2010 Namibia 2009 Thailand 2013 Slovak RepubIic 2009 Sierra Leone 2011 Georgia 2011 Belize 2009 Nepal 2010 Swaziland 2009 Timor-Leste 2011 Maldives 2009 Tajikistan 2011 Montenegro 2014 Honduras 2013 Guatemala 2014 0 10 20 30 40 50 60 70 80 90 100 Percent Poorest Q2 Q3 Q4 Richest Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 36); Europe and Central Asia (n = 13); Latin America and the Caribbean (n = 10); Sub-Saharan Africa (n = 7); South Asia (n = 4); East Asia and Pacific (n = 2); and Middle East and North Africa (n = 0). Social pensions include any of the following: noncontributory old-age pensions; disability pensions; and survivor pensions. Beneficiaries’ incidence is calculated as follows: (number of direct and indirect beneficiaries [people who live in a household where at least one member receives the transfer] in a given quintile)/(total number of direct and indirect beneficiaries). The sum of percentages across quintiles per given instrument equals 100 percent. Quintiles are calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; Q = quintile. this instrument’s bias in coverage toward richer somewhat propoor instrument. Public works are areas of the country. limited only to households with able-bodied, The beneficiary incidence of public works for unemployed members who are willing to work. the poorest quintile is 33 percent, on average. Because many public works rely on self-­ Yet, when looking at the two bottom quintiles, selection  and are oversubscribed, especially in all 10 public works programs analyzed have a low-income countries (more people want to beneficiary incidence rate of at least 45 percent work in these programs than they have employ- for the poorest 40 percent, which still makes it a ment positions), there is a lot of sharing and 48 The State of Social Safety Nets 2018 FIGURE 3.15  Distribution of Public Works Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare Mexico 2012 Argentina 2013 Nepal 2010 Rwanda 2013 India 2011 Niger2014 Afghanistan 2011 Ethiopia 2010 Malawi 2013 0 20 40 60 80 100 Percent Poorest Q2 Q3 Q4 Richest Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 9); Sub-Saharan Africa (n = 4); Latin America and the Caribbean (n = 2); South Asia (n = 3); East Asia and Pacific (n = 0); Europe and Central Asia (n = 0); and Middle East and North Africa (n = 0). Public works programs include the following: Afghanistan 2011: Cash-for-work, food-for-work programs, or income-generating program/projects. Argentina 2013: Plan de Empleo. Ethiopia 2010: Productive Safety Net Program (PSNP). India 2011: MGNREG and other public works. Malawi 2013: MASAF, PWP, Inputs-for-work program. Mexico 2012: Programa de Empleo Temporal (PET). Nepal 2010: Rural Community Infrastructure Works Program (RCIW) and other food-for-work and cash for work programs. Niger 2014: Public works. Rwanda 2013: Public works from the Vision 2020 Umurenge Program. Beneficiaries’ incidence is calculated as follows: (number of direct and indirect beneficiaries [people who live in a household where at least one member receives the transfer] in a given quintile)/(total number of direct and indirect beneficiaries). The sum of percentages across quintiles per given instrument equals 100 percent. Quintiles are calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; Q = quintile. capture of the program by the not so poor. 82 and 73 percent of the beneficiaries are drawn Mexico’s Programa de Empleo Temporal shows from the poorest quintile, respectively. All these the most propoor distribution, with 49 percent programs are rather narrow in coverage, and of beneficiaries coming from the poorest quin- stand out from a typical universal school feeding figure 3.15). tile (see ­ program, with different objectives that are often Fee waivers and targeted subsidies have a not focused on alleviating poverty. In addition, somewhat flatter beneficiary incidence along the in universal programs with high coverage, the welfare distribution, compared with other incidence of school feeding will depend on the instruments. On average, 33 percent of their access to schooling across quintiles, which may beneficiaries belong to the poorest quintile be skewed in favor of the nonpoor, and is not a and  24 percent to the second-poorest quintile. design feature of the program itself. In  Panama, most of the beneficiaries (82  per- In-kind transfers are generally quite cent) of food supplements and agricultural sub- propoor. In the sample of observed programs, sidies  belong to the poorest quintile, followed 34 percent of beneficiaries of in-kind transfers by  Vietnam, where nearly three-quarters belong to the poorest quintile and 23 percent (73 percent) of beneficiaries of housing, petro- to the second-poorest quintile. Food aid in leum, and kerosene subsidies and tuition fee Djibouti and food and nutritional programs in exceptions are in the poorest quintile. In South Uruguay have especially propoor distribu- Africa, by contrast, the rich capture most of tions, with 71 and 70 percent of the recipients these benefits (see figure 3.16). belonging to the poorest quintile, respectively Beneficiary incidence of school feeding pro- (see figure 3.18).17 grams varies substantially by country. On aver- age, 34 percent of beneficiaries of school feeding WHAT ARE THE BENEFIT LEVELS OF programs belong to the poorest quintile and SOCIAL SAFETY NET PROGRAMS? 24  percent to the second-­ poorest quintile (see The level of benefits is measured by two indica- ­ figure 3.17). The most propoor programs are tors: per capita average transfer (monetary found in the Slovak Republic and Latvia, where value) and share of benefits with respect to per Analyzing the Performance of Social Safety Net Programs 49 FIGURE 3.16  Distribution of Fee Waivers and Targeted Subsidies Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare Panama 2014 Vietnam 2014 Poland 2012 Latvia 2009 Costa Rica 2014 China 2013 Ukraine 2013 Slovak Republic 2009 Mauritius 2012 Romania 2012 Colombia 2014 Albania 2012 Tajikistan 2011 Chile 2013 Zimbabwe 2011 Belize 2009 Honduras 2013 Russian Federation 2016 Jamaica 2010 Dominican Republic 2014 Moldova 2013 South Africa 2010 0 10 20 30 40 50 60 70 80 90 100 Percent Poorest Q2 Q3 Q4 Richest Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 22); Europe and Central Asia (n = 10); Latin America and the Caribbean (n  =  8); Sub-Saharan Africa (n = 2); East Asia and Pacific (n = 2); Middle East and North Africa (n = 0); and South Asia (n = 0). Fee waivers and targeted subsidies include any of the following: energy products; education; utilities; housing or transportation fees waivers to specific households; or such fees discounted below the market cost. They do not include health benefits or subsidies, except for Zimbabwe. Beneficiaries’ incidence is calculated as follows: (number of direct and indirect beneficiaries [people who live in a household where at least one member receives the transfer] in a given quintile)/(total number of direct and indirect beneficiaries). The sum of percentages across quintiles per given instrument equals 100 percent. Quintiles are calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; Q = quintile. capita household income or consumption (ade- Program sets transfer levels based on the five- quacy ratio).18 The per-capita average transfer is year average price of cereals, whereas Zambia’s an absolute measure of benefit size and can Social Cash Transfer Program sets its benefits be  compared with social minimums, such as close to the price of a 50-pound bag of maize the poverty line or minimum wage. Benefits as (corn) monthly, which would allow a household a share of household welfare (income or to eat a second meal each day (Schüring 2010; ­ consumption), on the other hand, are a relative Garcia and Moore 2012). measure that allows the importance of the The size of the benefit is a determining factor transfers in proportion to household per capita to achieve positive impacts on household welfare to be assessed. well-being. The way program administrators set The level of benefits is set to achieve program benefit levels varies across countries; some pro- objectives within budget constraints. There are grams use flat benefits, whereas others adjust no standard rules to set benefits levels, given that benefits based on household size, number of they need to be calibrated to fulfill program objec- dependents, and so forth. Flat benefits raise the tives and meet budget constraints. Contributory issue that the per capita transfer will decrease old-age pensions, for example, are based on the with household size and program impacts will amount of contributions individuals make vary across beneficiary households. Inflation is during their active working life. In poverty reduc- another factor that can erode the real value of tion programs, the size of the transfers may be the transfers over time, unless regular adjust- calibrated to reduce the poverty gap of the target ment mechanisms are applied. In Kenya’s Cash population; programs that aim to address food Transfer Program for Orphans and Vulnerable security will set their benefits to meet nutritional Children, for example, the value of the transfer needs.19 For example, Kenya’s Hunger Safety Net decreased by almost 60 percent because of 50 The State of Social Safety Nets 2018 FIGURE 3.17  Distribution of School Feeding Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare Slovak Republic 2009 Latvia 2009 Uruguay 2012 Russian Federation 2016 Sri Lanka 2012 Morocco 2009 Peru 2014 Panama 2014 Belize 2009 Paraguay 2011 Costa Rica 2014 Chile 2013 Thailand 2013 Ecuador 2016 Colombia 2014 Honduras 2013 El Salvador 2014 Nicaragua 2014 Botswana 2009 Jamaica 2010 Bolivia 2012 Nigeria 2015 Guatemala 2014 Liberia 2014 Malawi 2013 Tanzania 2014 0 10 20 30 40 50 60 70 80 90 100 Percent Poorest Q2 Q3 Q4 Richest Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 26); Latin America and the Caribbean (n = 15); Sub-Saharan Africa (n = 5); Europe and Central Asia (n = 3); Middle East and North Africa (n = 1); East Asia and Pacific (n = 1); and South Asia (n = 1). School feeding programs encompass any type of meals or food items provided at school. Beneficiaries’ incidence is calculated as follows: (number of direct and indirect beneficiaries [people who live in a household where at least one member receives the transfer] in a given quintile)/(total number of direct and indirect beneficiaries). The sum of percentages across quintiles per given instrument equals 100 percent. Quintiles are calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; Q = quintile. inflation between 2007 to 2011 (Daidone et al. between the benefit amount needed to achieve 2016). Therefore, having flexibility to adjust objectives, needs for program coverage, and benefit levels is important to maximize positive the available budget. Given these consider- impacts on household well-being, granted there ations, the evidence gathered from impact is budget space to do so. evaluations suggests that most successful cash How large should a transfer be to achieve transfers programs, for example, transfer at meaningful impacts? The higher the share of percent of household consumption to least 20 ­ the transfers with respect to household welfare, beneficiaries (Handa et al. 2013). the greater the impacts on poverty and inequal- Empirical analysis indicates that the benefit ity are likely to be.20 Given budget constraints, level expressed as a share of beneficiary welfare however, larger transfers may imply fewer among recipients varies greatly across SPL beneficiaries. There are concerns that larger areas. Figures 3.19 and 3.20 show that on aver- transfer values can create disincentives to work. age the benefit level for social insurance pro- However, most impact evaluations have found grams is greater than the benefit level for SSN that transfers in general do not reduce labor programs. This is expected because social insur- supply, but they do influence the allocation of ance programs are designed to replace benefi- labor and time. Therefore, determining the size ciaries’ working earnings. The global average of the transfers is usually a delicate balance for social insurance programs as a share of Analyzing the Performance of Social Safety Net Programs 51 FIGURE 3.18  Distribution of In-Kind Transfer Beneficiaries, as Captured in Household Surveys, by Quintile of Pretransfer Welfare Djibouti 2012 Uruguay 2012 Lavia 2009 Vietnam 2014 Costa Rica 2014 Kosovo 2013 Turkey 2014 Croatia 2010 West Bank and Gaza 2009 Russian Federation 2016 Namibia 2009 Tunisia 2010 Nepal 2010 Nigeria 2015 Rwanda 2013 Sudan 2009 Peru 2014 Indonesia 2015 Panama 2014 Paraguay 2011 Georgia 2011 Kazakhstan 2010 Chile 2013 Sri Lanka 2012 Morocco 2009 Nicaragua 2014 El Salvador 2014 Côte d’lvoire 2014 Armenia 2014 Liberia 2014 Egypt, Arab Rep. 2008 Ecuador 2016 Romania 2012 Mauritania 2014 Guatemala 2014 Sierra Leone 2011 Iraq 2012 Haiti 2012 Ethiopia 2010 Tanzania 2014 China 2013 Senegal 2011 Malawi 2013 Niger 2014 South Sudan 2009 0 10 20 30 40 50 60 70 80 90 100 Percent Poorest Q2 Q3 Q4 Richest Source: ASPIRE database. Note: The number of countries per region is as follows: total (n = 45); Sub-Saharan Africa (n = 14); Latin America and the Caribbean (n = 11); Europe and Central Asia (n = 9); Middle East and North Africa (n = 6); East Asia and Pacific (n = 3); and South Asia (n = 2). In-kind transfers include any of the following: food aid; agricultural inputs; clothes; school supplies; and building materials. Beneficiaries’ incidence is calculated as follows: (number of direct and indirect beneficiaries [people who live in a household where at least one member receives the transfer] in a given quintile)/(total number of direct and indirect beneficiaries). The sum of percentages across quintiles per given instrument equals 100 percent. Quintiles are calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; Q = quintile. beneficiary welfare is 32 percent for the total population, increases with income, from population, whereas the share of SSN benefits is 18 percent in low-income countries to only 10 percent (see figure 3.19).21 49 ­percent in high-income countries. This is not The proportion of social insurance benefits the case for SSN programs, for which the rela- with respect to beneficiary welfare, for the total tive level of benefits differs less accros country 52 The State of Social Safety Nets 2018 FIGURE 3.19  Social Protection and Labor Transfer Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Total Population 60 49 50 40 39 32 31 Percent 30 20 18 10 11 11 10 9 7 0 World Low-income Lower-middle- Upper-middle- High-income countries income countries income countries countries Social assistance/social safety nets Social insurance Source: ASPIRE database. Note: The number of countries per country income group with monetary values for social assistance is as follows: total (n = 79), high-income countries (n = 5), upper-middle-income countries (n = 30), lower-middle-income countries (n = 30), and low-income countries (n = 14). The number of countries with monetary values for social insurance is as follows: total (n = 79), high-income countries (n = 6), upper-middle- income countries (n = 28); lower-middle-income countries (n = 29), and low-income countries ( n = 16). Transfers as a share of a beneficiary’s welfare can be generated only if monetary values are recorded in the household survey; for this reason, the sample of countries used in this figure is smaller than the one used to estimate coverage and beneficiary incidence. Labor market programs were not included because they encompass mostly active labor market programs for which only participatory variables (vs. monetary) are observed in the surveys. The sample of countries that include monetary variables (mostly for unemployment insurance) is too small to derive any meaningful conclusion (n = 18). The share of transfers is determined as follows: (transfer amount received by all direct and indirect beneficiaries in a population group)/(total welfare aggregate of the direct and indirect beneficiaries in that population group). Aggregated indicators are calculated using simple averages of country-level social assistance and social insurances transfers’ shares, across country income groups. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. income groups; SSN benefits as share of bene- poor, the share of social insurance benefits is ficiary welfare is 7 percent for high-­ income still higher than the share of SSNs, but there is countries, for example, whereas for low-­income not a clear correlation between the magnitude of countries and upper-middle-income countries this share and country income groups. For it is 11 percent (see figure 3.19). The relatively example, social insurance makes up 48 percent high level of SSN benefits for low-­ income of beneficiary welfare in low- and upper-­middle- countries could also reflect the fact that many income countries, higher than the share observed of the SSN programs in those countries are for lower-middle-income countries (39 percent). donor/externally funded. However, the avail- Likewise, SSN programs make  up 22 percent able data do not make it possible to test this of  beneficiary welfare in upper-middle-income hypothesis. countries, which is higher than the share SPL transfers make up a significant propor- in  ­lower-middle- and high-income countries tion of the welfare of individuals in the poorest (18 percent) (see figure 3.20). quintile. For the surveys included in ASPIRE, On average, SSN transfers account for the average share of the transfers in the welfare of 19 ­percent of the welfare of the poorest quintile. the poorest quintile is 46 percent for social insur- However, transfer levels vary greatly across SSN ance and 19 percent for SSNs (see figure 3.20), instruments and across countries. These differ- compared with 32 and 10 percent observed in ences reflect, in part, different program objec- the total population (see  figures  3.19). For the tives and the degree of transfer values captured Analyzing the Performance of Social Safety Net Programs 53 FIGURE 3.20  Social Protection and Labor Transfer Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile 80 70 68 60 48 48 50 46 Percent 39 40 30 22 19 18 18 20 13 10 0 World Low-income Lower-middle- Upper-middle- High-income countries income countries income countries countries Social assistance/social safety nets Social insurance Source: ASPIRE database. Note: The number of countries per country income group with monetary values for social assistance is as follows: total (n = 79), high-income countries (n = 5), upper-middle-income countries (n = 30), lower-middle-income countries (n = 30), and low-income countries (n = 14). The number of countries with monetary values for social insurance is as follows: total (n = 79), high-income countries (n = 6), upper-middle- income countries (n = 28), lower-middle-income countries (n = 29), and low-income countries (n = 16). Transfers as a share of a beneficiary’s welfare can be generated only if monetary values are recorded in the household survey; for this reason, the sample of countries used in this figure is smaller than the one used to estimate coverage and beneficiary incidence. Labor market programs were not included because they encompass mostly active labor market programs for which only participatory variables (vs. monetary) are observed in the surveys. The sample of countries that include monetary variables (mostly for unemployment insurance) is too small to derive any meaningful conclusion (n = 18). The share of transfers is determined as follows: (transfer amount received by all direct and indirect beneficiaries in a population group)/(total welfare aggregate of the direct and indirect beneficiaries in that population group). Aggregated indicators are calculated using simple averages of country-level social assistance and social insurance transfers’ shares, across country income groups. The poorest quintile is calculated using per capita posttransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. in household surveys. Figures 3.21 to 3.25 illus- various cash transfers implemented with differ- trate the proportion of SSN benefits with respect ent objectives may be aggregated: for example, to beneficiary welfare by type of SSN instru- poverty alleviation programs and universal ment and by country.22 Social pensions and family allowances. However, in a few countries, CCTs tend to make up, on average, a higher a single program is included in the UCT typol- proportion of beneficiary welfare—while the ogy; thus, results can be attributed to that par- share is much lower for public works, fee waiv- ticular program. ers, and targeted subsidies. However, this indi- Some UCT programs with a poverty allevia- cator varies greatly across countries within each tion objective tend to have a higher benefit level. SSN instrument. For example, UCT programs make up the largest On average, UCT transfers as a share of ben- share of the welfare of the poor (49 ­ percent) in eficiary welfare for the poorest quintile amount Georgia and Rwanda (­ figure 3.21). Several pro- to 19 percent. To estimate this indicator, 52 sur- grams are aggregated into the UCT categories veys with monetary values were used out of 63 for Georgia, but the results are mainly driven surveys with UCT information and out of a by the Targeted Social Assistance (TSA) pro- total of 79 surveys with monetary data for SSNs gram.23 In the case of Rwanda, the result cor- (see figure 3.21). As mentioned, in the case of responds to the Direct Support from the UCTs, this indicator could be imprecise because Vision 2020 Umurenge Programme (VUP), 54 The State of Social Safety Nets 2018 FIGURE 3.21  Unconditional Cash Transfer Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile Georgia 2011 Rwanda 2013 Maldives 2009 South Africa 2010 Bolivia 2012 Kosovo 2013 Armenia 2014 Belarus 2013 Indonesia 2015 Ukraine 2013 Mauritius 2012 Montenegro 2014 Serbia 2013 Poland 2012 Costa Rica 2014 Mongolia 2012 Jordan 2010 Belize 2009 Botswana 2009 Moldova 2013 West Bank and Gaza 2009 Jamaica 2010 Djibouti 2012 Colombia 2014 Kyrgyz Republic 2013 Albania 2012 Average, 18.6 Lithuania 2008 Romania 2012 Croatia 2010 Latvia 2009 China 2013 Mexico 2012 Egypt, Arab Rep. 2008 Iraq 2012 Slovak Republic 2009 El Salvador 2014 Liberia 2014 Russian Federation 2016 Tajikistan 2011 Pakistan 2013 Malaysia 2008 Sri Lanka 2012 Turkey 2014 Vietnam 2014 South Sudan 2009 Uganda 2012 Kazakhstan 2010 Nepal 2010 Chile 2013 Sudan 2009 0 10 20 30 40 50 60 70 80 90 100 Percent Source: ASPIRE database. Note: The number of countries per region with monetary values for unconditional cash transfers is as follows: total (n = 52); Europe and Central Asia (n = 20); Sub-Saharan Africa (n = 10); Latin America and the Caribbean (n = 8); East Asia and Pacific (n = 5); Middle East and North Africa (n = 5); and South Asia (n = 4). Unconditional cash transfers include any of the following: poverty alleviation and emergency programs; guaranteed minimum-income programs; and universal or poverty-targeted child and family allowances. They do not include social pensions or targeted subsidies in cash. Transfers as a share of a beneficiary’s welfare can be generated only if monetary values are recorded in the household survey; for this reason, the sample of countries used in this figure is smaller than the one used to estimate coverage and beneficiary incidence. The share of transfers is determined as follows: (transfer amount received by all direct and indirect beneficiaries in a population group)/(total welfare aggregate of the direct and indirect beneficiaries in that population group). Unconditional cash transfer average share is the simple average of unconditional cash transfers’ shares across countries. The poorest quintile is calculated using per capita posttransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. Analyzing the Performance of Social Safety Net Programs 55 the only UCT captured in the household Social pensions make up a higher proportion survey. On the other side of the spectrum, ­ of the welfare of the poor compared with other very little can be ascertained about Sudan, SSN instruments: 27 percent, on average. This where the household survey includes only a finding is expected because somewhat like con- single question about general government tributory pensions, social pensions are designed assistance, without specifying what type of to address the lack of earnings because of old assistance it may be. age and disability. Of 37 surveys with social On average, CCTs account for a 16 percent pension information, only 30 include transfer share of the beneficiary welfare of the poor. values that make it possible to calculate the Only 17 of 21 surveys with CCT information indicator. As shown in figure 3.23, the noncon- include a transfer value in the household survey tributory old-age pension, disability grants, that could be used to calculate this indicator. and the war veterans pension in South Africa Bolivia’s Bono Juancito Pinto and Bono Juana combined make up the largest proportion of Azurduy, combined, make up the largest share beneficiary welfare of the included countries of the beneficiary welfare (36 percent) and (77 percent), and Brazil’s Benefício de Prestação Honduras’ Bonos PRAF is second (32 p ­ ercent). Continuada follows (66 percent). The combined share of  CCTs aggregated for Public works programs have one of the low- Bangladesh is only 2.2  percent, the lowest of est shares among SSN programs with respect to all included countries (see figure 3.22). beneficiary welfare of the poorest quintile FIGURE 3.22  Conditional Cash Transfer Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile Bolivia 2012 Honduras 2013 Ecuador 2016 Mexico 2012 Argentina 2013 Brazil 2015 Paraguay 2011 Panama 2014 Peru 2014 Average, 15.6 Costa Rica 2014 Dominican Republic 2014 Colombia 2014 Chile 2013 Philippines 2015 Jamaica 2010 Timor-Leste 2011 Bangladesh 2010 0 20 40 60 80 100 Percent Source: ASPIRE database. Note: The number of countries per region with monetary values for conditional cash transfers is as follows: total (n = 17), Latin America and the Caribbean (n = 14), East Asia and Pacific (n = 2), South Asia (n = 1), Sub-Saharan Africa (n = 0), Middle East and North Africa (n = 0), and Europe and Central Asia (n = 0). Conditional cash transfer programs include the following: Argentina 2013: Asignación Universal por Hijo. Bangladesh 2010: Maternity allowance, Program for the Poor Lactating, Stipend for Primary Students (MOPMED), Stipend for drop out students, Stipend for Secondary and Higher Secondary/Female Student. Bolivia 2012: Bono Juancito Pinto and Bono Juana Azurduy. Brazil 2015: Bolsa Família. Chile 2013: Subsidio Familiar (SUF), Bono de Protección Familiar y de Egreso, Bono por control del niño sano, Bono por asistencia escolar and Bono por logro escolar. Colombia 2014: Familias en Acción. Costa Rica 2014: Avancemos. Dominican Republic 2014: Solidaridad Program and other transfers. Ecuador 2016: Bono de Desarrollo Humano. Honduras 2013: Asignaciones Familiares–Bonos PRAF and otro tipo de bonos. Jamaica 2010: Program of Advancement Through Health and Education (PATH)—Child 0–71 months, 6–17 years and pregnant and lactating women. Mexico 2012: Oportunidades. Panama 2014: Red de Oportunidades. Paraguay 2011: Tekopora. Peru 2014: Programa Juntos. Philippines 2015: Pantawid Pamilyang Pilipino Program (4Ps). Timor-Leste 2011: Bolsa da Mae. Transfers as a share of a beneficiary’s welfare can be generated only if monetary values are recorded in the household survey; for this reason, the sample of countries used in this figure is smaller than the one used to estimate coverage and beneficiary incidence. The share of transfers is determined as follows: (transfer amount received by all direct and indirect beneficiaries in a population group)/(total welfare aggregate of the direct and indirect beneficiaries in that population group). Conditional cash transfer average share is the simple average of conditional cash transfers’ shares across countries. The poorest quintile is calculated using per capita posttransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. 56 The State of Social Safety Nets 2018 FIGURE 3.23  Social Pensions’ Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile South Africa 2010 Brazil 2015 Maldives 2009 Georgia 2011 Mauritius 2012 Serbia 2013 Costa Rica 2014 Chile 2013 Panama 2014 Lithuania 2008 Paraguay 2011 Romania 2012 Botswana 2009 Montenegro 2014 Latvia 2009 Colombia 2014 Russian Federation 2016 Average, 27.3 Mexico 2012 Poland 2012 Rwanda 2013 Turkey 2014 Thailand 2013 Moldava 2013 Nepal 2010 Bangladesh 2010 Honduras 2013 Kazakhstan 2010 Tajikistan 2011 Sri Lanka 2012 Slovak Republic 2009 0 20 40 60 80 100 Percent Source: ASPIRE database. Note: The number of countries per region with monetary values for social pensions is as follows: total (n = 30), Europe and Central Asia (n = 13), Latin America and the Caribbean (n = 8), South Asia (n = 4), Sub-Saharan Africa (n = 4), East Asia and Pacific (n = 1), and Middle East and North Africa (n = 0). Social pensions include any of the following: noncontributory old-age pensions; disability pensions; and survivor pensions. Transfers as a share of a beneficiary’s welfare can be generated only if monetary values are recorded in the household survey; for this reason, the sample of countries used in this figure is smaller than the one used to estimate coverage and beneficiary incidence. The share of transfers is determined as follows: (transfer amount received by all direct and indirect beneficiaries in a population group)/(total welfare aggregate of the direct and indirect beneficiaries in that population group). Social pensions average share is the simple average of social pensions’ shares across countries. The poorest quintile is calculated using per capita posttransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. (7  percent). However, little can be concluded because they do not have monetary values in from the public works average indicator household surveys. because only 4 out of 10 surveys with public Fee waivers and targeted subsidies have the works information include monetary values lowest share of beneficiary welfare among SSN figure 3.24). This indicator cannot be esti- (see ­ programs: 6.7 percent, on average—much lower mated for some flagship public works pro- than the SSN global average of 19 percent (see grams (for example, the Productive Safety Net ­ figure 3.25). This finding is not surprising Program in Ethiopia or MGNREG in India), because fee waivers and targeted subsidies Analyzing the Performance of Social Safety Net Programs 57 FIGURE 3.24  Public Works’ Value Captured in Household Surveys as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile Mexico 2012 Rwanda 2013 Average, 6.9 Niger 2014 Malawi 2013 0 10 20 30 40 50 60 70 80 90 100 Percent Source: ASPIRE database. Note: The number of countries per region with monetary values for public works is as follows: total (n = 4), Sub-Saharan Africa (n = 3), Latin America and the Caribbean (n = 1), East Asia and Pacific (n = 0), Europe and Central Asia (n = 0), Middle East and North Africa (n = 0), and South Asia (n = 0). Public works programs include the following: Mexico 2012: Programa de Empleo Temporal. Malawi 2013: MASAF, PWP, Inputs for Work Program. Niger 2014: Public works. Rwanda 2013: Public works from the Vision 2020 Umurenge Program. Transfers as a share of a beneficiary’s welfare can be generated only if monetary values are recorded in the household survey; for this reason, the sample of countries used in this figure is smaller than the one used to estimate coverage and beneficiary incidence. The share of transfers is determined as follows: (transfer amount received by all direct and indirect beneficiaries in a population group)/(total welfare aggregate of the direct and indirect beneficiaries in that population group). Public works average share is the simple average of public works’ shares across countries. The poorest quintile is calculated using per capita posttransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. FIGURE 3.25  Fee Waivers and Targeted Subsidies Value Captured in Household Surveys, as a Share of Beneficiaries’ Posttransfer Welfare among the Poorest Quintile Poland 2012 Slovak Republic 2009 Mauritius 2012 Latvia 2009 Romania 2012 Ukraine 2013 Average, 6.7 Honduras 2013 Panama 2014 China 2013 Russian Federation 2016 Albania 2012 Tajikistan 2011 0 10 20 30 40 50 60 70 80 90 100 Percent Source: ASPIRE database. Note: The number of countries per region with monetary values for few waivers and targeted subsidies is as follows: total (n = 12); Europe and Central Asia (n = 8); Latin America and the Caribbean (n = 2); East Asia and Pacific (n = 1); Sub-Saharan Africa (n = 1); Middle East and North Africa (n = 0); and South Asia (n = 0). Fee waivers and targeted subsidies include any of the following: food, energy products, education, utilities, housing or transportation fees waivers to specific households, or discounted below the market cost. They do not include health benefits or subsidies. Transfers as a share of a beneficiary’s welfare can be generated only if monetary values are recorded in the household survey; for this reason, the sample of countries used in this figure is smaller than the one used to estimate coverage and beneficiary incidence. The share of transfers is determined as follows: (transfer amount received by all direct and indirect beneficiaries in a population group)/(total welfare aggregate of the direct and indirect beneficiaries in that population group). Fee waivers and targeted subsidies average share is the simple average of these programs’ shares across countries. The poorest quintile is calculated using per capita posttransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. 58 The State of Social Safety Nets 2018 included under this category are typically inequality. The reduction in poverty and aimed at helping the poor offset the cost of inequality is simulated by comparing beneficia- some services rather than support main earn- ries’ welfare recorded in the survey before and ings. In addition, the values captured in surveys after SSN transfers. Using information from and used to build this indicator are only a sub- surveys, the analysis computes what household set of this type of benefits: merely subsidies per capita income or consumption would be paid in cash. Many of these benefits are, by without SSN transfers (see box 3.2). In other design, in the form of fee waivers, and thus words, the transfer value is subtracted from the monetary values are not available in the sur- observed welfare (posttransfer welfare) to deter- veys. This indicator was calculated using only mine who would fall into poverty (using the 12 out of 24 surveys with fee waivers and tar- absolute or relative poverty lines) if the transfer geted subsidy information. is eliminated (pretransfer welfare). The propor- tion of individuals who are lifted out of poverty WHAT ARE THE POVERTY AND as a direct effect of the transfer is estimated and INEQUALITY IMPACTS OF SOCIAL illustrated in figure 3.26, along with the poverty SAFETY NET PROGRAMS? gap reduction.24 This section provides an analysis of the ability of On the basis of the information observed in SSN monetary transfers to reduce poverty and household surveys, the analysis shows that SSN BOX 3.2  Measuring the Impact of Social Protection and Labor Programs Household surveys provide a unique opportu- to the transfers. By replicating this calculation for nity to measure the impact of the social protec- the universe of households/individuals receiving tion and labor (SPL) programs on poverty or transfers and applying the population expansion inequality because they contain information on factors, the book is able to estimate the direct the household aggregated income or consump- poverty and inequality reduction rates from the tion (welfare). This information makes it possible transfers for any given country. to determine, in each country, who the poor are. This chapter uses three indicators to measure To allow international comparability, the Atlas of impacts on poverty and inequality: poverty Social Protection: Indicators of Resilience and headcount, poverty gap, and inequality reductions Equity (ASPIRE) database adopts two measures (in percent or relative terms). Poverty headcount of poverty: the international absolute poverty provides the number of people living under the line of US$1.90 a day per capita in purchasing poverty line in a given country. The poverty power parity (PPP) terms; and a relative poverty headcount reduction is the percentage reduction line set at the bottom 20  percent of the pre- in poverty because of the transfer. However, not transfer welfare distribution. all program beneficiaries become nonpoor after This chapter measures impacts on poverty receiving a transfer; this depends on their position and inequality by comparing households’ per under the poverty line and if the transfer helped capita welfare before and after the transfers. the beneficiary reach or surpass the poverty For example, assume that the threshold for line. For those who do not overcome poverty, the the bottom 20 percent in a given country is simple headcount does not provide information US$2.50 PPP per capita per day. The survey on how much “less poor” beneficiaries are after reports that a household has a per capita the transfer. For this reason, the poverty gap is income (or consumption) of US$3.00 PPP a day, an important indicator to estimate the depth of which includes a transfer of US$1.00 PPP a day. poverty or how far below, on average, the welfare The US$3.00 PPP welfare level constitutes the of poor individuals is from the poverty line. The posttransfer welfare for that household. Then poverty gap reduction thus provides information the analysis subtracts the transfer from that on the  ability of SPL programs to bridge this welfare (3.00 − 1.00 = 2.00). The US$2.00 PPP a gap. The Gini coefficient measures the inequality day amount constitutes the pretransfer welfare. among values of the welfare distribution. Because US$2.00 is lower than the quintile Inequality reduction thus provides the percentage threshold of US$2.50, it can be determined that reduction in the Gini index because of the SPL this household was lifted out of poverty thanks transfer (Yemtsov et al., forthcoming). Analyzing the Performance of Social Safety Net Programs 59 transfers are making a substantial contribution These results are remarkable considering that in the fight against poverty. Whether an abso- these figures are underestimated because house- lute poverty line (measured as US$1.90 per cap- hold surveys do not capture the whole universe ita per day in purchasing power parity [PPP] of SSN programs implemented in those coun- terms) or a relative poverty line (measured in tries. Therefore, it can be inferred that the real terms of the poorest 20 percent) is used, the impacts are likely to be even larger. analysis suggests that individuals are escaping The reductions in the poverty headcount, poverty or decreasing their depth of poverty poverty gap, and inequality by SSN transfers because of the SSN transfers.25 For the 79 coun- are  observed in all country income groups. tries that have monetary information, transfers In  the sample of 79 surveys, the relative pov- reduce the incidence of absolute p ­overty erty  headcount is reduced by 8 percent, the (US$1.90 PPP per day) by 36 percent, whereas ­poverty gap by 16 percent, and the Gini inequal- relative poverty (the bottom 20  ­ percent) is ity index by 2 percent (see figure 3.27). Across reduced by 8 percent (see figure 3.26). country income groups, the average reduc- On average, SSN transfers are reducing the tion  in relative poverty headcount is only poverty gap more than the poverty headcount. 2 ­ income countries, 7 percent in percent in low-­ In other words, even if SSN transfers are not lower-middle-­­income countries, 11 percent in lifting the poor and near-poor above the pov- upper-middle-income countries, and 15 per- erty line, they significantly reduce the poverty cent in high-income countries. In terms of gap. As shown in figure 3.26, SSN transfers the poverty gap, the average reduction is 3 per- reduce the absolute poverty gap by 45 percent cent for low-income countries,  14 percent for and the relative poverty gap by 16 percent. lower-­middle-income countries, 21 percent for FIGURE 3.26  World Reductions in Poverty from Social Safety Net Transfers, as Captured in Household Surveys, as a Share of Pretransfer Indicator Levels, by Relative and Absolute Poverty Lines 50 45 40 36 30 Percent 20 16 10 8 0 Poverty headcount Poverty gap Poverty headcount Poverty gap reduction reduction reduction reduction Under US$PPP 1.90 per day Poorest quintile Source: ASPIRE database. Note: The number of countries per region with monetary values for social safety nets is as follows: world (n = 79); Sub-Saharan Africa (n = 23); Europe and Central Asia (n = 20); Latin America and the Caribbean (n = 16); East Asia and Pacific (n = 10); Middle East and North Africa (n = 6); and South Asia (n = 4). This figure uses a relative measure of poverty defined as the poorest 20 percent of the welfare distribution (income or consumption) and absolute measure of poverty defined as US$1.90 PPP per day. Impacts on poverty and inequality can be estimated only if monetary values are recorded in the household survey; for this reason, the sample of countries used in this figure is smaller than the one used to estimate coverage and beneficiary incidence. Percentages of poverty and inequality reduction are calculated as follows: (poverty headcount pre-transfer – poverty headcount posttransfer)/(poverty headcount pretransfer). Same calculations apply for poverty gap percentage reductions. Aggregated indicators are calculated using simple averages of country-level percentage reductions of the indicator across country income groups. The reductions in poverty are underestimated because ASPIRE does not include data for every single country in the country income groups, and even for a given country the survey does not include all existing social safety net programs or provide monetary values for them. For example, India and the impact of its flagship program MNREGA are not included in the calculation because only participatory information is available for the program. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; PPP = purchasing power parity. 60 The State of Social Safety Nets 2018 FIGURE 3.27  Reductions in Poverty and Inequality from Social Safety Net Transfers, as Captured in Household Surveys, as a Share of Pretransfer Indicator Levels, by Country Income Group Using Relative Poverty Line 35 30 30 25 21 20 Percent 16 14 15 15 11 10 8 7 5 5 3 3 2 2 2 0.2 0 World Low-income Lower-middle- Upper-middle- High-income (n = 79) countries income countries income countries countries (n = 14) (n = 30) (n = 30) (n = 5) Inequality reduction Poverty headcount reduction Poverty gap reduction Source: ASPIRE database. Note: The total number of countries per country income group included in the analysis appears in parentheses. This figure uses a relative measure of poverty defined as the poorest 20 percent of the welfare distribution (income or consumption). Impacts on poverty and inequality can be estimated only if monetary values are recorded in the household survey; for this reason, the sample of countries used in this figure is smaller than the one used to estimate coverage and beneficiary incidence. Percentages of poverty and inequality reduction are calculated as follows: (poverty headcount pretransfer – poverty headcount posttransfer)/(poverty headcount pretransfer). The same calculations apply for the Gini index and poverty gap percentage reductions. Aggregated indicators are calculated using simple averages of country-level percentage reduction of the indicator across country income groups. The reductions in poverty and inequality are underestimated because ASPIRE does not include data for every single country in the country income groups, and even for a given country the survey does not include all existing social safety net programs or provide monetary values for them. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. upper-middle-­ income countries, and 30 percent lower than in other country income groups.26 for high-­income countries. The Gini inequality See appendix F.3 for a list of poverty and inequal- index is less affected by SSN transfers, but ity reductions from SSN programs by country. reductions are still observed, ranging from 0.2 percent in low-­income countries to 5 percent WHAT FACTORS AFFECT THE IMPACT OF in high-­income countries. SOCIAL SAFETY NET TRANSFERS ON The lower reduction in poverty and inequality POVERTY AND INEQUALITY? observed for low-income countries is likely The extent to which SSN transfers have an impact driven by several factors. First, fewer low-­income on poverty and inequality depends on factors countries have recent household survey data such as program coverage, transfer level, and the available compared with other country income beneficiary/benefit incidence. Policy makers groups. Second, of the 22 low-income country need to pay attention to the interaction of these surveys included in ASPIRE, only 14  surveys factors when designing policies to reduce pov- include monetary variables for SSN programs. erty/inequality. Figures 3.28 and 3.29 explore the Third, many low-income country surveys nei- reductions in poverty and inequality achieved by ther capture SSN-specific program information each country, given their degree of coverage of nor include the universe of programs that exist the poor and benefits levels. in the country. And fourth, less than 10 percent The analysis reveals, in general, that very of the global population live in low-income high  coverage levels paired with high benefit countries; therefore, the number of individu- ­ levels lead to higher outcomes in poverty and als moving out of poverty (in percent terms) is inequality reduction. For example, Georgia and Analyzing the Performance of Social Safety Net Programs 61 South Africa display the highest poverty head- allowances are among the SSN benefits with count reduction using the poorest quintile as high coverage of the poor; however, social pen- the poverty measure (see figure 3.28). Georgia’s sions constitute higher shares of beneficiary combination of high SSN coverage (93 percent welfare. In both countries, the cost of  SSN of the poorest quintile) and high level of bene- programs expressed as a percentage of GDP is fits (SSN transfers constitute 68 percent of the rather large (7 percent for Georgia and poor’s welfare) leads to the highest poverty 3.3 percent for South Africa). headcount reduction, nearly 43 percent. The Conversely, very low coverage levels paired universal old-age social pension drives these with low levels of benefits lead to negligible results because it covers 81 percent of the poor- results in reducing poverty and inequality. For est quintile and constitutes 56 percent of their example, figure 3.28 shows how Armenia, with a total welfare. The five programs included under combination of lower SSN coverage of the poor cash transfers programs for Georgia—­including and benefits level (46 and 32 percent, respec- the Targeted Social Assistance program—also tively), achieves a more modest reduction in the cover a high percentage of the poorest quintile poverty headcount (12 percent). SSNs in Liberia, (46 percent) and make up 49 percent of the with a much more modest combination of cov- beneficiary welfare. Likewise, South Africa also erage of the poor and ­ benefit level (10 and shows high coverage and benefit levels for the 17 percent, respectively), achieve a small ­poverty poor (96 and 72 percent, respectively), leading headcount reduction (2.5 percent). The estimated to a poverty headcount reduction of 40 percent. poverty headcount reduction for Chad is almost In the South African survey, family and other negligible (0.1 percent) because scholarships are FIGURE 3.28  Poverty Headcount Reduction from Coverage and Level of Social Assistance Benefits for the Poorest Quintile, as Captured in Household Surveys 100 Transfers as a share of beneficiary welfare (%) The size of the bubble indicates percentage of poverty Georgia, 80 headcount reduction. 42.6 South Africa, 40.0 60 40 Chad, 0.1 Armenia, 11.8 Liberia, 20 2.5 0 0 20 40 60 80 100 120 Coverage of poorest quintile (%) Low-income countries Lower-middle-income countries Upper-middle-income High-income countries countries Source: ASPIRE database. Note: The number of countries with monetary values for social safety nets is as follows: world (n = 79), high-income countries (n = 5), upper- middle-income countries (n = 30), lower-middle-income countries (n = 30), and low-income countries (n = 14). This figure uses a relative measure of poverty defined as the poorest 20 percent of the welfare distribution (income or consumption). Poverty headcount reductions can be estimated only if monetary values are recorded in the household survey; for this reason, the sample of countries used in this figure is smaller than the one used to estimate coverage and beneficiary incidence. Percentages of poverty headcount reduction are calculated as follows: (poverty headcount pretransfer – poverty headcount posttransfer)/(poverty headcount pretransfer). Poverty headcount reductions are underestimated because ASPIRE does not include data for every single country in the country income groups, and even for a given country the survey may not include all existing social safety net programs or provide monetary values for those programs. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. 62 The State of Social Safety Nets 2018 the only SSN program captured in the survey. (27  percent), producing a more modest esti- These scholarships have very low coverage of the mated poverty gap reduction (42 percent). In poor (0.2 percent), although their contribution Montenegro, SSN programs provide modest to beneficiary welfare is 23 percent. coverage of the poor (25 percent) and repre- In terms of the poverty gap reduction, the sent  a modest share of beneficiary welfare same interplay between coverage and benefit (28 percent); thus, Montenegro achieves a mod- size is observed. In third place behind est poverty gap reduction (23 percent). In the Georgia  and South Africa, Mauritius also Maldives, the household survey includes only shows  high coverage and benefit levels for the two SSN programs: unspecified government poorest quintile (84 percent and 55 percent, transfers and social pensions. Even though respectively), leading to a poverty gap reduction social pensions have a high benefit level, driving of 61 percent. The survey includes monetary the Maldives’ SSN average benefit to 76 percent information for eight programs, of which the of the beneficiary welfare, the coverage of both noncontributory basic retirement pension has types of programs is small (15 percent) and thus the highest coverage of the poor. Mauritius also the poverty gap reduction is modest (28 per- reports high social spending according the cent). In Burkina Faso, the poverty gap reduc- administrative database (3.5 percent of GDP). tion is 0.1 percent, mostly because the survey In Poland, SSN programs captured in the house- does not capture a monetary value for govern- hold surveys report a relatively high level of cov- ment transfers, meaning that the estimation erage of the poor (65 percent), but the transfer rests only on information on scholarships and as a share of beneficiary welfare is smaller other general transfers (figure 3.29). FIGURE 3.29  Poverty Gap Reduction from Coverage and Level of Social Assistance Benefits for the Poorest Quintile, as Captured in Household Surveys 100 Transfers as a share of beneficiary welfare (%) Maldives, The size of the bubble indicates 27.9 percentage of poverty 80 gap reduction. 60 Mauritius, 60.9 Burkina Faso, Montenegro, 40 0.1 23.1 Poland, 41.6 20 0 0 20 40 60 80 100 120 Coverage of poorest quintile (%) Low-income countries Lower-middle-income countries Upper-middle-income countries High-income countries Source: ASPIRE database. Note: The number of countries with monetary values for social safety nets is as follows: world (n = 79); high-income countries (n = 5); upper-middle-income countries (n = 30); lower-middle-income countries (n = 30); and low-income countries (n = 14). This figure uses a relative measure of poverty defined as the poorest 20 percent of the welfare distribution (in terms of income or consumption). Poverty gap reductions can be estimated only if monetary values are recorded in the household survey; for this reason, the sample of countries used in this figure (n = 79) is smaller than the one used to estimate coverage and beneficiary incidence (n = 96). Percentages of poverty gap reduction are calculated as follows: (poverty gap pretransfer – poverty gap post transfer)/(poverty gap pretransfer). Poverty gap reductions are underestimated because ASPIRE does not include data for every single country in the country income groups, and even for a given country, the survey may not include all existing social safety net programs or provide monetary values for those programs. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. Analyzing the Performance of Social Safety Net Programs 63 NOTES 11. In many countries, elderly people tend to live within 1. SPL programs encompass social safety nets/social larger extended multigenerational families, whose assistance, social insurance, and labor market members benefit indirectly from social pensions. programs. Moreover, these pensions provide sizable benefits, and because the measure of welfare used in this book 2. For example, the household survey for El Salvador is pretransfer, many of the recipient families of social includes programs under the three areas of SPL. In pensions tend to concentrate in the lowest quintile this case, even though the most iconic programs in once this benefit is subtracted. This helps explain the the country such as the CCT and social pensions are high coverage of all the poor by social pensions in not adequately captured in the household survey, the some countries featured in figure 3.6. in-kind transfers programs as well as the school feed- ing program drive the coverage of SSN programs up 12. For this analysis, universal health schemes and uni- to 53 percent of the total population and 72 percent versal general price subsidies are not included (most of the poorest quintile. fuel subsidies or price support schemes for food are excluded). Instead, for commodities, only those sub- 3. A source of potential bias is the fact that household sidies that are targeted to specific individuals and surveys do not collect information from all social families are included. One form of targeted subsidy is protection programs implemented in a country, but a lifeline (lower) tariff for a limited quantity of elec- big flagship programs are likely to be captured. tricity, which is available only for eligible groups (as 4. A factor influencing this low coverage is that very lit- in Albania or the Russian Federation). tle information on labor market interventions is cap- 13. The Rastra Program, formerly called Beras untuk tured in most household income and expenditure Rakyat Miskin (Raskin), is a rice subsidy introduced surveys; thus, it is difficult to draw meaningful con- in 1998 as an emergency food security program. clusions regarding this program type. It delivers rice for purchase at subsidized prices and 5. The terms “social safety nets” and “social assistance” prioritizes poor and near-poor households (World are used interchangeably in this book. Bank 2012). Even though the program reports the highest coverage of the poor, the benefit has been 6. See http://datatopics.worldbank.org/aspire/~​ diluted to 4 kg per household on average from the /­documentation. promised 15 kg. In light of this, the government is 7. Aggregated variables with too few observations proceeding with significant reforms. (less  than 0.5 percent of the survey sample) were 14. Benefits incidence is not included in this analysis excluded because they are not representative to derive because information on the transfer value is needed meaningful results. For example, the Livelihood to estimate it. Because not all the surveys include the Empowerment Against Poverty (LEAP), Ghana’s monetary value of the transfers, the report prefers to UCT program, was excluded under this criterion. use a larger sample of surveys by analyzing only the 8. Under ASPIRE, UCTs encompass interventions such distribution of beneficiaries. If the value of the trans- as poverty alleviation or emergency programs, guar- fers is similar across beneficiaries, the distribution of anteed minimum-income programs, and universal beneficiaries and benefit across income groups will or poverty-targeted child and family allowances; be similar. UCTs do not include social pensions, public works, 15. Whether a program is propoor can be determined or targeted subsidies paid in cash. by  the Coady-Grosh-Hoddinott (CGH) indicator, 9. The Child Money Program was introduced in which divides the share of the beneficiaries/benefits Mongolia as a targeted program in 2005 and became belonging to a certain population group (for exam- universal in July of 2006. Although the program was ple, the bottom 20 percent) by the population share discontinued temporarily, it was reintroduced in of this group. If the result is higher than 1, the pro- October 2012, with a benefit level of US$14.72 per gram is considered progressive toward this group. month per child (Yeung and Howes 2015). For instance, if the bottom 20 percent of the popula- tion accounts for 40 percent of beneficiaries, this 10. The BR1M program was launched in 2012 to p ­ rovide a indicator will be equal to 2, indicating that the pro- single transfer to the poor. Households earning below gram focuses on the poor. In the case of a universal RM3,000 monthly receive RM1,200; households earn- program where benefits are equally distributed, the ing between RM3,001 amd RM4,000 monthly receive CGH indicator will be 1. RM900; and single individuals, 21  years and older, earning less than RM2,000 per month, receive RM450 16. It is particularly interesting to observe that some (BR1M’s ­ official website: https://ebr1m.hasil.gov.my). ­programs with similar design features and institutional The Malaysian Household Income Survey does not structures—such as the Family Material Support (MOP) identify the name of the cash transfer program in and Family Allowances Program in Montenegro the  questionnaire. However, it is likely capturing the and Serbia, which have the same legacy of the Federal BR1M program. Republic of Yugoslavia, and preserve some core 64 The State of Social Safety Nets 2018 features—have quite different distributions of benefi- 26. There are 3 billion people living in lower-middle-­ ciaries. That implies that implementation matters a lot income countries and only 659 million in low-­ in how well the program is capable of targeting the income countries. In lower-middle-income countries, poor. 500 million people live below the US$1.90 PPP a day threshold, which is more than 75 percent of the total 17. In the case of Djibouti, these food distribution pro- population of low-income countries. In India alone, grams focus on rural areas where over half the popu- 285 million people are living on less than US$1.90 lation lives in poverty, according to the survey. PPP a day, as well as 120 million in Nigeria, and 49 18. Transfers as a share of beneficiaries’ welfare is esti- million in China. In other words, most of the world’s mated as the amount of transfers received by a quin- poor live outside low-income countries. tile divided by the total income or consumption of beneficiaries in that quintile. REFERENCES 19. The poverty gap is the distance below the poverty line Andrews, C., C. Del Ninno, C. Rodriguez Alas, and and the average income of the poor. It is typically K. Subbarao. 2012. Public Works as a Safety Net: expressed as the percentage shortfall in income of the Design, Evidence, and Implementation. Directions in poor with respect to the poverty line. Development. Washington, DC: World Bank. 20. At the same time, the greater is the concern about ASPIRE (Atlas of Social Protection: Indicators of potential disincentives to work. Resilience and Equity). 2017. Database, World Bank,  Washington, DC. http://datatopics.worldbank​ 21. To assess benefit levels, information on the actual value .org​/­aspire/. of the transfers is needed. Therefore, only 79 out of 96 surveys available in ASPIRE were analyzed because Daidone, S., S. Asfaw, B. Davis, S. Handa, and P. Winters. not all surveys include monetary information. 2016. “The Household and Individual-Level Economic Impacts of Cash Transfer Programmes in Sub-Saharan 22. This analysis was done only for the types of ­ programs Africa. Synthesis Report.” Food and Agricultural that involve a monetary transfer. Thus, the indicator Organization of the United Nations, Rome. was not calculated for school feeding programs nor Garcia, M., and C. Moore. 2012.  The Cash Dividend: The for in-kind transfers. The sample of countries is also Rise of Cash Transfer Programs in Sub-Saharan different per SSN instrument, compared with the Africa.  Directions in Development.  Washington, DC: sample used for coverage and beneficiary incidence, World Bank. depending on the availability of data about the trans- fer value in the survey for each of the SSN instruments Handa, S., M. Park, R. Osei Darko, I. Osei-Akoto, B. Davis, analyzed. and S. Daidone. 2013. “Livelihood Empowerment against Poverty Program: Impact Evaluation.” Carolina 23. These programs are Internally Displaced Persons Population Center, University of North Carolina at Assistance, social assistance to multichildren fami- Chapel Hill. lies, social assistance to orphans, social assistance to  children with disabilities, and Targeted Social Schüring, E. 2010. “Strings Attached or Loose Ends? The Assistance (TSA). Role of Conditionality in Zambia’s Social Cash Transfer Scheme.” Maastricht Graduate School of Governance, 24. Therefore, this is a first-level approximation of the Maastricht, Netherlands. direct impact of the monetary transfers observed in the surveys. It also does not assess any medium- or long- World Bank. 2012. “Raskin Subsidized Rice Delivery. term effects of SSN transfers on their specific objectives, Social Assistance Program and Public Expenditure which are better measured by impact evaluations. Review No. 3.” World Bank, Washington, DC. Yemtsov, R. M. Honorati, B. Evans, Z. Sajaia, and 25. The absolute poverty line of US$1.90 PPP a day may M. Lokshin. Forthcoming. “An Analytical Approach to not constitute a meaningful standard for many high-­ Assessing Social Protection Effectiveness: Concepts and income countries that have a very small population, Applications.” World Bank, Washington, DC. or may not even have anybody living under that threshold. In fact, countries such as Belarus, Croatia, Yeung, Y., and S. Howes. 2015. “Resources-to-Cash: Lithuania, Montenegro, and Romania are able to fully A Cautionary Tale from Mongolia.” Development Policy eradicate their poverty headcount at US$1.90 PPP a Centre Discussion Paper #42, Crawford School of Public day, according to information provided by household Policy, The Australian National University, Canberra., surveys (see appendix F). Australia. Analyzing the Performance of Social Safety Net Programs 65 HIGHLIGHT 1: Productive Outcomes of Social Safety Net Programs: Evidence from Impact Evaluations in Sub-Saharan Africa A defining characteristic of the poor in Sub- promote dependency and induce wasteful con- Saharan Africa is they are trapped in sumption. On the contrary, cash transfers have low-­ productivity or low-paying jobs. An economic impacts beyond their intended objec- analysis using a sample of 28 low-income-country tives that strengthen beneficiary livelihoods and, household surveys from the Atlas of Social if complemented with other instruments, can be Protection: Indicators of Resilience and Equity leveraged to produce long-lasting benefits for (ASPIRE) found that 73 percent of individuals in beneficiaries, paving their way out of poverty. the poorest 20 percent are employed, of which 69 percent work in agriculture.1 Of those employed, WHAT ARE THE ECONOMIC IMPACTS OF 49 percent are self-employed, whereas 30 ­ percent CASH TRANSFERS ON BENEFICIARIES are unpaid workers. Furthermore, many of the AND THEIR HOUSEHOLDS? poor live in a context of poorly functioning or Household Production nonexistent labor and insurance/credit markets Impact evaluations indicate that cash transfers that affect household economic decisions (Handa increase household crop production, lead to et al. 2017). In this context, cash transfers may changes in types of crops cultivated, and increase help households overcome these market ­ failures consumption and sales of homegrown produc- and may enable them to spend more and to make tion. The evaluation of programs in seven Sub- productive investments. This in turn may gener- Saharan African countries found significant ate household-level multiplier effects (Daidone impacts in all these areas, even though their et al. 2016), as well as spillover and income mul- magnitude varied across countries  (Daidone tiplier effects in local communities (Daidone et al. et  al. 2016). Crop production increased in 2015; Thome et al. 2016). Zambia and Lesotho. The value of the overall This highlight summarizes new research on the production in Zambia almost doubled, boosting broader impacts of unconditional cash transfers, postprogram per capita consumption to a level particularly regarding economic impacts and 25 percent higher than the transfer itself (Davis productive inclusion in the context of Sub- et al. 2016). In Ethiopia, Malawi, and Zimbabwe, Saharan Africa. The discussion draws largely on transfers led to changes in the types of crops papers that synthesize and analyze the results cultivated. In Kenya and Malawi-Mchinji, the from evaluations in eight Sub-Saharan African cash transfer increased consumption of the countries as part of The Transfer Project2 and the home-grown production (Davis et al. 2016). From Protection to Production (PtoP) Project.3,4 The narrative is complemented with the results Consumption and Productive Investments of a metadata analysis of the impact evaluations Robust evidence indicates that households use available for 20 programs in 10 African coun- transfers for basic needs and productive invest- tries (Ralston, Andrews, and Hsiao 2017); as well ments, dispelling myths of profligate spending. as a  systematic review of 56 cash transfers pro- A meta-analysis conducted on impacts for seven grams (including unconditional cash transfers, African countries found that on average, house- conditional cash transfers, noncontributory social hold consumption increases by US$0.74 for pensions, and enterprise grants) in low- and each US$1 transferred (Ralston, Andrews, and middle-income countries from 2000 to 2015 Hsiao 2017). The magnitude of this impact var- (Bastagli et al. 2016). ies across countries, with the largest impact Findings from these evaluations consistently experienced by programs targeted at the poor: show statistically positive impacts of cash transfer the Malawi Social Cash Transfer Program programs on productive outcomes such as crop reports the largest consumption impact, at production, productive investments, employ- 179 percent of the transfer value. In addition, a ment, and more effective risk-coping mecha- review of 19 programs and 11 studies using data nisms. The findings include income-multiplier from Africa, Asia, and Latin America found no effects for beneficiary and nonbeneficiary house- evidence that transfers, conditional or uncondi- holds and local economies. They debunk miscon- alcohol and tobacco tional, increase the use of ­ ceptions like cash transfers being handouts that (Evans and Popova 2014). 66 66 The State of Social Safety Nets 2018 The review by Bastagli et al. (2016) found that et al. 2015). In the context of informal rural labor cash transfers improved household consump- markets, casual wage labor is a last-resort activity tion in 25 of 35 studies looking at this effect. that households use to survive or when liquid Their review also found an increase in livestock funds are scarce. Therefore, for beneficiaries to ownership/purchase, agricultural assets and increase work and time on their own farms is a inputs, and savings (although not for all pro- preferred activity and a sign of improved eco- grams or for all types of livestock, assets, or nomic conditions. These findings are consistent inputs). The Transfer Project found that five out with other studies that have concluded that trans- of eight programs had significant impacts on the fers do not reduce the labor supply or create increase in livestock ownership. The effect of dependency. For example, Banerjee et al. (2015), households investing in diverse types of animals after conducting seven randomized control tri- was large in Malawi and Zambia, whereas more als  of cash transfers programs in six coun- limited effects were observed in Kenya, Lesotho, tries  (Honduras, Indonesia, Mexico, Morocco, and Zimbabwe, where small livestock were Nicaragua, and the Philippines), found no evi- acquired. Impacts were not found in Ghana. The dence that cash transfers have impacts on either meta-data analysis by Ralston, Andrews, and the propensity to work or the overall number of Hsiao (2017) confirms these findings by  esti- hours worked. mating a combined average increase of 34 per- cent in livestock ownership across ­ programs in Risk Management and Coping Strategies seven countries (four were significant). Cash transfers, impact evaluations find, allow Most programs had significant impacts on beneficiary households to manage risk more the purchase/use of agricultural inputs such as effectively  by  diversifying income-generating seeds, fertilizer, and pesticides, although the activities, increasing savings, and reducing detri- magnitude of these impacts varied across coun- mental coping strategies. In Zambia, the share of tries. In terms of agricultural assets (for exam- beneficiary households running nonfarm enter- ple, axes, hoes, picks, and other tools), positive prises increased 16  ­ percentage points, and the impacts were observed in Ethiopia, Malawi, businesses reported 1.4 more months in opera- Zambia, and Zimbabwe. Impacts were not tion, compared with the control group, after observed in Kenya, Lesotho, and Ghana. Even receiving cash transfers. In Zimbabwe, there was though all impacts were not always significant, an increase of 5 percentage points in the share of all countries reported positive significant results households operating these businesses and an for population subgroups, type of animal, or increase of 5 percentage points in the share asset (Daidone et al. 2016; Davis et al. 2016; reporting profits. Other countries do not report Handa et al. 2017). significant results (Daidone et al. 2016). In Ghana, a qualitative evaluation finds evidence of increased Labor and Time Allocation petty trading of small amounts of kerosene and Impact evaluations generally show no evidence the sale of cooked food (Barca et al. 2015). that cash transfers reduce the labor s ­ upply. In the Cash transfers often lead to increasing savings. African context, transfers have an impact on Ghana and Zambia reported increases in savings household decision making on labor allocation of 11 and 24 percentage points, respectively. In and time use, in terms of switching between dif- Ethiopia, Ghana, and Malawi transfers led to a ferent income-­ generating activities or between reduction of loans and debt repayments (Daidone labor, domestic tasks, or leisure (Handa et al. et al. 2016). In Zambia and Zimbabwe, transfers 2017). In other words, cash transfers give house- contributed to an increase in households’ credit- holds flexibility to allocate time and labor to these worthiness; however, households were still risk activities, leading to a switch from casual agricul- averse and reluctant to take on new credit (Davis tural labor to on-farm labor (David et al. 2016). et al. 2016). Ralston, Andrews, and Hsiao (2017) As a female beneficiary in Malawi expressed with found that beneficiary households are 20 percent- respect to doing casual labor or ganyu: “I used to age points more likely to save compared with be a slave to ganyu but now I’m a bit free” (Barca the control group; this translates into an average Analyzing the Performance of Social Safety Net Programs 67 HIGHLIGHT 1: Productive Outcomes of Social Safety Net Programs: Evidence from Impact Evaluations in Sub-Saharan Africa increase of 92 percent in the number of house- nominal income multipliers. Using the LEWIE holds setting aside savings. model, income multipliers  range from 1.27 in Studies also report a decrease in negative Malawi to 2.52 in Ethiopia (Hintalo region). The risk-coping strategies, such as distress sales of income spillover (­ multiplier minus 1) indicates assets, begging, eating less, or putting children to that for every US$1 transferred to the benefi- work. Ethiopia, Lesotho, and Malawi reported ciary household, the  local economy gains an less begging and changes in eating habits. additional US$0.27–1.25 (Thome et al. 2016). Beneficiary households were less likely to take After considering potential inflation, simula- children out of school in almost all the countries tions find that the real income multiplier, even analyzed by Davis et al. (2016) and Daidone et al. though lower than the nominal multiplier, is still (2016). As an elderly beneficiary in Ethiopia greater than 1 for all seven countries. If producers said: “Hunger pushed me to do this [beg]. Since in the local economy face constraints to increase I started to receive the cash transfer I no longer production in response to the higher demand of have to. I feel happier” (Barca et al. 2015). goods led by the transfers, this may put upward Qualitative evaluations also show that cash pressure on prices and lower income multipliers. transfers programs increase social capital. They However, even after accounting for inflation, help beneficiaries reenter social networks, income multipliers are still greater than 1, ranging strengthening informal social protection systems from 1.08 in Kenya (Nyanza Province) to 1.81 in and risk-sharing management arrangements Ethiopia (Hintalo). The elasticity of the supply of (Davis et al. 2016). The very poor faced fewer stig- local goods largely drives the differences between mas, participated more fully in the community, real and nominal multipliers. In economies that and supported other households or institutions can easily increase the supply of goods, the price because of the transfers. increase is small and nominal and real multipliers are similar (Thome et al. 2016). Spillover Effects in Local Economies Cash transfer programs not only bring economic impacts to beneficiary individuals and house- NOTES holds but also can create spillover effects that 1. ASPIRE generates country context indicators using benefit nonbeneficiary households and local 129 surveys standardized by the I2D2 Project economies. When beneficiaries spend transfers, (International Income Distribution Database). cash is injected into local markets, potentially 2. The Transfer Project is a joint effort of the United creating income multipliers. That is, for each Nations Children’s Fund (UNICEF), Save the Children, US$1 transferred to a beneficiary household, the and the University of North Carolina to support the total income of the local economy may increase implementation of impact evaluations of cash transfer programs in Sub-Saharan Africa. It focuses on the by more than US$1. These spillovers are difficult broad range of impacts of government-run cash trans- to measure by experimental methods because fer programs in Sub-Saharan Africa. they are second-order impacts that are diffused 3. The From Protection to Production (PtoP) Project is over a population greater than the beneficiary part of the Transfer Project and focuses on exploring population (Thome et al. 2016). However, the linkages between social protection, agriculture, researchers developed the Local Economy-Wide and rural development; http://www.fao.org/eco​ Impact Evaluation (LEWIE) to simulate the nomic​/ptop/home/en/. effect of cash transfers in local economies. 4. Countries and cash transfer programs included under LEWIE measures the impact of cash transfers on the PtoP Project include the following: Ethiopia: the production activities of beneficiary and non- Tigray Social Cash Transfer Pilot Program (SCTPP); beneficiary households, how these effects change Ghana: Livelihood Empowerment Against Poverty during programs scaled up, and the reasons these Program (LEAP); Kenya: Cash Transfer Program for Orphans and Vulnerable Children (CT-OVC); effects happen (Taylor 2012).5 Lesotho: Child Grants Program (CGP); Malawi: In Ethiopia, Ghana, Kenya, Lesotho, Malawi, Social Cash Transfer (SCT); Zambia: Program Child Zambia, and Zimbabwe, cash transfers have sig- Grant Program (CGP); and Zimbabwe: Harmonized nificant spillovers in the local economy and Social Cash Transfer Program (HSCT). 68 68 The State of Social Safety Nets 2018 5. To do this, LEWIE models link agricultural house- Daidone, S., S. Asfaw, B. Davis, S. Handa, and P. Winters. hold models into a general-equilibrium model of the 2016. “The Household and Individual-Level Economic local economy, which is a treated village (Thome Impacts of Cash Transfer Programmes in Sub-Saharan et al. 2016). Africa. Synthesis Report.” Food and Agricultural Organization of the United Nations, Rome. Davis, B., S. Handa, N. Hypher, N. Winder Rossi, REFERENCES P. Winters, and J. Yablonski, eds. 2016. From Evidence ASPIRE (Atlas of Social Protection: Indicators of Resilience to Action: The Story of Cash Transfers and Impact and Equity). 2017. Database, World Bank, Washington, Evaluation in Sub-Saharan Africa. Oxford, UK: Oxford DC. http://datatopics.worldbank.org /aspire/. University Press. Banerjee, A., R. Hanna, G. Kreindler, and B. A. Olken. Evans, D., and A. Popova. 2014. “Cash Transfers and 2015. “Debunking the Stereotype of the Lazy Temptation Goods: A Review of Global Evidence.” Welfare  Recipient: Evidence from Cash Transfer Policy Research Working Paper 6886, World Bank, Programs Worldwide.” Working Paper No. 308, Center Washington, DC. for International Development, Harvard University, Cambridge, MA. Handa, S., S. Daidone, A. Peterman, B. Davis, A. Pereira, T. Palermo, J. Yablonski. 2017. “Myth Busting? Barca, V., S. Brook, J. Holland, M. Otulana, and P. Pozarny. Confronting Six Common Perceptions about 2015. “Qualitative Research and Analyses of the Unconditional Cash Transfers as a Poverty Reduction Economic Impacts of Cash Transfer Programmes in Strategy in Africa.” Office of Research–Innocenti Sub-Saharan Africa. Synthesis Report.” Food and Working Paper 2017-11, UNICEF (United Nations Agricultural Organization of the United Nations, Children’s Fund). Rome. Bastagli, F., J. Hagen-Zanker, L. Harman, V. Barca, Ralston, L., C. Andrews, and A. Hsiao. 2017. “A Meta- G. Sturge, and T. Schmidt. 2016. “Cash Transfers: Analysis of Safety Net Programs in Africa.” Working What Does the Evidence Say? A Rigorous Review of Paper, World Bank, Washington, DC. Programme Impact and the Role of Design and Taylor, J. E. 2012. “A Methodology for Local Economy- Implementation Features.” Overseas Development Wide Impact Evaluation (LEWIE) of Cash Transfers.” Institute, London. http://www.odi.org/projects​ Food and Agricultural Organization of the United /2797​- social-protection-literature​-­review-poverty​ Nations, Rome. -­impact. Thome, K., J. Taylor, M. Filipski, B. Davis, and S. Handa. Daidone, S., L. Pellerano, S. Handa, and B. Davis. 2015. 2016. “The Local Economy Impacts of Social Cash “Is Graduation from Social Safety Nets Possible? Transfers: A Comparative Analysis of Seven Sub-Saharan Evidence from Sub-Saharan Africa.” IDS Bulletin Countries.” Food and Agricultural Organization of the 46 (2): 93–102. United Nations, Rome. Analyzing the Performance of Social Safety Net Programs 69 FPO PART II Special Topics CHAPTER 4 Social Assistance and Aging M ost individuals will reach old age. If aging trend will be sooner; for other regions, the effect is almost a certainty for so many individ- of the trend will be much greater (see figure 4.1). uals, how can elderly people mitigate wel- Today, the Europe and Central Asia region has fare risks or potential problems such as illness or the largest percentage of elderly people; in the disability? Social insurance programs can offer long term, Latin America and the Caribbean, coverage of illness, disability, and other potential South Asia, and East Asia and Pacific will expe- risks for a significant number of elderly people. rience the biggest increase. Although elderly However, in most developing countries, people people now represent 8 percent of the total pop- with low income are more exposed to these risks. ulation in Latin America and the Caribbean, Social insurance might be restricted to a small their share is projected to grow nearly four times group of workers, such as formal employees or by the year 2100. While the share of elderly peo- employees in certain sectors. For those individuals ple now represents 5 and 8 percent of the total who are not covered, the option of voluntary sav- populations in South Asia and East Asia and ings is close to impossible because a low income Pacific, respectively, their shares are projected to makes it difficult to save, and existing instruments grow over five times in South Asia (to 27 ­percent) for long-term savings are unavailable to them. and over 3.5 times in East Asia and Pacific (to 29 This chapter describes a key component that percent) by 2100  (United Nations Population countries are rapidly introducing to support the Research Council 2017). special needs of elderly people: old-age social Despite the evident aging trend, most coun- pensions.1 This type of program is becoming an tries do not have systems and benefits that can important policy tool to address issues of low fully cover elderly people or their special needs. social insurance coverage and, in some cases, to Many countries have social insurance pro- address aspects of poverty alleviation. This grams, but not everyone participates in those. chapter provides the general characteristics of In this context, very often the old-age social noncontributory old-age social pensions in dif- pension becomes a key instrument for provid- ferent regions, describes recent trends, gives ing social assistance coverage in old age. performance indicators, and includes a special Old-age social pensions are here defined as highlight on policy and implementation discus- noncontributory cash benefits targeted at elderly sions and considerations (see highlight 2). people, generally provided and financed by gov- ernments, and not linked to past contributions, WHAT ARE OLD-AGE SOCIAL PENSIONS, earnings, or years of service. Old-age social AND WHY ARE THEY ON THE RISE? pensions take different forms, but their main ­ An aging population is a common trend parameters for eligibility include age, citizenship, across regions. For some regions, the onset of the residency, and, in some cases, means testing. Social Assistance and Aging 71 FIGURE 4.1  Population, Age 64 Years and Older, as a Percentage of Total Population, by Region 35 30 25 20 Percent 15 10 5 0 20 30 40 60 70 50 80 90 00 15 45 25 95 35 65 75 85 55 20 20 20 20 20 20 20 20 20 20 20 20 20 20 21 20 20 20 East Asia and Pacific Europe and Central Asia Latin America and the Caribbean Middle East and North Africa South Asia Sub-Saharan Africa Source: Estimates based on UN Population Division data (https://www.un.org/development/desa/publications/world-population-prospects​ -the-2017-revision.html). Three main characteristics of old-age social social pension for poor citizens 60 years of pensions set them apart from social insurance age  and older. New Zealand (1898), Australia and some social assistance programs (Palacios (1908), Iceland (1909), the United Kingdom and Knox-Vydmanov 2014). First, there is an (1909), and Sweden (1913) then followed this important distinction between social pensions policy. The first social pensions were means- and minimum pensions. Minimum pensions tested. It was not until 1938 that New Zealand, are set up as the minimum guaranteed benefit for example, introduced a universal pension for provided by social insurance systems, while individuals age 65 years and older.2 Figure 4.2 social pensions in countries with social insur- presents the evolution of old-age social pension ance systems tend to be set below the minimum systems around the world. Between 1940 and pension and to be accessible to those who have 1990, many countries introduced this type of not made any contributions; have not partici- social security benefit. Yet, around half of all pated in any mandatory scheme; and/or have an old-age social pension programs have existed extremely low income. Second, old-age social only since 1990. pensions tend to be publicly financed out of Old-age social pensions have proliferated in government revenues (through general taxa- the past two decades. Since 2001, 29 economies tion). The third core distinction is between have introduced or expanded this social assis- ­ old-age social pensions and broader social tance/social safety net instrument (figure 4.3). assistance. Old-age social pensions are social Latin America and the Caribbean have led assistance benefits exclusively targeted at older the  trend, followed by East Asia and several people. Social assistance programs can include African economies. In addition, economies that and cover, but are not necessarily targeted at, already had a social pension system (mostly con- elderly people. tributory systems) introduced parallel benefits Old-age social pensions were introduced in aimed at covering different groups (for example, some countries almost as early as social insur- the rural programs and “70 and Up” in Mexico). ance. In 1891, Denmark introduced a local By 2014, an estimated 101 economies had 72 The State of Social Safety Nets 2018 FIGURE 4.2  Number of Countries with Old-Age Social Pensions, 1898–2012 80 70 60 Number of countries 50 40 30 20 10 0 04 10 16 8 22 28 34 40 46 52 58 64 70 76 82 88 94 00 06 12 9 20 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 18 19 20 Source: HelpAge Social Pensions Database. FIGURE 4.3  Introduction of Old-Age Social Pensions, 2001–13 35 Nigeria 30 China Peru Fiji El Salvador Philippines Nigeria Panama Uganda 25 Papua Number of countries New Maldives Guinea 20 Korea, Rep. Timor- Zambia Leste Mexico 15 Georgia Indonesia Guatemala Belize Swaziland 10 Kiribati Lesotho Germany Belgium Jamaica Kosovo 5 Mexico 0 10 02 04 08 09 03 12 13 06 01 05 11 07 20 20 20 20 20 20 20 20 20 20 20 20 20 Source: HelpAge Social Pensions Database. Notes: The height of each column indicates the cumulative number of countries with a social pension during this period. Germany: needs- based pension supplement (Grundsicherung im Alter); Mexico: “70 y Mas” regional scheme was introduced in 2001; Nigeria: Osun Elderly Persons Scheme. Papua New Guinea: Only one province (New Ireland) has an old-age social pension. ­ntroduced old-age social pensions.3 Almost all i WHY DO COUNTRIES INTRODUCE Latin American countries have them, whereas OLD-AGE SOCIAL PENSIONS? Sub-Saharan Africa economies have some of Old-age social pensions are introduced on the the  largest old-age social pensions systems in basis of an economy’s needs and capacity, in terms of the share of the elderly population particular to alleviate poverty, establish the covered. main component of a pension system, or Social Assistance and Aging 73 MAP 4.1  Countries with Old-Age Social Pensions and Their Main Purpose IBRD 43440 | JANUARY 2018 Poverty program Main element of pension system Addressing coverage gap No data Source: World Bank 2017. address a coverage gap in an existing pension those who reach a certain age and fulfill citi- system (see map 4.1). Bangladesh, India, zenship or residency criteria. Old-age social Kenya, Myanmar, and Vietnam, for example, pensions can be considered a type of uncondi- have introduced old-age social pensions as tional cash transfer.4 Means-tested or targeted poverty alleviation programs. Australia and programs provide benefits to the poor, who New Zealand (pioneer economies) and tend not to be covered by other (contributory) Bolivia, Maldives, and Timor-Leste (new- elements of the pension system. Means-tested comer economies) have introduced old-age benefits have the potential to be a main source social pensions as the main component of of income for elderly people, and thus have their pension systems, in the form of universal the  capacity to be pension-tested (that is, pensions. Other economies have used old-age the  capacity to exclude beneficiaries of other social pensions to address the coverage gap pension schemes).5 Figure 4.4 presents, for left by existing mandatory pension schemes each region, the share of economies that had a (see highlight 2 for a policy discussion on social pension by 2014, the type of social pen- introducing old-age social pensions and the sion, and the average total cost to GDP by special considerations that inform their region. Nearly 90 percent of Organisation for design). Among those, some have mature con- Economic Co-operation and Development tributory schemes but insufficient coverage (OECD) economies, 70 percent of Latin America (for example, Chile and Mexico), while others and the Caribbean economies, and nearly have immature contributory schemes for sig- 65 percent of Europe and Central Asian econ- nificant aging population trends (for example, omies have old-age social pensions (panel a). Hong Kong SAR, China; the Republic of Means-tested pensions are  most common Korea; and Thailand). among all regions, except for Europe and Old-age social pensions reflect the economy Central Asia, where pension-tested schemes context and take one of two forms: universal or dominate (panel b). means-tested. Universal pensions provide flat- The design of each program in terms of the rate benefits to all elderly people, generally eligibility age is independent of the type of 74 The State of Social Safety Nets 2018 FIGURE 4.4  Distribution of Old-Age Pension Programs a. Share of economies that have an old-age social pension, by region 100 88 75 73 Percentage of programs 65 60 50 50 29 25 10 0 OECD Latin America Europe East Asia South Sub-Saharan Middle East and the and and Asia Africa and North Caribbean Central Asia Pacific Africa b. Composition of social pensions by the targeting type and spending, by region 100 1.80 Average spending (percentage of GDP) Percentage of programs, by types 75 1.20 50 0.60 25 0 0 OECD Latin Europe East Asia South Asia Sub-Saharan Middle East America and and Africa and North and the Central Asia Pacific Africa Caribbean Means-tested Universal Pension (contributory) tested Not defined None (no program in place) Regional average spending Source: Calculations based on HelpAge International Social Pensions. Note: Data are as of 2014. GDP = gross domestic product; OECD = Organisation for Economic Co-operation and Development. Social Assistance and Aging 75 FIGURE 4.5  Age of Eligibility for Pension Programs a. By region b. By gender and region 80 80 70 77 70 70 70 Years of age Years of age 67 65 60 60 62 60 60 60 60 60 50 50 CD be nd sia d c ia ia a ric d CD be nd sia d ci ia ia a ric ric lA n Af an lA n Pa As As Pa As As Ce ur n So ific a Ce Eur n So fic rib a a ra e a rib a a ra e a OE OE a a Af Af h t d st th d st th rt as nt op nt op Ca ic Ca ic n n an Ea an Ea u u e er e er No e E ra ra th Am th m ha ha E dl A id Sa Sa tin tin M b- b- La La Su Su Mean Male Female Sources: ASPIRE team, using HelpAge International Social Pensions Database. Note: In panel a, for each region, there are three numbers: minimum, mean, and maximum. Because of insufficient data on Middle East and North Africa, some of the indicators could not be estimated. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; OECD = Organisation for Economic Co-operation and Development. pension. Figure 4.5 shows the diversity of the coverage of old-age social pensions (measured design across regions by analyzing the age of as a percentage of the population age 60 years eligibility. Panel a shows the minimum, maxi- and older).6 In contrast, a few Middle Eastern mum, and mean by region. In most regions, the and North African countries have introduced mean ­ eligibility age tends to be 65 years, except social pensions. for Sub-Saharan Africa and the Middle East Old-age social pensions have expanded rap- and North Africa. Panel b presents the average idly in certain countries. Between 2010 and age of eligibility by gender. Old-age social pen- 2015, the number of beneficiaries in Chile, sions across regions often include age differen- Mexico, the Philippines, and Vietnam grew by tiation between men and women, with the latter more than 70 percent (lowering the eligibility usually eligible five years earlier. age often drives such expansion). In Bolivia, Mauritius, and Namibia, coverage of the popu- WHAT HAVE OLD-AGE SOCIAL PENSIONS lation age 60  years and older has become ACCOMPLISHED? universal.7 Old-age social pensions provide an alternative Elderly people in the first (poorest) quintile source of income for elderly people who are not have benefited the most from old-age social covered by contributory schemes. Social pen- pensions, no matter the program design. Using sions cover close to 35 percent of the popula- household survey data from the Atlas of Social tion age 60 years and older in Organisation for Protection: Indicators of Resilience and Equity Economic Co-operation and Development (ASPIRE) database, the distribution of benefi- countries and in the Europe and Central Asia, ciaries (or beneficiary incidence) of old-age East Asia and Pacific, Latin America and the social pension programs was estimated.8 Caribbean, and South Asia regions, according Figure 4.7 presents the distribution of old-age to estimations from HelpAge International data social pension beneficiaries by quintile of per (figure 4.6). The Africa region has the largest capita pretransfer welfare and by type of 76 The State of Social Safety Nets 2018 FIGURE 4.6  Old-Age Pension Coverage of Population Age 60 Years and Older, by Region 160 Percentage of 60+ population covered by pensions 120 80 40 0 D be nd sia d cifi ia ia a ric d ic lA n Af an EC Pa As As an c a rib a a ra e a r Af h t O d st h rt as nt op C a ric ut an Ea n N o le E Ce Eur e e ra So th Am ha d id Sa tin M b- La Su Mean Sources: ASPIRE database, using data from HelpAge International. Note: For each region, there are three numbers: minimum, mean, and maximum. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; OECD = Organisation for Economic Co-operation and Development. ­argeting method, based on available data. In t Furthermore, the effect of old-age social pen- Bulgaria, Latvia, Lithuania, and Turkey, more sions on inequality (as a reduction in the Gini than 50 percent of the old-age social pension coefficient for the o ­ verall population) is less than beneficiaries are in the poorest quintile, 10 percent (see ­ figure 4.9), except for the same while in Swaziland and Guatemala, more than three countries (where coverage and benefit 50  percent of the beneficiaries of social pen- levels are high). ­ sions are in the wealthiest (fourth and fifth) In some African and Latin American coun- quintiles. tries, elderly people are not necessarily among Old-age social pensions provide income the poorest, so the benefit’s impact on poverty security and dignity in old age. To analyze the alleviation might be marginal. While Mali, relative importance of the old-age social pen- Mauritius, and Namibia have large shares sion, Figure 4.8 presents the level of pension of  elderly people in the poorest households benefits as a share of beneficiary welfare for the (32 percent, 38 percent, and 43 percent, respec- poorest and second-poorest quintiles. In Brazil, tively), they have even larger shares of children Mauritius, and South Africa, old-age social who live there (92 percent, 64 percent, and pensions represent more than 50 percent of the 95  percent, respectively), Guven and Leite total welfare for elderly people (and their house- (2016) note. However, in Argentina, Brazil, and holds as indirect beneficiaries) in the poorest Chile, mandatory schemes and social pensions quintile. have contributed to lowering poverty among Old-age social pensions have helped benefi- elderly people; they also had a larger overall ciaries reduce or altogether escape poverty. In a poverty reduction impact than programs tar- sample of 18 countries (see figure 4.9), the effect geted at children (Acosta, Leite, and Rigolini of old-age social pensions on the poverty head- 2011). BÖger et al. (forthcoming) find that more count and poverty gap reduction is significant than 50 percent of countries worldwide with (10–40 percent) in only three (Mauritius, South old-age social pensions have raised elderly Africa, and Thailand). In the other 16 countries, ­ people above the international poverty line but the poverty impact is much less pronounced. not above national poverty lines. Social Assistance and Aging 77 FIGURE 4.7  Distribution of Old-Age Social Pension Beneficiaries, by Income Quintile a. Means-tested Bulgaria Turkey Brazil Costa Rica Bangladesh South Africa Cabo Verde Paraguay Chile Colombia Belize Guatemala 0 10 20 30 40 50 60 70 80 90 100 Percent b. Pensions-tested Lithuania Latvia Mexico Panama Thailand Nepal Tajikistan 0 10 20 30 40 50 60 70 80 90 100 Percent c. Universal Lesotho Mauritius Namibia Timor-Leste Swaziland 0 10 20 30 40 50 60 70 80 90 100 Percent Q1 Q2 Q3 Q4 Q5 Sources: ASPIRE database; selected surveys. Note: Countries were selected based on the availability of reliable household survey data (ASPIRE). Beneficiaries’ incidence is determined as follows: (number of direct and indirect beneficiaries [people who live in a household where at least one member receives the transfer] in a given quintile)/(total number of direct and indirect beneficiaries). The sum of percentages across quintiles per given instrument equals 100 percent. Quintiles are calculated using per capita pretransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; Q = quintile. 78 The State of Social Safety Nets 2018 FIGURE 4.8  Old-Age Social Pensions as a Share of Beneficiaries’ Welfare, Poorest and Second-Poorest Quintiles South Africa Brazil Mauritius Chile Costa Rica Lithuania Panama Paraguay Latvia Poland Mexico Cabo Verde Rwanda Slovak Republic Bulgaria Colombia Turkey Thailand Romania Nepal Bangladesh Tajikistan Jamaica Timor-Leste Sri Lanka 0 10 20 30 40 50 60 70 80 Percent Q1 Q2 Sources: ASPIRE database; selected surveys. Note: Countries are selected based on the availability of reliable household survey data (ASPIRE). Pensions as a share of beneficiary welfare is defined as the ratio of the pension size to the average household per capita consumption or income of the respective welfare  quintile (Q1:  bottom 20 percent; Q2: bottom 21–40 percent). The poorest first and second quintiles are calculated using per  capita  posttransfer welfare (income or consumption). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; Q = quintile. Old-age social pensions have mixed results US$1 spent on old-age social pensions reduces in reducing poverty, as seen in the benefit– the poverty gap by 32 cents. Only Latvia and cost ratio, measured as the reduction in the Lithuania have a ratio above 0.5. The design of poverty gap obtained for each US$1 spent.9 the old-age social pension (universal or On average, among the social pensions means-tested) likely affects this indicator, but and  countries included in this analysis, the more evidence is needed to better understand benefit–cost ratio is 0.32, meaning that every the potential link. Social Assistance and Aging 79 FIGURE 4.9  Impact of Old-Age Social Pensions on Poverty Headcount, Poverty Gap, and Gini Inequality Index Reduction, as a Share of Pretransfer Indicator Levels, Using Relative Poverty Line (Poorest 20 Percent) 60 12 Poverty headcount and gap reduction (%) 50 10 Inequality reduction (%) 40 8 30 6 20 4 10 2 0 0 i L ia Ro ka Ja nia th a ra a lo y a ng via Tu h Ti Me y -L o te Po l nd st zil Pa ica a bo hile Th rde A d au a s pa iu Co ua e c Li aic Pa ani bi s m M fric ut ilan ar Co ra es rk de m xi an la rit Ba at R m Ne a na Ve Ca C g lg m B m u la So a a L Bu or h Sr Poverty headcount reduction Poverty gap reduction Gini reduction Source: Calculations based on the ASPIRE database. Note: Countries were selected based on the availability of reliable household survey data in the ASPIRE database. The impacts on poverty and inequality reduction were calculated as follows: (poverty headcount pretransfer – poverty headcount post transfer)/(poverty headcount pretransfer). Same calculations apply for the Gini index and poverty gap percentage reductions. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. FIGURE 4.10  Benefit–Cost Ratio of Old-Age Social Pensions 0.6 0.5 0.4 $PPP 0.3 0.2 0.1 0 th a or an a M h Bu ia st ia Tu a o te a s a a Pa le Co de ay ey l bo nd l iu i Ba Ric s pa ic c ni m ric M zi tv b Co ar i es Ch de ai Ti kist r gu rk ex So urit a m ua na Ca aila Ve Ne lg Af La m Br -L la lo ra a Ja ji ng a Th h Pa Ta Li ut m Source: Calculations based on the ASPIRE database. Note: Countries were selected based on the availability of reliable household survey data in the ASPIRE database. The benefit–cost ratio is the poverty gap reduction in US$ for each unity (US$1) spent in the social program. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; PPP = purchasing power parity. 80 The State of Social Safety Nets 2018 NOTES 9. More precisely, the benefit–cost ratio is estimated 1. In this book, old-age social pensions and social as the poverty gap before the transfer minus the pov- ­pensions are defined differently. Social pensions erty gap divided by the total amount spent in the (­ discussed in part I) are defined using the ASPIRE program. database and include old-age social pensions, non- contributory disability benefits, noncontributory benefits to war victims or war veterans, and noncon- REFERENCES tributory survivorship benefits. Acosta, P., P. Leite, and J. Rigolini. 2011. “Should Cash Transfers Be Confined to the Poor? Implications 2. The 1938 Social Security Act lowered the age for the for Poverty and Inequality in Latin America.” means-tested pension to 60 and introduced a uni- Policy  Research Working Paper 5875, World Bank, versal (not means-tested) superannuation from age Washington, DC. 65 years (https://en.wikipedia.org/wiki/Welfare_in​ _New_Zealand). ASPIRE (Atlas of Social Protection: Indicators of Resilience and Equity). 2017. “Data Sources and Methodology.” 3. HelpAge’s Social Pensions Database (http://www​ Database, World Bank, Washington, DC. http://datatop- .pension-watch.net/about-social-pensions/about​ ics​.worldbank.org/aspire/~/documentation/. social-pensions/social-pensions-database). For 20 -­ of these 101 countries, there is no information on Böger, T., and L. Leisering. Forthcoming. “Social when the social pension was established. Citizenship for Older Persons? Measuring the Social Quality of Social Pensions in the Global South and 4. More specifically, they are conditional only on an age Explaining Their Spread.” World Bank Social Protection threshold. Working Paper, World Bank, Washington, DC. 5. Countries like South Africa include an income test Guven, M., and P. Leite. 2016. Benefits and Costs of Social and an asset test. Beneficiaries cannot have assets Pensions in Sub-Saharan Africa. Washington, DC: worth more than R1,056,000 (US$77,728) if single or World Bank. R2,112,000 (US$155,456) if married. Palacios, R., and C. Knox-Vydmanov. 2014. “The Growing Role of Social Pensions: History, Taxonomy, and Key 6. Coverage in Africa ranges from universal programs Performance Indicators.” Public Administration and in Southern Africa to no programs in many countries Development 34 (4): 251–64. across the continent. Hence, a measure of “average” coverage could be misleading. Pallares-Miralles, M., C. Romero, and E. Whitehouse. 2012. “International Patterns of Pension Provision II: 7. Coverage rates above 100 percent are possible. A  Worldwide Overview of Facts and Figures.” Beneficiaries under universal old-age social pensions Social  Protection and Labor Discussion Paper 1211, might outnumber the potential population of recipi- World Bank, Washington, DC. ents because of identification issues (such as weak death registration systems) or fraud. United Nations Population Research Council. 2017. World Population Prospects: The 2017 Revision. New York: 8. Beneficiary incidence is the percentage of program United Nations. https://www.un.org/development/desa​ beneficiaries belonging to each quintile of the welfare /publications/world-population-prospects-the​-2017​ distribution. -revision.html. Social Assistance and Aging 81 HIGHLIGHT 2.  Policy Considerations for Introducing Old-Age Social Pensions M ost pension systems are mandatory, earn- FIGURE H2.1  Elderly and Labor Force Coverage ings-related, contributory programs. The first and second pillars of the pension sys- Central African Republic Congo, Dem. Rep. tem are mandatory (based on the World Bank’s Congo, Rep. typology), either publicly or privately managed, Ethiopia Liberia and linked to workers’ length of service or previ- Madagascar Mauritania ous contributions (Pallares-Miralles, Romero, and Nigeria Whitehouse 2012). These schemes tend to cover São Tomé and Príncipe Sierra Leone public sector employees, the military, occupational Chad schemes, and, in some cases, the private sector, as Gambia, The Benin well as facilitating the participation of the self-­ Zambia employed. In the Middle East and North Africa Rwanda Burkina Faso region and in many countries in Asia, most man- Zimbabwe datory schemes cover only public sector employ- Burundi Cameroon ees and occupational schemes. Coverage as a Malawi percentage of the labor force of the working-age Tanzania Sudan population and coverage of those age 60 years and Ghana older or 65 years and older (relative to the size of Mali Niger the population group) have decreased over time. Togo Uganda Different reasons have prevented mandatory Côte d’Ivoire schemes from reaching full coverage. Even though Kenya Guinea these schemes (first and second pillars) have been Mozambique in place in low- and middle-income countries for Senegal Cabo Verde decades, the labor force share (mainly in the for- South Africa mal sector) eligible to participate in them is low Swaziland Lesotho and has remained almost constant over time. The Botswana Latin America and the Caribbean region, for Mauritius Namibia example, has high levels of informality as pen- Seychelles sions systems are largely designed for salaried 0 20 40 60 80 100 workers (Bosch, Melguizo, and Pagés 2013). In Percent many Sub-Saharan African countries, pension Elderly coverage Labor force coverage schemes do not cover employees in the private sector and the combination of high contribution Source: Dorfman 2015. rates for social insurance, low wages, and high informality restrict coverage (Dorfman 2015). The rural poor are very unlikely to be able to par- simply not possible through social insurance or ticipate in earnings-related pension schemes, mandatory pension schemes. Guven and Leite (2016) find. In addition, many The short- and long-term fiscal implications of civil servants and occupational schemes tend to old-age social pensions need to be considered care- be  unfunded while providing high benefit fully. Guven and Leite (2016) find that certain Sub- levels.   This limits the financial capacity of gov- Saharan African countries expend most of the social ernments to introduce and implement other pro- protection budget on old-age programs; 29–61 per- grams targeted at elderly people and the poor. cent of total social assistance spending goes to old- To this end, old-age social pensions are fast age social pension programs. Bosch, Melguizo, and becoming a tool to meet the urgent needs of Pagés (2013) describe how, in Latin America, uni- expanding ­ coverage and alleviating poverty versal old-age social pensions provide coverage for (Palacios and Knox-Vydmanov 2014). In Africa, elderly people but require considerable resources; in full coverage of elderly people is (nearly) attained some cases, governments transfer resources from only in countries that have old-age social pensions infrastructure, health, and/or education budgets to (see figure H2.1). These high coverage rates are meet pension obligations. 82 82 The State of Social Safety Nets 2018 Analyzing the long-term costs of old-age pro- social pensions in Latin America and the Caribbean, grams, given either slow or fast demographic particularly when they exist in parallel with contrib- changes, is imperative. In several regions, the popu- utory pension systems, can affect labor market lation age 65 years and older as a percentage of the dynamics and generate incentives for informal total population will double by 2050, jeopardizing employment. the fiscal sustainability of old-age social pensions. An analysis at the microeconomic level, such as Understanding the relationship between demo- household characteristics, is also necessary to inform graphic trends, macroeconomic and labor market the instrument design. The poverty gap of elderly conditions, and key program parameters is neces- people living alone can be misconstrued. Guven and sary to maintain the fiscal sustainability of pro- Leite (2016) find that for a set of countries in Africa, grams not only now but also in the medium and the poverty headcount is significantly higher for long terms. households with children than for any other popula- Introducing old-age social pensions should take tion group. Dorfman (2015) shows that many into consideration factors beyond closing the cov- African countries have high co-residency levels and erage gap. Key elements of program design and that elderly people rarely live alone. Thus, social impact should be carefully analyzed before imple- assistance programs targeted at poor households menting such programs. This analysis should also can benefit elderly people as much as other mem- take into consideration the overall social assistance bers, including children. programs in place in a country to avoid an overlap The context for old-age social pension design of benefits and to ensure an efficient use of limited and implementation matters. Old-age social pen- resources. sions are both a form of social assistance/safety net From a design perspective, universal and means- for alleviating poverty and a potential component tested programs have clear trade-offs in efficiency, of pension systems for addressing coverage gaps. cost, and effectiveness of implementation. In addi- Considering the totality of existing social assistance tion, parameters such as age of eligibility, benefit programs, the priorities and needs of the overall levels, and benefit indexations can all affect people’s population, and budget constraints when analyzing behavior during their working life, creating incen- their introduction, are essential to designing an tives or disincentives to participate or not in the effective old-age social pension system. pension system and/or in the labor force. Although a universal program might be perceived as easier to implement (that is, it leverages existing adminis- REFERENCES Bosch, M., Á. Melguizo, and C. Pagés. 2013. Better Pensions trative capacities and reduces potential errors of Better Jobs: Towards Universal Coverage in Latin inclusion and/or exclusion), it can be more expen- America and the Caribbean. Washington, DC: Inter- sive. Dorfman (2015) shows that despite the chal- American Development Bank. lenges of implementing an old-age social pension Dorfman, M. 2015. “Pension Patterns in Sub-Saharan system targeted exclusively at the elderly poor, the Africa.” World Bank Social Protection and Labor policy has been found to reduce national poverty at Discussion Paper 1503, World Bank, Washington, DC. almost twice the rate of a universal approach. In Guven, M., and P. Leite. 2016. Benefits and Costs of Social addition, targeting those age 65 years and older, Pensions in Sub-Saharan Africa. Washington, DC: rather than 60 years and older, has a significantly World Bank. greater impact. Palacios, R., and C. Knox-Vydmanov. 2014. “The Growing Setting the eligibility age, the benefit level, and the Role of Social Pensions: History, Taxonomy, and Key benefit indexation are key elements in the design and Performance Indicators.” Public Administration and performance of an old-age social pension program. Development 34 (4): 251–64. For example, if the benefit level is high and the eligi- Pallares-Miralles, M., C. Romero, and E. Whitehouse. 2012. bility age is close to or the same as the main social “International Patterns of Pension Provision II: insurance, an old-age social pension might discour- A  Worldwide Overview of Facts and Figures.” age labor force participation. Bosch, Melguizo, and Social  Protection & Labor Discussion Paper 1211, Pagés (2013) find that certain parameters of old-age World Bank, Washington, DC. Social Assistance and Aging 83 CHAPTER 5 The Emergence of Adaptive Social Protection WHY DOES THE WORLD NEED ADAPTIVE book details. As a result, safety nets are better SOCIAL PROTECTION? positioned than ever to help households man- Today’s global landscape is fraught with multi- age the risks associated with the multiplicity ple, interconnected, and often devastating and complexity of shocks. Indeed, SSNs and the shocks. Between 1980 and 2012, the annual broader social protection suite of policies, pro- ­ frequency of natural disasters increased by grams, and instruments are widely recognized 250 percent and the number of people affected as successful tools for building the resilience of increased 140 percent (figure 5.1). Climate the poor and most vulnerable. Specifically, the change is expected to exacerbate these trends World Bank Social Protection and Labor and, without climate-informed development, to Strategy (2012b) emphasizes that social protec- push an additional 100 million people into tion builds the resilience “of the vulnerable extreme poverty by 2030 (World Bank 2016b). through insuring against the impact of drops in Forced displacement also has hit record highs; well-being from a range of shocks.” Safety nets on average, 20 persons were estimated to have can provide cash, food, insurance, and other fled their homes every 60 seconds in 2016 means to smooth income and consumption (UNHCR 2016). In total, more than 64 million when shocks occur, increasing the resilience of people were displaced worldwide by the end of households. When combined with complemen- 2015 (figure 5.2). Furthermore, the worst eco- tary interventions, safety nets can enhance nomic and financial shock in recent history household resilience in the long term by materialized less than a decade ago, and the promoting human capital development and ­ 2014 Ebola outbreak reawakened the global income-generating activities (World Bank community to the potential devastation of pan- 2012b). demics. Such shocks, their trends, and associ- However, limitations in SSN coverage and ated risks are deeply interconnected (see, for design restrict the ability for safety nets to pro- example, WEF 2017), creating an environment tect households that are vulnerable to shocks. of heightened complexity for households, pol- Generally, the poor are particularly vulnerable icy makers, and practitioners alike to navigate. to shocks for multiple reasons, which include a Never has the challenge been more acute for lack of savings and limited access to finance social safety nets (SSNs) to build household and formal insurance (see, for example, World resilience and to respond to shocks across the Bank 2016a). To protect their short-term life cycle. Significant progress has been made in well-being and consumption after a shock, the past decade in terms of introducing new poorer households may instead turn to such SSN programs and scaling up existing programs “negative coping” strategies as removing chil- to expand the coverage of the poorest, as this dren from school to work for extra household 84 The State of Social Safety Nets 2018 FIGURE 5.1  Total Number of Disasters and Affected People, 1980–2012 600 700 Number of affected persons (millions) Number of natural disasters per year 600 500 500 400 400 300 300 200 200 100 100 0 0 80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12 20 20 19 19 19 19 19 20 19 20 20 19 19 19 19 20 20 Occurrence Total affected (millions) Linear (occurrence) Linear (Total affected (millions)) Source: EM-DAT database. FIGURE 5.2  Total Number of Displaced People, 1951–2015 70 60 Global number of displaced persons (millions) 50 40 30 20 10 0 51 55 59 63 67 71 75 79 83 87 91 95 99 03 07 11 15 20 19 19 20 19 19 19 19 19 20 19 19 19 19 19 20 19 Source: United Nations High Commission on Refugees Population Statistics Database. The Emergence of Adaptive Social Protection 85 income, availing high-interest loans, and sell- Program and Kenya’s rapidly growing Hunger ing productive assets. However, such short- Safety Nets Program. Countries in Group C have term coping strategies can work to the lower safety net coverage and higher humanitar- household’s longer term detriment. Receiving ian spending—including countries mired in cri- assistance from safety nets can lessen the need ses and fragility such as Afghanistan, Democratic for such negative coping strategies after shocks Republic of Congo and Haiti—and may be less occur. But the persistent undercoverage of the well prepared to institute government-led safety poorest and most vulnerable to shocks means nets and more reliant on humanitarian funding that those most in need of such support may and related programming. Countries in Group B have no access to SSNs. Furthermore, rigid pro- have both low safety net coverage and low gram design can hamper attempts to adjust humanitarian spending, indicating they may be parameters to meet changed needs on the less beset by persistent crises than those coun- ground, reaching beyond a core SSN caseload tries in Group C. For such countries, it may be after a shock has occurred. particularly beneficial to further invest in safety For example, empirical evidence suggests nets and their use for building household resil- that countries at high risk of natural disasters ience to shocks. often have lower safety net coverage. Figure 5.3, In this context, Adaptive Social Protection panel a, measures the coverage of all SSN pro- (ASP) has emerged in recent years. At the out- grams within a country (based on the latest-­ set, ASP was conceptualized as “a series of year data in the Atlas of Social Protection: measures which aim to build resilience of the Indicators of Resilience and Equity [ASPIRE] poorest and most vulnerable people to climate database) against a country’s risk from natural change by combining elements of social pro- disasters (as ranked by the 2016 World Risk tection, disaster risk reduction and climate Report). While there is a significant degree of change” (Arnall et al. 2010; see also IDS 2012). variance, most disaster-prone countries have Since then, the term “adaptive” has come to be large coverage gaps, leaving those most at risk, ­ understood by social protection policy makers in many cases, unreachable by safety net pro- and practitioners as entailing the need to bet- gramming. Furthermore, the lack of coverage is ter adapt  social protection to all types of even more evident among the poorest quintile shocks. This recognition has resulted in many (figure 5.3, panel b). In both cases, the South complex questions, including precisely how Asia and Africa regions, home to the world’s best can SSNs and social protection be largest share of poor, have safety net coverage equipped to help households manage diverse well below ­ levels commensurate with their types of shocks across myriad country contexts disaster risk. (Groups A, B, and C)? Because is a nascent There is a large degree of heterogeneity in area, this question is not fully answered; but it terms of the readiness and suitability of national has begun to crystalize around two interrelated safety net programs to play more prominent approaches focused on building household roles where shocks are concerned. Figure 5.4 resilience and increasing the responsiveness of again looks at SSN coverage, this time alongside programming. a measurement of humanitarian aid received, per capita (from Gentilini 2016). Safety nets do FOCUS AREA 1: BUILDING HOUSEHOLD not exist in a vacuum, and national systems and RESILIENCE BEFORE SHOCKS OCCUR humanitarian programming coexist to varying The first of these interrelated approaches cen- degrees, depending on the context. The data are ters on boosting the role of social protection somewhat porous, but three broad country and safety nets in building the resilience of groupings can be drawn. Countries in Group A the most vulnerable households before shocks have higher safety net coverage and lower occur. By doing so, this resilience-building humanitarian spending, indicating greater read- approach seeks to break the deleterious cycle of iness and suitability for their safety nets to poverty and vulnerability that may otherwise address the risk of shocks; examples include the occur. In short, a more resilient household will Philippines’ Pantawid Conditional Cash Transfer be better able to withstand shocks if household 86 The State of Social Safety Nets 2018 FIGURE 5.3  Ranking of Natural Disasters and Safety Net Coverage a. By total population and region 100 90 80 Social safety net coverage (%) 70 60 50 40 30 20 10 0 0 50 100 150 World Risk Index, 2016 (169 = most at risk; 1 = least at risk) b. By poorest quintile and region 100 Social safety net coverage (%) 80 60 40 20 0 0 50 100 150 World Risk Index, 2016 (169 = most at risk; 1 = least at risk) Sub-Saharan Africa East Asia and Pacific Europe and Central Asia Latin America and the Caribbean Middle East and North Africa South Asia Expon. (ALL) Sources: Garschagen et al. 2016; and ASPIRE database. Note: Social safety net coverage is based on the latest year for the ASPIRE database (all programs). ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. members have more human capital and are able Livelihood Empowerment Against Poverty to access job opportunities, accumulate physi- (LEAP), Kenya’s Hunger Safety Net Program, cal capital, and diversify their livelihoods. and Zambia’s Child Grant Program (World Significant evidence confirms that SSNs, Bank 2016c). For example, Hoddinott et al. adaptive or otherwise, help improve resilience (2015) examined distress sales of livestock at the household level. Impact evaluations indi- between 2010 and 2014 among beneficiaries of cate that beneficiaries of cash transfer programs Ethiopia’s Productive Safety Net Program are more likely to save, as seen in Ghana’s (PSNP), compared with a control group. The Emergence of Adaptive Social Protection 87 FIGURE 5.4  Social Safety Net Coverage of the Poor and Humanitarian Spending, 2010–15 100 MNG RWA Safety net coverage of the poorest (%) IRQ 80 UGA SWZ UKR LSO LBR 60 TUR PHL NPL LKA 40 KEN SLE BFA MRT DJI TUN 20 PAK ZWE MWI ETH YEM KGZ AFG TJK SLB CIV SEN ZAR NER 0 CMR COG GMB CAF HTI 0 10 20 30 40 Humanitarian aid (US$per capita, average 2010–15) Group A: Countries with higher safety net coverage and lower humanitarian spending Group B: Countries with both low safety net coverage and low humanitarian spending Group C: Countries with lower safety net coverage and higher humanitarian spending Source: Gentilini 2016. In 2010, 54 percent of public works households It  combined the transfer of a productive asset reported making a distress sale of assets to meet with consumption support, training, and coach- food needs, and 26 percent did so to obtain cash ing, as well as efforts to encourage savings and for nonfood emergency cash needs. By 2014, access to health and education s­ ervices. The trial these proportions had fallen to 25 and 13 per- found statistically significant, cost-effective cent, respectively. Brazil’s Bolsa Família delayed impacts on consumption (fueled mostly by the entry of  children into the labor market. increases in self-employment income) and the Children in beneficiary families of the Programa psychosocial status of the targeted households, de Asignacion Familiar in Honduras are less with impacts on the poor households lasting at likely to work. Children in the Philippines’ least a year after all implementation had ended Pantawid conditional cash transfer program (Banerjee et al. 2015). work six fewer days per month than a control This productive, inclusive approach is group (World Bank 2016c). These are selected being implemented in many countries across examples from a proliferating body of evidence West and East Africa, where similar ASP- reporting similar findings. focused initiatives look to boost household The evidence base for the impact of produc- resilience in the face of repeated and chronic tive inclusion interventions (“graduation drought, along with other shocks. A recent models”) that support sustainable exits from ­ World Bank publication, Social Protection poverty—and by extension, resilience-­building— Programs for Africa’s Drylands (Del Ninno is also growing. A primary example of this comes and Coll-Black 2016), describes resilience from a randomized control trial for a similar building as a process of “improving house- “integrated approach” in six countries (Ethiopia, holds’ or communities’ economic and social Ghana, Honduras, India, Pakistan, and Peru). stability by addressing their structural 88 The State of Social Safety Nets 2018 vulnerabilities and increasing their access to and “vertical” expansion (Oxford Policy services while helping them prepare against Management 2015). A commonly cited exam- future crises. . . . This is achieved at the house- ple of an SSN with these c ­apabilities is hold level, for instance, through the regular Ethiopia’s PSNP, as witnessed by its response distribution of cash transfers accompanied by to the 2011 drought (see box 5.1). training activities to help diversify livelihoods Increasing grant amounts to existing SSN away from climate-dependent activities.” In beneficiaries following shocks (vertical expan- this context, resilience may be the product of sion) is a pragmatic and increasingly common (i) diversified livelihood strategies and access safety net response. Leveraged in this way, to markets; (ii) access to financial, social, existing programs such as cash transfers and human, physical, and natural capital; (iii) public works can can be used as conduits to access to quality basic social services; (iv) rapidly inject assistance to pretargeted and access to social protection programs, includ- enrolled poor households in affected areas. ing safety nets, particularly in difficult peri- Recently, this approach reached existing benefi- ods; (v) access to the information and skills ciaries that were affected by disasters in Fiji and needed to adapt to shocks; and (vi) local and the Philippines (see box 5.2). Preparedness national institutions able to adapt to chang- measures for SSNs can be advanced even fur- ing realities. ther through additional investments to make In the form of public works, ASP programs programs more flexible and capable of expand- can reduce the sources of risk from a shock in ing horizontally to reach additional households, rural areas, as has been done in Ethiopia, as in the case of the PSNP. Rwanda, and across the Sahel. A well-known Specifially, horizontal expansion can be example is the public works component of achieved by investing in more dynamic deliv- Ethiopia’s PSNP. It helps increase household ery systems. Safety nets designed to address and community resilience to droughts by creat- chronic poverty in times of relative calm ing community assets that reverse the severe and  stability  adopt methodologies and degradation of watersheds and provide a more supply-driven approaches to delivery for a ­ reliable water supply under different climatic fixed period. These may include time-bound conditions. Similarly, the Rwanda Vision 2020 approaches to targeting (that is, a “census Umurenge Program targets public works for sweep” approach, repeated again only after creating anti-erosive ditches and terracing hill- several years have passed) and fixed, central- sides, improving soil productivity, and expand- ized lists of beneficiaries. This approach is typ- ing the area of cultivable land (IDS 2012). ically easier to administer, but its rigidity often produces unintended effects (for example, FOCUS AREA 2: INCREASING THE household exclusion errors), which are magni- CAPABILITY OF SAFETY NETS TO fied under the influence of shocks when needs, RESPOND TO SHOCKS AFTER THEY poverty status, well-being, and vulnerability OCCUR can change rapidly. In this sense, these deliv- The second interrelated approach to ASP ery systems are static; they are unable to focuses on increasing the capability of safety administratively respond to changes in house- nets to respond to shocks after they occur hold needs. The hallmark of an adaptive safety by introducing greater flexibility and scalabil- net is dynamic delivery systems that enable the ity in program design. Such design features required flexibility and scalability to achieve enable faster adjustment to postshock needs. horizontal and/or vertical expansion, depend- Conceptually, a program becomes capable of ing on postshock needs. “scaling out” to nonregular social protection In addition, information systems tied to beneficiaries that have been affected by a understanding risks and vulnerabilities, along shock and/or “scaling up” to increase benefit with pre-positioned risk financing, can imbue amounts at an acute time of need to existing safety net programs with the capability to social protection beneficiaries (see figure 5.5). ­horizontally expand and to reach more affected This process is also referred to as “horizontal” households. Early warning and related risk The Emergence of Adaptive Social Protection 89 FIGURE 5.5  Program Scalability to Enable Responsiveness to Shocks Benefit amount Vertical expansion Additional benefit amount Core/regular Regular safty net Horizontal expansion benefit program Population Regular program Those not in regular beneficiaries (subset program but affected and of the poor) targeted for assistance Source: World Bank 2017. BOX 5.1  Horizontal and Vertical Expansions through Ethiopia’s Productive Safety Net Program Ethiopia’s Productive Safety Net Program (PSNP) most vulnerable to shocks and climate change. is a large, national SSN program. It is designed to These investments in more dynamic targeting for respond to the impacts of chronic drought, food the PSNP and other preparedness measures insecurity, and climate change on Ethiopia’s enabled the program to extend the duration of its poorest households. To do so, the PSNP incorpo- regular support for 6.5 million existing beneficia- rates public works activities that improve climate providing an extra three months of assis- ries, ­ resilience and promote community-level adapta- tance (vertical expansion), while also extending tion; provide a federal contingency budget to programming to an additional 3.1 million people help poor households and communities better who were not in the core PSNP caseload (hori- cope with transitory shocks when they occur; and zontal expansion) in response to the droughts of target methods to identify those ­ communities 2011 (White and Ellis 2012). information (e.g.,  hazard mapping, market systems often work in silos without coordina- monitoring, meteorological monitoring, con- tion, integration, and direct linkages to social flict mapping, climate variance mapping, and safety net programming. The Dominican geospatial data),  along  with information on Republic, Kenya, and the Republic of Yemen household composition and characteristics, have all developed innovative and integrated can provide vital information about the information systems, looking to overcome nature, location, and depth of a shock as well these limitations (see box  5.3). Alongside as the appropriateness and type of responses. these  information systems, pre-­ positioned However, where they exist, these information financing is of critical importance for more 90 The State of Social Safety Nets 2018 BOX 5.2  Responding Rapidly to Disasters through Vertical Expansions in Fiji and the Philippines In response to Typhoon Yolanda in 2013, the previously agreed-on legislation to remove the government of the Philippines released the conditionalities of the regular program during equivalent of US$12.5 million between November states of emergency (Bowen 2015). 2013 and February 2014—three months after the In Fiji, following Tropical Cyclone Winston in disaster struck—in unconditional cash transfers 2016, the government disbursed F$19.9  million to existing beneficiaries of the national condi- (US$39.6 million) in the form of top-up grants tional cash transfer program, Pantawid. In addi- to beneficiaries of existing safety net programs, tion, the existing Pantawid cash delivery in order to reach vulnerable groups and platform and national targeting systems helped inject  much-needed liquidity into the economy. the World Food Programme and the United A recent impact evaluation found that the Nations Children’s Fund (UNICEF) provide top-up transfers were received in a timely fashion and benefit amounts to Pantawid households in that those receiving the transfers recovered affected areas. Emergency support was ­provided faster than those who did not (Mansur, Doyle, for two months and included the activation of and Ivaschenko 2017). BOX 5.3  Investing in Risk and Vulnerability Information and Tying It to Safety Net Programming in the Dominican Republic, Kenya, and the Republic of Yemen Kenya’s Hunger Safety Nets Program is an helps predict and map potential vulnerable unconditional, poverty-targeted cash transfer areas and coordinate disaster responses. program that can expand horizontally and ver- The Republic of Yemen used an adaptive tically, acting as an emergency cash transfer in approach to respond to a humanitarian crisis times of drought. In response to drought events due to armed conflict. There, existing SSN and specifically, the scaling up is determined by social protection programs were reoriented to objective triggers and thresholds in terms of help manage food insecurity, address the lack environmental deterioration, measured by a of critical basic services, and deal with losses Vegetation Condition Index. The predeter- of employment and livelihoods. The approach mined triggers are used to set the benefit level introduced a conflict-sensitive monitoring and the eligibility of households (NDMA 2016). arrangement that uses GPS technology, real- The Dominican Republic’s safety net time data flows, and third-party monitoring. systems use a single beneficiary system called Targeting also complemented a poverty SIUBEN, which contains socioeconomic and approach with measures to identify conflict- demographic information on poor populations. related vulnerabilities, such as internally The information corresponds to a quality-of-life displaced people and their host communities, index that determines beneficiary eligibility for female-headed households, and youth. safety net programming. Recently, innovative In addition, the allocation of assistance steps have integrated vulnerability with climate adopted a conflict-sensitive approach, ensuring change into SIUBEN. The integrated approach predefined objectives as well as transparent estimates the probability of a household being and data-based criteria that could translate vulnerable to hurricanes, storms, and flooding, into a “distress index” and be used in a fund- given its socioeconomic characteristics; this allocation formula. predictable and timely responses (see, for in response to  shocks, based on predefined example, Decron and Clarke 2016). With triggers for dispersal. direct linkages to safety nets, risk financing In summary, ASP is an emerging agenda in can mobilize funds quickly in support of the the field of social protection. Given the sheer rapid scaling up of social protection programs degree of complexity associated with the issues The Emergence of Adaptive Social Protection 91 that ASP seeks to address—multiple risks and Garschagen, M., M. Hagenlocher, M. Comes, R. Sabelfeld, shocks, vulnerability, uncertainty, and their Y. J. Lee, L. Grunewald, M. Lanzendörfer, P. Mucke, O.  Neuschäfer, S. Pott, J. Post, S. Schramm, interconnectedness—a neat and comprehen- D.  Schumann-Bölsche, B. Vandemeulebroecke, sive framing of all elements of the growing ASP T.  Welle, and J. Birkmann. 2016.  World Risk Report agenda is somewhat elusive. However, it is clear 2016: The Importance of Infrastructure. Bonn: Bündnis in the current global context that social protec- Entwicklung Hilft  and UNU-EHS. http://collections​ tion and SSN practitioners and policy makers .unu.edu/eserv​ /­U NU:5763/WorldRiskReport2016​ _­small.pdf. must begin to factor such issues into their thinking more fully and undertake greater ­ Gentilini, U. 2016. “Sorting through the Hype: Exploring preparedness for shocks. ASP is a recognition ­ the Interface between Humanitarian Assistance and of this necessity. The approach outlined in this Safety Nets.” Social Protection and Labor Policy Note 19, World Bank, Washington, DC. chapter—building the resilience of the most vulnerable before shocks occur, and increasing Hoddinott, J., J. Lind, G. Berhne, K. Hirvonen, N. Kumar, the preparedness of SSNs to respond to the B. Biniyam, R. Sabates-Wheler, A. Strickland, A. Taffesse, M. Tefera, and Y. Yahanns. 2015. “PSNP- shocks of the future—will likely serve to make HABP Final Report, 2014”. http://essp.ifpri.info​ social protection more adaptive in the long /files/2017/07/PSNP-HABP_Final_Report_2014​ run, and enable it to more effectively protect .pdf. the well-being of the most vulnerable against IDS (Institute for Development Studies). 2012. “Realising the impacts of all manner of shocks. the Potential of Adaptive Social Protection.” IDS in Focus Policy Briefing Note 28. http://www.ids.ac.uk​ /­files/dmfile/IF28.pdf.   REFERENCES Arnall, A., K. Oswald, M. Davies, T. Mitchell, and Mansur, A., J. Doyle, and O. Ivaschenko. 2017. “Social C. Coirolo. 2010. “Adaptive Social Protection: Mapping Protection and Humanitarian Assistance Nexus for the Evidence and Policy Context in the Agriculture Disaster Response: Lessons Learnt from Fiji’s Tropical Sector in South Asia.” Working Paper No. 345, Institute Cyclone Winston.” Social Protection and Labor of Development Studies, Brighton, UK. http://www​ Discussion Paper 1701, World Bank, Washington, DC. .ids.ac.uk/files/dmfile​/­Wp345.pdf. NDMA (National Disaster Management Authority of ASPIRE (Atlas of Social Protection: Indicators of Resilience Kenya). 2016. “Hunger Safety Net Program Scalability and Equity). 2017. Database, Washington, DC: World Guidelines: Standard Operating Procedures for Bank. http://datatopics.worldbank.org​/aspire/. Scaling-Up HSNP Payments.” http://www.hsnp.or.ke​ /­i ndex.php/news/current-news​/12-current-news​ Banerjee, A., E. Duflo, N. Goldberg, D. Karlan, R. Osei, /112-new-hsnp-scalability-guidelines. W.  Pariente, J. Shapiro, B. Thuysbaert, and C. Udry. 2015. “A Multifaceted Program Uses Lasting Progress Oxford Policy Management. 2015. “Conceptualizing for the Very Poor: Evidence from Six Countries.” Shock-Responsive Social Protection.” https://www​ Science 348 (6236): 1260799. http://www.econ.yale​ .irishaid.ie/media/irishaid/allwebsitemedia/20new​ .edu/~cru2/pdf/Science-2015-TUP.pdf. sandpublications/publicationpdfsenglish/s-2947​/­IA-S ocial-Protection-Strategy-2017.pdf. Bowen, T. 2015.  “Social Protection and Disaster Risk Management in the Philippines: The Case of Typhoon UNHCR (United Nations High Commission on Refugees). Yolanda (Haiyan).” Policy Research Working Paper 2016. Global Trends: Forced Displacement in 2016. 7482, World Bank, Washington, DC. http://www.unhcr.org/globaltrends2016/. Del Ninno, C., and S. Coll-Black. 2016. Social Protection WEF (World Economic Forum). 2017. The Global Risks Programs for Africa’s Drylands. Washington, DC: Report, 2017. Geneva: WEF. http://www3.weforum​ World Bank. .org/docs/GRR17​_Report_web.pdf. Dercon, S., and D. Clarke. 2016. Dull Disasters? How White, P., and F. Ellis. 2012. “Ethiopia’s Productive Safety Planning Ahead Will Make a Difference. Oxford, UK: Net Program, 2010–2014: A Value for Money Oxford University Press. Assessment.” DFID/International Development UEA. EM-DAT (Emergency Events Database). Database, Centre for Research on the Epidemiology of Disasters, School World Bank. 2012a. “An Evaluation of World Bank of Public Health, Universite Catholique de Louvain, Support, 2000–2010.” Independent Evaluation Group, Brussels, Belgium. http://www.emdat.be/database/. World Bank, Washington, DC. 92 The State of Social Safety Nets 2018 ———. 2012b. “World Bank 2012–2022 Social Protection ———. 2016b. Shock Waves: Managing the Impacts and Labor Strategy: Resilience, Equity, and of Climate Change on Poverty. Washington, DC: Opportunity.” World Bank, Washington, DC. World Bank. ———. 2016a. “Closing the Gap: Building Resilience to ———. 2016c. Unbreakable: Building the Resilience of the Natural Disasters and Man-Made Shocks through Poor in the Face of Natural Disasters. Washington, DC: Social Safety Nets.” Brief, World Bank, Washington, DC. World Bank. The Emergence of Adaptive Social Protection 93 APPENDIX A Methodological Framework, Definitions, and Data Sources The Atlas of Social Protection: Indicators of total years and amount of contributions. Resilience and Equity (ASPIRE) database of the Examples of social insurance programs World Bank Group is the primary source of this include contributory old-age, survivor, and book. This appendix provides ASPIRE’s defini- disability pensions; sick leave and maternity/ tions, methodology, and data sources for gener- paternity benefits; and health insurance ating public expenditure and performance coverage. indicators. Further information is available in 3. Labor market programs can be contributory Data Sources and Methodology of the ASPIRE or noncontributory and are designed to help website.1 protect individuals against loss of income from unemployment (passive labor market SOCIAL PROTECTION AND LABOR policies) or help individuals acquire skills PROGRAMS CLASSIFICATION and connect them to labor markets (active As discussed in Chapter 1, social protection labor market policies). Unemployment and labor (SPL) generally fall into three main insurance and early retirement incentives categories: are examples of passive labor market policies, while training, employment 1. Social safety net (SSN)/social assistance (SA) intermediation services, and wage subsidies programs are noncontributory interventions are examples of active policies. that are designed to help individuals and households cope with chronic poverty, For cross-country comparability, this book destitution, and vulnerability. Potential adheres to the ASPIRE harmonized classifica- beneficiaries are not required to pay a tion of SPL programs. ASPIRE groups SPL pro- premium (contribute) to access benefits. SSN/ grams into three program areas (social safety SA programs target the poor and vulnerable. nets social assistance, social insurance, and 2. Social insurance is a contributory intervention labor markets) with 12 harmonized categories that is designed to help individuals manage based on program objectives. While Chapter 1 sudden changes in income due to old discusses the SSN/SA classification, the infor- age,  sickness, disability, or natural disaster. mation in this appendix is more extensive. Individuals pay insurance premiums to This standardization is applied to each coun- be eligible for coverage or contribute a try in the ASPIRE database to generate compa- percentage of their earnings to an insurance rable expenditure and performance indicators scheme to access benefits, which link to the (see table A.1). 94 APPENDIXES TABLE A.1  ASPIRE Social Protection and Labor Program Classification Program category Program subcategory Social safety net/social assistance Unconditional cash transfers Poverty-targeted cash transfers and last-resort programs Family, children, and orphan allowances (including benefits for vulnerable children) Noncontributory funeral grants, burial allowances Emergency cash support (including support to refugees and returning migrants) Public charity, including zakāt Conditional cash transfers Conditional cash transfers Social pensions (noncontributory) Old-age social pensions Disability benefits, war victim noncontributory benefits Survivorship benefits Food and in-kind transfers Food stamps, rations, and vouchers Food distribution programs Nutritional programs (therapeutic, supplementary feeding, and people living with HIV) School supplies (free textbooks and uniforms) In-kind and nonfood emergency support Other in-kind transfers School feeding School feeding Public works, workfare, and direct Cash for work job creation Food for work (including food for training and for assets) Fee waivers and targeted subsidies Health insurance exemptions and reduced medical fees Education fee waivers Food subsidies Housing subsidies and allowances Utility and electricity subsidies and allowances Agricultural inputs subsidies Transportation benefits Other social assistance Scholarships, education benefits Social care services, transfers for caregivers (care for children, youth, family, disabled, and older persons) Tax exemptions Other Social insurance Contributory pensions Old-age pension (all schemes: national, civil servants, veterans, and other special categories) Survivors pension (all schemes: national, civil servants, veterans, and other special categories) Disability pension (all schemes: national, civil servants, veterans, and other special categories) Other social insurance Occupational injury benefits Paid sick leave Health Maternity and paternity benefits Contributory grants (insurance) Other social insurance (Table continues next page) APPENDIXES 95 TABLE A.1  ASPIRE Social Protection and Labor Program Classification (Continued) Program category Program subcategory Labor market Labor market policy measures Labor market services and intermediation through public employment services (active labor market programs) Training (vocational, life skills, and cash for training) Employment incentives and wage subsidies Employment measures for the disabled Entrepreneurship support and startup incentives (cash and in-kind grants, microcredit) Job rotation and job sharing Other active labor market programs Labor market policy support Out-of-work income maintenance (contributory unemployment benefits) (passive labor market programs) Out-of-work income maintenance (noncontributory employment benefits) Benefits for early retirement Source: ASPIRE. Note: ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. DATA SOURCES DATA HARMONIZATION METHODOLOGY The ASPIRE dataset has two main sources of Harmonization Methodology for information: administrative data and house- Administrative Data hold survey data. The ASPIRE administra- The book uses only the latest year data available tive  database sources include primary and for each country and only if the data are not secondary sources of data. Program-level older than 2010. For four countries (Bhutan, ­ administrative data are used to generate pub- Jordan, Marshal Islands, and Vanuatu), the lat- lic  expenditure indicators and the number of est available year is 2009 and includes only total beneficiaries. SSN spending. To ensure comparability and Household survey data for 96 countries are aggregation in spending across programs, the used to generate the performance indicators book first divides program-level spending by (see appendix B for a full list of the surveys). the corresponding year GDP for which the pro- Those cover national representative surveys gram data are available (only active/ongoing with information on income, con­ sumption, programs are used). Overall spending for SSN and  SPL programs, including the following: programs in the country is then approximated Household  Income and Expenditure/Budget by summing up all program shares as a percent- Surveys, Living Standard Measurement age of GDP. For each country, the book then Surveys, Multiple Indicator Cluster Surveys, provides the total SSN spending (as percent- Surveys on Income and Living Conditions, age of GDP), total spending without health fee and Welfare Monitoring Surveys. The book waivers (as a percentage of GPD), and the only uses the latest year for each country and spending (as a percentage of GDP) on the SSN/ only if the data are not older than 2008; under SA specific component (see appendix D). this criterion, 20  countries were excluded. To calculate program expenditures, the book In  addition, countries were excluded where considers all program spending no matter the surveys did not have social safety nets whether it is government- or donor-financed. information. When calculating averages (global or regional), As of October of 2017, the ASPIRE adminis- all countries are assumed to have equal weight trative database included information on the (in other words, simple averages are calculated). number of beneficiaries for 142 countries Country GDP is taken from the World (appendix C) and program-level spending data Development Indicators database. Timor-Leste for 124 countries (see appendix D). data on GDP is taken from the World Economic 96 APPENDIXES Outlook database. Average spending on SSNs Household weights are used to expand the for member countries of the Organisation results to the total population of each country. for  Economic Co-operation and Development For cross-country comparability, all mone- (OECD) (Hungary, Slovak Republic, and tary variables are expressed in 2011 prices Slovenia) and for OECD averages are based on and  daily purchasing power parity (PPP) in the OECD Social Expenditure Database (SOCX) U.S. dollars. The consumption or income aggre- by combining “family” and “other social policy gates used to rank households by their welfare functions” as the closest approximation to non- distribution are validated by the World Bank contributory safety nets, as defined in this book. regional poverty teams. The book uses two definitions of poverty: rel- Harmonization Methodology for Household ative poverty (individuals in the poorest Survey Data 20 percent of the welfare distribution), and Household surveys are reviewed to identify SPL absolute poverty (individuals living on less than program information. Individual variables are $1.90  purchasing power parity a day). Pre- generated for each SPL program that is captured transfer welfare (income or consumption with- in the survey; the individual variables are then out the SPL transfer) is used to generate the grouped into the 12 SPL harmonized program indicators by quintile, except for the adequacy categories. Performance indicators are gener- of benefits indicator; posttransfer welfare is ated using the harmonized program variables. used to generate the adequacy indicator. APPENDIXES 97 APPENDIX B Household Surveys Used in the Book TABLE B.1  Household Surveys Used in the Book Country/economy/ territory Year Region Survey name Afghanistan 2011 SA National Risk and Vulnerability Assessment (NRVA) 2011–2012, Living Conditions Survey Albania 2012 ECA Living Standards Measurement Survey Argentina 2013 LAC Encuesta Permanente de Hogares Continua Armenia 2014 ECA Integrated Living Conditions Survey Bangladesh 2010 SA Household Income and Expenditure Survey Belarus 2013 ECA Household Budget Survey Belize 2009 LAC Living Standards Measurement Survey (LSMS) Bhutan 2012 SA Bhutan Living Standards Survey Bolivia 2012 LAC Encuesta de Hogares Botswana 2009 SSA Core Welfare Indicators Survey Brazil 2015 LAC Pesquisa Nacional por Amostra de Domicílios Contínua Burkina Faso 2014 SSA Enquete Multisectorielle Continue Cameroon 2014 SSA Quatrième Enquête Camerounaise Auprès des Ménages 2014 (ECAM4) Central African Republic 2008 SSA Enquête Centrafricaine pour le Suivi-Evaluation du Bien-être Chad 2011 SSA Troisième Enquete Sur La Consommation et le Secteur Informel China 2013 EAP Chinese Household Income Project 2013–2014 Chile 2013 LAC Encuesta de Caracterización Socioeconómica Nacional (CASEN) Colombia 2014 LAC Encuesta Nacional de Calidad de Vida (ENCV) Congo, Dem. Rep. 2012 SSA Troisieme Enquete Sur La Consommation et le Secteur Informel Costa Rica 2014 LAC Encuesta Nacional de Hogares (ENAHO) Côte d’Ivoire 2014 SSA Enquete sur le Niveau de Vie des Menages de Côte d’Ivoire 2014–2015 Croatia 2010 ECA Household Budget Survey Djibouti 2012 MENA Enquete Djiboutienne Aupres des Menages (EDAM 3-IS) Dominican Republic 2014 LAC Encuesta Nacional de Fuerza de Trabajo Ecuador 2016 LAC Encuesta Nacional de Empleo Desempleo y Subempleo Egypt, Arab Rep. 2008 MENA Household Income, Expenditure, and Consumption Survey 2008–2009 El Salvador 2014 LAC Encuesta de Hogares de Propósitos Múltiples Ethiopia 2010 SSA Household Income, Consumption and Expenditures Fiji 2008 EAP Household Income and Expenditure Survey Gambia 2010 SSA Integrated Household Survey 2010–2011 (Table continues next page) 98 APPENDIXES TABLE B.1  Household Surveys Used in the Book (Continued) Country/economy/ territory Year Region Survey name Georgia 2011 ECA Welfare Monitoring Survey Ghana 2012 SSA Living Standards Survey V 2012–2013 Guatemala 2014 LAC Encuesta Nacional de Condiciones de Vida Guinea 2012 SSA Enquête Légère pour l’Evaluation de la Pauvreté Haiti 2012 LAC Enquête sur les Conditions de Vie des Ménages après Séisme 2012–2013 Honduras 2013 LAC Encuesta Permanente de Hogares de Propósitos Múltiples India 2011 SA National Sample Survey 2011–2012 (68th round) - Schedule 10 - Employment and Unemployment Indonesia 2015 EAP Survei Sosial Ekonomi Nasional 2015, Maret (SUSENAS) Iraq 2012 MENA Household Socio Economic Survey Jamaica 2010 LAC Survey of Living Conditions Jordan 2010 MENA Household Income and Expenditure Survey Kazakhstan 2010 ECA Household Budget Survey Kosovo 2013 ECA Household Budget Survey Kyrgyz Republic 2013 ECA Kyrgyz Integrated Household Survey Latvia 2009 ECA Household Budget Survey Liberia 2014 SSA Household Income and Expenditure Survey 2014–2015 Lithuania 2008 ECA Household Budget Survey Madagascar 2010 SSA Enquete Periodique Aupres Des Menages (EPM 2010) Malawi 2013 SSA Integrated Household Panel Survey 2013 Malaysia 2008 EAP Household Income Survey Maldives 2009 SA Household Income and Expenditure Survey 2009–2010 Mauritania 2014 SSA Enquête Permanente sur les Conditions de Vie des ménages 2014 Mauritius 2012 SSA Household Budget Survey Mexico 2012 LAC Encuesta Nacional de Ingresos y Gastos de los Hogares Moldova 2013 ECA Household Budget Survey Mongolia 2012 EAP Household Socio-Economic Survey 2012 Montenegro 2014 ECA Household Budget Survey Morocco 2009 MENA Household and Youth Survey Mozambique 2008 SSA Inquerito Sobre Orçamento Familiar 2008–2009 Namibia 2009 SSA National Household Income and Expenditure Survey 2009–2010 Nepal 2010 SA Living Standards Survey 2010–2011, Third Round Nicaragua 2014 LAC Encuesta Nacional de Hogares sobre Medición de Nivel de Vida Niger 2014 SSA Enquête Nationale sur les Conditions de Vie des Ménages et l’Agriculture Nigeria 2015 SSA General Household Survey Panel Pakistan 2013 SA Social and Living Standards Measurement (PSLM_HIES) 2013–2014 Panama 2014 LAC Encuesta de Mercado Laboral Papua New Guinea 2009 EAP Household Income and Expenditure Survey 2009–2010 Paraguay 2011 LAC Encuesta Permanente de Hogares (Table continues next page) APPENDIXES 99 TABLE B.1  Household Surveys Used in the Book (Continued) Country/economy/ territory Year Region Survey name Peru 2014 LAC Encuesta Nacional de Hogares Philippines 2015 EAP Family Income and Expenditure Survey 2015–2016 Poland 2012 ECA Household Budget Survey Romania 2012 ECA Household Budget Survey Russian Federation 2016 ECA Statistical Survey of Income and Participation in Social Programs Rwanda 2013 SSA Integrated Household Living Conditions Survey 2013–2014 Senegal 2011 SSA Enquete de Suivi de la Pauvrete au Senegal 2011 Serbia 2013 ECA Household Budget Survey Sierra Leone 2011 SSA Integrated Household Survey (SLIHS)–Main Survey Slovak Republic 2009 ECA Household Income and Living Conditions Survey South Africa 2010 SSA Income and Expenditure Survey South Sudan 2009 SSA National Baseline Household Survey 2009, First Round Sri Lanka 2012 SA Household Income and Expenditure Survey 2012–2013 Sudan 2009 SSA National Baseline Household Survey 2009 Swaziland 2009 SSA Household Income and Expenditure Survey 2009–2010 Tajikistan 2011 ECA Panorama Tanzania 2014 SSA LSMS - National Panel Survey 2014–2015, Wave 4 Thailand 2013 EAP Household Socio-Economic Survey Timor-Leste 2011 EAP Household Income and Expenditure Survey 2011–2012 Tunisia 2010 MENA Enquete Nationale Sur le Budget la Consommation et le Niveau de Vie des Menage Turkey 2014 ECA Household Income and Consumption Expenditures Survey Uganda 2012 SSA National Household Survey 2012–2013 Ukraine 2013 ECA Household Living Conditions Survey Uruguay 2012 LAC Encuesta Continua de Hogares Vietnam 2014 EAP Household Living Standard Survey West Bank and Gaza 2009 MENA Expenditure and Consumption Survey Zambia 2010 SSA Living Conditions Monitoring Survey VI (LCMS VI) Zimbabwe 2011 SSA Income, Consumption and Expenditure Survey 2011–2012 Source: ASPIRE. Note: The total number of countries/economies/territories for this analysis is 96. ASPIRE = Atlas of Social Protection: Indicators for Resilience and Equity; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; SA = South Asia; SSA = Sub-Saharan Africa. 100 APPENDIXES APPENDIX C Global Program Inventory Appendix C presents information available in fewer than 10–15 (as in Bolivia, Croatia, and the ASPIRE database on the largest programs Timor-Leste), whereas in high-number cases (in  terms of beneficiary numbers) existing in there might be more than 50 programs (as in 142  countries/economies/territories by aggre- Chile and Burkina Faso). Thus, for some more gate program categories. Countries/economies/ program-fragmented countries/­economies/​terri- territories differ significantly in their number of tories, appendix C does not show the full picture operating social safety net (SSN) programs. In of coverage or versatility of programs and should some cases, the number of programs might be be treated with caution. APPENDIXES 101 TABLE C.1  Conditional Cash Transfers and Unconditional Cash Transfers   Conditional cash transfer Unconditional cash transfer Number of Number of individual individual beneficiaries Estimated beneficiaries Country/ (unless number of (unless economy/ Program specified Beneficiary beneficiaries Program specified territory name otherwise) unit is HH (individuals) Year name otherwise) Albania — — — — — Ndihme 106,635 Ekonomike Algeria — — — — — — — Angola — — — — — Cartão 200,000 Kikuia–Kikuia Card Cash Transfer Programme Argentina Asignación 3,560,704 — 3,560,704 2015 — — Universal por Hijo para la Protección Social Armenia — — — — — Family Poverty 104,131 Benefit Azerbaijan — — — — — Targeted Social 530,670 Assistance Bangladesh Stipend for 7,800,000 — 7,800,000 2013 Allowances for 1,113,200 primary students Widows, Deserted and Destitute Women Belarus — — — — — Child care 348,261 benefit, for children up to 3 years old Belize Building 8,600 — 8,600 2012 — — Opportunities for Our Social Transformation (BOOST) Benin — — — — — Decentralized 13,000 Services Driven by Communities Bhutan — — — — — — — Bolivia Bono Juancito 2,228,000 — 2,228,000 2015 — — Pinto Bosnia and — — — — — Child Protection 105,844 Herzegovina Allowance Botswana — — — — — Destitute persons 35,441 program Brazil Bolsa Familia 41,810,373 — 41,810,373 2015 — — Bulgaria — — — — — Family or child 777,726 allowance Burkina Faso Social safety net 88,500 — 88,500 2016 Unconditional 27,000 project cash distribution “Burkin-Nong- operations Saya” (supplement to cereal distribution) 102 APPENDIXES Unconditional cash transfer Social pension Number of Estimated individual Estimated number of beneficiaries number of Beneficiary beneficiaries Program (unless specified Beneficiary beneficiaries unit is HH (individuals) Year name otherwise) unit is HH (individuals) Year HH 415,877 2014 — — — — — — — — — — — — — — 200,000 2015 Old-age, survivor and — — — — disability — — — Pensión no contributiva 1,068,959 — 1,068,959 2015 por discapacidad HH 374,872 2014 — — — — — HH 2,600,283 2014 Old-age allowance 13,833 — 13,833 2015 (persons who are not entitled for pension) — 1,113,200 2015 Old-age allowance 3,000,000 — 3,000,000 2015 — 348,261 2015 Benefit for taking care 42,575 — 42,575 2015 of disabled category 1 or seniors above 80 — — — Social pension 3,711 — 3,711 2015 — 13,000 2015 — — — — — — — — — — — — — — — — Renta Dignidad 934,748 — 934,748 2015 HH 324,941 2010 — — — — — — 35,441 2016 The Old-Age Pension 105,754 — 105,754 2016 (OAP) — — — Old-age social pensions 2,323,808 — 2,323,808 2015 (Beneficio de Prestacao Continuada– Idosos) HH 1,710,997 2014 — — — — — HH 135,000 2015 — — — — — (Table continues next page) APPENDIXES 103 TABLE C.1  Conditional Cash Transfers and Unconditional Cash Transfers (Continued)   Conditional cash transfer Unconditional cash transfer Number of Number of individual individual beneficiaries Estimated beneficiaries Country/ (unless number of (unless economy/ Program specified Beneficiary beneficiaries Program specified territory name otherwise) unit is HH (individuals) Year name otherwise) Burundi — — — — — Take a Step 2,000 Forward (Terintambwe) Cabo Verde — — — — — Support for 550 orphans and other vulnerable children Cambodia MoEYS 150,655 — 150,655 2015 — — Scholarships for Primary and Secondary Education Cameroon Social Safety 21,500 — 21,500 2016 Program 559: 118,710 Nets–Cash National transfers Solidarity and Social Justice Central — — — — — Emergency 46,168 African livelihood Republic support to conflict-affected populations in southwestern Central African Republic Chad Projet ECHO6 11,833 — 11,833 2016 Protection 308,862 - CARE awaiting solutions of Sudanese refugees settled in eastern Chad Chile Subsidio unico 2,015,393 — 2,015,393 2015 — — familiar China — — — — — Dibao 69,040,000 Colombia Mas Familias en 13,672,125 — 13,672,125 2015 — — Accion Comoros — — — — — Cash transfers 2,537 Congo, Dem. — — — — — Alternative 64,343 Rep. Responses for Communities in Crisis cash transfer programme (ARCC2) Congo, Rep. FSA project 40,000 — 40,000 2015 — — Costa Rica Avancemos 167,029 — 167,029 2015 — — Côte d’Ivoire — — — — — Programme 5,000 National des Filets Sociaux Productifs Croatia — — — — — Child and family 216,013 benefits  104 APPENDIXES Unconditional cash transfer Social pension Number of Estimated individual Estimated number of beneficiaries number of Beneficiary beneficiaries Program (unless specified Beneficiary beneficiaries unit is HH (individuals) Year name otherwise) unit is HH (individuals) Year HH 10,000 2013 — — — — — — 550 2010 Basic Pension 23,000 — 23,000 2011 — — — MoSAVY Diability — — — — Grant Pilot Project (starting May 2016) — 118,710 2016 — — — — — — 46,168 2015 — — — — — — 308,862 2016 — — — — — — — — Old-age solidarity 399,049 — 399,049 2015 pensions — 69,040,000 2015 Government programs 12,638,000 — 12,638,000 2009 for disabled persons — — — Programa Colombia 1,473,690 — 1,473,690 2014 Mayor HH 13,700 2016 — — — — — — 64,343 2015 — — — — — — — — — — — — — — — — Social pension 104,141 — 104,141 2015 — 23,000 2016 — — — — — — 216,013 2011 — — — — — (Table continues next page) APPENDIXES 105 TABLE C.1  Conditional Cash Transfers and Unconditional Cash Transfers (Continued)   Conditional cash transfer Unconditional cash transfer Number of Number of individual individual beneficiaries Estimated beneficiaries Country/ (unless number of (unless economy/ Program specified Beneficiary beneficiaries Program specified territory name otherwise) unit is HH (individuals) Year name otherwise) Czech — — — — — Benefit in 71,153 Republic material need Djibouti Programme 16,344 HH 91,526 2015 Distribution de 6,740 National de zakāt Solidarité Famille (PNSF) Dominica — — — — — — — Dominican Progresando con 2,542,384 — 2,542,384 2015 — — Republic Solidaridad (PROSOLI) Ecuador Bono de 444,562 HH 1,640,434 2014 — — Desarollo Humano Egypt, Arab Takāful and  1,964,895 HH 8,347,320 2017 — — Rep. Karama El Salvador Comunidades 75,000 HH 324,000 2014 — — Solidarias Rurales Estonia — — — — — Child Allowance 251,075 Ethiopia — — — — — Pilot social cash 17,705 transfer – Tigray Fiji iTaukei 3,500 — 3,500 2010 Family 22,826 Assistance Program (FAP) Gabon — — — — — National Social — Action Fund (HIF) Gambia, The — — — — — Cash transfers — Georgia — — — — — Targeted social 428,492 assistance Ghana — — — — — Livelihood 213,414 Empowerment Against Poverty (LEAP) Grenada Support for 7,368 — 7,368 2015 Child assistance 68 Education, (Carriacou) Empowerment & Development (SEED) Guatemala Mi Bono 1,021,959 HH 5,150,673 2013 — — Seguro – Bono Seguro Escolar Guinea Cash transfer for 10,000 — 10,000 2012 — — nutrition and for girl’s education Guinea-Bissau — — — — — Cash transfer 200,000 program Haiti Ti Manman 86,234 — 86,234 2014 — — Cheri 106 APPENDIXES Unconditional cash transfer Social pension Number of Estimated individual Estimated number of beneficiaries number of Beneficiary beneficiaries Program (unless specified Beneficiary beneficiaries unit is HH (individuals) Year name otherwise) unit is HH (individuals) Year HH 142,306 2008 — — — — — — 6,740 2015 — — — — — — — — — — — — — — — — Suplemento 111,389 — 111,389 2015 Alimenticio del Programa de Protección a la Vejez en Extrema Pobreza — — — Bono matrícula para la 3,015,199 — 3,015,199 2010 eliminación del aporte voluntario — — — Social solidarity 7,000,000 — 7,000,000 2014 pension — — — Universal basic pension 28,200 — 28,200 2013 for the elderly HH 502,150 2014 National pensions 6,516 — 6,516 2014 — 17,705 2014 — — — — — HH 107,282 2015 Social Pension Scheme 22,073 — 22,073 2015 (SPS) — — — — — — — — — — — — — — — — — 428,492 2013 Old-age pension 684,301 — 684,301 2013 HH 939,022 2016 — — — — — HH 202 2015 Public assistance 206 — 206 2015 — — — Social Pension program 108,664 — 108,664 2013 for elderly — — — — — — — — — 200,000 2015 Program for the 800 — 800 2015 handicapped — — — — — — — — (Table continues next page) APPENDIXES 107 TABLE C.1  Conditional Cash Transfers and Unconditional Cash Transfers (Continued)   Conditional cash transfer Unconditional cash transfer Number of Number of individual individual beneficiaries Estimated beneficiaries Country/ (unless number of (unless economy/ Program specified Beneficiary beneficiaries Program specified territory name otherwise) unit is HH (individuals) Year name otherwise) Honduras Bono Vida Mejor 259,879 — 259,879 2015 — — Hungary For the Road 26,000 — 26,000 2008 Regular social 269,000 assistance India Janani Suraksha 1,946,858 — 1,946,858 2016 National Family — Yojana Benefit Scheme (NFBS) Indonesia Program 6,000,000 HH 23,400,000 2016 Bantuan 15,800,000 Keluarga Langsung Harapan Sementara Masyrakat (BLSM) Iran, Islamic — — — — — Compensatory 6,100,000 Rep. cash transfer Iraq — — — — — Social Protection 877,520 Network Jordan — — — — — National Aid 250,000 Fund Kazakhstan BOTA 135,000 — 135,000 2010 State payment to 562,614 foundation CCT families with children under 18 years Kenya Cash transfer for 365,232 HH 1,765,000 2016 Hunger Safety 101,630 Orphans and net program vulnerable (HSNP) children (OVC) Kiribati — — — — — — — Kosovo — — — — — Social Assistance 29,506 Scheme (Ndihma I and II) Kuwait Families with — — — — Physical — students grant Disability Grant Kyrgyz — — — — — Monthly Benefit 361,500 Republic for Poor Families with Children (MBPF) Lao — — — — — — — Latvia — — — — — Family State 306,300 Benefit Lebanon — — — — — Family and — education allowance Lesotho — — — — — Child Grants 24,500 Program (CGP) Liberia — — — — — Social cash 8,000 transfer program – income transfer plus Lithuania — — — — — Social benefit 87,898 108 APPENDIXES Unconditional cash transfer Social pension Number of Estimated individual Estimated number of beneficiaries number of Beneficiary beneficiaries Program (unless specified Beneficiary beneficiaries unit is HH (individuals) Year name otherwise) unit is HH (individuals) Year — — — Bono Tercera Edad 36,919 — 36,919 2012 — 269,000 2009 — — — — — — — — Indira Gandhi National 24,243,753 — 24,243,753 2016 Old-Age Pension Scheme (IGNOAPS) HH 61,620,000 2015 Elderly Social Security 26,500 — 26,500 2014 Programme (Pilot) — 6,100,000 2009 — — — — — — 877,520 2013 — — — — — — 250,000 2011 — — — — — — 562,614 2014 Base Pension (Pillar 1,848,469 — 1,848,469 2014 Zero) HH 507,190 2016 Older Persons Cash 320,636 — 320,636 2016 Transfer (OPCT) — — — Elderly Pension 2,090 — 2,090 2010 — 29,506 2014 Basic pension 125,883 — 125,883 2014 — — — — — — — — — 1,662,900 2012 — — — — — — — — Cash and in-kind — — — — transfer for veterans and elderly (Decree 343/PM) — 306,300 2014 — — — — — — — — — — — — — HH 117,600 2015 Old-age pension 85,087 — 85,087 2015 HH 41,086 2016 — — — — — — 87,898 2016 Social pension — — — — (Table continues next page) APPENDIXES 109 TABLE C.1  Conditional Cash Transfers and Unconditional Cash Transfers (Continued)   Conditional cash transfer Unconditional cash transfer Number of Number of individual individual beneficiaries Estimated beneficiaries Country/ (unless number of (unless economy/ Program specified Beneficiary beneficiaries Program specified territory name otherwise) unit is HH (individuals) Year name otherwise) Macedonia, CCT – increased 7,679 — 7,679 2014 Social financial 31,085 FYR child allowance assistance Madagascar Filets Sociaux de 26,500 HH 127,272 2016 Travaux HIMO 6,660 Sécurité (FSS) TMDH Malawi — — — — — Social Cash 782,561 Transfer Scheme (SCTS) Malaysia — — — — — Bantuan Rakyat 15,300,000 1 Malaysia (BR1M) scheme Maldives — — — — — Single Parents’ — Allowance Mali — — — — — Jigisemejiri 60,715 Marshall — — — — — — — Islands Mauritania Tekavou – 7,100 HH 31,110 2016 — — conditional cash transfers Mauritius — — — — — Basic Widow’s 19,619 Pension Mexico Prospera 6,168,900 HH 23,441,820 2015 — — Moldova — — — — — Ajutor Social 53,605 Mongolia — — — — — The Child Money 960,300 Programme Montenegro — — — — — Family material 12,830 support and benefits based on social care Morocco — — — — — INJAZ 22,627 Mozambique — — — — — Basic Social 379,850 Subsidy Programme Myanmar Stipends 37,000 — 37,000 2014 Cash and in-kind 318,157 Program support to (Ministry of internally Education) displaced people Namibia — — — — — Provision of 175,659 Social Assistance Nepal Basic Education 1,075,260 — 1,075,260 2013 Basic education 1,985,657 for Dalits for girls Nicaragua Mi Beca familiar — — — — — — Niger Projet de Filets 68,737 — 309,317 2016 Family allowance — Sociaux – Cash transfer with disaster risk management component 110 APPENDIXES Unconditional cash transfer Social pension Number of Estimated individual Estimated number of beneficiaries number of Beneficiary beneficiaries Program (unless specified Beneficiary beneficiaries unit is HH (individuals) Year name otherwise) unit is HH (individuals) Year — 31,085 2014 — — — — — — 6,660 2016 — — — — — — 782,561 2016 — — — — — — 15,300,000 2014 Senior citizen aid 140,000 — 140,000 2013 — — — Old-age Pension — — — — Scheme HH 321,790 2016 — — — — — — — — — — — — — — — — — — — — — — 19,619 2016 Basic Retirement 195,591 — 195,591 2016 Pension (BRP) zero pillar retirement only — — — Pension para adultos 5,701,662 — 5,701,662 2015 mayores HH 160,815 2015 Social Pension (“for 6,222 — 6,222 2015 elderly”) — 960,300 2013 Social Welfare Pension 63,423 — 63,423 2013 HH 42,339 2008 — — — — — — 22,627 2013 — — — — — HH 1,671,340 2015 — — — — — — 318,157 2014 — — — — — — 175,659 2013 Provision of Social 146,482 — 146,482 2013 Assistance – Old-Age Grant — 1,985,657 2013 Old-age pension 922,741 — 922,741 2014 scheme — — — Defensa Civil 403,016 — 403,016 2013 — — — — — — — — (Table continues next page) APPENDIXES 111 TABLE C.1  Conditional Cash Transfers and Unconditional Cash Transfers (Continued)   Conditional cash transfer Unconditional cash transfer Number of Number of individual individual beneficiaries Estimated beneficiaries Country/ (unless number of (unless economy/ Program specified Beneficiary beneficiaries Program specified territory name otherwise) unit is HH (individuals) Year name otherwise) Nigeria Kano 16,271 — 16,271 2014 Eradication of 47,746 Conditional Extreme Poverty Cash Transfer for and Hunger/cash Girls’ Education transfer Pakistan Benazir Income 51,000 — 51,000 2014 Benazir Income 5,042,032 Support Program Support Program (BISP), CCT (BISP) component Panama Red de 67,385 HH 222,371 2015 — — Oportunidades Papua New — — — — — — — Guinea Paraguay Tekoporâ 722,377 — 722,377 2015 — — Peru Juntos 771,970 HH 2,933,486 2015 — — Philippines Pantawid 4,400,000 HH 20,240,000 2015 National — Pamilyang Comission of Pilipino Program Indigenous (4Ps) Peoples (NCIP) cash program Poland — — — — — Child allowance 3,820,000 500+ Qatar — — — — — — — Romania Money for High 79,810 — 79,810 2014 Universal Child 3,727,859 School Allowance (UCA) Russian — — — — — Child allowances 8,423,000 Federation Rwanda — — — — — Vision 2020 86,772 Umurenge (VUP) Samoa — — — — — — — São Tomé and Needy Mothers 1,224 HH 5,018 2017     Príncipe Saudi Arabia Support 428,028 — 428,028 2012 Regular 370,846 assistance for assistance: school bags and divorced/ uniforms widowed women Senegal National cash 300,000 HH 2,400,000 2016 Cash transfer — transfer nutritional programme programs Serbia — — — — — Child allowances 394,557 Seychelles — — — — — Social Welfare 2,978 Assistance Sierra Leone Social Safety 136,768 — 136,768 2016 The National — Nets Program Commission for Social Action (NaCSA) Slovak Motivation 31,000 — 31,000 2011 Material need 111,000 Republic allowance benefit Slovenia — — — — — Child benefits 371,000 112 APPENDIXES Unconditional cash transfer Social pension Number of Estimated individual Estimated number of beneficiaries number of Beneficiary beneficiaries Program (unless specified Beneficiary beneficiaries unit is HH (individuals) Year name otherwise) unit is HH (individuals) Year — 47,746 2013 Ekiti State Social — — — 2013 Security Scheme HH 31,260,598 2015 — — — — — — — — 120-65 117,940 — 117,940 2015 — — — — — — — — — — — Old-age social pensions 100,272 — 100,272 2015 — — — Pension 65 501,681 — 501,681 2015 — — — Social Pension/ 939,606 — 939,606 2015 Value-Added-Tax- Assisted Cash Subsidy to grandparents — 3,820,000 2016 — — — — — — — — — — — — — HH 10,065,219 2014 Social indemnity for 495,005 HH 1,336,514 2014 pensioners — 8,423,000 2013 Social pension — — — — HH 246,009 2015 Direct support for 2,821 — 2,821 2015 disabled former combatants — — — Senior citizens benefit 8,700 — 8,700 2010 —     Social pension –subsidy 2,024 — 2,024 2014 to the unknown — 370,846 2012 — — — — — — — — — — — — — HH 1,144,215 2013 — — — — — HH 11,019 2015 Retirement pension 9,496 — 9,496 2015 — — — — — — — — — 111,000 2011 — — — — — — 371,000 2007 Social pension — — — — (Table continues next page) APPENDIXES 113 TABLE C.1  Conditional Cash Transfers and Unconditional Cash Transfers (Continued)   Conditional cash transfer Unconditional cash transfer Number of Number of individual individual beneficiaries Estimated beneficiaries Country/ (unless number of (unless economy/ Program specified Beneficiary beneficiaries Program specified territory name otherwise) unit is HH (individuals) Year name otherwise) Solomon — — — — — — — Islands Somalia Building 15,000 HH 87,000 2016 Short-Term 5,925 Resilience Humanitarian through Social Transfers – Safety Nets in Unconditional Somalia Cash Transfers South Africa — — — — — Child Support 11,703,165 Grant South Sudan — — — — — Juba urban poor 42,000 cash response pilot Sri Lanka Free scholarship 85,000 — 85,000 2012 Divineguma 1,400,000 programs for Subsidy Program school children (ex Samurdhi) –Grade 5 St. Kitts and — — — — — — — Nevis St. Lucia — — — — — Public assistance 2,396 program St. Vincent — — — — — Public assistance 6,000 and the relief Grenadines Sudan National Student 200,000 — 200,000 2016 Zakāt 15,327,539 Welfare Fund Suriname — — — — — — — Swaziland — — — — — Public assistance 5,075 Syrian Arab — — — — — Social Welfare — Republic Fund Tajikistan Conditional cash — — — — Targeted social 11,184 payments, assistance (pilot) allowances to large families and children Tanzania Productive Social 1,098,856 HH 5,164,623 2016 — — Safety Net (PSSN) – conditional cash transfer Thailand — — — — — Baan Mankong — Program Timor-Leste Bolsa da Mae 54,488 HH 316,030 2012 Benefits for 31,852 veterans and survivor families Togo CCT with 12,079 HH 50,732 2015 Prise en charge — conditions on des enfants nutrition victimes de la traite 114 APPENDIXES Unconditional cash transfer Social pension Number of Estimated individual Estimated number of beneficiaries number of Beneficiary beneficiaries Program (unless specified Beneficiary beneficiaries unit is HH (individuals) Year name otherwise) unit is HH (individuals) Year — — — — — — — — HH 34,365 2016 — — — — — — 11,703,165 2015 Old-age grant 3,086,851 — 3,086,851 2015 — 42,000 2016 — — — — — HH 5,880,000 2016 Public Assistance 580,720 — 580,720 2015 Monthly Allowance (PAMA) — — — Assistance pensions 1,000 — 1,000 2008 HH 6,685 2014 — — — — — — 6,000 2009 — — — — — — 15,327,539 2016 — — — — — — — — Old-age social pensions 44,739 — 44,739 2015 — 5,075 2011 Old Age Grant (OAG) 63,500 — 63,500 2014 — — — — — — — — HH 26,842 2012 Social pension — — — — — — — — — — — — — — — Old-age allowance 5,698,414 — 5,698,414 2011 — 31,852 2015 Transfers for the elderly 84,569 — 84,569 2012 — — — — — — — — (Table continues next page) APPENDIXES 115 TABLE C.1  Conditional Cash Transfers and Unconditional Cash Transfers (Continued)   Conditional cash transfer Unconditional cash transfer Number of Number of individual individual beneficiaries Estimated beneficiaries Country/ (unless number of (unless economy/ Program specified Beneficiary beneficiaries Program specified territory name otherwise) unit is HH (individuals) Year name otherwise) Tonga — — — — — — — Tunisia Programme 89 626 — 89 626 2017 Programme 242,000 d’allocations National d’Aide scolaires du aux Familles PNAFN Nécessiteuses (PNAFN) – cash transfers Turkey CCT education 1,965,633 HH 7,076,279 2013 Socio-economic 62,256 Support for the Children of Needy Families Uganda Compassion 94,457 — 94,457 2016 Nothern Uganda 108,540 International Social Action Child Fund (II) – Development Household Programme Income Support Programme Ukraine Social assistance 564,062 HH 1,410,155 2014 Child birth 500,700 for low-income benefit families Uruguay Asignaciones 375,734 — 375,734 2015 — — Familiares Uzbekistan — — — — — Social assistance 600,000 to poor families Vanuatu — — — — — Family 38,493 Assistance Support Program Venezuela, RB — — — — — — — Vietnam — — — — — Subsidies for Tet 808,581 holiday expenditures for poor households West Bank — — — — — Cash Transfer 115,951 and Gaza Program (CTP) Yemen, Rep. Basic Education 39,791 — 39,791 2014 Social Welfare 1,500,000 Support for Girls Fund (SWF) CCT Zambia — — — — — Social Cash 240,000 Transfer Scheme Zimbabwe — — — — — Harmonised 52,000 Cash Transfer Source: ASPIRE database. Note: ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; CCT = conditional cash transfer; HH = household. 116 APPENDIXES Unconditional cash transfer Social pension Number of Estimated individual Estimated number of beneficiaries number of Beneficiary beneficiaries Program (unless specified Beneficiary beneficiaries unit is HH (individuals) Year name otherwise) unit is HH (individuals) Year — — — Social Welfare Scheme — — — — for the Elderly HH 992,200 2017 — — — — — — 62,256 2013 2022 Sayili Kanun 632,407 — 632,407 2013 Kapsaminda Yapilan (old-age social pension) HH 510,138 2016 Social Assistance 91,843 — 91,843 2013 Grants for Empowerment – Senior Citizens Grant — 500,700 2014 Social pension 65,569 — 65,569 2014 — — — Noncontributory 86,939 — 86,939 2015 pensions for old age and disability — 600,000 2011 — — — — — — 38,493 2009 — — — — — — — — Old-age social pensions 531,546 — 531,546 2015 HH 3,056,436 2014 Monthly subsidy for 1,500,844 — 1,500,844 2015 elderly according to Decree 136 HH 695,706 2014 The Disabled 2,813 — 2,813 2014 Rehabilitation Fund/ Economic Empowerment program (DEEP) HH 7,500,000 2017 Disability Fund 361,514 — 361,514 2014 HH 1,248,000 2016 Disability Benefits — — — — (ZAPD, NTFPD) HH 218,400 2015 National Heroes 582 HH 2,444 2015 dependants assistance APPENDIXES 117 TABLE C.2  Food and In Kind and School Feeding Food and in-kind Country/ Number of individual Estimated number economy/ beneficiaries (unless Beneficiary of beneficiaries territory Program name specified otherwise) unit is HH (individuals) Year Albania — — — — — Algeria — — — — — Angola Programa de Apoio Social 600,000 — 600,000 2011 Argentina Plan Nacional de Seguridad 3,302,235 — 3,302,235 2014 Alimentaria Armenia — — — — — Azerbaijan — — — — — Bangladesh Vulnerable Group Feeding 9,960,101 HH 48,306,490 2014 Belarus — — — — — Belize Women’s Iron and Folic Acid 9,000 — 9,000 2009 Distribution Program Benin — — — — — Bhutan — — — — — Bolivia Assistance to drought-affected 893,696 — 893,696 2015 populations in Bolivia Bosnia and — — — — — Herzegovina Botswana Vulnerable Group Feeding 383,392 — 383,392 2013 Program Brazil Cestas de Alimentos 827,109 — 827,109 2015 Bulgaria Assistance for pupils and 47,096 HH 103,611 2014 students Burkina Faso Treatment of Acute 153,499 — 153,499 2016 Malnutrition Moderate Burundi WFP food distribution (all 111,301 — 111,301 2013 programs) Cabo Verde Nutritional support to 1,900 — 1,900 2011 vulnerable groups and people living with HIV Cambodia — — — — — Cameroon Programme PAM/Cameroun — — — — (stock cereal) – Nutrition (treatment of moderate acute malnutrition) Central Distribution de vivres et 80,000 — 80,000 2015 African protection de semences Republic Chad Food aid to vulnerable/ 422,457 — 422,457 2016 food-insecure households Chile Programa nacional de 689,984 — 689,984 2015 alimentación complementaria China Wubao 5,300,000 — 5,300,000 2014 Colombia Raciones Alimentarias de 481,362 — 481,362 2012 emergencia del ICBF Comoros — — — — — Congo, Dem. WFP food distribution (WFP’s 3,233,000 — 3,233,000 2016 Rep. PPRO 200832) Congo, Rep. — — — — — Costa Rica Cen-cinai 31,184 — 31,184 2013 118 APPENDIXES School feeding Number of individual Estimated number of beneficiaries (unless Beneficiary beneficiaries Program name specified otherwise) unit is HH (individuals) Year School feeding — — — — School feeding 31,000 — 31,000 2011 School feeding program 418,733 — 418,733 2011 Comedores Escolares 1,687,785 — 1,687,785 2015 School feeding 38,000 — 38,000 2011 — — — — — School Feeding Programme in 3,003,124 — 3,003,124 2015 poverty-prone areas School feeding — — — — School feeding — — — — School feeding 103,440 — 103,440 2009 School feeding 30,345 — 30,345 2014 School feeding 2,162,921 — 2,162,921 2012 School feeding — — — — School feeding 430,690 — 430,690 2013 National School Feeding Program 42,236,234 — 42,236,234 2014 School feeding — — — — Government school feeding program 2,906,000 — 2,906,000 2016 (primary education) School feeding 316,315 — 316,315 2013 School lunch 3,168 — 3,168 2015 School feeding by WFP 296,007 — 296,007 2015 Programme PAM/Cameroun (stock 55,000 — 55,000 2016 cereal) – school feeding — — — — — School meals – WFP 126,000 — 126,000 2016 Programa Nacional de Alimentacion 1,828,556 — 1,828,556 2015 Escolar School feeding 26,000,000 — 26,000,000 2011 Programa de Alimentación Escolar 4,000,000 — 4,000,000 2010 — — — — — — — — — — — — — — — Programa de alimentacion y nutricion 691,294 — 691,294 2014 escolar (Comedores escolares) – MEP (Table continues next page) APPENDIXES 119 TABLE C.2  Food and In Kind and School Feeding (Continued) Food and in-kind Country/ Number of individual Estimated number economy/ beneficiaries (unless Beneficiary of beneficiaries territory Program name specified otherwise) unit is HH (individuals) Year Côte d’Ivoire — — — — — Croatia Child care (both cash and 391,836 HH 1,097,141 2011 in-kind) Czech — — — — — Republic Djibouti Programme de distribution de 13,000 HH 72,800 2015 vivres au camp de réfugiés Dominica — — — — — Dominican Provisión Alimentaria – 10,999,125 — 10,999,125 2015 Republic Comedores Económicos Ecuador Alimentate Ecuador 935,061 — 935,061 2010 Egypt, Arab Ration cards — — — — Rep. El Salvador Programa de Agricultura 570,000 HH 2,462,400 2014 Familiar Estonia — — — — — Ethiopia Emergency Food Aid 2,550,579 — 2,550,579 2013 Fiji Food Voucher Program (FVP) 26,394 — 26,394 2015 Gabon Maternity grant in kind — — — — Gambia, The Emergency support to 50,100 — 50,100 2015 vulnerable people affected by floods (WFP) Georgia — — — — — Ghana Targeted supplementary — — — — feeding for malnourished children Grenada Uniform and Transportation 4,532 — 4,532 2015 Guatemala Mi Bolsa Segura (MIDES) 196,341 — 196,341 2013 Guinea — — — — — Guinea-Bissau — — — — — Haiti Unconditional food transfer 300,000 — 300,000 2014 relief assistance Honduras Comedores Solidarios 39,000 — 39,000 2011 Hungary — — — — — India Public Distribution System 152,074,000 — 152,074,000 2014 (PDS) Indonesia Rastra (ex. Raskin) — — — — Iran, Islamic — — — — — Rep. Iraq Food rations from Public — — — — Distribution System (PDS) Jordan Urban Targeted Food 115,000 — 115,000 2011 Assistance Kazakhstan — — — — — Kenya WFP Kenya Rural Resilience   — 333,000 2016 Kiribati — — — — — Kosovo — — — — — Kuwait — — — — — 120 APPENDIXES School feeding Number of individual Estimated number of beneficiaries (unless Beneficiary beneficiaries Program name specified otherwise) unit is HH (individuals) Year Programme Intégré de Pérennisation 1,086,721 — 1,086,721 2016 des Cantines Scolaires (PIPCS) School feeding 152,000 — 152,000 2011 — — — — — School feeding 16,814 — 16,814 2014 — — — — — Programa de Alimentacion Escolar/ 1,710,620 — 1,710,620 2015 Desayuno Escolar Programa de Alimentación Escolar 293,303 — 293,303 2014 School feeding 13,500,000 — 13,500,000 2016 Programa de Alimentacion Escolar 1,453,118 — 1,453,118 2013 School feeding — — — — Food for Education 681,195 — 681,195 2014 — — — — — — — — — — The School Feeding Program and 100,000 — 100,000 2016 take-home rations (WFP) — — — — — Ghana School Feeding Programme 1,700,000 — 1,700,000 2014 School feeding 7,051 — 7,051 2012 School feeding 3,052,000 — 3,052,000 2011 — — — — — School feeding program 145,000 — 145,000 2014 School feeding (cantines scolaire) – 818,828 — 818,828 2013 nombre d’élève Programa Escuela Saludables 1,460,000 — 1,460,000 2011 School feeding — — — — School feeding 104,500,000 — 104,500,000 2014 School feeding (PMTAS) 1,400,000 — 1,400,000 2011 School feeding 3,000 — 3,000 2011 School feeding 555,000 — 555,000 2011 School nutrition 115,000 — 115,000 2011 School feeding — — — — Home Grown School Meals (HGSM) 907,659 — 907,659 2016 — — — — — — — — — — School feeding — — — — (Table continues next page) APPENDIXES 121 TABLE C.2  Food and In Kind and School Feeding (Continued) Food and in-kind Country/ Number of individual Estimated number economy/ beneficiaries (unless Beneficiary of beneficiaries territory Program name specified otherwise) unit is HH (individuals) Year Kyrgyz Wheelchairs, assistive 800 — 800 2012 Republic appliances for persons with disabilities Lao Community-based social — — — — proteciton: Livelihood Opportunities and Nutritional Gains Latvia — — — — — Lebanon — — — — — Lesotho Targeted Supplementary 134,000 — 134,000 2011 Feeding Liberia Libera Safety Nets Emergency 30,000 — 30,000 2015 Support refugees/returning migrants Lithuania — — — — — Macedonia, — — — — — FYR Madagascar Réponse aux chocs et 21,359 HH 115,339 2016 protection contre les risques sociaux – Prévention et gestion de sinistres Malawi WFP – Food Aid Program — — — — Malaysia Milk program — — — — Maldives     —     Mali EMOP (opération d’urgence) 427,048 — 427,048 2016 et PRRO: intervention de secours prolongée et de redressement (lutte contre la malnutrition chez les enfants de 6–59 mois et les femmes enceintes ou allaitantes) Marshall — — — — — Islands Mauritania Emergency Relief Program 50,000 HH 300,000 2015 Mauritius Corrugated Iron Sheet 3,000 — 3,000 2014 Housing Mexico Programa Social de Abasto de 6,432,853 — 6,432,853 2015 Leche a cargo de Liconsa Moldova — — — — — Mongolia Free school textbooks 343,700 — 343,700 2013 Montenegro — — — — — Morocco Un million de cartables 3,906,948 — 3,906,948 2014 Mozambique Direct Social Assistance 258,940 — 258,940 2015 (Apoio Social Directo) Myanmar Provision of food and 40,399 — 40,399 2013 micronutrient supplements for pregnant and lactating mothers, fortified food for children 122 APPENDIXES School feeding Number of individual Estimated number of beneficiaries (unless Beneficiary beneficiaries Program name specified otherwise) unit is HH (individuals) Year School feeding 400,000 — 400,000 2012 School feeding 179,297 — 179,297 2013 School feeding — — — — School feeding — — — — School feeding program 389,000 — 389,000 2014 WFP school feeding 648,000 — 648,000 2011 School meal 635,500 — 635,500 2016 — — — — — School feeding – WFP 237,000 — 237,000 2011 World Food Program – Malawi 2,230,000 — 2,230,000 2016 Government School Meals Programme (SMP) School feeding — — — —     —     Programe cantine scolaires (CNCS) 479,465 — 479,465 2016 School feeding — — — — School feeding 346,164 — 346,164 2016 School feeding program 75,000 — 75,000 2011 School feeding 5,164,000 — 5,164,000 2011 School feeding — — — — School feeding 280,400 — 280,400 2009 — — — — — School feeding program (various 1,267,109 — 1,267,109 2014 programs) School Feeding (Alimentação Escolar) 427,000 — 427,000 2011 School feeding 583,271 — 583,271 2014 (Table continues next page) APPENDIXES 123 TABLE C.2  Food and In Kind and School Feeding (Continued) Food and in-kind Country/ Number of individual Estimated number economy/ beneficiaries (unless Beneficiary of beneficiaries territory Program name specified otherwise) unit is HH (individuals) Year Namibia — — — — — Nepal Fortified flour distribution — — — — Nicaragua Programa de Seguridad 54,217 — 54,217 2013 Alimentaria Nutricional Niger PAM Récupération 1,178,830 — 1,178,830 2016 Nutritionnelle Nigeria Save the Children 7,000 — 7,000 2012 Pakistan — — — — — Panama Bono Familiar para la compra 9,200 HH 30,360 2009 de alimentos Papua New — — — — — Guinea Paraguay Programme to Progressively 7,700 — 7,700 2015 Decrease Child Work in the Streets: Food and Health Services Peru Vaso de Leche 1,768,049 — 1,768,049 2010 Philippines Supplemental Feeding — — — — Program Poland Food benefit (in-kind and 554,400 — 554,400 2013 cash) Qatar — — — — — Romania School supplies for pupils 680,260 — 680,260 2014 Russian — — — — — Federation Rwanda Girinka: MINAGRI’s One 203,000 HH 872,900 2015 Cow One Family Samoa — — — — — São Tomé and Cantine for elderly poor 280 — 280 2014 Príncipe Saudi Arabia — — — — — Senegal Food insecurity ripost (CSA) 927,416 — 927,416 2015 Serbia — — — — — Seychelles — — — — — Sierra Leone Caregiver and Supplementary 110,000 — 110,000 2012 Feeding (government and WFP) Slovak — — — — — Republic Slovenia — — — — — Solomon — — — — — Islands Somalia Short-Term Humanitarian 4,122 HH 24,732 2016 Transfers – Food Voucher South Africa Social Relief of Distress — — — — South Sudan Emergency Operation EMOP 2,208,005 — 2,208,005 2016 200859 for IDPs and returnees 124 APPENDIXES School feeding Number of individual Estimated number of beneficiaries (unless Beneficiary beneficiaries Program name specified otherwise) unit is HH (individuals) Year National School Feeding Programme 300,000 — 300,000 2013 to Orphans and Vulnerable Children School feeding (various programs) 666,378 — 666,378 2014 Programa Integral de Nutrición 1,050,000 — 1,050,000 2013 Escolar School feeding (different programs): 168,000 — 168,000 2011 Cantine scolaire School feeding 155,000 — 155,000 2011 — — — — — School feeding 461,000 — 461,000 2011 — — — — — School feeding 10,000 — 10,000 2011 Qali Warma 2,398,480 — 2,398,480 2015 Breakfast feeding program 562,000 — 562,000 2013 School feeding 730,000 — 730,000 2011 School feeding — — — — School feeding — — — — School feeding 2,647,000 — 2,647,000 2011 School feeding 25,000 — 25,000 2014 — — — — — School feeding 41,000 — 41,000 2014 School feeding — — — — School Lunch Program School Feeding 344,706 — 344,706 2015 (government) School feeding — — — — — — — — — School feeding (different programs) 125,000 — 125,000 2012 School feeding — — — — School feeding — — — — — — — — — — — — — — National School Nutrition Programme 9,200,000 — 9,200,000 2013     —     (Table continues next page) APPENDIXES 125 TABLE C.2  Food and In Kind and School Feeding (Continued) Food and in-kind Country/ Number of individual Estimated number economy/ beneficiaries (unless Beneficiary of beneficiaries territory Program name specified otherwise) unit is HH (individuals) Year Sri Lanka Free school uniform material 3,973,909 — 3,973,909 2013 program St. Kitts and Uniforms and shoes 2,000 — 2,000 2008 Nevis St. Lucia — — — — — St. Vincent and Nutrition Support Program 1,000 — 1,000 2009 the Grenadines Sudan General food distribution 2,095,568 — 2,095,568 2015 program Suriname — — — — — Swaziland Food distribution 88,511 — 88,511 2010 Syrian Arab — — — — — Republic Tajikistan Food for tuberculosis patients 45,000 — 45,000 2011 Tanzania Disaster relief food response 910,653 — 910,653 2016 Thailand — — — — — Timor-Leste Ad hoc in-kind support — — — — Togo Nutrition program by 25,914 — 25,914 2011 UNICEF Tonga — — — — — Tunisia — — — — — Turkey Food assistance (GIDA 2,442,599 HH 8,793,356 2013 YARDIMI) Uganda Intergrated Management of 55,000 — 55,000 2015 Acute Malnutrition Ukraine — — — — — Uruguay Tarjeta Uruguay social 69,162 — 69,162 2013 Uzbekistan Support for breastfeeding 475,000 — 475,000 2008 Vanuatu — — — — — Venezuela, RB — — — — — Vietnam Food subsidy for hunger 2,092,170 — 2,092,170 2015 according to Decree 136 West Bank and Food rations, in-kind 876,497 — 876,497 2015 Gaza assistance Yemen, Rep. Emergency Food and 4,313,631 — 4,313,631 2013 Nutrition Support to Food Insecure and Conflict- Affected People. Zambia Food Security Pack 30,100 HH 156,520 2015 Zimbabwe Amalima – Response to 266,277 HH 1,118,363 2015 Humanitarian Situation Source: ASPIRE database. Note: ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; OVC = Orphans and Vulnerable children; CCT = conditional cash transfer; DFID = U.K. Department for International Development; HH = household; IPD = internally displaced persons; UNICEF = United Nations Children’s Fund; WFP = World Food Programme. 126 APPENDIXES School feeding Number of individual Estimated number of beneficiaries (unless Beneficiary beneficiaries Program name specified otherwise) unit is HH (individuals) Year School Meal Program – Mid-day Meal 890,404 — 890,404 2014 Program School feeding — — — — School feeding 7,500 — 7,500 2014 School feeding — — — — School Feeding Programme 974,099 — 974,099 2016 — — — — — National school meal program 328,000 — 328,000 2011 School feeding 46,000 — 46,000 2011 School feeding 330,000 — 330,000 2011 Fee-free Basic Education – School 127,118 — 127,118 2016 meal susidy School feeding 1,677,000 — 1,677,000 2011 School feeding program 288,000 — 288,000 2011 School feeding (different programs) 40,000 — 40,000 2011 School feeding — — — — School feeding 240,000 — 240,000 2011 School Milk Project 6,182,368 — 6,182,368 2013 — — — — — School feeding — — — — School feeding program 256,000 — 256,000 2011 School feeding — — — — — — — — — School feeding 4,031,000 — 4,031,000 2011 School feeding — — — — School feeding 65,000 — 65,000 2014 School feeding 65,000 — 65,000 2011 School feeding program 1,052,760 — 1,052,760 2016 — — — — — APPENDIXES 127 TABLE C.3  Public Works and Fee Waivers Public works Number of   individual Estimated Country/ beneficiaries number of economy/ (unless specified Beneficiary beneficiaries territory Program name otherwise) unit is HH (individuals) Year Albania Employment program 834 — 834 2013 Algeria — — — — — Angola — — — — — Argentina Plan de Empleo Comunitario (PEC) 187,282 — 187,282 2015 Armenia Partial wage subsidy/Relocation 4,161 HH 14,980 2014 Allowances/Work practice for unemployed and disabled/Public Work Azerbaijan Public works 1,605 — 1,605 2014 Bangladesh Employment Generation Program 1,400,000 — 1,400,000 2014 for the Poorest (EGPP) Belarus Public works 55,300 — 55,300 2015 Belize — — — — — Benin Community works 12,000 — 12,000 2015 Bhutan — — — — — Bolivia Empleo Digno e Intensivo de Mano — — — — de Obra Bosnia and — — — — — Herzegovina Botswana Ipelegeng (self-reliance) 65,000 — 65,000 2014 Brazil Economia Solidaria – Programa 534,053 — 534,053 2012 Economia Solidaria em Desemvolvimento Bulgaria Direct job creation 44,222 — 44,222 2013 Burkina Faso Cash for work 25,619 HH 128,095 2016 Burundi WFP different PW (excluding the 91,480 — 91,480 2013 ones through IFAD) Cabo Verde Project insertion — — — — female heads of families and disabled in the job market Cambodia Productive Assets and Livelihoods 28,680 HH 131,928 2015 Support (food for work component) Cameroon Social Safety Nets – labor-intensive 5,000 — 5,000 2016 public works (THIMO) Central African Support to the stabilisation and 60,000 — 60,000 2015 Republic early recovery of communities at risk in CAR (SIRIRI) Phase 2 Chad Food Assistance for Assets 10,000 — 10,000 2016 (Volunteer cooks) – WFP Chile Programa de Apoyo al Empleo 1,913 — 1,913 2015 Sistema Chile Solidario (part of PROEMPLEO programs) China Food-for-Work Program of Poverty — — — — Alleviation Fund Colombia Programa de Empleo Temporal 6,049 — 6,049 2015 Comoros Productive safety net 24,756 — 24,756 2016 128 APPENDIXES Fee waivers Number of individual Estimated beneficiaries number of (unless specified Beneficiary beneficiaries Program name otherwise) unit is HH (individuals) Year Energy benefit 45,833 HH 178,749 2014 — — — — — — — — — — PROGRESARE 708,029 — 708,029 2015 Health benefits and reduced medical 38,951 — 38,951 2013 fees for vulnerable groups — — — — — Construction of Houses for Landless — — — — and Insolvent Freedom Fighters Subsidies for housing and utilities 1,490,000 — 1,490,000 2011 — — — — — Health fund for the poor 10,932 — 10,932 2008 — — — — — — — — — — — — — — — — — — — — — — — — — Energy benefit 254,012 — 254,012 2014 Additional subsidy of emergency 702,083 — 702,083 2014 obstetric and neonatal care for indigent women Fee waivers for indigents 7,846 — 7,846 2015 Medical assistance 153,326 — 153,326 2010 Health Equity Fund 2,956,305 — 2,956,305 2015 Centre Pasteur Case Management 46,449 — 46,449 2016 Shelter and Food Security in Ouham 50,000 — 50,000 2015 Province — — — — — Subsidio para la Prueba de Seleccion 187,619 — 187,619 2015 Universitaria Medical assistance 91,190,000 — 91,190,000 2014 — — — — — — — — — — (Table continues next page) APPENDIXES 129 TABLE C.3  Public Works and Fee Waivers (Continued) Public works Number of   individual Estimated Country/ beneficiaries number of economy/ (unless specified Beneficiary beneficiaries territory Program name otherwise) unit is HH (individuals) Year Congo, Dem. World Bank – Eastern Recovery 588,359 — 588,359 2016 Rep. Project Congo, Rep. — — — — — Costa Rica Programa Nacional de Empleo – 9,225 — 9,225 2014 MTSS Côte d’Ivoire — — — — — Croatia — — — — — Czech Republic — — — — — Djibouti Social Safety Net Project 6,740 HH 37,744 2015 Dominica — — — — — Dominican — — — — — Republic Ecuador Mi Primer Empleo 1,222 — 1,222 2013 Egypt, Arab Rep. Labor-Intensive Investment Project 38,308 — 38,308 2014 for Egypt El Salvador Temporary Income Support 5,500 — 5,500 2014 Program – Urban Estonia Public works, workfare, and direct 143 — 143 2013 job creation, including community development programs Ethiopia Productive Safety Net (PSNP) 7,997,218 — 7,997,218 2016 Fiji — — — — — Gabon — — — — — Gambia, The — — — — — Georgia — — — — — Ghana Labour-Intensive Public Works 164,785 — 164,785 2016 (LIPW) programme Grenada Debushing Program 33,392 — 33,392 2015 Guatemala — — — — — Guinea Labor-intensive public works 24,005 — 24,005 2013 program with a focus on women and youth, and life skills development – urban areas Guinea-Bissau — — — — — Haiti National Project of Community 450,000 — 450,000 2009 Participation Development (PRODEP, in French) Honduras Public works 13,000 — 13,000 2011 Hungary Public work 329,000 — 329,000 2015 India Mahatma Gandhi National Rural 75,287,000 — 75,287,000 2016 Employment Guarantee (MGNREG) Indonesia Cash for work/disaster risk 106,810 HH 416,559 2013 reduction program 130 APPENDIXES Fee waivers Number of individual Estimated beneficiaries number of (unless specified Beneficiary beneficiaries Program name otherwise) unit is HH (individuals) Year UNICEF – Projet d’Appui a la Mise en 47,580 — 47,580 2016 Oeuvre des Mesures de Protection Sociales pour la Scolarisation des Eleves Vulnerables — — — — — Education scholarships (from El Fondo 135,895 — 135,895 2015 Nacional de Becas) — — — — — — — — — — — — — — — Educational program for poor 600 — 600 2015 children/ orphans — — — — — Seguro Familiar de Salud – Regimen — — — — Subsidiado Programa Textos Escolares 6,206,416 — 6,206,416 2015 — — — — Becas Escolares para Estudiantes de 181,171 — 181,171 2013 Educación Media Subsistence benefit to cover expenses 56,948 — 56,948 2014 for standard allotted living space — — — — — Poverty Alleviation Scheme 5,877 HH 27,622 2010 Health insurance plan for economically 483,000 — 483,000 2014 weak Gabonese — — — — — Domestic subsidies (household 59,741 HH 203,119 2013 allowance) National Health Insurance Scheme 6,700,000 — 6,700,000 2014 (NHIS) indigent exemptions — — — — — — — — — — Health grants — — — — Fee waivers and scholarships — — — — Fee waiver for primary education 1,399,173 — 1,399,173 2013 — — — — — — — — — — Rashtriya Swasthya Bima Yoja (RSBY)   —     PBI-JKN (Penerima Bantuan 92,000,000 — 92,000,000 2016 Luran – Jaminan Kesehatan Nasional) (Table continues next page) APPENDIXES 131 TABLE C.3  Public Works and Fee Waivers (Continued) Public works Number of   individual Estimated Country/ beneficiaries number of economy/ (unless specified Beneficiary beneficiaries territory Program name otherwise) unit is HH (individuals) Year Iran, Islamic — — — — — Rep. Iraq — — — — — Jordan Rural Food for Assets 42,000 — 42,000 2011 Kazakhstan “Road Map” program 94,500 — 94,500 2014 Kenya WFP cash for assets CFA 60,000 HH 300,000 2016 Kiribati — — — — — Kosovo — — — — — Kuwait — — — — — Kyrgyz Republic Public works — — — — Lao Poverty Reduction Fund — — — — Latvia Public works 7,223 — 7,223 2011 Lebanon — — — — — Lesotho Integrated Watershed Management 115,000 — 115,000 2012 Public Works Program Liberia Youth, Employment, Skills (YES) 58,581 — 58,581 2016 Lithuania Direct job creation 3,076 — 3,076 2015 Macedonia, FYR — — — — — Madagascar PUPIRV et PURSAPS 69,848 HH 377,179 2013 Malawi Public Works Program – 2,623,702 — 2,623,702 2014 conditional cash transfer Malaysia — — — — — Maldives     —     Mali Assistance Alimentaire pour la 91,038 — 91,038 2016 création d’actifs (3A) Marshall Islands — — — — — Mauritania National integration program and   —     support for microenterprises Mauritius Workfare Programme 1,107 — 1,107 2009 Mexico Programa de Empleo Temporal 1,440,640 — 1,440,640 2014 Ampliado Moldova Moldova Social Investment Fund 112,000 — 112,000 2009 Mongolia — — — — — Montenegro — — — — — Morocco Promotion Nationale 50,000 — 50,000 2009 Mozambique Productive Social Action Program 12,498 HH 282,480 2015 Myanmar Asset Creation Program food and 225,511 — 225,511 2014 cash for work (WFP only) Namibia — — — — — Nepal Karnali Employment Program 323,600 — 323,600 2014 Nicaragua — — — — — 132 APPENDIXES Fee waivers Number of individual Estimated beneficiaries number of (unless specified Beneficiary beneficiaries Program name otherwise) unit is HH (individuals) Year — — — — — — — — — — Housing for the Poor — — — — — — — — — Health Insurance Subsidy Programme 186,462 HH 932,310 2016 (HISP) — — — — — — — — — — Housing Conditions Grant (permanent — — — — to temporary) Electricity compensation 532,300 HH 2,448,580 2012 Health Equity Funds 626,180 — 626,180 2014 Housing benefit 185,146 HH 444,350 2012 — — — — — OVC Bursaries 13,000 — 13,000 2014 Basic Package of Health and Social — — — — Welfare Services (BPHS) Utility allowance (compensation for 111,000 — 111,000 2009 heating expenses) Fee waivers for health insurance 5,653 — 5,653 2014 Subvention aux écoles — — — — — — — — — Rental assistance — — — — Welfare Assistance for Medical   —     Services within Maldives and Abroad     —     — — — — — Indigent Health Coverage 603 — 603 2015 Preprimary school project 517 — 517 2009 Programa Atencion a la Demanda de 2,984,153 — 2,984,153 2015 Educacion para Adultos Heating allowance 123,375 — 123,375 2015 Free public transportation 103,000 — 103,000 2013 Electricity bill subsidy 20,829 HH 68,736 2007 Villes Sans Bidonvilles 324,000 HH 1,684,800 2010 — — — — — Support to compulsory primary 5,200,000 — 5,200,000 2013 education — — — — — Healthcare subsidies – free medicine, — — — — free surgery, food supplements Paquetes educativos soldarios 300,000 — 300,000 2013 (Table continues next page) APPENDIXES 133 TABLE C.3  Public Works and Fee Waivers (Continued) Public works Number of   individual Estimated Country/ beneficiaries number of economy/ (unless specified Beneficiary beneficiaries territory Program name otherwise) unit is HH (individuals) Year Niger Projet de Filets Sociaux – Public 14,510 — 14,510 2016 Works Nigeria Inputs For Work Programme 720,000 — 720,000 2015 (FADAMA) Pakistan PPAF: CPI (Community Physical 502,976 HH 3,118,451 2016 Infrastructure) + WECC (Water, Energy and Climate Change) Panama Public works, training programs 110,095 — 110,095 2011 Papua New Public works program — — — — Guinea Paraguay — — — — — Peru Programa para la Generación de 46,936 — 46,936 2014 Empleo Social Inclusivo “Trabaja Perú” Philippines Cash for Work — — — — Poland Direct job creation 9,070 — 9,070 2013 Qatar — — — — — Romania Solidarity contracts for young 2,812 — 2,812 2012 people with difficulties and at risk of professional exclusion Russian Organization of temporary 811,900 — 811,900 2013 Federation employment Rwanda Vision 2020 Umurenge (VUP) 106,041 HH 296,915 2015 Samoa — — — — — São Tomé and — — — — — Príncipe Saudi Arabia — — — — — Senegal — — — — — Serbia Public works 6,127 — 6,127 2012 Seychelles — — — — — Sierra Leone — — — — — Slovak Republic — — — — — Slovenia — — — — — Solomon Islands Rapid Employment Program — — — — Somalia Resilience Building 880 HH 5,104 2016 South Africa Extended Public Works Programme 350,068 — 350,068 2013 (EPWP) South Sudan Safety Nets and Skills Development 4,864 HH 29,184 2015 Project – public works Sri Lanka Emergency Northern Recovery — — — — Project (ENReP) St. Kitts and — — — — — Nevis St. Lucia Short-term Employment 9,487 — 9,487 2013 Programme 134 APPENDIXES Fee waivers Number of individual Estimated beneficiaries number of (unless specified Beneficiary beneficiaries Program name otherwise) unit is HH (individuals) Year — — — — — — — — — — Child Domestic Labor Basic Education — — — — Enabling Programme Beca universal 554,953 — 554,953 2015 — — — — — — — — — — — — — — — Philhealth-sponsored program 38,640,000 — 38,640,000 2013 Health premium for caregivers 188,650 — 188,650 2013 — — — — — Heating allowance 3,592,213 — 3,592,213 2009 Housing and heating subsidies 9,076,000 — 9,076,000 2009 — — — — — — — — — — — — — — — — — — — — Universal health coverage 792,985 — 792,985 2015 — — — — — — — — — — School fees subsidy — — — — Parent fees for full time care in 314 HH 848 2013 preschool institutions Housing subsidy 4,500 — 4,500 2007 — — — — — — — — — — — — — — — — — — — — Kerosene oil stamp 977,463 HH 4,105,345 2007 — — — — — Education assistance 3,000 — 3,000 2008 (Table continues next page) APPENDIXES 135 TABLE C.3  Public Works and Fee Waivers (Continued) Public works Number of   individual Estimated Country/ beneficiaries number of economy/ (unless specified Beneficiary beneficiaries territory Program name otherwise) unit is HH (individuals) Year St. Vincent and Road Cleaning Program 3,000 — 3,000 2009 the Grenadines Sudan Food for assets 108,362 — 108,362 2015 Suriname — — — — — Swaziland Pilot food for work — — — — Syrian Arab Public works program — — — — Republic Tajikistan Direct job creation — — — — Tanzania Productive Social Safety Net (PSSN) 298,970 HH 1,405,159 2016 –Public Works Thailand Income-generation activities — — — — Timor-Leste Cash-for-Work 55,000 — 55,000 2008 Togo Public Works with High Labor 25,000 — 25,000 2011 Instensity Tonga — — — — — Tunisia — — — — — Turkey Community Services Program 197,182 — 197,182 2013 (TYCP) Uganda Northern Uganda Social Action 33,085 — 33,085 2015 Fund II Karamoja/Karamoja Productive Assets Programme (KPAP) Ukraine Direct job creation 45,500 — 45,500 2012 Uruguay Uruguay Trabaja 3,081 — 3,081 2015 Uzbekistan Public Works Employment Program 100 HH 560 2009 Vanuatu — — — — — Venezuela, RB — — — — — Vietnam Public Works Program for Poor — — — — Unemployed or Underemployed Labourers West Bank and Cash for Work Program 20,550 HH 123,300 2014 Gaza Yemen, Rep. Labor-intensive works by Social 400,000 — 400,000 2017 Fund for Development (SFD) Zambia Public Works Programs in Rural — — — — Area Zimbabwe Food deficit mitigation program 180,000 HH 756,000 2015 Source: ASPIRE. Note: ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; CCT = conditional cash transfer; HH = household; OVC = Orphans and vulnerable children; WFP = World Food Progamme. — = not available. 136 APPENDIXES Fee waivers Number of individual Estimated beneficiaries number of (unless specified Beneficiary beneficiaries Program name otherwise) unit is HH (individuals) Year — — — — — Heath insurance 15,725,537 — 15,725,537 2016 — — — — — Fee waivers for health care — — — — — — — — — — — — — — Fee-free basic education – transport 48,717 — 48,717 2016 benefits The Universal Coverage Scheme (UCS) 48,142,994 — 48,142,994 2011 Food Security Fund — — — — — — — — — — — — — — — — — — — Genel Sağlik Sigortasi Prim Ödemeleri 9,403,251 HH 33,851,704 2013 (green card project) — — — — — Housing and utility allowances 1,845,300 HH 4,613,250 2012 Programa Maestros Comunitarios 14,875 — 14,875 2015 (PMC) — — — — — — — — — — — — — — — Support production land, housing and 43,000 HH 162,540 2013 clean water according to Decision 134 Waivers on school fees 250,000 HH 1,500,000 2014 — — — — — OVC Bursary Program 20,676 — 20,676 2013 Basic Education Assistance Module 118,408 — 118,408 2015 (BEAM) primary APPENDIXES 137 APPENDIX D Spending on Social Safety Net Programs TABLE D.1  Spending on Social Safety Net Programs   Annual spending as a percentage of GDP Country/Economy/ Social School Territory Region Year Total CCT UCT pension feeding Albania ECA 2014 1.57 – 0.46 1.03 – Angola SSA 2015 2.30 – 1.96 0.32 0.03 Argentina LAC 2015 2.05 0.98 0.02 0.78 0.01 Armenia ECA 2014 1.37 – 1.03 0.25 – Azerbaijan ECA 2014 0.84 – 0.39 0.45 – Bangladesh SA 2015 0.73 0.10 0.05 0.17 0.03 Belarus ECA 2015 3.06 – 2.10 0.66 – Benin SSA 2014 2.95 – 2.90 .. 0.04 Bhutan SA 2009 0.33 – – – – Bolivia LAC 2015 2.18 0.28 – 1.21 0.29 Bosnia and Herzegovina ECA 2010, 2011 3.89 – 0.27 3.62 – Botswana SSA 2014–16 1.66 – 0.31 0.28 – Brazil LAC 2015 1.35 0.44 0.03 0.70 0.06 Bulgaria ECA 2014 1.39 – 0.61 0.27 – Burkina Faso SSA 2016 1.99 0.03 0.25 – 0.38 Burundi SSA 2015 2.28 – 0.31 0.04 0.06 Cabo Verde SSA 2010 2.50 – 0.05 1.13 0.17 Cambodia EAP 2015 0.90 0.06 – – 0.04 Cameroon SSA 2016 0.04 0.03 0.01 – – Central African Republic SSA 2015 2.79 – 0.23 – – Chad SSA 2014–16 0.69 – 0.03 – – Chile LAC 2015 3.49 0.07 0.99 0.47 0.31 China EAP 2014 0.76 – 0.26 – – Colombia LAC 2015 3.01 0.29 0.10 0.15 0.09 Comoros SSA 2016 0.67 – 0.01 – – Congo, Dem. Rep. SSA 2016 0.72 – – – – Congo, Rep. SSA 2015 0.05 0.05 – – – Costa Rica LAC 2013 0.74 – – – – Côte d’Ivoire SSA 2016 0.01 – .. – 0.01 Croatia ECA 2014 3.38 – 1.74 1.63 – Djibouti MENA 2013–15 0.18 0.07 0.07 – – Dominican Republic LAC 2015 1.18 0.06 0.15 – 0.33 Ecuador LAC 2010, 2015 1.49 0.26 0.26 0.47 0.21 Egypt, Arab Rep. MENA 2010 0.17 – 0.17 – – El Salvador LAC 2014 0.81 0.07 0.09 0.01 0.10 Estonia ECA 2014 2.60 – 1.46 0.35 – 138 APPENDIXES Annual absolute spending per capita (2011 $PPP) Total excluding Total excluding Public works In kind All fee waivers Other SA health fee waivers Total health fee waivers 0.01 – 0.03 0.04 1.57 163 163 – .. – – 2.30 117 117 0.10 0.03 0.12 – 2.05 278 278 – – 0.09 .. 1.28 105 98 .. – – – 0.84 144 144 0.28 0.09 .. 0.01 0.73 23 23 .. – 0.10 0.20 2.96 472 456 – – 0.01 .. 2.94 58 58 – – – – 0.33 20 20 – 0.40 – – 2.18 129 129 – – – – 3.89 371 371 0.47 0.58 – 0.01 1.66 232 232 – .. 0.08 0.03 1.35 204 204 0.15 0.02 0.11 0.23 1.39 239 239 0.30 0.96 0.06 0.01 1.94 31 30 0.97 0.65 0.21 0.04 2.07 17 15 .. 0.06 0.63 0.45 1.88 154 115 0.01 – 0.10 0.69 0.80 28 25 .. .. .. .. 0.04 1 1 1.05 1.02 – 0.48 2.79 11 11 0.02 0.56 0.08 – 0.62 10 9 .. 0.17 0.75 0.74 3.49 771 770 – 0.03 0.48 – 0.72 96 91 0.02 0.12 1.86 0.37 1.19 378 149 0.66 – – – 0.67 11 11 0.04 0.69 – – 0.72 6 6 – – – .. 0.05 2 2 – – – – 0.46 102 63 – – – .. 0.01 0 0 – – – – 3.38 661 661 .. 0.04 .. – 0.18 6 6 – 0.30 0.23 0.11 0.97 156 129 .. 0.09 0.07 0.13 1.46 151 148 – – – – 0.17 17 17 0.05 0.14 0.31 0.05 0.81 66 66 .. 0.33 – 0.46 2.60 716 716 (Table continues next page) APPENDIXES  139 TABLE D.1  Spending on Social Safety Net Programs (Continued)   Annual spending as a percentage of GDP Country/Economy/ Social School Territory Region Year Total CCT UCT pension feeding Ethiopia SSA 2013–16 0.97 – .. – 0.03 Fiji EAP 2015 1.14 0.02 0.65 0.09 – Gabon SSA 2014 0.20 – 0.06 – – Georgia ECA 2013 6.99 – 1.14 4.59 – Ghana SSA 2014–16 0.58 – 0.29 – 0.15 Grenada LAC 2015 1.98 0.40 0.49 – 0.12 Guatemala LAC 2013 0.19 0.13 0.01 – – Guinea SSA 2015 1.55 0.08 – – – Guinea-Bissau SSA 2015 0.02 – – – 0.01 Honduras LAC 2014 0.77 0.65 0.07 0.01 0.04 Hungary ECA 2013 3.06 – – – – India SA 2016 1.52 – .. 0.06 0.06 Indonesia EAP 2013–15 0.84 0.17 0.14 .. .. Iraq MENA 2012–13 2.56 – 0.36 – – Jordan MENA 2009 0.68 – – – – Kazakhstan ECA 2014 1.62 0.03 0.46 0.98 – Kenya SSA 2016 0.37 0.12 0.07 0.11 0.02 Kiribati EAP 2012 0.69 – – 0.69 – Kosovo ECA 2014 2.84 – 0.48 2.32 – Kuwait MENA 2010 0.80 0.02 0.19 0.18 – Kyrgyz Republic ECA 2014 3.08 – 2.53 0.37 0.15 Lao PDR EAP 2011 0.16 – – 0.06 – Latvia ECA 2012–14 0.77 – 0.50 0.13 – Lebanon MENA 2013 1.04 – 0.40 0.04 – Lesotho SSA 2010, 2013 7.09 – 2.62 2.03 1.37 Liberia SSA 2010, 2016 2.64 – 0.15 – 0.98 Lithuania ECA 2016 0.45 – 0.37 – 0.04 Macedonia, FYR ECA 2014 1.22 0.55 0.03 0.64 – Madagascar SSA 2016 0.16 0.07 .. – 0.04 Malawi SSA 2015–16 1.50 – – – 1.09 Malaysia EAP 2013 0.72 – 0.51 0.06 0.11 Maldives SA 2010–11 1.21 – 0.02 1.02 – Mali SSA 2016 0.60 – 0.11 – 0.05 Marshall Islands EAP 2009 1.05 – – – – Mauritania SSA 2015–16 2.49 1.65 – – 0.08 Mauritius SSA 2014–15 3.46 – 0.25 3.19 – Mexico LAC 2015 1.67 0.39 0.34 0.22 – Moldova ECA 2015 1.25 – 0.83 0.42 – Mongolia EAP 2010–13 2.00 – 1.60 0.28 – Montenegro ECA 2013 1.76 – 1.04 0.57 – Morocco MENA 2014–16 1.09 0.10 0.01 – 0.10 Mozambique SSA 2010, 2015 1.27 – 0.55 – 0.09 Myanmar EAP 2013–15 0.27 .. – – – Namibia SSA 2014 3.19 – 0.29 2.82 0.08 Nepal SA 2010, 2014 1.32 – 0.85 0.28 0.16 Nicaragua LAC 2013 2.22 – 0.19 – 0.20 140 APPENDIXES Annual absolute spending per Annual spending as a percentage of GDP capita (2011 $PPP) Total excluding Total excluding Public works In kind All fee waivers Other SA health fee waivers Total health fee waivers 0.65 0.30 – – 0.97 16 16 – 0.02 0.34 0.02 1.05 104 96 – – 0.14 – 0.06 31 9 – – 1.24 0.02 5.75 588 483 0.01 – 0.11 0.03 0.54 26 24 0.77 .. .. 0.21 1.98 275 275 – 0.05 – – 0.19 13 13 0.29 – 1.18 – 1.49 16 15 – – 0.01 – 0.01 0 0 – .. – .. 0.77 34 34 – – – – 3.06 698 698 0.25 1.03 0.06 0.06 1.51 77 77 .. – 0.51 0.01 0.65 83 64 – 2.20 – .. 2.56 368 368 – – – – 0.68 68 68 0.04 – – 0.10 1.62 390 390 0.03 0.01 0.02 – 0.35 11 10 – – – – 0.69 12 12 – – – 0.04 2.84 255 255 – – 0.41 – 0.80 525 525 0.01 0.02 – – 3.08 101 101 0.05 0.03 – 0.01 0.16 7 7 0.07 0.01 0.06 .. 0.77 178 177 – – 0.61 – 0.44 157 66 0.63 0.27 0.18 – 7.09 188 188 1.00 0.50 – – 2.64 23 23 0.05 – – – 0.45 124 124 – – .. 0.01 1.22 152 151 0.03 0.01 – .. 0.16 2 2 0.41 – – – 1.50 16 16 – 0.03 – – 0.72 164 164 – – 0.10 0.07 1.11 135 124 0.07 0.31 – 0.06 0.60 11 11 – – – – 1.05 33 33 – 0.70 0.07 – 2.43 28 28 – 0.02 – – 3.46 626 626 0.01 0.07 0.62 0.02 1.13 269 181 – – – – 1.25 62 62 – – – 0.12 2.00 202 202 – 0.07 – 0.09 1.76 246 246 – 0.02 0.14 0.72 0.95 80 70 0.13 0.44 0.04 0.02 1.27 15 15 – – 0.27 .. 0.03 12 1 – – – – 3.19 334 334 – – – 0.03 1.32 29 29 – .. 1.70 0.14 1.03 98 46 (Table continues next page) APPENDIXES 141 TABLE D.1  Spending on Social Safety Net Programs (Continued)   Annual spending as a percentage of GDP Country/Economy/ Social School Territory Region Year Total CCT UCT pension feeding Niger SSA 2016 0.67 0.11 – – 0.02 Nigeria SSA 2014–16 0.28 – – .. – Pakistan SA 2011, 2016 0.58 .. 0.43 – – Panama LAC 2014–15 1.52 0.08 0.57 0.33 0.05 Papua New Guinea EAP 2015 0.01 – – 0.01 – Peru LAC 2015 1.43 0.49 – 0.13 0.20 Philippines EAP 2013–14 0.67 0.49 0.01 0.02 0.00 Poland ECA 2013 1.98 – 1.48 0.22 – Romania ECA 2014 1.06 0.14 0.46 0.39 – Russian Federation ECA 2015 1.89 – – – – Rwanda SSA 2015–16 1.50 – 1.21 0.09 – Samoa EAP 2014 0.76 – – 0.76 – São Tomé and Príncipe SSA 2014 .. .. – .. – Saudi Arabia MENA 2012 0.71 .. 0.35 0.36 .. Senegal SSA 2015 0.99 0.21 0.49 – 0.03 Serbia ECA 2013 1.96 – 1.50 0.45 – Seychelles SSA 2015 2.57 – 0.45 2.06 – Sierra Leone SSA 2011, 2016 0.90 0.14 0.03 – 0.45 Slovak Republic ECA 2013 2.43 – – – – Slovenia ECA 2013 2.61 – – – – Somalia SSA 2015–16 0.18 0.18 – – – South Africa SSA 2015 3.31 – 1.27 1.68 0.13 South Sudan SSA 2016 10.10 – – – – Sri Lanka SA 2013–15 0.66 0.01 0.39 0.04 0.04 St. Lucia LAC 2013–14 0.48 – 0.20 – 0.03 Sudan SSA 2016 1.02 0.02 0.50 – 0.01 Swaziland SSA 2010–11 1.71 – 0.77 0.47 0.09 Tajikistan ECA 2014 0.56 0.02 0.10 0.33 – Tanzania SSA 2016 0.46 0.25 – – 0.13 Thailand EAP 2010–11 0.47 – – 0.30 0.07 Timor-Leste EAP 2015 6.48 0.31 0.02 5.71 0.12 Togo SSA 2010, 2015 0.18 – .. – 0.04 Tunisia MENA 2013–15 0.76 0.03 0.54 – – Turkey ECA 2013 1.14 0.04 0.22 0.22 0.02 Uganda SSA 2014–16 0.77 0.17 0.01 0.10 – Ukraine ECA 2014 4.36 0.39 2.32 0.44 – Uruguay LAC 2015 1.15 0.29 0.14 0.54 – Vanuatu EAP 2009 0.28 – – – – Vietnam EAP 2015 1.02 .. 0.87 .. 0.03 West Bank and Gaza MENA 2013–14 2.34 – 1.13 .. – Zambia SSA 2016 0.25 – 0.01 0.02 0.02 Zimbabwe SSA 2015 0.43 – 0.11 .. – Source: ASPIRE database, except for Hungary, Slovak Republic, and Slovenia, for which the data come from ESSPROS (European System of Integrated Social Protection Statistics). Note: The total social safety net spending for the Russian Federation is provided by the Research Institute of Finance of the Ministry of Finance of the Russia Federation in collaboration with the ASPIRE team. ASPIRE = Atlas of Social Protection: Indicators of Reselience and Equity; CCT = conditional cash transfer; ECA = Europe and Central Asia; EAP = East Asia and Pacific; FW = fee waivers; IK = in-kind; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; PPP = purchasing power parity; PW = public works; Other SA = other social assistance; SA = South Asia; SSA = Sub-Saharan Africa; UCT = unconditional cash transfer. — = not available; .. = value was very close to zero (less than 0.001 percent). 142 APPENDIXES Annual absolute spending per Annual spending as a percentage of GDP capita (2011 $PPP) Total excluding Total excluding Public works In kind All fee waivers Other SA health fee waivers Total health fee waivers 0.03 0.50 – – 0.67 6 6 0.28 – – – 0.28 13 13 .. – .. 0.14 0.58 27 27 .. 0.03 0.45 0.01 1.52 325 325 – – – – 0.01 0 0 – 0.06 0.29 0.27 1.14 159 127 .. 0.03 0.11 0.01 0.56 43 36 0.03 0.10 .. 0.14 1.98 501 501 .. 0.01 – 0.06 1.06 212 212 – – – – 1.89 437 437 0.13 .. – 0.08 1.50 25 25 – – – – 0.76 43 43 – – – – .. 0 0 – – – – 0.71 352 352 – 0.13 0.10 0.03 0.89 22 20 0.01 – .. – 1.96 253 253 – 0.02 – 0.04 2.57 671 671 – 0.18 0.02 0.08 0.90 13 13 – – – – 2.43 626 626 – – – – 2.61 698 698 – – – – 0.18 2 2 0.22 0.01 – .. 3.31 408 408 0.05 10.05 – – 10.10 92 92 – 0.12 0.01 0.04 0.66 76 76 0.23 – – 0.03 0.48 49 49 0.01 0.13 0.35 – 0.67 42 28 0.07 0.10 0.22 .. 1.49 130 114 – – – 0.12 0.56 18 18 0.06 0.02 – .. 0.46 11 11 – 0.08 – 0.01 0.47 64 64 0.33 – – – 6.48 116 116 – 0.02 0.12 – 0.07 3 1 – – 0.16 0.02 0.59 79 62 0.04 0.09 0.32 0.19 0.82 207 150 0.20 0.04 – 0.25 0.77 14 14 .. .. 0.62 0.59 4.34 395 393 0.02 0.12 – 0.04 1.15 231 231 – – – – 0.28 8 8 – 0.03 0.07 0.01 1.02 57 57 0.17 0.90 0.15 – 2.24 106 102 – 0.02 0.18 .. 0.25 9 9 0.06 0.20 0.05 0.01 0.43 8 8 APPENDIXES 143 APPENDIX E Monthly Benefit Level Per Household TABLE E.1  Monthly Benefit Level Per Household for Selected Programs $PPP 2011 Monthly transfer amount in Country Year Income group Program name $PPP 2011 Argentina 2016 UMIC Asignación Universal por Hijo para la Protección Social 468 Ukraine 2014 LMIC Social assistance for low-income families 362 Mauritius 2016 UMIC Basic Widow’s Pension 255 South Africa 2008 UMIC Disability grant 226 Cambodia 2015 LMIC MoEYS Scholarships for Primary and Secondary Education 141 China 2016 UMIC Urban Dibao 129 Romania 2014 UMIC Minimum guaranteed income 118 Peru 2013 UMIC Juntos 112 Brazil 2016 UMIC Bolsa Familia 101 Namibia 2014 UMIC Provision of Social Assistance 95 Philippines 2016 LMIC Pantawid Pamilyang Pilipino Program (4Ps) 66 Fiji 2016 UMIC Poverty Benefit Scheme (PBS) 61 Cameroon 2016 LMIC Pilot CCT with Productive Aspects 59 Mexico 2015 UMIC Prospera 58 Malaysia 2014 UMIC Bantuan Rakyat 1 Malaysia (BR1M) scheme 48 Pakistan 2015 LMIC Benazir Income Support Program (BISP) 48 Angola 2015 LMIC Cartão Kikuia—Kikuia Card Cash Transfer Programme 48 Indonesia 2017 LMIC Program Keluarga Harapan (PKH) 44 Kenya 2016 LMIC Cash transfer for OVC (CT-OVC) 43 Burkina Faso 2015 LIC Social Safety net project “Burkin-Nong-Saya” 43 Niger 2011 LIC Projet de Filets Sociaux 43 Lesotho 2016 LMIC Child Grants Program (CGP) 40 Senegal 2015 LIC National cash transfer programme 33 Ghana 2015 LMIC Livelihood Empowerment Against Poverty (LEAP) 28 Mozambique 2015 LIC Basic Social Subsidy Programme 26 Sierra Leone 2011 LIC Social Safety Nets Program 25 Congo, Dem. Rep. 2016 LIC UNICEF – Femme et Homme Progressons Ensemble 25 Mongolia 2015 LMIC The Child Money programme 23 Malawi 2016 LIC Social Cash Transfer Scheme (SCTS) 19 Kazakhstan 2014 UMIC Conditional Targeted Social Assistance (CTSA) 19 Swaziland 2011 LMIC Public assistance 15 Mauritania 2013 LMIC Tekavou – conditional cash transfers 13 Bangladesh 2016 LMIC Allowances for Widows, Deserted and Destitute Women 12 Timor-Leste 2011 LMIC Bolsa da Mãe 11 Madagascar 2016 LIC Filets Sociaux de Sécurité (FSS) 2 Zambia 2016 LMIC Social Cash Transfer Scheme 0 Source: ASPIRE. Note: ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; LIC = low-income country; LMIC = lower-middle-income country; OVC = orphans and vulnerable children; PPP = purchasing power parity; UMIC = upper-middle-income country; UNICEF = United Nations Children’s Fund. 144 APPENDIXES APPENDIX F Performance Indicators TABLE F.1  Key Performance Indicators of Social Protection and Labor Programs Percent Transfer as a share of Benefit Beneficiary beneficiary welfare Coverage incidence incidence (adequacy) Country/economy/ Poorest Poorest Poorest Poorest territory Year quintile Total quintile quintile quintile Total Afghanistan 2011 12.47 8.82 – 28.27 – – Albania 2012 71.81 54.89 25.61 26.14 34.80 27.99 Argentina 2013 70.18 46.36 11.43 30.27 34.76 38.15 Armenia 2014 88.08 66.76 33.46 26.37 59.81 33.63 Bangladesh 2010 28.81 17.77 46.25 32.42 5.40 9.12 Belarus 2013 97.46 75.05 34.35 25.89 73.16 48.08 Belize 2009 47.02 39.39 51.04 23.79 53.07 23.35 Bhutan 2012 4.68 2.92 54.42 32.04 17.20 27.64 Bolivia 2012 91.55 76.75 17.66 23.85 41.14 13.50 Botswana 2009 94.93 73.77 32.76 25.71 22.53 9.53 Brazil 2015 78.84 53.81 9.24 29.30 36.09 44.76 Burkina Faso 2014 1.96 4.28 3.51 9.16 32.18 17.15 Cameroon 2014 1.17 5.01 7.11 4.67 2.91 3.13 Central African Republic 2008 0.87 1.39 11.28 12.54 27.87 4.82 Chad 2011 1.96 2.97 16.97 13.17 41.18 24.52 Chile 2013 97.34 88.54 18.68 21.98 20.29 18.76 China 2013 81.67 63.05 27.84 25.90 17.03 36.83 Colombia 2014 83.86 65.68 4.40 25.51 13.06 25.30 Congo, Dem. Rep. 2012 15.08 11.06 37.61 27.26 64.40 38.26 Costa Rica 2014 85.51 66.64 9.18 25.66 27.53 28.58 Croatia 2010 93.48 70.72 25.56 26.41 45.34 39.52 Côte d’Ivoire 2014 42.29 32.86 36.75 25.73 75.38 47.54 Djibouti 2012 40.05 20.94 51.77 38.23 28.90 28.95 Dominican Republic 2014 46.60 34.53 12.91 26.98 13.31 12.72 Ecuador 2016 88.99 72.70 12.02 24.47 34.04 32.67 Egypt, Arab Rep. 2008 67.96 55.45 19.77 24.51 14.96 21.03 El Salvador 2014 73.43 58.09 4.98 25.25 11.76 25.86 Ethiopia 2010 16.21 13.25 – 24.46 – – Fiji 2008 24.90 14.33 51.76 34.73 38.12 24.20 Gambia, The 2010 6.11 8.31 13.75 14.66 27.56 5.22 Georgia 2011 92.95 64.65 38.82 28.72 68.49 29.25 Ghana 2012 64.63 63.73 56.26 20.28 62.48 91.72 Guatemala 2014 72.27 63.25 1.54 22.84 25.20 22.32 Guinea 2012 2.07 3.89 7.59 10.64 29.46 10.27 (Table continues next page) APPENDIXES 145 TABLE F.1  Key Performance Indicators of Social Protection and Labor Programs (Continued) Transfer as a share of Benefit Beneficiary beneficiary welfare Coverage incidence incidence (adequacy) Country/economy/ Poorest Poorest Poorest Poorest territory Year quintile Total quintile quintile quintile Total Haiti 2012 21.53 19.46 – 22.11 33.08 34.60 Honduras 2013 72.34 55.79 11.42 25.92 24.58 13.33 India 2011 30.45 29.71 – 20.49 – – Indonesia 2015 83.07 57.41 52.67 28.94 27.32 15.94 Iraq 2012 91.65 83.56 18.72 21.92 9.63 9.52 Jamaica 2010 76.67 58.26 25.32 26.20 11.02 11.59 Jordan 2010 89.02 73.19 22.52 24.32 20.78 18.49 Kazakhstan 2010 64.11 48.31 17.16 26.54 9.85 9.10 Kosovo 2013 67.06 44.53 27.44 30.06 26.37 17.81 Kyrgyz Republic 2013 81.45 57.86 37.01 28.15 57.57 42.16 Latvia 2009 96.51 80.30 32.45 24.02 59.08 32.37 Liberia 2014 9.66 7.33 28.35 26.31 16.88 9.49 Lithuania 2008 97.53 81.91 37.82 23.82 80.09 33.19 Madagascar 2010 1.25 5.90 96.40 4.23 2.63 39.07 Malawi 2013 43.19 42.46 21.47 17.64 1.83 1.92 Malaysia 2008 94.78 84.09 14.73 22.54 8.36 4.51 Maldives 2009 16.32 14.85 20.88 18.47 74.50 24.21 Mauritania 2014 47.46 45.22 – 20.98 – – Mauritius 2012 85.62 46.63 42.59 36.71 63.01 45.87 Mexico 2012 83.38 72.23 20.10 23.08 31.92 28.59 Moldova 2013 79.60 56.59 30.89 28.09 51.06 34.33 Mongolia 2012 99.83 99.90 25.69 19.97 40.99 20.71 Montenegro 2014 86.60 56.44 29.49 30.68 46.68 50.78 Morocco 2009 52.47 41.01 – 25.57 – – Mozambique 2008 9.81 6.98 29.90 28.03 188.17 43.10 Namibia 2009 26.52 16.15 – 32.85 – – Nepal 2010 54.27 43.49 21.86 24.96 3.78 5.95 Nicaragua 2014 77.04 68.15 4.96 22.60 42.30 17.14 Niger 2014 17.28 21.38 7.03 16.14 3.53 3.38 Nigeria 2015 6.06 6.41 20.11 18.88 7.40 18.37 Pakistan 2013 25.08 16.80 13.38 29.86 8.40 19.53 Panama 2014 86.52 63.39 9.11 27.28 20.63 19.72 Papua New Guinea 2009 2.10 4.24 4.47 9.87 1.28 0.46 Paraguay 2011 74.33 52.27 3.72 28.43 24.02 29.23 Peru 2014 88.68 64.85 6.53 27.34 13.46 17.83 Philippines 2015 67.56 40.62 15.32 33.26 9.08 7.75 Poland 2012 97.03 64.41 35.59 30.13 65.61 60.56 Romania 2012 98.47 83.69 31.98 23.52 45.78 49.18 Russian Federation 2016 94.18 77.88 19.18 24.19 30.29 25.58 Rwanda 2013 37.96 31.40 28.05 24.17 8.33 4.93 Senegal 2011 10.58 16.79 9.30 12.59 4.81 11.79 Serbia 2013 90.37 63.18 29.72 28.57 58.80 54.96 Sierra Leone 2011 34.57 30.20 15.05 22.89 0.00 0.00 (Table continues next page) 146 APPENDIXES TABLE F.1  Key Performance Indicators of Social Protection and Labor Programs (Continued) Transfer as a share of Benefit Beneficiary beneficiary welfare Coverage incidence incidence (adequacy) Country/economy/ Poorest Poorest Poorest Poorest territory Year quintile Total quintile quintile quintile Total Slovak Republic 2009 98.69 88.62 29.63 22.26 31.85 32.30 South Africa 2010 96.10 62.76 39.97 30.61 72.36 23.87 South Sudan 2009 3.42 6.24 8.67 10.96 3.85 1.53 Sri Lanka 2012 53.61 33.13 21.39 32.36 7.85 15.46 Sudan 2009 13.10 7.45 18.40 35.18 1.07 1.05 Swaziland 2009 77.28 51.65 44.84 29.90 21.71 15.70 Tajikistan 2011 50.92 39.29 21.99 25.83 10.86 7.60 Tanzania 2014 13.32 17.21 9.82 15.47 11.88 13.35 Thailand 2013 87.11 79.38 32.79 21.94 13.45 16.62 Timor-Leste 2011 41.71 35.33 7.55 23.58 16.75 20.62 Tunisia 2010 20.07 14.43 20.58 27.79 17.99 3.80 Turkey 2014 74.85 49.87 16.39 30.01 31.87 37.85 Uganda 2012 76.01 60.71 11.78 25.02 5.22 13.51 Ukraine 2013 96.92 71.71 36.12 27.03 67.70 51.16 Uruguay 2012 96.00 79.82 8.26 24.06 42.06 45.24 Vietnam 2014 58.29 34.86 16.13 33.44 7.70 22.52 West Bank and Gaza 2009 25.91 10.72 56.45 48.31 21.59 20.67 Zambia 2010 1.61 1.62 60.30 19.89 37.61 34.02 Zimbabwe 2011 38.53 29.55 12.93 26.07 25.08 21.26 Source: ASPIRE. Note: The information presented is for 96 countries/economies/territories for which household survey data are available. The poorest quintile is calculated using per capita post transfer welfare (income or consumption) except for the indicator that expresses the social transfers as a share of total beneficiary welfare, which includes transfers. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; — = not available. TABLE F.2  Key Performance Indicators of Social Safety Nets Programs Percent Transfer as a share of Benefit Beneficiary beneficiary welfare Coverage incidence incidence (adequacy) Country/economy/ Poorest Poorest Poorest Poorest territory Year quintile Total quintile quintile quintile Total Afghanistan 2011 12.47 8.82 – 28.27 – – Albania 2012 31.34 19.08 39.34 32.82 10.88 6.02 Argentina 2013 49.19 19.76 54.41 49.78 21.92 11.31 Armenia 2014 45.67 28.40 50.92 32.15 32.39 16.99 Bangladesh 2010 24.81 13.09 47.26 37.90 4.39 4.00 Belarus 2013 70.40 54.10 31.97 25.95 23.04 11.09 Belize 2009 20.12 16.28 32.63 24.63 23.43 8.62 Bhutan 2012 3.95 2.25 – 34.98 – – Bolivia 2012 91.33 75.18 30.12 24.29 40.70 7.76 Botswana 2009 94.93 73.77 32.76 25.71 22.53 9.53 (Table continues next page) APPENDIXES 147 TABLE F.2  Key Performance Indicators of Social Safety Nets Programs (Continued) Transfer as a share of Benefit Beneficiary beneficiary welfare Coverage incidence incidence (adequacy) Country/economy/ Poorest Poorest Poorest Poorest territory Year quintile Total quintile quintile quintile Total Brazil 2015 64.15 23.72 57.02 54.10 24.52 17.27 Burkina Faso 2014 1.58 2.29 1.87 13.82 12.82 5.50 Cameroon 2014 0.12 0.87 32.02 2.79 – 2.61 Central African Republic 2008 0.39 0.57 – 13.77 – – Chad 2011 0.15 0.57 12.20 5.40 23.18 10.48 Chile 2013 93.40 74.69 38.09 25.00 14.12 7.66 China 2013 60.99 43.81 33.00 27.84 6.02 2.34 Colombia 2014 83.32 59.25 40.79 28.10 12.36 5.10 Congo, Dem. Rep. 2012 14.11 9.99 38.42 28.24 65.10 37.45 Costa Rica 2014 79.43 45.85 50.03 34.65 22.88 13.31 Croatia 2010 46.34 24.57 48.36 37.69 15.69 8.92 Côte d’Ivoire 2014 36.09 27.23 33.27 26.50 9.76 14.74 Djibouti 2012 32.71 9.53 71.33 68.65 20.92 11.91 Dominican Republic 2014 44.75 29.96 27.77 29.87 10.67 5.25 Ecuador 2016 87.48 67.19 50.21 26.03 29.49 13.73 Egypt, Arab Rep. 2008 59.09 44.88 28.87 26.33 5.00 3.56 El Salvador 2014 71.86 53.14 44.15 27.01 9.38 7.39 Ethiopia 2010 16.21 13.25 – 24.46 – – Fiji 2008 16.47 9.55 57.00 34.47 30.28 14.02 Gambia, The 2010 5.73 7.08 – 16.15 – – Georgia 2011 92.95 64.64 38.78 28.72 68.43 29.18 Ghana 2012 1.60 1.51 52.89 21.14 74.90 23.55 Guatemala 2014 71.48 59.08 – 24.19 – – Guinea 2012 1.29 1.71 – 14.99 – – Haiti 2012 21.53 19.13 – 22.49 – – Honduras 2013 72.05 54.22 31.60 26.56 23.80 5.65 India 2011 26.56 17.27 – 30.75 – – Indonesia 2015 82.05 48.74 52.67 33.67 27.32 15.94 Iraq 2012 87.15 75.77 34.26 22.98 3.45 2.75 Jamaica 2010 73.88 54.97 45.70 26.75 9.62 4.63 Jordan 2010 86.42 65.68 47.67 26.31 6.89 3.98 Kazakhstan 2010 43.99 30.58 25.49 28.77 5.23 3.34 Kosovo 2013 42.66 14.05 72.18 60.60 22.95 12.99 Kyrgyz Republic 2013 15.66 7.18 48.37 43.63 17.61 11.15 Latvia 2009 58.11 50.81 36.93 22.85 14.85 7.08 Liberia 2014 9.66 7.33 28.35 26.31 16.88 9.49 Lithuania 2008 57.80 58.67 33.07 19.70 18.59 6.50 Madagascar 2010 0.22 0.24 96.40 17.93 2.63 39.07 Malawi 2013 42.79 41.74 13.56 17.78 1.76 1.03 Malaysia 2008 94.25 82.81 25.54 22.76 6.49 1.75 Maldives 2009 15.40 13.49 22.21 19.19 75.63 24.76 (Table continues next page) 148 APPENDIXES TABLE F.2  Key Performance Indicators of Social Safety Nets Programs (Continued) Transfer as a share of Benefit Beneficiary beneficiary welfare Coverage incidence incidence (adequacy) Country/economy/ Poorest Poorest Poorest Poorest territory Year quintile Total quintile quintile quintile Total Mauritania 2014 47.46 45.22 – 20.98 – – Mauritius 2012 83.52 44.85 45.59 37.23 54.90 28.76 Mexico 2012 75.38 37.05 46.43 40.69 30.10 10.21 Moldova 2013 32.09 23.79 36.39 26.94 21.93 9.04 Mongolia 2012 99.83 99.83 22.36 19.98 26.10 11.07 Montenegro 2014 24.81 8.08 55.47 61.45 27.98 22.09 Morocco 2009 50.11 36.59 – 27.37 – – Mozambique 2008 7.75 5.40 – 28.65 – – Namibia 2009 26.08 15.18 – 34.37 – – Nepal 2010 53.24 40.14 24.58 26.53 3.60 2.51 Nicaragua 2014 74.27 59.75 – 24.86 – – Niger 2014 17.27 20.11 16.97 17.15 3.53 1.63 Nigeria 2015 4.34 3.76 8.71 23.02 1.24 2.26 Pakistan 2013 22.45 11.20 31.61 40.10 6.09 7.91 Panama 2014 83.78 51.12 35.03 32.76 17.52 6.78 Papua New Guinea 2009 1.92 3.36 7.70 11.44 0.17 0.04 Paraguay 2011 74.07 47.75 60.65 31.01 19.64 13.45 Peru 2014 88.02 56.10 65.18 31.38 12.49 7.99 Philippines 2015 66.28 33.83 33.20 39.18 8.85 5.43 Poland 2012 65.49 38.75 52.03 33.80 27.46 10.24 Romania 2012 74.67 61.83 41.30 24.14 19.70 10.91 Russian Federation 2016 85.30 67.89 29.69 25.13 10.47 6.80 Rwanda 2013 23.00 20.09 35.09 22.90 9.94 4.59 Senegal 2011 6.06 8.24 18.37 14.69 6.96 7.22 Serbia 2013 27.00 11.72 52.91 46.00 27.17 19.00 Sierra Leone 2011 34.57 30.20 – – – – Slovak Republic 2009 97.49 83.20 40.81 23.42 14.45 4.37 South Africa 2010 96.10 60.81 39.97 31.60 72.36 23.87 South Sudan 2009 3.42 6.24 8.67 10.96 3.85 1.53 Sri Lanka 2012 48.29 26.19 46.02 36.88 5.83 3.65 Sudan 2009 13.10 7.45 18.40 35.18 1.07 1.05 Swaziland 2009 77.28 51.65 44.84 29.90 21.71 15.70 Tajikistan 2011 13.70 9.75 8.52 28.00 1.26 2.43 Tanzania 2014 9.22 10.94 42.95 16.85 5.82 3.69 Thailand 2013 83.75 59.32 35.01 28.23 13.08 6.21 Timor-Leste 2011 41.71 35.21 13.87 23.66 3.72 2.06 Tunisia 2010 20.07 14.43 20.58 27.79 17.99 3.80 Turkey 2014 44.50 17.68 45.56 50.35 7.78 5.77 Uganda 2012 75.93 60.47 16.95 25.10 3.66 9.46 Ukraine 2013 71.91 48.07 39.25 29.91 21.28 11.76 Uruguay 2012 86.28 59.21 – 29.14 – – (Table continues next page) APPENDIXES 149 TABLE F.2  Key Performance Indicators of Social Safety Nets Programs (Continued) Transfer as a share of Benefit Beneficiary beneficiary welfare Coverage incidence incidence (adequacy) Country/economy/ Poorest Poorest Poorest Poorest territory Year quintile Total quintile quintile quintile Total Vietnam 2014 48.41 17.51 73.04 55.29 4.64 2.89 West Bank and Gaza 2009 25.91 10.72 56.45 48.31 21.59 20.67 Zambia 2010 1.02 0.57 – 35.78 – – Zimbabwe 2011 37.92 27.73 42.71 27.34 19.79 19.24 Source: ASPIRE. Note: The information presented here is for 96 countries for which household survey data are available. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. The poorest quintile is calculated using per capita post-transfer welfare (income or consumption) except for the indicator that expresses the social transfers as a share of total beneficiary welfare, which includes transfers. — = not available. TABLE F.3  Poverty and Inequality Reduction as a Result of Social Safety Nets Programs Poverty headcount Poverty gap Gini reduction reduction inequality Benefit-cost (% population) (% population) reduction (%) ratio ($PPP) $1.90 $1.90 $1.90 Country/economy/ Poorest poverty Poorest poverty Total Poorest poverty territory Year quintile line quintile line population quintile line Afghanistan 2011 – – – – – – – Albania 2012 4.62 38.20 10.97 65.63 1.62 0.33 0.03 Argentina 2013 6.46 60.09 16.69 79.12 2.13 0.52 0.04 Armenia 2014 11.79 49.02 28.69 67.50 4.24 0.43 0.13 Bangladesh 2010 3.21 4.11 6.26 8.17 0.71 0.34 0.26 Belarus 2013 24.19 100.00 43.37 100.00 9.78 0.38 0.01 Belize 2009 0.71 58.74 2.24 88.79 0.18 0.26 0.13 Bhutan 2012 – – – – – – – Bolivia 2012 17.35 33.62 28.86 43.56 5.30 0.29 0.13 Botswana 2009 20.02 28.64 38.36 49.19 3.92 0.19 0.14 Brazil 2015 10.90 39.39 23.55 59.89 2.78 0.44 0.11 Burkina Faso 2014 0.00 0.12 0.10 0.06 −0.08 0.02 0.00 Cameroon 2014 0.12 0.10 0.35 0.29 0.02 0.17 0.20 Central African Republic 2008 – – – – – – – Chad 2011 0.07 0.16 0.16 0.11 −0.01 0.05 0.13 Chile 2013 14.42 55.09 25.50 65.89 3.48 0.31 0.02 China 2013 5.03 16.60 10.03 24.83 1.09 0.29 0.07 Colombia 2014 6.54 15.74 10.63 20.17 0.95 0.38 0.11 Congo, Dem. Rep. 2012 8.89 1.98 18.55 4.40 0.50 0.10 0.40 Costa Rica 2014 8.92 38.82 16.88 49.04 1.90 0.42 0.04 Croatia 2010 9.00 100.00 24.04 100.00 3.87 0.49 0.01 Côte d’Ivoire 2014 0.21 0.18 0.34 0.29 0.01 0.10 0.12 Djibouti 2012 2.42 3.52 7.60 9.46 0.90 0.61 0.56 Dominican Republic 2014 6.10 20.50 10.25 19.40 1.32 0.26 0.01 Ecuador 2016 7.58 24.82 17.23 36.52 2.22 0.43 0.15 Egypt, Arab Rep. 2008 5.78 22.83 11.69 50.13 1.35 0.22 0.01 El Salvador 2014 1.90 7.40 3.14 13.60 0.36 0.36 0.07 (Table continues next page) 150 APPENDIXES TABLE F.3  Poverty and Inequality Reduction as a Result of Social Safety Nets Programs (Continued) Poverty headcount Poverty gap Gini reduction reduction inequality Benefit-cost (% population) (% population) reduction (%) ratio ($PPP) $1.90 $1.90 $1.90 Country/economy/ Poorest poverty Poorest poverty Total Poorest poverty territory Year quintile line quintile line population quintile line Ethiopia 2010 – – – – – – – Fiji 2008 5.75 9.79 11.22 29.54 1.04 0.21 0.09 Gambia, The 2010 – – – – – – – Georgia 2011 42.63 61.18 68.39 80.73 19.06 0.33 0.23 Ghana 2012 0.10 0.00 0.36 0.07 0.02 0.23 0.57 Guatemala 2014 – – – – – – – Guinea 2012 – – – – – – – Haiti 2012 – – – – – – – Honduras 2013 5.93 7.35 11.10 11.56 1.19 0.27 0.25 India 2011 – – – – – – – Indonesia 2015 16.87 45.02 38.17 67.97 4.62 0.47 0.22 Iraq 2012 7.88 81.18 14.77 96.63 1.55 0.21 0.02 Jamaica 2010 10.18 16.39 9.42 43.44 1.20 0.40 0.05 Jordan 2010 10.35 94.95 24.78 98.56 3.02 0.35 0.02 Kazakhstan 2010 5.48 32.49 8.93 42.89 1.11 0.24 0.00 Kosovo 2013 7.53 59.23 21.77 76.49 3.83 0.62 0.10 Kyrgyz Republic 2013 4.74 14.58 10.57 32.82 1.39 0.44 0.16 Latvia 2009 11.44 69.48 26.44 84.53 4.05 0.30 0.03 Liberia 2014 2.45 – 4.94 – 0.69 0.20 – Lithuania 2008 14.64 100.00 29.83 100.00 4.45 0.30 0.00 Madagascar 2010 0.04 0.02 0.18 0.03 −0.11 0.02 0.08 Malawi 2013 0.55 0.23 1.03 0.58 0.13 0.14 0.41 Malaysia 2008 6.26 59.79 13.34 68.05 1.32 0.24 0.01 Maldives 2009 11.74 47.40 27.93 76.06 3.86 0.29 0.11 Mauritania 2014 – – – – – – – Mauritius 2012 36.88 89.05 60.93 96.20 13.83 0.32 0.06 Mexico 2012 13.18 36.18 29.70 54.04 3.38 0.42 0.19 Moldova 2013 9.61 91.54 23.80 96.29 3.24 0.37 0.03 Mongolia 2012 34.75 90.11 52.85 95.26 10.28 0.24 0.01 Montenegro 2014 3.94 100.00 23.08 100.00 2.59 0.53 0.01 Morocco 2009 – – – – – – – Mozambique 2008 – – – – – – – Namibia 2009 – – – – – – – Nepal 2010 4.79 6.26 7.16 8.73 0.72 0.20 0.13 Nicaragua 2014 – – – – – – – Niger 2014 1.71 0.29 1.56 0.98 0.28 0.15 0.53 Nigeria 2015 0.26 0.06 0.26 0.16 0.02 0.08 0.32 Pakistan 2013 3.16 11.54 7.21 20.49 0.67 0.23 0.08 Panama 2014 12.33 41.87 20.45 52.37 2.77 0.31 0.05 Papua New Guinea 2009 0.00 0.00 0.00 0.04 0.00 0.08 0.03 Paraguay 2011 2.34 2.91 3.10 5.56 0.30 0.42 0.07 (Table continues next page) APPENDIXES 151 TABLE F.3  Poverty and Inequality Reduction as a Result of Social Safety Nets Programs (Continued) Poverty headcount Poverty gap Gini reduction reduction inequality Benefit-cost (% population) (% population) reduction (%) ratio ($PPP) $1.90 $1.90 $1.90 Country/economy/ Poorest poverty Poorest poverty Total Poorest poverty territory Year quintile line quintile line population quintile line Peru 2014 2.55 20.28 7.65 30.59 0.80 0.64 0.13 Philippines 2015 8.75 19.10 15.24 27.81 1.51 0.32 0.11 Poland 2012 17.03 98.76 41.58 99.76 7.25 0.53 0.03 Romania 2012 23.07 100.00 43.51 100.00 9.25 0.36 0.02 Russian Federation 2016 16.85 67.94 25.50 76.70 4.76 0.22 0.00 Rwanda 2013 3.99 0.58 8.12 2.12 0.61 0.21 0.51 Senegal 2011 0.91 0.67 1.63 1.00 −0.11 0.06 0.10 Serbia 2013 7.38 91.63 21.19 97.06 3.40 0.48 0.04 Sierra Leone 2011 – – – – – – – Slovak Republic 2009 15.77 81.20 28.92 90.25 7.15 0.44 0.03 South Africa 2010 40.04 56.82 66.85 79.61 7.36 0.36 0.25 South Sudan 2009 0.06 0.00 0.09 0.07 0.00 0.06 0.17 Sri Lanka 2012 4.07 31.14 9.11 43.15 0.93 0.32 0.04 Sudan 2009 0.00 0.34 0.07 0.13 0.01 0.18 0.10 Swaziland 2009 11.72 6.12 27.75 14.34 2.79 0.20 0.35 Tajikistan 2011 0.35 1.83 0.67 2.05 −0.04 0.08 0.01 Tanzania 2014 0.90 0.01 0.38 0.27 0.05 0.24 0.54 Thailand 2013 11.55 87.79 21.14 96.55 2.62 0.30 0.00 Timor-Leste 2011 0.00 0.00 0.35 0.26 0.04 0.14 0.42 Tunisia 2010 0.27 – 1.73 – 0.14 0.20 – Turkey 2014 4.18 77.62 8.49 84.94 0.97 0.43 0.01 Uganda 2012 0.35 0.23 0.54 0.44 −0.02 0.08 0.12 Ukraine 2013 23.29 99.49 44.57 99.99 9.39 0.39 0.02 Uruguay 2012 – – – – – – – Vietnam 2014 0.90 15.86 4.19 31.32 0.43 0.68 0.19 West Bank and Gaza 2009 4.56 42.63 10.84 74.37 1.30 0.36 0.03 Zambia 2010 – – – – – – – Zimbabwe 2011 0.93 – 1.89 – 0.15 0.18 – Source: ASPIRE. Note: The information presented here is for 96 countries for which household survey data are available. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. — = not available. 152 APPENDIXES APPENDIX G Old-Age Social Pensions TABLE G.1  Old-Age Social Pensions around the World Country/economy/ territory Region Year Name of scheme Location Algeria MENA 2009–13 Allocation forfaitaire de solidarite National Antigua and Barbuda LAC 2011 Old-Age Assistance Programme National Argentina LAC 2011–13 Pensiones Asistenciales National Armenia ECA – Old-Age Social Pension National Australia OECD 2009–11 Age Pension National Azerbaijan ECA 2008 Social Allowance (Old-Age) National Bangladesh SA 2011–12 Old-Age Allowance National Barbados LAC 2008–12 Noncontributory Old-Age Pension National Belarus ECA – Social Pension National Belgium OECD 2011–13 IGO/GRAPA (Income Guarantee for the National Elderly) Belize LAC 2011 Non-Contributory Pension National Programme (NCP) Bolivia LAC 2013–14 Renta Dignidad or Renta Universal de National Vejez (previously Bonosol) Botswana SSA 2010 State old-age pension National Brazil LAC 2011–13 Previdencia Rural (Rural Pension) Regional/Rural Brazil (2) LAC 2012 Beneficio de Prestacao Continuada National (BPC/Continuous Cash Benefit) Brunei Darussalam EAP 2009 Old-age pension National Bulgaria ECA – Social Old-Age Pension National Cabo Verde SSA 2011 Pensao Social Minima (Minimum Social National Pension) Canada OECD 2011 Pension de la Securite Vieillesse (S.V.) National (Old-Age Security Pension) Chile LAC 2012–14 Sistema de pensiones solidarias National (vejez) –includes Pensión Básica Solidaria de Vejez (PBS-Vejez) and Aporte Previsional Solidario de Vejez (APS-Vejez) China EAP 2012 Rural social pension National Colombia LAC 2012–14 Programa Colombia Mayor National Costa Rica LAC 2009–10 Programa Regimen No Contributivo National Denmark OECD 2012–13 Folkepension (national pension) National Ecuador LAC 2012–13 Pension para Adultos Mayores (Pension National for Older People) El Salvador LAC 2009–13 Pension Basica Universal National (Universal Basic Pension) Estonia ECA 2012–13 National Pension National Fiji EAP 2013–14 Social Pension Scheme (SPS) National 154 APPENDIXES Beneficiaries Age of eligibility Benefit level (% of (% of population Total cost Targeting (years) GDP per capita) over eligible age) (% of GDP) Means-tested 60 21 14 0.06 Means-tested 77 8 – 0.02 Means-tested 70 25 1 0.04 Pensions-tested 65 8 – – Means-tested 65 26 71 2.23 Pensions-tested Men: 67 10 – – Women: 62 Means-tested Men: 65 5 35 0.08 Women: 62 Means-tested 65 and 6 months 23 30 0.74 Pensions-tested Men: 60 5 – – Women: 55 Means-tested 65 35 5 0.30 Means-tested Men: 67 12 30 0.13 Women: 65 Universal 60 15 155 1.08 Universal 65 4 133 0.27 Means-tested Men: 60 31 42 0.98 Women: 55 Means-tested 65 33 12 0.26 Universal 60 6 143 0.02 Means-tested 70 11 – – Means-tested 60 17 84 0.93 Universal 65 12 96 1.45 Means-tested 65 12 55 0.05   60 1 75 0.11 Means-tested Men: 59 5 40 0.13 Women: 54 Means-tested 65 15 29 0.37 Means-tested 65 21 101 5.82 Means-tested 65 7 61 0.24 Means-tested 70 15 6 0.07 Pensions-tested 63 12 3 0.06 Pensions-tested 66 4 22 0.05 (Table continues next page) APPENDIXES  155 TABLE G.1  Old-Age Social Pensions around the World (Continued) Country/economy/ territory Region Year Name of scheme Location Finland OECD 2010 Kansanelake (Old-Age Pension) National France OECD 2007–12 Allocation de Solidarité aux Personnes National Agées (ASPA) Georgia ECA 2010–12 Old-Age Pension National Germany OECD 2012 Grundsicherung im Alter (Needs-Based National Pension Supplement) Greece OECD 2008 Pension to uninsured elderly National Guatemala LAC 2010–12 Programa de aporte economico o del National Adulto Mayor (Economic Contribution Program for Older People) Guyana LAC 2014 Old-age Pension National Hong Kong SAR, China EAP 2013 Normal/higher old-age allowance) National Hungary ECA 2012 Idoskoruak jaradeka (Old-Age Allowance) National Iceland OECD 2011 Lífeyristryggingar Almannatrygginga National (National Basic Pension) India SA 2006–14 Indira Gandhi National Old-Age National Pension Scheme Indonesia EAP 2010 Program Jaminan Sosial Lanjut Usia National (JSLU) (Elderly Social Security Program) (pilot) Ireland OECD 2010 State Pension (noncontributory) National Israel OECD – Special old-age benefit National Italy OECD – Assegno sociale (Social Allowance) National Kazakhstan ECA – State Basic Pension National Kenya SSA 2011 Older Persons Cash Transfer National Kiribati EAP 2010–12 Elderly fund National Korea, Rep. of  EAP 2009–11 Basic old-age pension National Kosovo ECA 2011 Old-age “basic pension” National Kyrgyzstan ECA – Social assistance allowance (old-age) National Latvia ECA – State social security benefit National Lesotho SSA 2009 Old-Age Pension National Lithuania ECA – Old-age social assistance pension National Malaysia EAP 2010 Bantuan Orang Tua (Elderly National Assistance Scheme) Maldives SA 2011–12 Old-Age Basic Pension National Malta MENA 2009 Age Pension National Mauritius SSA 2011–14 Basic Retirement Pension National Mexico LAC 2013 Pensión para Adultos Mayores National Moldova ECA 2009 State Social Allocation for Aged Persons National Mongolia EAP 2007 Social welfare pension National Mozambique SSA 2013 Programa de Subsido Social Basico National (PSSB) (Basic Social Subsidy Program) Namibia SSA 2007–10 Old-Age Pension (OAP) National Nepal SA 2010 Old-Age Allowance National Netherlands OECD 2011–13 Old-Age pension National 156 APPENDIXES Beneficiaries Benefit level (% of (% of population Total cost (% of Targeting Age of eligibility (years) GDP per capita) over eligible age) GDP) Pensions-tested 65 20 53 0.90 Means-tested 65 29 6 0.25 Universal Men: 65 18 106 2.96 Women: 60 Means-tested 65 12 3 – Means-tested 60 24 3 0.18 Means-tested 65 18 16 0.13 Universal 65 18 151 1.06 Means-tested 65 4 61 0.38 Means-tested 62 9 0 0.01 Means-tested 67 6 66 0.60 Means-tested 60 2 25 0.03 Means-tested 70 9 0 – Means-tested 66 33 18 0.60 Means-tested Men: 65–67 13 – – Women: 60–64 Means-tested 65 and 3 months 21 – – Pensions-tested Men: 63 5   – Women: 58 Means-tested 65 24 5 0.02 Universal 67 36 56 0.74 Means-tested 65 4 70 0.30 Universal 65   91 1.19 Pensions-tested Men: 63 17 – – Women: 58 Pensions-tested 67 6 – – Universal 70 39 93 1.31 Pensions-tested Men: 62.5 2   – Women: 60 Means-tested 60 10 9 0.04 Pensions-tested 65 21 91 1.03 Means-tested 60 30 8 0.24 Universal 60 14 159 2.18 Pensions-tested 65 5 63 0.20 Pensions-tested Men: 62 4 1 0.12 Women: 57 Means-tested Men: 60 6 3 0.02 Women: 55 Means-tested Men: 60 13 32 0.19 Women: 55 Universal 60 12 200 0.56 Pensions-tested 70 12 47 0.32 Universal 65 34 110 6.49 (Table continues next page) APPENDIXES 157 TABLE G.1  Old-Age Social Pensions around the World (Continued) Country/economy/ territory Region Year Name of scheme Location New Zealand OECD 2010 Superannuation National Nigeria SSA 2013 Ekiti State Social Security Scheme Regional/Rural Nigeria (2) SSA 2012 Osun Elderly Persons Scheme Regional/Rural Norway OECD 2011 Grunnpensjon (Basic Pension) National Panama LAC 2011 100 a los 70 National Papua New Guinea EAP 2009 Old-Age and Disabled Pension Scheme Regional/Rural (New Ireland Province) Paraguay LAC 2014 Pensión Alimentaria para las Personas National Adultas Mayores Peru LAC 2014 Pension 65 National Philippines EAP 2013–14 Social Pension National Portugal OECD 2012 Pensao Social de Velhice National (Old-Age Social Pension) Samoa EAP 2010–14 Senior Citizens Benefit National Seychelles SSA 2006–12 Old-age pension (social security fund) National Slovenia ECA 2010 State Pension National South Africa SSA 2011–13 Older Persons Grant National Spain OECD 2011–13 Pension no Contributiva de Jubilacion National (Noncontributory Pension for Retirement) St. Vincent and the LAC 2009–12 Elderly Assistance Benefit National Grenadines Suriname LAC 2012 Algemene Oudedags Voorzieningsfonds National (AOV) (State Old-Age Pension) Swaziland SSA 2009 Old-Age Grant National Sweden OECD 2011 Garantipension (Guaranteed Pension) National Switzerland OECD 2012 Extraordinary pension National Tajikistan ECA 2011 Old-age pension National Thailand EAP 2011 Old-Age Allowance National Timor-Leste EAP 2011–12 Support allowance for the elderly National Trinidad and Tobago LAC 2012 Senior Citizens’ Pension National Turkey ECA – Means-Tested Old-Age Pension National Turkmenistan ECA – Social Allowance National Uganda SSA 2012–13 Senior Citizens Grant Regional/Rural (pilot in 14 districts) Ukraine ECA – Social pension and social pension National supplement United Kingdom OECD 2011–13 Pension credit (guarantee credit) National United States OECD 2014 Old-Age Supplementary Security Income National Uruguay LAC 2013–14 Programa de Pensiones No-Contributivas National Uzbekistan ECA 2012 Social pension National Venezuela, RB LAC 2012–13 Gran Mision Amor Mayor National Vietnam EAP 2008–11 Social assistance benefit (category 1) National Zambia SSA 2009 Social Cash Transfer Programme, Regional/Rural Katete (pilot) Source: HelpAge International and ASPIRE. Note: ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. EAC = Europe and Central Asia; EAP = East Asia and Pacific; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; OECD = Organisation of Economic Cooperation and Development; SA = South Asia; SSA = Sub-Saharan Africa; — = not available. 158 APPENDIXES Beneficiaries Benefit level (% of (% of population Total cost (% of Targeting Age of eligibility (years) GDP per capita) over eligible age) GDP) Universal 65 34 97 3.87 Pensions-tested 65 23 1 0.00 Means-tested – 45 – 0.01 Means-tested 67 12 95 4.51 Pensions-tested 70 5 32 0.17 Universal 60 7 3 0.01 Means-tested 65 27 26 0.44 Means-tested 65 8 16 0.11 Means-tested 77 5 7 0.03 Means-tested 66 15 – – Universal 65 19 93 0.89 Universal 63 17 116 1.52 Means-tested 68 13 5 0.10 Means-tested 60 23 100 1.15 Means-tested 65 20 2 0.12 Means-tested 67 10 77 – Universal 60 19 154 1.61 Universal 60 8 134 0.41 Pensions-tested 65 25 41 0.52 Pensions-tested Men: 65 18 – – Women: 60 Pensions-tested Men: 65 12 36 – Women: 60 Pensions-tested 60 4 94 0.32 Universal 60 7 149 2.20 Means-tested 65 27 68 1.41 Means-tested 65 6 – – Pensions-tested Men: 62 7 – – Women: 57 Not defined 65 (60 in Karamoja Region) 17 7 0.03 Means-tested Men: 63 26 – – Women: 59 Means-tested 65 27 11 0.44 Means-tested 65 16 5 0.07 Means-tested 70 22 7 0.24 Pensions-tested Men: 60 26 – – Women: 55 Means-tested Men: 60 18 28 0.60 Women: 55 Pensions-tested 80 5 16 0.01 Not defined 60 10 0 – APPENDIXES 159 TABLE G.2  Old-Age Social Pensions Captured in ASPIRE Household Surveys Country Year Region Program name Bangladesh 2010 SA Old-Age Allowance (MOSW) Belize 2009 LAC Noncontributory pension for women (SSB) Brazil 2012 LAC Beneficio de Prestacao Continuada (BPC) Bulgaria 2007 ECA Social pension Cabo Verde 2007 SSA Minimum social pension Chile 2013 LAC Pensión Básica Solidaria (PBS) de vejez Colombia 2012 LAC Programa de adultos mayores Costa Rica 2012 LAC Pensiones del Régimen no Contributivo Guatemala 2011 LAC Programa Adulto Mayor Honduras 2011 LAC Bono por tercera edad Lithuania 2008 ECA Social pension for persons after retirement age Mauritius 2012 SSA Old-age pension (Basic Retirement Pension) Mexico 2012 LAC Programa 70 y mas Namibia 2009 SSA State old-age pension Nepal 2010 SA Social pension Panama 2012 LAC 100 a los 70 Paraguay 2011 LAC Adulto Mayor Poland 2012 ECA Social pension Romania 2012 ECA Social assistance pension Rwanda 2010 SSA Old-age grant Slovak Republic 2009 ECA Other old-age repeated monetary allowances and benefits South Africa 2010 SSA Old-age pension Sri Lanka 2012 SA Elderly payment Swaziland 2009 SSA Pension Tajikistan 2011 ECA Social pension Thailand 2013 EAP Social pension for the elderly and disable Timor-Leste 2011 EAP Elderly pensions Turkey 2012 ECA Old-age benefits paid by Turkish Pension Fund to those individuals who are older than 65 years of age (Yasli) Source: ASPIRE household surveys. Note: ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity. EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; SA = South Asia; SSA = Sub- Saharan Africa. 160 APPENDIXES APPENDIX H Basic Characteristics of Countries Included in the Book TABLE H.1  Basic Characteristics of Countries Included in the Book Country/ economy/ Income GNI per GDP per Total population territory Code Region classification capita, PPP capita, PPP (million) Afghanistan AFG SA LIC 1,900 1,877 34.7 Albania ALB ECA UMIC 11,929 11,880 2.9 Algeria DZA MENA UMIC 15,075 14,720 40.6 Angola AGO SSA LMIC 6,499 6,220 28.8 Argentina ARG LAC UMIC 19,934 19,480 43.8 Armenia ARM ECA LMIC 8,818 9,000 2.9 Azerbaijan AZE ECA UMIC 17,253 16,130 9.8 Bangladesh BGD SA LMIC 3,581 3,790 163.0 Belarus BLR ECA UMIC 18,060 17,210 9.5 Belize BLZ LAC UMIC 8,448 8,000 0.4 Benin BEN SSA LIC 2,168 2,170 10.9 Bhutan BTN SA LMIC 8,744 8,070 0.8 Bolivia BOL LAC LMIC 7,236 7,090 10.9 Bosnia and Herzegovina BIH ECA UMIC 12,075 12,140 3.5 Botswana BWA SSA UMIC 16,735 16,380 2.3 Brazil BRA LAC UMIC 15,128 14,810 207.7 Bulgaria BGR ECA UMIC 19,199 19,020 7.1 Burkina Faso BFA SSA LIC 1,720 1,680 18.6 Burundi BDI SSA LIC 778 770 10.5 Cabo Verde CPV SSA LMIC 6,553 6,220 0.5 Cambodia KHM EAP LMIC 3,735 3,510 15.8 Cameroon CMR SSA LMIC 3,286 3,250 23.4 Central African Republic CAF SSA LIC 699 700 4.6 Chad TCD SSA LIC 1,991 1,950 14.5 Chile CHL LAC HIC 23,960 23,270 17.9 China CHN EAP UMIC 15,535 15,500 1378.7 Colombia COL LAC UMIC 14,158 13,910 48.7 Comoros COM SSA LIC 1,522 1,520 0.8 Congo, Dem. Rep. ZAR SSA LIC 801 730 78.7 Congo, Rep. COG SSA LMIC 5,719 5,380 5.1 (Table continues next page) APPENDIXES 161 TABLE H.1  Basic Characteristics of Countries Included in the Book (Continued) Country/ economy/ Income GNI per GDP per Total population territory Code Region classification capita, PPP capita, PPP (million) Costa Rica CRI LAC UMIC 16,614 15,750 4.9 Croatia HRV ECA UMIC 23,596 22,880 4.2 Czech Republic CZE ECA HIC 34,711 32,710 10.6 Côte d’Ivoire CIV SSA LMIC 3,720 3,610 23.7 Djibouti DJI MENA LMIC 3,342 2,200 0.9 Dominica DMA LAC UMIC 10,975 10,610 0.1 Dominican Republic DOM LAC UMIC 15,209 14,480 10.6 Ecuador ECU LAC UMIC 11,286 11,070 16.4 Egypt, Arab Rep. EGY MENA LMIC 11,132 11,110 95.7 El Salvador SLV LAC LMIC 8,619 8,220 6.3 Estonia EST ECA HIC 29,365 28,920 1.3 Ethiopia ETH SSA LIC 1,735 1,730 102.4 Fiji FJI EAP UMIC 9,561 9,140 0.9 Gabon GAB SSA UMIC 18,108 16,720 2.0 Gambia, The GMB SSA LIC 1,689 1,640 2.0 Georgia GEO ECA LMIC 9,997 9,450 3.7 Ghana GHA SSA LMIC 4,294 4,150 28.2 Grenada GRD LAC UMIC 13,928 13,440 0.1 Guatemala GTM LAC LMIC 7,947 7,750 16.6 Guinea GIN SSA LIC 1,311 1,200 12.4 Guinea-Bissau GNB SSA LIC 1,582 1,580 1.8 Guyana GUY LAC UMIC 7,819 7,860 0.8 Haiti HTI LAC LIC 1,784 1,790 10.8 Honduras HND LAC LMIC 4,738 4,410 9.1 Hungary HUN ECA HIC 26,681 25,640 9.8 India IND SA LMIC 6,572 6,490 1324.2 Indonesia IDN EAP LMIC 11,612 11,220 261.1 Iran, Islamic Rep. IRN MENA UMIC 17,046 17,370 80.3 Iraq IRQ MENA UMIC 17,353 17,240 37.2 Jamaica JAM LAC UMIC 8,835 8,500 2.9 Jordan JOR MENA LMIC 9,050 8,980 9.5 Kazakhstan KAZ ECA UMIC 25,264 22,910 17.8 Kenya KEN SSA LMIC 3,156 3,130 48.5 Kiribati KIR EAP LMIC 2,047 3,240 0.1 Kosovo KSV ECA LMIC 10,066 10,200 1.8 Kuwait KWT MENA HIC 73,817 83,420 4.1 Kyrgyz Republic KGZ ECA LMIC 3,551 3,410 6.1 Lao PDR LAO EAP LMIC 6,186 5,920 6.8 Latvia LVA ECA HIC 26,031 26,090 2.0 Lebanon LBN MENA UMIC 13,996 13,860 6.0 Lesotho LSO SSA LMIC 3,029 3,390 2.2 Liberia LBR SSA LIC 813 700 4.6 Lithuania LTU ECA HIC 29,966 28,840 2.9 (Table continues next page) 162 APPENDIXES TABLE H.1  Basic Characteristics of Countries Included in the Book (Continued) Country/ economy/ Income GNI per GDP per Total population territory Code Region classification capita, PPP capita, PPP (million) Macedonia, FYR MKD ECA UMIC 15,121 14,480 2.1 Madagascar MDG SSA LIC 1,506 1,440 24.9 Malawi MWI SSA LIC 1,169 1,140 18.1 Malaysia MYS EAP UMIC 27,681 26,900 31.2 Maldives MDV SA UMIC 13,199 11,970 0.4 Mali MLI SSA LIC 2,117 2,040 18.0 Marshall Islands MHL EAP UMIC 4,072 5,280 0.1 Mauritania MRT SSA LMIC 3,854 3,760 4.3 Mauritius MUS SSA UMIC 21,088 20,980 1.3 Mexico MEX LAC UMIC 17,862 17,740 127.5 Moldova MDA ECA LMIC 5,334 5,670 3.6 Mongolia MNG EAP LMIC 12,220 11,290 3.0 Montenegro MNE ECA UMIC 16,854 17,090 0.6 Morocco MAR MENA LMIC 7,838 7,700 35.3 Mozambique MOZ SSA LIC 1,217 1,190 28.8 Myanmar MMR EAP LMIC 5,773 5,070 52.9 Namibia NAM SSA UMIC 10,585 10,550 2.5 Nepal NPL SA LIC 2,468 2,520 29.0 Nicaragua NIC LAC LMIC 5,541 5,390 6.1 Niger NER SSA LIC 978 970 20.7 Nigeria NGA SSA LMIC 5,867 5,740 186.0 Pakistan PAK SA LMIC 5,249 5,580 193.2 Panama PAN LAC UMIC 23,015 20,990 4.0 Papua New Guinea PNG EAP LMIC 2,761 2,700 8.1 Paraguay PRY LAC UMIC 9,577 9,060 6.7 Peru PER LAC UMIC 13,022 12,480 31.8 Philippines PHL EAP LMIC 7,806 9,400 103.3 Poland POL ECA HIC 27,811 26,770 37.9 Qatar QAT MENA HIC 127,523 124,740 2.6 Romania ROM ECA UMIC 23,626 22,950 19.7 Russian Federation RUS ECA UMIC 23,163 22,540 144.3 Rwanda RWA SSA LIC 1,913 1,870 11.9 Samoa WSM EAP UMIC 6,345 6,200 0.2 Saudi Arabia SAU MENA HIC 54,431 55,760 32.3 Senegal SEN SSA LIC 2,568 2,480 15.4 Serbia SRB ECA UMIC 14,512 13,680 7.1 Seychelles SYC SSA HIC 28,391 28,390 0.1 Sierra Leone SLE SSA LIC 1,473 1,320 7.4 Slovak Republic SVK ECA HIC 30,632 29,910 5.4 Slovenia SVN ECA HIC 32,885 32,360 2.1 Solomon Islands SLB EAP LMIC 2,236 2,150 0.6 Somalia SOM SSA LIC – – 14.3 (Table continues next page) APPENDIXES 163 TABLE H.1  Basic Characteristics of Countries Included in the Book (Continued) Country/ economy/ Income GNI per GDP per Total population territory Code Region classification capita, PPP capita, PPP (million) South Africa ZAF SSA UMIC 13,225 12,860 55.9 South Sudan SSD SSA LIC 1,925 1,700 12.2 Sri Lanka LKA SA LMIC 12,316 11,970 21.2 St. Kitts and Nevis KNA LAC HIC 26,686 25,940 0.1 St. Lucia LCA LAC UMIC 11,546 11,370 0.2 St. Vincent and the Grenadines VCT LAC UMIC 11,606 11,530 0.1 Sudan SDN SSA LMIC 4,730 4,290 39.6 Suriname SUR LAC UMIC 14,146 13,720 0.6 Swaziland SWZ SSA LMIC 8,343 7,980 1.3 Syrian Arab Republic SYR MENA LMIC – – 18.4 São Tomé and Príncipe STP SSA LMIC 3,229 3,240 0.2 Tajikistan TJK ECA LMIC 2,980 3,500 8.7 Tanzania TZA SSA LIC 2,787 2,740 55.6 Thailand THA EAP UMIC 16,916 16,070 68.9 Timor-Leste TMP EAP LMIC 2,290 4,340 1.3 Togo TGO SSA LIC 1,491 1,370 7.6 Tonga TON EAP UMIC 5,752 5,760 0.1 Trinidad and Tobago TTO LAC HIC 31,908 30,810 1.4 Tunisia TUN MENA LMIC 11,599 11,150 11.4 Turkey TUR ECA UMIC 24,244 23,990 79.5 Uganda UGA SSA LIC 1,849 1,820 41.5 Ukraine UKR ECA LMIC 8,272 8,190 45.0 Uruguay URY LAC HIC 21,625 21,090 3.4 Uzbekistan UZB ECA LMIC 6,514 6,640 31.8 Vanuatu VUT EAP LMIC 3,081 3,050 0.3 Venezuela, RB VEN LAC UMIC 18,281 17,700 31.6 Vietnam VNM EAP LMIC 6,424 6,050 92.7 West Bank and Gaza WBG MENA LMIC 2,943 3,290 4.6 Yemen, Rep. YEM MENA LMIC 2,508 2,490 27.6 Zambia ZMB SSA LMIC 3,922 3,790 16.6 Zimbabwe ZWE SSA LIC 2,006 1,920 16.2 Source: World Development Indicators. Note: The inventory includes the list of countries for which administrative and/or household data on social protection and labor programs are available and used in this book. GNI per capita, GDP per capita, and population values were collected from the World Development Indicators for 2016. For the cases in which data are not available, they are replaced by the most recent available data. Specifically, Djibouti 2015; Iran, Islamic Rep. 2015; Kuwait 2015; Papua New Guinea 2014; South Sudan 2015; Timor-Leste 2015; and Venezuela, RB 2013 are used to replace the missing GNI per capita cells. Djibouti 2005; Iran, Islamic Rep. 2014; Kuwait 2015; Myanmar 2015; Papua New Guinea 2014; Qatar 2015; South Sudan 2015; Timor-Leste 2015; Vanuatu 2014; and Venezuela, RB 2013 are used to replace missing GDP per capita cells. Blank cells (for example, Syria) mean no information is available. GDP per capita is in current international dollars. ASPIRE = Atlas of Social Protection: Indicators of Resilience and Equity; ECA = Europe and Central Asia; EAP = East Asia and Pacific; GNI = gross national income; HIC = high-income country; LAC = Latin America and the Caribbean; LIC = low-income country; LMIC = lower-middle-income country; MENA = Middle East and North Africa; OECD = Organisation of Economic Co-operation and Development; PPP = purchasing power parity; SSA = Sub-Saharan Africa; UMIC = upper-middle-income country; — = not available. 164 APPENDIXES Glossary Term Definition Adequacy of benefits The total transfer amount received by all beneficiaries in a quintile as a share of the total welfare of beneficiaries in that quintile. Specifically, adequacy of benefits is the amount of transfers received by a quintile divided by the total income or consumption of beneficiaries in that quintile. Average per capita transfer For each beneficiary household, the per capita average transfer is estimated as the total amount of transfers received divided by the household size. Beneficiary incidence Percentage of program beneficiaries in a quintile relative to the total number of beneficiaries in the population. Specifically, the beneficiary incidence is the number of individuals in each quintile who live in a household where at least one member participates in a social protection and labor program divided by the number of individuals participating programs in the population. in social protection and labor ­ Benefit–cost ratio Reduction in poverty gap obtained for each US$1 spent on social protection and labor programs. The indicator is estimated for the entire population and by program type. Specifically, the benefit–cost ratio is estimated as the poverty gap before transfer minus the poverty gap after transfer divided by the total transfer amount. Benefit incidence Percentage of benefits going to each group or quintile of the posttransfer (or pretransfer) welfare distribution relative to the total benefits going to the population. Specifically, benefit incidence is equal to the sum of all transfers received by all individuals in the quintile divided by the sum of all transfers received by all individuals in the population. The indicator usually includes both direct and indirect beneficiaries. Coverage Percentage of the population or population group participating in the social protection and labor program. Specifically, coverage of a given group or quintile is the number of benefit recipients in the group or quintile divided by the number of individuals in that quintile. The coverage includes both direct and indirect beneficiaries. Gini inequality reduction Simulated change (percentage) in the Gini inequality ­ coefficient because of social protection and labor programs. Specifically, the Gini inequality reduction is computed as the inequality pretransfer minus the inequality posttransfer divided by the inequality pretransfer. Poverty gap reduction Simulated change (percentage) in poverty gap because of social protection and labor programs. The poverty gap index is the average percentage shortfall in the income of the poor. Specifically, the poverty gap reduction is computed as the poverty gap pretransfer minus the poverty gap posttransfer divided by the poverty gap pretransfer. Poverty headcount reduction Simulated change (percentage) in the poverty headcount because of social protection and labor programs. The poverty headcount ratio is the percentage of the population below the poverty line. Specifically, the poverty headcount reduction is computed as the poverty headcount pretransfer minus the poverty headcount posttransfer divided by the poverty headcount pretransfer. Program beneficiaries, number of Number of program beneficiaries (households or individuals) as reported in administrative data. The data indicate the original beneficiary unit (household or individual). For the household-level benefit, the data also report the respective number of individuals benefiting from the program. This information is presented in appendix C. Program duplication and overlap Percentage of beneficiaries who receive one or more benefits from different social protection and labor programs. Public expenditures Total program expenditures, including spending on benefits and administrative costs. The indicator captures both the recurrent and capital program budget and is usually based on administrative program records. Program-level expenditures are presented as a percentage of gross domestic product for the respective year, and is aggregated by harmonized social safety net program categories. Total program expenditures without health fee waivers are also presented. GLOSSARY 165 Environmental Benefits Statement The World Bank Group is committed to reducing its environmental footprint. In support of this commitment, we leverage electronic publishing options and print-on-demand technology, which is located in regional hubs worldwide. Together, these initiatives enable print runs to be lowered and shipping dis- tances decreased, resulting in reduced paper consumption, chemical use, greenhouse gas emissions, and waste. We follow the recommended standards for paper use set by the Green Press Initiative. The majority of our books are printed on Forest Stewardship Council (FSC)–certified paper, with nearly all containing 50–100 percent recycled ­ ­ content. The recycled fiber in our book paper is either unbleached or bleached using totally chlorine-free (TCF), processed chlorine–free (PCF), or enhanced elemental chlorine–free (EECF) processes. More information about the Bank’s environmental philosophy can be found at http://www.worldbank.org/corporateresponsibility. T he State of Social Safety Nets 2018 examines global trends in the social safety net/social assistance coverage, spending, and program performance based on the World Bank Atlas of Social Protection: Indicators of Resilience and Equity (ASPIRE) updated database. The book documents the main social safety net programs that exist globally and their use to alleviate poverty and to build shared prosperity. This 2018 edition expands on the 2015 version in administrative and household survey data coverage. This edition is distinctive in that for the first time, it describes what happens with social safety net/social assistance programs spending and coverage over time, when the data allow us to do so. The State of Social Safety Nets 2018 also features two special themes—social assistance and aging, focusing on the role of old-age social pensions; and adaptive social protection, focusing on what makes social safety net systems and programs adaptive to various shocks. SKU 211254