Accelerating Poverty Reduction Africa in Kathleen Beegle Luc Christiaensen Editors Accelerating Poverty Reduction in Africa Accelerating Poverty Reduction in Africa Kathleen Beegle Luc Christiaensen Editors © 2019 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 22 21 20 19 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 Execu- tive 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 immuni- ties 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: Beegle, Kathleen, and Luc Christiaensen, eds. 2019. Accelerating Poverty Reduction in Africa. Washington, DC: World Bank. doi:10.1596/978-1-4648-1232-3. 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 attribu- tion: 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 attribu- tion: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adap- tation 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 responsibil- ity 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 (paper): 978-1-4648-1232-3 ISBN (electronic): 978-1-4648-1233-0 DOI: 10.1596/978-1-4648-1232-3 Cover art: Courtesy of Leen Boon. Used with the permission of Leen Boon; 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 Editors and Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix Key Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Poverty Reduction in Africa: A Global Agenda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Poverty in Africa: Stylized Facts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Africa’s Slower Poverty Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Growth Fundamentals and Poverty Financing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Earning More on the Farm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Moving Off the Farm: Household Enterprises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Managing Risks and Conflict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Mobilizing Resources for the Poor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Way Forward: Four Primary Policy Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 v vi   C o n t e n t s 1. Poverty in Africa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Poverty Today and Tomorrow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Africa’s Poverty in Profile. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Lessons from Recent Experience. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 2. Africa’s Demography and Socioeconomic Structure. . . . . . . . . . . . . . . . 51 High Fertility Holds Back Poverty Reduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 Poor Initial Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 More and Better Income-Earning Opportunities for the Poor . . . . . . . . . . . . . . . . . . . . . . . 63 A Way Forward. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Fundamentals 1  Africa’s Human Development Trap. . . . . . . . . . . . . . . . . . . . . . . . . 83 The Health Poverty Trap. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 The Education Poverty Trap. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Escaping the Human Development Poverty Trap. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 3. Earning More on the Farm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Largely Favorable Conditions for Agricultural Development . . . . . . . . . . . . . . . . . . . . . . . . 96 Not All Agricultural Growth Is Equally Poverty Reducing. . . . . . . . . . . . . . . . . . . . . . . . . 101 An Integrated Approach Is Needed. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Inclusive Value Chain Development as Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 The Need for Complementary Public Goods, Especially for Staples. . . . . . . . . . . . . . . . . . 123 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 Fundamentals 2  The Nexus of Gender Inequality and Poverty. . . . . . . . . . . . . . . 145 Gender Gaps in Human Endowments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Glaring Differences in the Time Use of Men and Women. . . . . . . . . . . . . . . . . . . . . . . . . . 147 Differences in Asset Ownership and Control between Women and Men. . . . . . . . . . . . . . 147 Gender Gaps Exacerbated by Formal and Informal Institutions and Norms . . . . . . . . . . . 148 Mobility and Safety Challenges for Women. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Policy Levers to Address Gender Gaps and Reduce Poverty. . . . . . . . . . . . . . . . . . . . . . . . 150 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 4. Moving to Jobs Off the Farm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 A Profile of Africa’s Off-Farm Work. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156 The Prospect of Formal Wage Jobs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 Key Traits of Household Enterprises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 Better Household Enterprises for the Poor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170 Fostering Demand: The Role of Towns and Regional Trading. . . . . . . . . . . . . . . . . . . . . . 174 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 Fundamentals 3  Leapfrogging with Technology (and Trade) . . . . . . . . . . . . . . . . 187 Trends, Challenges, and Leapfrogging Opportunities. . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 How Can the Poor Benefit from These Technological Advances?. . . . . . . . . . . . . . . . . . . . 190 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194 C o n t e n t s   vii 5. Managing Risks and Conflict. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 The Urgency of Risk Management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 Risk and Conflict Increase Poverty and Keep People Poor. . . . . . . . . . . . . . . . . . . . . . . . . 199 Prevalence of Shocks and Conflict in Africa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 Reducing Exposure to Shocks in Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212 How Do Households Manage Shocks? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Better Insurance for the Poor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Addressing Constraints to Investment in Risk Prevention and Management . . . . . . . . . . . 227 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234 Fundamentals 4  Politics and Pro-Poor Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Varying Politics and Incentives, Varying Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Channels for Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 6. Mobilizing Resources for the Poor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 Africa’s Large Poverty Financing Gap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248 Fiscal Systems in Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248 Mobilizing More (and Less-Harmful) Revenues. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 Toward Better Spending for the Poor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278 Boxes O.1 Investments in human capital are critical to alleviate poverty. . . . . . . . . . . . . . . . . . . . . 8 O.2 Gender inequality is a hurdle to poverty reduction in Africa. . . . . . . . . . . . . . . . . . . . . . 9 O.3 Leapfrogging technology holds promise for poverty reduction in Africa. . . . . . . . . . . . 11 1.1 Efforts to improve Africa’s poverty data are starting to pay off. . . . . . . . . . . . . . . . . . . 37 1.2 To measure gender and age gaps in poverty, you need to get into the household. . . . . . 38 1.3 Africa’s poverty-to-growth elasticity is low because Africa is poor. . . . . . . . . . . . . . . . 43 1.4 High fertility and initial poverty reduce Africa’s poverty-to-growth elasticity. . . . . . . 44 2.1 The fertility transition has not begun in much of Africa, and where it has, it bypasses the poorest. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 2.2 New insights from the psychology of poverty can inform project design. . . . . . . . . . . . 58 2.3 Formal safety nets and commitment savings devices can help households overcome the investment trap. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 2.4 Should low- and middle-income countries go BIG?. . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 F1.1 Child marriage and early childbearing trap girls in poverty. . . . . . . . . . . . . . . . . . . . . 84 3.1 Unprocessed staples make up much of Africa’s rapidly growing food demand . . . . . . . 97 3.2 African staple crop supply responds to price incentives and reductions in transaction costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 3.3 Shifts in agriculture bring new terminology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 4.1 Can industrial policy drive poverty reduction? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 4.2 Are household enterprises created of necessity or opportunity?. . . . . . . . . . . . . . . . . . 161 4.3 Job creation by nonfarm enterprises in Rwanda shows a high churn rate and the importance of location for market access. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 4.4 The farm economy and nonfarm employment are strongly linked. . . . . . . . . . . . . . . . 166 4.5 How not to do it: Governments’ approaches to household enterprises have ranged from wishing them away to outright harassment. . . . . . . . . . . . . . . . . . . . . . . 176 F3.1 Rules matter for mobile money adoption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 5.1 Displaced Malians suffered substantially but less than those staying behind. . . . . . . . 200 viii   C o n t e n t s 5.2 Forced displacement is a poverty trap in Africa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 5.3 In Africa, shocks affect income more often than assets. . . . . . . . . . . . . . . . . . . . . . . . 204 5.4 Farmer-managed natural regeneration of trees and land holds promise for reducing drought risk. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 5.5 Safety nets and subsidized insurance help protect pastoralists in Kenya . . . . . . . . . . . 225 5.6 A new humanitarian-development paradigm emerges for managing long-term displacement crises. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228 6.1 Fiscal incidence analysis offers a way to estimate the distributional impacts of taxes and transfers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 6.2 Tobacco taxes can provide a poverty win-win. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262 6.3 Three stories illustrate African countries’ losses of corporate tax revenue. . . . . . . . . . 264 6.4 Farm input subsidies are less effective than alternative policies in reducing poverty. . 269 6.5 Fertilizer markets are often not competitive. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 Figures O.1 Natural resource dependence has increased substantially in most African countries. . . . . 4 O.2 In Africa, fertility is less responsive to conventional parameters of development than in other LDCs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 O.3 Africa’s food import bill has tripled since the mid-2000s . . . . . . . . . . . . . . . . . . . . . . . 12 O.4 ODA is a large share of GNI in low-income countries. . . . . . . . . . . . . . . . . . . . . . . . . . 21 I.1 More than half of the world’s extreme poor live in Africa. . . . . . . . . . . . . . . . . . . . . . . 30 1.1 The poverty rate in Africa has gone down, but the number of African people living in poverty has increased. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 1.2 Africa cannot eradicate poverty by 2030 but can accelerate poverty reduction. . . . . . . 35 B1.1.1 African countries’ poverty status can now be estimated from recent household surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 2.1 Fertility has declined much more slowly in Africa than elsewhere. . . . . . . . . . . . . . . . . 54 2.2 In Africa, fertility is less responsive to conventional parameters of development than in other LDCs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 2.3 Natural resource dependence has increased substantially in most African countries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 2.4 Fiscal accounts and government debt in Africa have deteriorated since the global crisis of 2008–09 and have yet to recover. . . . . . . . . . . . . . . . . . . . . . . 66 2.5 The relative advantage of agricultural growth in reducing poverty declines with development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 F1.1 In Africa, poor children are much more likely to be stunted. . . . . . . . . . . . . . . . . . . . . 86 F1.2 In Africa, poor children learn less. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 3.1 Africa’s food import bill has tripled since the mid-2000s . . . . . . . . . . . . . . . . . . . . . . . 97 3.2 Poverty rates in Africa are highest in more-remote areas with better agroecological potential. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 3.3 Rural individuals in Malawi work less, and more seasonally, than urban residents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 3.4 Land productivity in Africa grows faster in countries with lower agricultural endowments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 3.5 Agriculture spending in Africa substantially lags spending in East Asia and the Pacific . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 F2.1 The literacy gap between men and women in Africa is shrinking but still large . . . . . 146 F2.2 Across African countries, women carry most of the burden of unpaid domestic and care work. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 F2.3 Far fewer African women than men own land or housing. . . . . . . . . . . . . . . . . . . . . . 148 C o n t e n t s   ix F2.4 Norms constrain women’s physical mobility, especially in western and central Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 4.1 Household enterprise work is far more common than wage employment for the poor in Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 4.2 The poor’s household enterprises tend to be smaller than those of the nonpoor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 4.3 Urban household enterprises and those with better-educated owners tend to grow over time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 B4.4.1 In Ethiopia, rural nonfarm activity peaks soon after the main harvest. . . . . . . . . . . . 166 4.4 Enterprise profits rise with household wealth and take a big jump in the top quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 4.5 The contribution of household enterprises to income is higher among wealthier households. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 F3.1 Mobile internet is expanding throughout Africa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188 5.1 The share of nonpoor in Africa who fall into poverty is about the same as the share of poor people who move out of poverty . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 5.2 Life in African countries is riskier than in other regions. . . . . . . . . . . . . . . . . . . . . . . 203 B5.3.1 Income losses are twice as prevalent as asset losses in Africa. . . . . . . . . . . . . . . . . . . . 205 5.3 The nature of risk in Africa varies by country, type of shock, and poverty level. . . . . 206 5.4 Unexpected income losses are reported by rich and poor alike . . . . . . . . . . . . . . . . . . 208 5.5 Rural life is particularly risky in Africa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 5.6 Female-headed households often face more risk. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 5.7 The risk of conflict has been increasing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 5.8 Conflict events have recently increased in Kenya and Nigeria. . . . . . . . . . . . . . . . . . . 212 5.9 Savings, family, and friends help households cope with shocks. . . . . . . . . . . . . . . . . . 218 5.10 Out-of-pocket health care payments increase and deepen poverty in Africa. . . . . . . . 220 5.11 Managing health and weather shocks requires a mix of tools. . . . . . . . . . . . . . . . . . . 221 5.12 Formal health insurance coverage in Africa is low, and concentrated among the better-off . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 5.13 The number of safety net programs in Africa is increasing, but their coverage is low. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 5.14 Take-up of preventive health care products drops precipitously in response to very small fees. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229 6.1 High poverty levels imply high tax rates on the nonpoor to cover need. . . . . . . . . . . . 249 6.2 Natural resource revenues are not sufficient to eliminate the poverty gap. . . . . . . . . . 249 6.3 Most African countries have a domestic revenue deficit. . . . . . . . . . . . . . . . . . . . . . . . 250 6.4 ODA is a large share of GNI in low-income countries . . . . . . . . . . . . . . . . . . . . . . . . 252 6.5 African countries vary in spending by sector, but education spending dominates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 6.6 Not all African countries are reaching spending targets in education . . . . . . . . . . . . . 255 6.7 Fiscal policy in Africa frequently increases poverty. . . . . . . . . . . . . . . . . . . . . . . . . . . 258 6.8 Fiscal systems in Africa create net losses for the poor even when the incidence of poverty is reduced . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 6.9 Indirect taxes outweigh subsidy and transfer benefits for the bottom 40 percent of most African populations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 6.10 Corporate tax rates in Africa have declined . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266 B6.4.1 Direct transfers have greater poverty impact than subsidies . . . . . . . . . . . . . . . . . . . . 269 6.11 Greater concentration of political influence can result in more subsidies . . . . . . . . . . 270 B6.5.1 African and world urea prices show a large gap. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 6.12 Inefficiency in the primary years of education remains a challenge for many African countries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 x   C o n t e n t s Maps O.1 Some parts of Africa are hit harder by risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 1.1 Africa’s poverty and poor are concentrated in a limited number of (often landlocked) countries and regions within these countries . . . . . . . . . . . . . . . . . . 40 3.1 Remote, high-potential areas are concentrated in central Africa, eastern Ethiopia, and Madagascar. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 4.1 Agriprocessing firms concentrate along the borders in Zambia. . . . . . . . . . . . . . . . . . 177 BF3.1.1 Mobile money account penetration in Sub-Saharan Africa varies widely by country. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 5.1 Some parts of Africa are hit harder by risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 5.2 Many cost-effective risk-reducing strategies are not well used. . . . . . . . . . . . . . . . . . . 215 Tables B1.3.1 Africa’s elasticity and semi-elasticity of poverty to per capita income growth are not different from the rest of the world when controlling for initial poverty. . . . . . . . . 43 B1.4.1 Substantial poverty reduction could result from increasing Africa’s human and physical capital indicators to the global median . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.1 In Tanzania, there are larger income gains per grower in fruits and vegetables but much larger employment gains in staples, especially rice. . . . . . . . . . . . . . . . . . . . 104 3.2 Staple crop productivity growth is more poverty reducing than export crop productivity growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 6.1 Service providers are often absent at schools and health clinics in Africa . . . . . . . . . . 275 Foreword O ur goal is a world free of poverty. To improving well-being, they are not s ­ ufficient. get there, we must accelerate pov- Despite economic growth in Africa, the erty reduction in Africa. Although region’s persistently rapid population growth, the share of Africa’s population living in structural impediments (low human capital, extreme poverty has come down substan- persistent gender inequality, and large infra- tially, from 54 percent in 1990 to 41 percent structure deficits), and increasing reliance in 2015, more Africans are living in poverty on natural resources continue to hold back today than in 1990, in part because of pop- ­poverty reduction. ulation growth. In fact, the world’s poor are This report revisits the challenges and increasingly concentrated in Africa. opportunities to tackle Africa’s poverty, draw- Tackling this challenge begins with ing on the latest evidence. It focuses on the being able to measure it robustly. Following income opportunities of the poor, the policies Poverty in a Rising Africa —the precursor needed to support these opportunities, and the to this report, which mapped the data land- resources needed to finance pro-poor invest- scape—efforts to improve Africa’s poverty ments. A pro-poor agenda means generating data are starting to pay off. More and better more formal jobs while working to increase the household surveys are now available to track incomes of smallholder farmers and informal and analyze poverty. And Africa’s Statistical workers in secondary towns and strengthening Capacity Indicator—which grades country their capacity to manage risks. This approach statistical systems on the quality, frequency, is how the poor will likely benefit the most. and timeliness of core economic and social The report advances a poverty-reduction data—has been improving. agenda for Africa that rests on four pillars: The key features of Africa’s poverty, and accelerating Africa’s fertility transition; lever- its causes, have been widely documented. aging the food system, both on and off the But some of the challenges, such as cli- farm; mitigating fragility; and addressing mate change, fragility, and debt pressures, the poverty financing gap. The report fur- are gaining in importance. And although ther calls for integrated approaches in these ­ m acroeconomic stability and growth are areas—simultaneously addressing supply- critical components for reducing poverty and and demand-side constraints—and highlights xi xii   F o r e w o r d the promise of technological leapfrogging for we are able to paint a more accurate picture poverty reduction in Africa. of both the complexity of the issue and how The World Bank is committed to help- best to address it. Thanks to this report, we ing Africa build a better future for its people are one step closer to achieving our twin and to alleviating poverty in all its forms. goals of eradicating extreme poverty and Through comprehensive data and analysis, boosting shared prosperity. Hafez Ghanem Vice President, Africa Region The World Bank Acknowledgments T his report has been prepared by a and inputs from Andrew Dabalen, Markus team led by Kathleen Beegle and Goldstein, and Johannes Hoogeveen. Luc Christiaensen, with a core team The team benefited from feedback from comprising Tom Bundervoet, Alejandro participants at workshops and presenta- de la Fuente, Lionel Demery, Patrick Eoz- tions at the African Center for Economic enou, Isis Gaddis, Ruth Hill, Siddhartha Transformation (ACET) in Accra, Ghana; Raja, Joachim Vandercasteelen, Philip the ACET African Transformation Forum Verwimp, and Eleni Yitbarek. Georgina in 2018; the Centre for Social Policy Studies Maku Cobla, Moctar N’Diaye, and Kwame (CSPS) Second International Conference ­ Twumasi-Ankrah served as research assis- at the University of Ghana in 2018; the tants. Thomas Sohnesen also contributed. Households in Conflict Network (HiCN) The team is grateful to Albert Zeufack for 13th A nnual Workshop in 2017; the his overall guidance throughout the process. International Union for the Scientific Study of The team has also benefited greatly from Population (IUSSP) International Population extensive consultations, discussions, and sug- Conference in Cape Town in 2017; the gestions involving many colleagues through- United Nations University-World Institute for out the preparation of the report. These Development Economics Research (UNU- include the inputs and guidance on specific WIDER) Think Development Conference in chapters from Javier Baez, Umberto Cattaneo, 2018; and the University of Guelph, Ontario. Nabil Chaherli, Daniel Clarke, David Coady, The thoughtful comments of the peer Aline Coudouel, Julie Dana, Chris Delgado, reviewers—Stephan Klasen, Peter Lanjouw, Sunita Dubey, Patrick Eozenou, Louise Fox, Jacques Morisset, and an anonymous Ugo Gentilini, Stephane Hallegatte, Bernard reviewer—are greatly appreciated. Haven, Ruth Hill, Gabriela Inchauste, Jon This task received financial support from Jellema, Nora Lustig, Rose Mungai, Nga the Office of the Chief Economist of the Thi Viet Nguyen, Nadia Piffaretti, Marco World Bank Group’s Africa Region. Ranzani, Emmanuel Skoufias, Andre Marie The findings, interpretations, and conclu- Taptue, and Dominique van de Walle. And sions are those of the authors and do not nec- we thank Nga Thi Viet Nguyen for her essarily reflect the views of management, the analysis on direct dividend payments. The reviewers, and other colleagues consulted or team received valuable cross-cutting advice engaged in the preparation of the report. xiii About the Editors and Contributors Kathleen Beegle is a lead economist in the co-leads several lending operations in sta- World Bank’s Gender Group. She was previ- tistical capacity building and urban social ously a human development program leader protection. Before joining the World Bank based in Accra, Ghana, covering Ghana, in 2012, Tom worked in humanitarian envi- Liberia, and Sierra Leone. She co-led the ronments in Burundi and the Democratic World Bank regional studies Realizing Republic of Congo. Prior to that, he was the Full Potential of Social Safety Nets in engaged as an academic in Belgium. Tom Africa (2018) and Poverty in a Rising Africa has a doctorate in economics from the (2016) and was deputy director of World University of Brussels and has published his Development Report 2013: Jobs. As part of research in several peer-reviewed academic the World Bank’s Research Group for more journals. than a decade, Kathleen’s research focused on poverty, labor, and economic shocks. She Luc Christiaensen is a lead agricultural was also a lead member of the World Bank economist in the World Bank’s Jobs Group. Living Standards Measurement Study team, He has written extensively on poverty, where she led the design and implementa- structural transformation, and second- tion of national household surveys, as well ary towns in Africa and East Asia. He as methodological studies on survey design. led the team that produced Agriculture in Before joining the World Bank, she worked Africa: Telling Myths from Facts and was at RAND Corporation. Kathleen holds a a core member of the team for the World doctorate in economics from Michigan State Development Report 2008: Agriculture University. for Development. He also co-led the World Bank regional study, Poverty in a Rising Tom Bundervoet is a senior economist in Africa, the precursor to this report. He was the World Bank’s Poverty and Equity Global a Senior Research Fellow at UNU-WIDER Practice, based in Addis Ababa, Ethiopia, in Helsinki, Finland, during 2009 –10. and previously in Rwanda. His work focuses He is an honorary research fellow at the on policy analysis of poverty, employment, Maastricht School of Management and the and human development. He also leads or Catholic University of Leuven. Luc holds xv xvi   About the Editors and Contributors a PhD in agricultural economics from Sub-Saharan Africa. Before joining the Cornell University. World Bank, Patrick was a postdoctoral fellow at the International Food Policy Alejandro de la Fuente is a senior econo- Research Institute (IFPRI) in Washington, mist in the World Bank’s Poverty and Equity DC. His current work focuses on health Global Practice. His current work involves equ it y a nd healt h f i na nci ng. Pat rick providing policy advice and operational has published in the Briti sh Jour n al and technical support on poverty analysis, of Nutrition , Health Affairs , Journ al food and nutrition security, and program of Development Economics , Journal of evaluation to Liberia and Sierra Leone. Development Studies, The Lancet Global Previously, he worked on similar issues in Health, Oxford Re vie w of Economic Malawi, Zambia, and Zimbabwe. He has Policy, and PLoS Medicine . He holds a also worked on and led projects on poverty, doctorate in economics from the European natural disasters, and weather insurance in University Institute in Florence. countries in East Asia and Latin America and the Caribbean. Before joining the World Isis Gaddis is a senior economist in the Bank, Alejandro worked for the Human World Bank’s Gender Group. She was pre- Development Report Office at the United viously based in Dar es Salaam, working as Nations Development Programme, the a poverty economist for Tanzania. Isis was International Strategy for Risk Reduction a member of the core team for the Poverty Secretariat of the United Nations Office for and Shared Prosperity (World Bank 2018) Disaster Risk Reduction, the Inter-American report and coauthored the regional study, Development Bank, and in various positions Poverty in a Rising Africa (World Bank at the Ministry of Social Development and 2016). Her main research interest is empir- the Office of the President in Mexico. He ical microeconomics, with a focus on the holds a doctorate in development studies measurement and analysis of poverty and and development economics from Oxford inequality, gender, labor, and public ser- University. vice delivery. She holds a doctorate in eco- nomics from the University of Göttingen, Lionel Demery is an independent consul- where she was a member of the develop- tant, specializing in development econom- ment economics research group from 2006 ics. Previously, he was a lead economist in to 2012. the Africa Region of the World Bank. He has taught in the economics departments of Ruth Hill is a lead economist and the global the University of Warwick and University lead for the spatial and structural transfor- College Cardiff. He has also worked for mation global solutions group in the World the International Labour Organization in Bank’s Poverty & Equity Global Practice. Bangkok and the Overseas Development Previously, she was a senior economist in the Institute in London. He has published South Asia region working on Bangladesh widely, focusing recently on poverty in and Nepal and, before that, Ethiopia, Africa. Lionel holds a master’s degree from Somalia, and Uganda. In addition to con- the London School of Economics. ducting analytical work on poverty and labor markets, she co-led the World Bank’s Patrick Eozenou is a senior economist at Systematic Country Diagnostics for Nepal the World Bank in the Health, Nutrition, and Ethiopia, as well as the Ethiopian Urban and Population Global Practice. He has Productive Safety Net Program. Before join- more than 10 years of experience in devel- ing the World Bank in 2013, she was a senior opment economics, microeconomics, and research fellow at IFPRI. Ruth has published health economics, with a strong focus on in the American Journal of Agricultural A b o u t t h e E d i t o r s a n d C o n t r i b u t o r s    xvii Economics, Economic Development and Philip Verwimp is a professor of development Cultural Change, Experimental Economics, economics at Université Libre de Bruxelles, Jour n al of De velopme nt E conomic s, where he teaches in the Solvay Brussels Wo rl d Bank E c on o mic R e v i e w , a nd School of Economics and Management; World Development. She received her he is also a fellow at the European Center doctorate in economics from the University ­ for Advanced Research in Economics and of Oxford. Statistics (ECARES). Philip is cofounder and codirector of the Households in Conflict Siddhartha Raja is a senior digital devel- Network (HiCN), which undertakes collab- opment specialist with the World Bank orative research into the causes and effects Group. He works with governments across of violent conflict at the household level. Asia and Europe to connect more people to Between 1999 and 2017, he spent 30 months information, markets, and public services. of f ield work i n Bu r u nd i, E t h iopia , Siddhartha’s work has led to the expan- Morocco, Rwanda, and Tanzania. Philip has sion of broadband connectivity, to people offered policy advice and worked on reports developing their digital skills and working for the World Bank and the United Nations online, to public agencies getting online Children’s Fund (UNICEF), as well as vari- to deliver services to more people, and to ous countries’ ministries of foreign affairs. exponential improvements in international He has 30 peer-reviewed publications on the connectivity in countries across Europe and ­ economics of conflict, poverty and undernu- Asia. He has published regularly with the trition, child health and education, political World Bank on telecommunications policy economy, and migration in leading journals, and the future of work. Siddhartha has a including the American Economic Review, bachelor’s degree in telecommunications Economic Development and Cultural engineering from the University of Bombay Change, Journal of Conflict Resolution, and a master’s degree in infrastructure Jour n al of De velopme nt E conomic s, policy studies from Stanford University. Journal of Human Resources, and World He has studied media law and policy at Bank Economic Review. Philip’s academic the University of Oxford and has a doctor- studies have encompassed economics, soci- ate in telecommunications policy from the ology, and political science at the University University of Illinois. of Antwerp (bachelor’s degree), KU Leuven (bachelor’s and master’s degrees), University Joachim Vandercasteelen is an indepen- of Göttingen (master’s degree), and Yale dent consultant and a postdoctoral fellow University (predoctoral work). He was a at the LICOS Centre for Institutions and postdoctoral fellow at Yale with a Fulbright Economic Performance at the University of Fellowship and a visiting researcher at Leuven (KU Leuven), Belgium. Currently, the University of California, Berkeley. He he is working on the impact evaluation of obtained his doctorate in development eco- different value chain interventions in the nomics at KU Leuven, with a dissertation on agricultural subsectors of Côte d’Ivoire, the political economy of development and Ethiopia, and Tanzania. He is working as genocide in Rwanda. a consultant for the World Bank, the World Food Programme, IFPRI, and the Weather Eleni Yitbarek is a postdoctoral research Index-Based Risk Services (WINnERS) proj- fellow at the University of Pretoria and a ect based at Imperial College London. His fellow in applied development finance at research interests include themes crossing the European Investment Bank and Global rural development and agricultural econom- Development Network. Eleni’s research ics. Joachim holds a doctorate in economics focuses on applied research in poverty from KU Leuven. dynamics, the socioeconomic effects of xviii   About the Editors and Contributors idiosyncratic and transient shocks, and Organisation (SNV). She has a master’s gender-based social mobility in Africa. Eleni ­ degree from Maastricht University in pub- was a World Bank Africa Fellow while in lic policy and human development, special- graduate school. Before pursuing her doctor- izing in social policy financing, and earned ate, Eleni worked for the National Bank of her doctorate in economics from Maastricht Ethiopia and the Netherlands Development University. Abbreviations AEP agroecological potential AGI Adolescent Girls Initiative APG aggregate poverty gap ARC Africa Risk Capacity ATAF African Tax Administration Forum BEPS Base Erosion and Profit Shifting BIG basic income guarantee CAADP Comprehensive Africa Agriculture Development Programme Cat DDO Catastrophe Deferred Drawdown Option CEQ Commitment to Equity CO2 carbon dioxide DAC Development Assistance Committee (of the OECD) DC direct current DDP direct dividend payment DDR disarmament, demobiliza­ tion, and reintegration DHS Demographic and Health Survey EC Establishment Census ECOWAS Economic Community of West African States EFA Education for All EITI Extractive Industries Transparency Initiative EU European Union FDI foreign direct investment FGP fiscal gains to the poor FI fiscal impoverish­ment FIA fiscal incidence analysis FISP Farm Input Subsidy Program FIV flexible input voucher FMNR farmer-managed natural regeneration GDP gross domestic product GER gross enrollment rate xix xx   A b b r e v i a t i o n s GIS geographic information system GNI gross national income GSMA Global System for Mobile Association HIV/AIDS human immunodeficiency virus/acquired immune deficiency syn­ drome HSNP Hunger Safety Net Program ICT information and communication technology ICTD International Centre for Tax and Development IDA International Development Association (of the World Bank Group) IDP internally displaced person ILO International Labour Organization IMF International Monetary Fund IPCC Intergovernmental Panel on Climate Change IRS indoor residual spraying ITN insecticide-treated bed net LDCs least developed countries LED light-emitting diode LIC low-income country LSMS-ISA Living Standards Measurement Study-Integrated Surveys on Agriculture MA market access MDG Millennium Development Goal MIC middle-income country MSMEs micro, small, and medium enterprises MVP Millennium Villages Project NCD noncommunicable disease NEET not in employment, education, or training NFE nonfarm enterprise NGO nongovernmental organization NRA nominal rate of assistance NRP nominal rate of pro­ tection NSDS National Strate­ gies for the Development of Statistics ODA official development assistance OECD Organisation for Economic Co-operation and Development PEF Pandemic Emergency Fund P4P Purchase for Progress PPP pur­chasing power parity PV photovoltaic R&D research and development RFID radio frequency iden­ tification RRA relative rate of assistance SARA semi­autonomous revenue authority SCI Statistical Capacity Indicator SDG Sustainable Development Goal SDI Service Delivery Indicators SEZ special economic zone SIGI Social Institutions and Gender Index SMEs small and medium enter­ prises SMS short mes­ sage service SPEED Statistics on Public Expenditures for Economic Development SPS sanitary and phytosanitary STEM science, technol­ ogy, engineering, and mathematics A b b r e v i a t i o n s   xxi TFP total factor productivity TFR total fertility rate UN United Nations UNCTAD United Nations Conference on Trade and Development UNU-WIDER United Nations University-World Institute for Development Economics Research VAT value added tax VCD value chain development WASH water, sanitation, and hygiene WFP World Food Programme Currencies CFAF CFA franc R South African rand RF Rwanda franc U Sh Uganda shilling Key Messages Poverty in Africa Today and Tomorrow •  Poverty in Africa has fallen substantially—from 54 percent in 1990 to 41 percent in 2015— but the number of poor has increased, from 278 million in 1990 to 413 million in 2015. •  Under a business-as-usual scenario, the poverty rate is expected to decline to 23 percent by 2030, rendering global poverty primarily an African phenomenon. Main Features of African Poverty •  Most of the poor (82 percent) live in rural areas, earning their living primarily in farm- ing. Nonwage microenterprises are the main source of nonagricultural employment and income for the poor and near poor. Strikingly, rural poverty is higher in areas with better agroecological potential. •  Poverty is a mix of chronic and transitory poverty. Fragile and conflict-affected states have notably higher poverty rates. •  Low human capital and high gender inequality impede poverty-reduction efforts. Four Primary Areas for Policy Action •  Accelerate the fertility transition. Rapid population growth and high fertility are features of many countries on the continent. They hold back poverty reduction through multiple channels. Family planning programs will play an important, cost-effective role in acceler- ating the fertility transition, which will complement the effect of increasing female educa- tion, and empowering women (including by offering life skills, addressing social norms around gender, and reducing child marriage). •  Leverage the food system. Raising smallholder agricultural productivity, especially in sta- ple crops, increases the incomes of the poor directly and addresses rising urban demand for higher-value agricultural products. Complementary public investment (in agricultural research and extension, irrigation, and rural infrastructure) remains key. Inclusive value chain development and technological leapfrogging can bring previously unattainable markets and production techniques (such as irrigation and mechanization) within reach of ­ the poor. •  Mitigate fragility. Uninsured risks and conflict entrap people or push them back into pov- erty. Many risk management solutions already exist, with roles for both the private and public sectors, but an important hurdle remains incentivizing the public and private actors to act now, before the shocks and conflict occur. •  Address the poverty financing gap. More, and more efficient, public financing focused on the poor is needed to finance this poverty-reduction policy agenda. In addition to the continued need for official development assistance (ODA), domestic tax compliance and international tax avoidance need to be addressed, as well as making public spending more pro-poor and more efficient. This is especially important in resource-rich countries, where poverty reduction and human development indicators are often relatively worse. 1 Overview Poverty Reduction in Africa: continue to be in double d ­ igits. Slowing economic growth in recent years has also A Global Agenda slowed poverty r ­ eduction. And notably, the Africa’s turnaround over the past couple number of poor in Africa is rising (from of decades has been ­ d ramatic.1 After many 278 million in 1990 to 413 million in years in decline, the continent’s economy 2015), in part because of high population picked up in the mid-1990s, expanding at growth (World Bank 2018c). Africa will a robust annual average of 4.5 percent into not reach the United Nations Sustainable the early 2010­ s . People became healthier Development Goal (SDG) of eradicating and ­b etter nourished, youngsters attended poverty by 2030. 2 schools in much greater numbers, and the Globally, there is a shifting concentration of poverty rate declined from 54 percent in poverty from South Asia to ­ Africa. Forecasts 1990 to 41 percent in 2015 (World Bank suggest that poverty will soon become a pre- 2018c). The region has also benefited from dominantly African ­ phenomenon. The non- decreased conflict (although simmering monetary dimensions of poverty (nutritional in some countries and notwithstanding and health status, literacy, personal security, pressing numbers of displaced persons), an empowerment), while improving, are still the expansion of political and social freedoms, lowest in the world in many countries (Beegle and progress in the legal status of women al. 2016­ et ­ ). The world’s bifurcating demog- (Hallward-Driemeier, Hasan, and Rusu raphy, inequality and climate change, and the 2013; World Bank 2019b). The availability resulting migratory pressures, add further and quality of poverty data to record this global interest to address poverty in ­ A frica. progress have also ­ improved. But the rapid spread of digital ­ technologies D e spit e t h e s e ac c o mpl i s h m e nt s — and solar power and increasing South-South described in detail in the precursor to this trade also provide new opportunities to report, Poverty in a Rising Africa (Beegle tackle this pressing challenge (Dixit, Gill, al. 2016)—the poverty and shared pros- et ­ and Kumar 2018; Gill and Karakülah 2018; perity challenges remain daunting: Poverty World Bank 2019a). How Africa can accel- rates in many African countries are the erate its poverty reduction is now a global highest in the world and are forecast to ­ preoccupation—and the focus of this ­ report. 3 4   Acceler ating Pover t y Reduc tion in Afric a Of course, Africa comprises many coun- well as the slowest poverty reduction, even tries with quite varying poverty rates and long after the conflict e ­ nded. This pattern divergent socioeconomic and agroecologi- emphasizes the debilitating role that con- cal ­c onditions. Half of Africa’s poor live flict plays in improving well-being as well in 5 countries; 10 countries account for as the critical importance of tackling pov- 75 percent of Africa’s ­poor. 3 Yet the poor- erty in fragile states to advance Africa’s est countries, and regions within coun- poverty ­agenda. tries (those with the highest poverty rates), Many African countries depend heavily on are not necessarily the same countries or natural ­resources. Resource dependence has regions housing most of the ­ p oor. This only grown since the commodity boom of the poses a challenge as to where to target the 1990s and 2000s (figure O.1) and is increas- poverty-reduction efforts, at least from a ingly the environment within which Africa’s global ­perspective. poverty reduction must take place. Yet, Fragility and resource abundance are resource dependence often undermines insti- key country features to account for in tutional quality and erodes long-run growth the design of poverty-reduction policies. potential and poverty r ­ eduction. Spending Historically, neglect of regions and coun- on human capital in these countries, and the tries with high poverty rates, even when efficiency of that spending, is ­systematically not densely populated, has often bred con- lower than in non-resource-dependent coun- flict, which easily spreads to the surround- tries (de la Brière et ­ a l. 2017­). In extreme ing ­ a reas. Fragile and conflict-affected cases, resource abundance may even lead to states have notably higher poverty rates as conflict (Collier and Hoffler 2004). FIGURE O.1  Natural resource dependence has increased substantially in most African countries a. Change in share of mineral exports in total exports, 1996–2013 60 Change in share of exports (percentage points) 40 20 0 –20 i c Er nea m a Lib bia Ca n eria an U roo a ga n rín nda Gu ial T ipe in Gu ogo Se -Bi ea he u C es Ghhad g a bo en a Ve ya Es abo e M wa n Courit ni Et or s Cô M iop s te al ia Co C N ’Ivo wi ng on am ire So em, Re a ut . R p. Af . Se ig a Ga Soneg r m m al Si Bot , Tha er sw e Zi Le na ba ne Bu uda e n so di ad Be o M augas in am tan r Ta biq ia Bu R zan e rk wa ia aF a Mo i h ep al e oz ri ca m iu h o Gu bli Za itre e l Ni an Ca K eri i N ic a i in nd G rd S w n u th as yc ssa m go o, go ib bi al a ti Ma n Le run ll d a ea in ra a m o r b c pu A Re P D d n r M ca to ua fri é m lA Eq To ra o nt Sã Ce (Figure continues next page) O v e r v i e w   5 FIGURE O.1  Natural resource dependence has increased substantially in most African countries (continued) b. Change in share of oil and gas exports in total exports, 1996–2013 100 Change in share of exports (percentage points) 50 0 –50 a s ng ab o o, on bo run . Ve di ad Be de er a n Le ar e ut rit a A a d al a So cip i Co ma e Na or a M mi os a so s Ni tho m sw ia Gu ia, Tna in Ug he -B da u w al M i e li l A o Aibe a fri , D ng ria n . la Ta ub . nz lic N ia Et and r te To ia Chan Cô iop a M d’Iv go Zi urit ire Ca ba nia er e Gu oon M ial ha a oz Gu na bi ea Su ue ad Ca Bu Rep p p Pr aw in Rw ige in lle Le ritiu a So E mbi h re an M fric m li au bi L ny r G e Za on m bw Co G Fas Se issa Si ag ni GaBot ger Es neg ca em o an Re Re ra sc in am in at d b a ea an q r a o m a rk he ín K h Bu eyc S M to ra ng ua é m nt Co Eq To o Sã Ce Source: Calculations based on United Nations Conference on Trade and Development (UNCTAD) ­data. Note: There is a close correlation between the export and government revenue shares of natural resources. Data on the latter, although arguably the better indicator of resource dependence, are patchy. Poverty in Africa: Stylized Facts well as effective risk management strategies are both important for poverty ­ reduction. Across countries, poverty manifests itself also They often also interact with each ­other. in many similar ­ways. First, poverty remains Third, about half of Africa’s poor are predominantly rural—82 percent of Africa’s younger than 15 years old, showing the poor are rural—with the poor earning their need for greater attention to reach ­children. living primarily in farming or, when working Measured gender gaps in monetary pov- off the farm, in agriculture-​related activities erty are modest, though the data underpin- (Allen, Heinrigs, and Heo 2018; Beegle et a ­ l. ning these numbers assume equal sharing 2016; Castañeda et ­ ). Although this al. 2018­ in ­households. Numerous other nonmon- does not mean the solution lies automatically etary indicators show large structural gender in agricultural or rural development, it does ­inequalities. indicate a policy entry point—either to rein- Fourth, the poor have weak links to the force the income-earning opportunities of state. They have weak access to good-quality ­ the poor in situ or to help them connect with public goods (infrastructure) and services, income-earning opportunities ­ elsewhere. and they have limited voice in public policy Second, poverty is a mix of chronic and ­making. transitory: about 60 percent of Africa’s poor Moreover, Africa’s poverty rate has not are chronically poor, and 40 percent are in only been higher than in most other low- and poverty. Therefore, asset building transitory ­ middle-income countries; it has also declined and the generation of income opportunities as more ­slowly. 6   Acceler ating Pover t y Reduc tion in Afric a Africa’s Slower Poverty High Fertility, Slow Poverty Reduction Reduction At 2.7 percent per year on average, rapid Three notable factors have contributed to population growth remains a defining fea- Africa’s slower poverty reduction: ture for many countries on the ­ continent. It follows from continuing high fertility (5.1 children per woman in 2010–15 com- •  Persistently high fertility and population pared with 6.7 in 1950–55) despite a rapid ­growth. Although Africa’s gross domestic decline in under-five child mortality (from product (GDP) growth has been robust 307 deaths per thousand in 1950–55 to 91 over the past couple of decades (except in in 2010 –15) (World Bank 2019­ c). High recent years), economic output has grown population growth poses a substantial bur- more slowly in per capita terms than in den on African governments, families, and other low- and middle-income c ­ ountries. especially women through several ­ channels. African countries’ higher fertility and It elevates the fiscal needs for social services, faster population growth have left their which only pay off much ­ later. High ­fertility populations with much lower income per has also been an important direct contribu- ­person. tor to Africa’s explosive urban growth, not •  Poor initial ­ c onditions. Less of Africa’s simply the result of rural-urban migration (rather modest) per capita household (Jedwab, Christiaensen, and Gindelsky income growth has translated into pov- 2017­ ). Rapid urban growth makes it hard erty reduction than in other countries, for urban centers to keep up the infra- simply because of the high initial pov- structure base to remain productive, create erty in the region. The lack of assets and employment, and be an effective force for access to public goods and services, as poverty reduction (Lall, Henderson, and well as the limited availability of good Venables 2017­). income-earning opportunities for a large With rural populations often clustered on share of the population, limit the ability a small share of the arable rural land, high of many to contribute to and participate population growth is further increasing land in economic growth. It is poverty, rather pressures in several African countries, with- than inequality per se, that has been hold- out concomitant agricultural intensification ing back poverty reduction in many Afri- to compensate thus far (Jayne, Chamberlin, countries. When compared with other can ­ and Headey 2014­ ). And, not least, the bur- equally poor countries in other regions, den on women of care and domestic work African countries have not been less effec- increases with more children and reduces tive at converting per capita household their income-earning ­ opportunities. This is income growth into poverty ­ reduction. especially hard on poor women, who often •  The composition of Africa’s g ­ rowth. Afri- begin childbearing at much younger ages and ca’s poverty reduction has been slower also have more children (on average at least because of the composition of Africa’s twice as many [5–7] as women in wealthy growth—in particular, the increasing reli- ­households). ance on natural resources and the modest Fertility reduction, on the other hand, is performance of its agriculture and manu- associated with faster economic growth (the facturing sectors. demographic dividend) and faster poverty reduction. A 1 percent fall in the dependency ­ Accelerating the fertility transition, rate is associated with a 0.75 percentage point addressing key facets of Africa’s poor initial fall in headcount poverty (Cruz and Ahmed conditions, and shifting to a pro-poor growth 2016­ ). Accelerating fertility reduction is and policy agenda will go a long way toward therefore an important entry point for accel- accelerating poverty ­reduction. erating Africa’s poverty ­ reduction. Africa’s O v e r v i e w   7 fertility rate per woman of childbearing age it has remained at two (Günther and Harttgen is, on average, one birth higher than in other 2016), suggesting a large latent demand for least developed countries (LDCs), controlling contraception. Limited provision and poor ­ for conventional demographic and socioeco- implementation of family planning pro- nomic factors (figure ­O.2) (Bongaarts 2017­ ). grams explains much of the delayed decline In addition to female education, much in Africa’s fertility rate (de Silva and Tenreyro greater attention to family planning program- 2017­ ). Other entry points to accelerate the ming is n ­ eeded. Outside Africa the average demographic transition include empower- number of unwanted births per woman of ing women, including providing life skills for childbearing age has decreased from one to women and girls, addressing social gender ­ ecades. In Africa zero over the past couple of d norms, and focusing on child ­ marriage. FIGURE O.2  In Africa, fertility is less responsive to conventional parameters of development than in other LDCs a. TFR by GDP per capita b. TFR by educational attainment 10 8 9 7 8 6 7 Births per woman Births per woman 6 5 5 4 4 3 3 2 2 1 1 0 0 100 1,000 10,000 100,000 0 20 40 60 80 100 Annual GDP per capita (US$) Women with primary education or more (%) c. TFR by urban share of population d. TFR by life expectancy 10 10 9 9 8 8 7 7 Births per woman Births per woman 6 6 5 5 4 4 3 3 2 2 1 1 0 0 0 50 100 40 60 80 100 Urban share of population (%) Life expectancy (years) Africa Other LDCs Source: World Bank calculations, adapted from Bongaarts (2017), using latest data from the World Development Indicators 2019 d ­ atabase. Note: LDCs = least developed countries (as defined by the United Nations Committee for Development Policy); TFR = total fertility rate (total number of children born to a woman in her lifetime). Data used are for the years 1990 and 2018. Last available year chosen when data were ­missing. 8   Acceler ating Pover t y Reduc tion in Afric a Poor Initial Conditions exercise self-control, and aspire—behaviors associated with escaping poverty (Haushofer Poor initial conditions also hold Africa back and Fehr 2014; World Bank 2015). in addressing ­ p overty. These include not Gender inequality also drives poorer eco- only the low levels of human capital and nomic growth outcomes by reducing total access to infrastructure but also the more productivity—in addition to its influ- factor ­ deep-seated structural impediments such ence on gender gaps in education, employ- as natural resource dependence (discussed ment, and governance (Ferrant and Kolev earlier), gender inequality, and social redis- 2016). This is particularly the case in low- tributive ­pressures. income ­ countries. Dismantling gender-based At the individual level, poor educational discrimination in social institutions could attainment reduces the prospect of escap- increase global growth by as much as 0.6 per- ing ­poverty.4 Where the gap in educational centage points per year over the next 15 years attainment is large, as in much of Africa, (Branisa, Klasen, and Ziegler 2009, 2013, much growth and poverty reduction can 2014; Yoon and Klasen 2018). Reducing already be expected from widespread, qual- gender gaps would also raise the growth ­ ity basic education (box ­ O.1­). A severe lack prospects of African economies—and hence of infrastructure exacerbates things. The low also reduce poverty (box ­O.2­). returns to the poor’s land, labor, and skills Finally, with poverty widespread, shocks arise partly also from their inability to access frequent, and insurance absent, people often and afford information and communication hold back from investing for fear of redistrib- technology, energy, and transport services utive consequences (Platteau 2014). (Christiaensen, Demery, and Paternostro 2003; Grimm et ­ al. 2017; James 2016­ ). More recent insights on the psychology of poverty More and Better Jobs for the Poor further show how the lack of human capital, physical assets, and access to basic infrastruc- Finally, the scope and need for pro-poor ture not only reduce the earning capacity of growth policies to accelerate poverty reduc- the poor but also tax their mental “band- large. Although Africa will tion in Africa is ­ width” and undermine their ability to plan, not be able to eradicate poverty by 2030, BOX O.1  Investments in human capital are critical to alleviate poverty Human capital investments yield substantial long-run Raising human capital in Africa is a pressing benefits and are critical in the agenda to reduce pov- poorest. Children in poor issue, and more so for the ­ erty in A ­ frica. A range of evidence shows that children households have worse childhood outcomes across who have a disadvantaged start in life face a greater many dimensions of w ­ ell-being. The scale of under- lifelong risk of being trapped in p ­ overty. A human nutrition in Africa is staggering, with children in development trap initiates a cycle of poverty that runs poor households having much higher rates (World across generations and traps families in poverty (for Bank 2018­ b). And poor children (and poor parents) example, low education and poor health result in low in Africa have starkly unequal access to critical adult income, poor human development for children, services that influence children’s h­ ealth. Although and so on) (Bhalotra and Rawlings 2013; Bhutta et ­ al. universal education access has greatly shrunk the 2013; Victora et a ). Because the economic ­ l. 2008­ enrollment gap between poor and n ­ onpoor chil- benefits of public investments in human development dren at least at the primary level, poor children are realized far into the future (a decade or longer), are learning much less than their peers in nonpoor they may lack appeal to governments, given the many households (World Bank 2018­ d). immediate demands on public ­ finances. O v e r v i e w   9 BOX O.2  Gender inequality is a hurdle to poverty reduction in Africa African women continue to encounter disadvantages g ender gaps include the ­ from closing ­ f ollowing in education, health, empowerment, and income- (Klasen 2006): generating ­ a ctivities. They tend to have signifi- cantly lower human capital endowments than men •  A growth strategy that raises the demand for (although, among the youngest cohort, this gap female labor (such as the export-led growth strat- has narrowed, with girls having caught up to boys egies of East Asia) in some countries); worse access to labor markets; •  Addressing gender gaps in education, especially lower wages; more limited access or title to produc- in poorer households where school enrollment tive assets (such as land, credit, and other inputs); rates tend to be much lower than in the rest of the fewer political and legal rights; and more stringent population constraints on mobility and socially acceptable •  Actions to improve women’s access to productive activities. As a result, gender inequality can trap ­ assets—more secure property rights and access women in poverty and generate a vicious cycle for to land as well as better access to credit, modern their ­children. inputs, and other means of production (including Beyond the intrinsic value of equal opportuni- land) ties, gender equality will bring with it economic •  Policies that help poorer couples reduce their growth and greater poverty reduction for c ­ ountries. ­fertility. Four entry points to reap the economic returns the poverty projections show that 50 ­ million the structural, spatial, and institutional more people could be lifted out of poverty by transformations needed to raise the incomes then if the incomes of the poor were to grow of the poor and ­ v ulnerable, in particular, 2 percentage points faster annually (while on sectoral and subsectoral policies and keeping constant each country’s historical i nvestments—on agriculture, on off-farm ­ per capita annual growth rate over the past employment, and on managing risk and 15 years) (Cattaneo 2017­ ). Combined with conflict—to broker these transformations. ­ lower population growth and addressing What these are is far from obvious, because poor initial conditions, pro-poor growth— just as not all growth policies are equally growth whereby the incomes of the poor also poverty reducing, neither are all agricultural grow substantially as the economy devel- growth or urbanization models equally good ops—will go a long way in accelerating pov- for the poor (Christiaensen and Kanbur erty reduction now and in the ­ future. ­ l. 2012; Dorosh and Thurlow 2017; Diao et a A pro-poor policy agenda requires get- 2018; Pauw and Thurlow 2011­ ). ting the growth fundamentals right as well as increasing growth where the poor work and live (so that they can contribute and ben- Growth Fundamentals and efit directly), while addressing the many risks to which households are ­ exposed. With the Poverty Financing scope for redistribution to solve Africa’s pov- Macroeconomic stability, regional integration erty limited in most countries, the focus is and trade facilitation as well as a conducive squarely on the productivity and livelihoods business environment are fundamental for of the poor and ­ vulnerable—that is, what it economic g ­ rowth (Bah and Fang 2015; Sakyi will take to increase their e­ arnings. As such, et al. 2017). They also affect poverty (Antoine, this report views its task through a “jobs” Singh, and Wacker 2017; Dollar and Kraay lens. This naturally focuses the report on ­ 2002; Le Goff and Singh 2014; Rodrik 1998­ ). 10   Acceler ating Pover t y Reduc tion in Afric a Particularly, three macroeconomic indicators Earning More on the Farm have emerged as statistically important in the cross-country growth regressions: Leveraging Africa’s food system, on and off the ­farm, is key to bringing poverty down •  The rate of price inflation, reflecting mon- and raising living standards. Agriculture etary policy has historically proven to be particularly •  The exchange rate, reflecting openness to poverty reducing, especially at low income trade and other trade policies levels (Christiaensen and Martin 2018­ ). •  The level of government consumption Rapid urbanization and income growth add expenditure, or the size of the fiscal defi- opportunities for agribusiness development cit, reflecting fiscal ­ policy. and employment generation in agriculture’s value chains, off the ­farm. But not all agri- When these indicators deteriorate, pov- cultural growth is equally poverty reducing, erty is likely to rise (Antoine, Singh, and with smallholder staple crop productivity Wacker 2017; Christiaensen, Demery, and and livestock development continuing to Paternostro 2003; Dollar and Kraay 2002; demand particular attention for poverty Rodrik 2016­). ­ r eduction. More integrated approaches The evolution of inflation and exchange are needed, leveraging the private sector rates in Africa has been mostly ­ favorable. through value chain d ­ evelopment. But pub- Yet, rapidly rising fiscal deficits in many lic investment focused on the provision of countries pose c ­ oncern. Gross govern- public goods (for example, irrigation) and ment debt in Africa increased from about services (for example, extension) remains 32 percent of GDP in 2012 to 56 percent equally vital, especially to boost smallholder of GDP in 2016. Fourteen countries were staple crop and livestock ­ productivity. considered at high risk of debt distress at the end of 2017, compared with seven in 2012 (World Bank 2018a). Looking at Favorable Conditions for Leveraging debt dynamics—the growing difference the Food System between real interest and growth rates, and The conditions for leveraging the food sys- widening primary deficits—adds further tem for poverty reduction in Africa today urgency to reining in public debt (Gill and are particularly favorable: Karakülah 2018­). In addition to implementing the pol- •  Food demand is robust, though mainly icy frameworks needed to broker pro- driven by population ­growth. poor growth, financing the accompanying •  World food prices are still about 70 per- p overty-reducing investments—many of ­ cent higher than before the 2008 world which only pay off over time, such as human terms). food crisis (40 percent in real ­ capital—within a tightening fiscal space, •  Urbanization and income growth add is the other important challenge to t ­ackle. opportunities for product differentiation More resource mobilization is needed as well and value addition, and thus for off-farm as more, and more efficient, spending on employment opportunities in ­ agribusiness. areas important for the poor, such as health, •  The domestic agricultural policy and trade education, agriculture (for example, exten- environment (including intraregional) have infrastructure. sion and irrigation), and rural ­ ­improved. Here there is a considerable role for making •  Political leadership remains largely maximum use of leapfrogging technologies ­supportive. to bring hitherto inaccessible (and tradition- ally expensive) communication, energy, and Against this background, supply has also transport services within the reach of the r ­ esponded. But not enough, and Africa’s poor (box ­O.3­). food import bill has still risen steeply, O v e r v i e w   11 BOX O.3  Leapfrogging technology holds promise for poverty reduction in Africa Most of the poor in rural areas (and to a lesser African rural towns and households might simi- extent in urban areas) remain deprived of access larly leapfrog straight to cheap renewable electric- to a f ford able a nd rel iable i n for m at ion a nd ity provided by solar panels and minigrids based on communication, energy, and transport infrastruc- ­ shared solar photovoltaic (PV) systems and direct ture (and services). Without these, it is hard to current (DC) distribution ­ l ines. Tanzania has been access markets and public services, increase produc- a front-runner in the rollout of microgrid electrifica- tivity, and raise income in either farm or off-farm tion programs; other countries have started to fol- activities. By reducing fixed costs and thus the tra- ­ low suit (including Kenya, N ­ igeria, Rwanda, and ditional economies of scale in infrastructure provi- ­Uganda). sion, technology is helping Africa address this ­ gap. The poor can benefit from these leapfrogging Prepayment and per unit payment business models, technologies directly, as adopters, through greater facilitated by mobile-phone technology, are further access to productivity-enhancing capital goods bringing services within the reach of the ­poor. This (for example, solar power) as well as better mar- holds great promise for poverty ­ reduction. ket access to buy and sell their goods and ­ services. Perhaps the most dramatic of these technologi- But, more often than not, they mainly benefit indi- cal changes has been in telecommunication services, rectly, through the wider and cheaper availabil- with 73 percent of Africa’s population now having ity of goods and services following adoption by a mobile-phone subscription (World Bank 2018a). ­others. And the trend is not just about phone ­ c alls. The Importantly, however, these technologies will development of the M-Pesa mobile money applica- deliver on the promise of accelerating poverty reduc- tion in Kenya (“M” for mobile, “pesa” for “money” tion only when deliberate complementary public in Swahili) put a rudimentary “bank account” in policies are taken in three areas: (a) the removal of everyone’s ­pocket. And Hello Tractor in Nigeria, an barriers to the technologies’ adaptation and diffu- app for renting tractors, reduces search and match- sion to rural areas where the poor live and work; ing costs, bringing the economies of scale of high- (b) investment in skill formation (foundational as productivity, lumpy capital goods within the reach well as digital); and (c) the creation of an appro- of smallholders (Jones 2018­ ). The next frontier is priate enabling ecosystem to run and maintain the widespread penetration of high-speed ­ internet. technologies. by US$30 billion over the past 20 years Yet, the expected climatic changes are not f igure ­ (­ ). Many of these imports could O.3­ unequivocally ­ d etrimental. Maize yields, be competitively produced ­ d omestically. for example, are predicted to increase in the Output growth in cassava and maize, and Sahel and many parts of eastern and central partly also in rice, including through yield Africa (Jalloh et ­ a l. 2013; Waithaka et ­a l. growth, confirm the potential for a more 2013­ ). And agriculture also plays an impor- robust supply ­response. Africa’s rising food tant role in the prevention of conflict—which import bill poses a burden on the exter- often finds its origins in climate-related agri- nal balances and signifies an important cultural shocks—as well as in the recovery missed ­ opportunity. This holds even more of fragile states (Martin-Shields and Stojetz in Africa’s oil-rich countries, where public 2019­ ). A climate-resilient and remunerative investment in agriculture is lower and poul- agriculture provides a viable alternative to try imports are ­higher. illicit and mercenary activities for individuals Climate change and resurging conflict who otherwise see a low opportunity cost to pose challenges to reap these ­opportunities. participating in ­conflict. 12   Acceler ating Pover t y Reduc tion in Afric a FIGURE O.3  Africa’s food import bill has tripled since the mid-2000s 50 45 40 35 US$ (billions) 30 25 20 15 10 5 0 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 19 19 19 19 19 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Food imports (excluding fish) Agricultural exports Source: FAOSTAT 2018 database, Food and Agriculture Organization (FAO), ­http://www.fao.org/faostat/. Most important, brokering the supply smallholders (Otte et ­ ). Development al. 2012­ response will require sustained political of Africa’s agricultural exports (old and new) ­ attention. The recent decline in the agricul- complements the staple crop a ­ genda. It also tural share of total spending to pre-2008 lev- does not have to compete with public invest- els, despite declared political commitment, ment in staples, because private sector inter- will need to be reversed. ests can be ­ leveraged. The challenge is to balance policy ­attention. Larger poverty-reducing effects come fur- Not All Agricultural Growth Is Equally ther from supporting slightly larger, commer- Poverty Reducing cially oriented smallholders, with the poorest Raising smallholder staple crop produc- and least productive farmers in the village tivity (the so-called Green Revolution) (often also those with less land) benefiting demands particular ­ attention. 5 Low labor primarily through lower food prices and the productivity in staple crops still locks many local labor markets (in and outside agricul- people into staple crop ­ agriculture. Because al. 2010; Mellor 2017). ture) (Hazell et ­ of this, as well as more widespread income Poorer farmers may further benefit from (including via the price channel) and link- better access to technology and inputs as age effects, raising staple crop productivity well as ­markets. Such positive spillovers are has larger growth multipliers and greater less likely however when farms become large poverty-to-growth elasticities than an equal (more than 100 hectares) or even of medium amount of productivity growth in cash scale (more than 10 ­ hectares). These enti- crops (Diao et ­al. 2012­). ties tend to use less agricultural wage labor Unfortunately, staple crops attract less and yield smaller local consumption linkages public and private sector attention than cash for the poor (that is, more of the revenues crops, as does smallholder livestock holding, are spent on urban [and imported] goods which is the second income source for many and services) (Chamberlin and Jayne 2017; O v e r v i e w   13 Deininger and Xia 2016, 2018; Pauw and illustrative. The experience of Ethiopia is ­ Thurlow 2011). The government simultaneously and sustain- Larger (“estate”) farm entities may how- ably focused on ever be needed for certain crops, to ensure consistent volumes of high-quality crops •  Increasing smallholder staple crop pro- in compliance with standards to access the ductivity by deploying 45,000 extension more-demanding export ­ markets. Examples agents (three per district), facilitating include labor-intensive exports of high-value access to credit, and improving water and fruits and vegetables, flowers, and f ­ish. land management; Less clear is the necessity of such an agrar- •  Improving market connectivity through ian structure to supply the domestic urban rural road investment; and ­markets. •  Providing a form of insurance through the Productive Safety Net Program, one of the largest social protection programs An Integrated Approach Is Needed in ­A frica. So, what are the entry points to raise Africa’s agricultural labor ­ p roductivity? Since the mid-1990s, smallholder cereal A myriad of input, factor, and prod- yields in Ethiopia have more than doubled; uct market constraints hold agricultural extreme poverty has more than ­ halved. intensification back, with pockets of land Ev idenc e f rom de t a i led m ic ro e co - scarcity emerging and the seasonality of nom ic st ud ies suppor ts the ex istence agricultural labor calendars too often of ­i mportant synergies from integrated ­ ignored. The latter often leads to u ­ nderuse ­ a gricultural interventions (A mbler, de of agricultural labor and the perception Brauw, and Godlonton 2018; Daidone agriculture as an intrinsically less pro- of ­ a l. 2017; Pace et ­ et ­ a l. 2018). Yet, success ductive ­activity. This only holds, however, of an integrated approach is not ­ a ssured. when agricultural labor productivity is With integration comes complexity, which expressed as agricultural output per worker, challenges effective implementation, espe- not when expressed per hour of work cially in low-capacity, poor-governance (McCullough 2017­). ­e nvironments. Mechanization and better water man- agement can ­ h elp. Less than 2 percent of Inclusive Value Chain Development, the cultivated area and less than 5 per- but Also Public Goods cent of households in six African coun- tries (which together cover 40 percent of Value chain development (VCD), often Africa’s population) use any form of water facilitated by external agents such as gov- control (Sheahan and Barrett 2014­ ). Small- ernments as well as nongovernmental and scale, simple, affordable, self-managed international organizations, increasingly irrigation systems that are rolled out at emerges as a market-based, institutional scale hold hope if access to complemen- solution to simultaneously address the tary inputs and markets are developed multiple market constraints (Swinnen and ­simultaneously. Kuijpers 2017­ ). Smallholder farmers can be Yet, too often, singularly focused interven- linked to higher-value domestic and export tions are pursued, or interventions are poorly markets by (a) supplying raw agricultural ­ c oordinated. Africa’s Green Revolution, products (gains stemming from reduced mechanization, and irrigation efforts each production and price risk, higher premium need an integrated approach that simulta- prices, and access to previously unattain- neously addresses supply- and demand-side able input and output markets and agro- poverty. constraints to tackle ­ nomic knowledge); or (b) indirectly through 14   Acceler ating Pover t y Reduc tion in Afric a employment ­ opportunities. Buyers gain by goods and services ­ g rows. About a third of securing a consistent volume of high-quality this employment will still be linked to agri- crops as well as the standards compliance culture, up and down the value chain, in needed to access these m ­ arkets. The poorest agricultural input production and provision often benefit through localized s ­ pillovers. as well as food processing, marketing, and Horizontal coordination of smallholder services (Allen, Heinrigs, and Heo 2018; farmers is often important to make value al. 2015­ Tschirley et ­ ). chains more ­ i nclusive. It reduces the trans- Over the short to medium term, for many action costs of involving small farmers and of Africa’s poor, moving to work opportuni- can increase bargaining power and thus ties off the farm will largely mean moving their share of the value ­ added. into informal household enterprises (typically Although VCD holds promise for tra- with no hired workers) but unlikely into wage ditional and new cash crops as well as for ­ employment (be it formal or informal wage livestock and livestock products, contract work). Even in countries where wage employ- enforcement is inherently more difficult in ment is growing fast (for example, through staple marketing because of the risk of either increasingly challenged, labor-­ i ntensive (opportunistic) side-selling by smallhold- exports), the low base of wage employment ers or strategic contract breach by buyers and the pace at which youth enter the labor (Swinnen, Vandeplas, and Maertens 2010).6 force imply that wage employment will Experimentation with VCD for staples has absorb only a small share of the job seekers begun, however, along with the growing over the coming 10–15 ­ years. demand for consistent volumes and quality Only a few household enterprises fall as well as opportunities for value addition into the category of “opportunity” entrepre- in Africa’s domestic staple markets (rice and neurship, “constrained gazelles,” or “trans- teff for urban markets, feedstock maize for formational” ­ entrepreneurs. Nonetheless, livestock, barley for beer)—a space to be household enterprises are an important part ­watched. of the broader economic transition—and a Nonetheless, to raise smallholder sta- particularly important one at that for pov- ple crop productivity, the need for public erty r­ eduction. They typically have low good provision remains ­ undiminished. This productivity, remain small and informal requires increased public spending in agricul- throughout their life cycle, are managed and ture, which has started to falter, as well as operated by household members, and only a a shift in its composition away from private few create paid jobs for nonhousehold work- (input subsidies) to public goods, including ers (Nagler and Naudé 2017­ ). (a) agricultural research and development These enterprises are often started from (R&D) and extension for both staples and necessity. The lack of wage jobs and the ­ livestock, and (b) investment in irrigation and absence of formal unemployment insurance rural ­infrastructure. The latter also benefits push people to jump-start self-­ employment the broader rural ­ economy, and new technol- as a survival s ­trategy. Therein also lies ogies hold promise. their strength for the ­ poor. They are read- ily available, and with little skills and capi- Moving Off the Farm: Household tal required, easy to enter and exit, and often critical in complementing the income, Enterprises thus helping households cope and smooth In addition to raising incomes on the farm, ­ consumption. They are often also an impor- employment opportunities off the farm will tant source of cash for financing modern become increasingly important as agricul- input purchases and thus for developing tural productivity and incomes rise, coun- other activities (Adjognon, Liverpool-Tasie, tries urbanize, and the demand for nonfood and Reardon 2017­ ). O v e r v i e w   15 The importance of the informal or semi- To reach the poorest and most vulnerable, formal nonfarm sector as a provider of jobs an emerging and promising approach is to and livelihoods for Africa’s burgeoning combine safety net interventions with pack- labor force means it cannot be neglected by ages of support (including skills, finance, ­ policy. The choice of focusing on the formal advisory services, working space, and so on) or informal sector or on small and medium to facilitate entry into self-employment and enterprises (SMEs) and large firms or house- raise the labor earnings of social protection hold enterprises is, however, not simply beneficiaries (Banerjee et ­ ). These a l. 2015­ an “either-or” ­proposition. Investments in combined “protection and promotion” inter- human capital, infrastructure, and a trans- ventions are currently being implemented parent regulatory framework will benefit the on a large scale in several African countries, spectrum of ­ enterprises. But not all invest- with ongoing impact evaluations examining ments cut across, and investments can also their ­effects. be made that more directly benefit nonfarm Much remains to be learned, including businesses run by poor ­households. with respect to agricultural value chains linking SMEs with ­ m icroenterprises. Few More Profitable Household studies have focused specifically on poor or Enterprises for the Poor near-poor households, which may face dif- ferent constraints than vocational or trans- Because most household enterprises do formational entrepreneurs or may lack any not grow, they mainly create employment ambition to grow their businesses in the first through ­ e ntry. Available evidence sug- ­ place. In addition, most studies have focused gests that job creation through entry can on urban settings, though most of Africa’s be achieved by relatively small amounts areas. poor live in rural ­ of financing, which can be combined with skills training, though the addition of train- ing tends to make the interventions less Fostering Demand: The Roles of cost-effective. As in agriculture, stand-alone ­ Towns, Regional Trading, and Digital interventions addressing one single con- Technology straint (such as skills or finance) tend to be less successful than interventions that target Most interventions targeting the entry or multiple constraints at the same time, high- growth of household enterprises focus on lighting the importance of packaging differ- alleviating the supply-side constraints (such ent interventions in ­ one. skills). Although these supply- as finance or ­ In many African countries, access to side interventions can help entry into self- finance is difficult, especially for youth from employment and, to some extent, increase less well-off families without ­ c ollateral. earnings, the survival and growth of these Although several countries have attempted small enterprises is ultimately determined by to improve access to finance, especially for the demand for the goods and services they the politically sensitive demographic segment ­ provide. Household enterprises are rarely of unemployed youth, financing modali- a source of job creation beyond the house- ties have not always been flexible enough to hold members, but data show that those make a big impact (entailing short repayment better connected to markets (in urban areas periods without grace periods, high inter- and towns) and owned by a better-educated est rates, requirements to borrow in groups, person nevertheless appear to have the abil- and so ­ on). Creating jobs by facilitating entry ity to grow and hire workers (Nagler 2017; of household enterprises will require the Nagler and Naudé 2017­ ). design of flexible and affordable financing From this perspective, Africa’s ongo- mechanisms as part of a broader enabling ing urbanization and the increasing edu- ­environment. cation level of its youth could increase the 16   Acceler ating Pover t y Reduc tion in Afric a potential for job creation in future household Integrating household enterprises or the enterprises. In rural areas, improving con- ­ informal economy in general into urban or nectivity with nearby markets and towns has national development plans would be a start the potential to improve earnings and spur toward leveraging their p ­ otential. It would welfare-enhancing ­ d iversification. Such an provide a framework for the government and improvement entails not only investment in the informal sector to start discussing the rural infrastructure but also policies to foster design of supportive policies that facilitate better transport ­ services.7 the operation of household enterprises while Critical within this agenda is how govern- still protecting the public ­interest. ments manage their urban ­ spaces. Not all The demand for the poor’s goods and urban development has shown equal poverty- services often also finds itself just across reducing ­ potential. Cross-country research the ­border. This is vividly illustrated by the and case country evidence from India, concentration of (agriprocessing) enterprises Mexico, and Tanzania suggest that, for pov- along the eastern and northern borders of erty reduction, growing towns m ­ atters more Zambia, catering to Lilongwe in Malawi than growing cities (Berdegué and Soloaga and Lubumbashi in the Democratic Republic 2018; Christiaensen, De Weerdt, and Kanbur of Congo, ­ respectively. Cross-border trade 2019; Christiaensen and Todo 2014; Gibson is often also an important driver of town et ­al. 2017­).8 Secondary towns in rural areas development (the so-called border towns) provide local centers of economic activity and (Eberhard-Ruiz and Moradi 2018­ ). demand and are more accessible to the poor Finally, digital technology holds promise because of their proximity and the lower to connect the enterprises of the poor with threshold for migration (Rondinelli and expanding urban and foreign demand for Ruddle 1983­ ). This accessibility facilitates goods and ­ s ervices. Recent evidence from especially the first move, which is often the China shows the potential: e-commerce most difficult (Ingelaere et ­ a l. 2018), and penetration (typically clustered in so-called their proximity makes it ­ e asier to return Taobao villages) is associated with higher home, when things fail, which is especially consumption growth, with the effects stron- important in the absence of formal safety ger for the rural sample, inland regions, and nets. The type of employment available in ­ poorer households (Luo, Wang, and Zhang towns (unskilled and semiskilled) also tends 2019­ ). Capitalizing on this trend will require to be more compatible with the skill sets of equipping youth from poor households the ­poor. with at least basic education and digital Public investments to help rural towns skills while also making internet connectiv- grow can increase demand for agricultural ity affordable, reliable, and widely available products produced in surrounding rural (see box ­ O.3­earlier in this overview). areas, thus increasing rural incomes, which in turn would increase demand for the nonfarm goods and services produced by Managing Risks and Conflict household ­ enterprises. Unfortunately, more Risk and conflict are higher in Africa than often than not, governments view house- in other ­ r egions and exacerbate poverty hold enterprises, which are mostly informal, challenges. Civil war is prevalent; the domi- as a detriment to urban spaces rather than nant livelihood, rainfed agriculture, is risky; as a critical source of income for the poor markets are poorly integrated, making and many nonpoor, especially in the larger prices volatile; and health, water, and sani- urban ­ centers. For example, efforts to “sani- tation systems are ­weak. Price, weather, and tize” city centers may well lead to impover- health shocks have large impacts on welfare, ishment of vulnerable workers who depend especially given the inadequacy of financial on dense foot traffic for their livelihoods markets, social protection, and humanitar- (Resnick 2017­). ian systems, as well as the continued reliance O v e r v i e w   17 on costly coping ­ mechanisms. Conflict has and support for disaffected youth and ex-­ far-reaching consequences, including forced combatants could help reduce the risk of displacement and migration of those able to conflict (Blattman and Annan 2016­). More ­migrate. needed. evidence is ­ The direct impact of a calamity on well- being is the visible, headline-grabbing way Better Insurance for the Poor that conflict or poorly managed disasters set back ­ p rogress. However, the persis- When prevention is not possible, a mix of tent impact of uninsured risk on household safety nets and financial instruments can behavior every year—regardless of whether help households manage in the aftermath of the feared event occurs—is arguably the a­shock. Both are needed to manage ­ shocks. larger constraint to accelerating poverty Savings and regular safety net transfers help reduction in ­Africa. Poor households choose households manage small shocks, while safer, less remunerative activities that limit larger shocks are better managed by insur- income growth and poverty ­ reduction. ance or by scaling up safety net ­ support. Better-off households are more likely than poorer households to rely on financial mar- Addressing Risk and Conflict through kets to manage risk, but poor households Prevention still need access to financial markets to help Much can be done to reduce risks and to them manage smaller shocks and to enable help households manage risks ex ­ post. The them to secure more “insurance” than could most prevalent shocks in Africa—relating be provided through public safety nets to price, weather, health, and conflict—are ­alone. slow in onset; affect incomes more than Public finances spent on insurance sub- assets; and tend to be covariate, affecting sidies and shock-responsive safety nets may many households in the same area at o ­ nce. target different households or different risks Risk is higher in poorer areas and in rural and may substitute for each other depending ­ a reas. The prevalence of different shocks on the relative strength of public delivery and varies across the continent (map ­ O.1­ ). private markets in the local ­ context. During In many cases, the cost of prevention is conflict, financial market development that lower than the cost of managing the ­ event. reduces the cost of sending and receiving Development of markets is the best way to remittances can also help, because private reduce price risk in Africa, and this requires transfers and migration are predominant addressing tariff policies as well as investing coping ­strategies. in infrastructure and transport ­ services. To However, financial markets are often reduce health risks and improve child health, weak, and safety net investments are too improving water, sanitation, and hygiene often made after shocks o ­ ccur. Moreover, (WASH); fighting malaria; and achieving countries continue to rely on ex post humani- mass immunizations are ­ key. And targeted tarian aid to help households, which by its investments in irrigation, natural resource nature is neither timely nor ­ p redictable. management, and improved seeds can reduce Reforming humanitarian financing—from exposure to weather ­ risks. In general, there is reducing reliance on ex post appeals to using underinvestment in these cost-effective risk- ex ante financing instruments with predict- reducing ­interventions. able and timely payout mechanisms (like As for conflict, a discussion on address- the World Bank’s Pandemic Emergency ing the sources of fragility that under- Financing Facility)—is ­ essential. But it will lie specific conflicts in Africa is beyond not improve support to households on the the scope of this report, but some emerg- ground unless it is combined with invest- ing evidence has highlighted that well- ments in contingency planning for support targeted aid focused around job creation service ­delivery. 18   Acceler ating Pover t y Reduc tion in Afric a MAP O.1  Some parts of Africa are hit harder by risk ­ https://map.ox.ac.uk/); panel e: World Development Indicators d Sources: Panels a–c: Fisker and Hill 2018; panel d: the Malaria Atlas Project ( ­ atabase, maternal mortality ratio. Note: Panel c: A drought year is defined as a year in which at least half the growing period months are recorded to have a predicted greenness anomaly value below the 10th ­percentile of predicted greenness. Panel d: Each 5 km2 pixel on the map shows the predicted Plasmodium falciparum (Pf) prevalence rate as a proportion of all children ages 2–10. The Time to Act Is Now disasters using humanitarian aid is much cheaper (that is, free) than predisaster invest- Addressing risk and conflict—through ments in prevention and ­ p reparedness. either risk reduction or risk management— Building capacity within governments to requires action before shocks ­ o ccur. There invest in risk reduction and risk management is room for more technological innovation ­ ecessary. is also n and ­better information systems, but funda- For individuals, this will require induc- mentally encouraging action before shocks ing households to overcome behavior that occur will require addressing the incentives limits household investment in risk reduc- currently keep postponing action until that ­ tion and management: a scarcity-induced after shocks ­occur. focus on the present, resignation, and For governments, this requires addressing ambiguity ­ aversion. This can be done by the perverse political incentives that reward reducing the cost to households of investing them for big postdisaster gestures rather than in risk reduction and management while for planning for a rainy ­ d ay. Coping with households learn about new strategies to O v e r v i e w   19 reduce or manage r ­ isk. In addition, there is development progress (Gaspar, Jaramillo, a need to expand mandates and regulations and Wingender 2016­ ). The Organisation for to address adverse selection in health insur- Economic Co-operation and Development ance markets, to increase trust in financial (OECD) average in 2015, for comparison, was institutions, and to reduce fixed-cost insur- 34.3 percent (OECD 2017­ ). ance ­markets. W hile low on average, the level of And finally, as with many aspects of revenue collection in Africa has shown improving policies and programs, there is ­ i mprovement. The region experienced the a data ­ agenda. Better data on disasters as largest increase in tax revenue across the they unfold and on ex ante risk exposure globe since the turn of the century, albeit will help improve financial market devel- starting from a very low point (IMF 2015­ ). opment and the design of shock-responsive But IMF projections find that the countries safety ­nets. with the lowest domestic resource mobili- zation levels are also expected to grow at lower rates, further widening the ­ g ap. To Mobilizing Resources for turn this around, countries need to continue the Poor to improve tax compliance; start focusing more on local large taxpayers, corporate The agenda to address poverty in Africa taxes, and transfer (mis)pricing (which has extends beyond shifting programs and a global agenda); and expand excise and p olicies. It will also require a careful ­ property tax ­ collection. revisiting of a range of domestic revenue Some countries in Africa also gener- and spending ­ patterns. Within the region, ate substantial revenues from natural some countries have the means to address r esources. Out of 37 countries for which ­ the poverty gap (the income needed for a data are available, 22 are considered poor household to just escape poverty), resource-rich—from oil-rich countries be it through theoretical tax rates on the like Chad and the Democratic Republic of nonpoor or through transfers of natural Congo to those with lucrative mining oper- resource revenues directly to citizens, such ations such as Botswana (diamonds) and as through “direct dividend payments” Mauritania or Niger ( ­minerals). In these ­(DDPs). countries, revenues make up 10–20 percent For most African countries, however, of ­GDP. Low- and middle-income countries closing the poverty gap (as a theoretical with substantial natural resources also tend exercise) would mean implausibly high tax to have higher tax revenues than countries rates on the rich or implausible natural at the same income level that lack such resource ­ revenues. Current domestic rev- ­resources. enues are not enough to tackle poverty in Therefore, in principle, resource reve- the short term, let alone to improve Africa’s nues can enhance spending on agriculture, poor initial conditions in human capital— rural infrastructure, and social sectors (for investments that only pay off a genera- example, health and education as well as tion ­later. What is the path to tackle these social protection programs) and thus con- challenges? tribute to poverty e ­ radication. These rev- enues notwithstanding, poverty reduction is slower and multiple human develop- The Domestic Revenue Imperative ment indicators are worse in resource-rich Several low-income African countries have tax countries in Africa than in other African revenues relative to GDP of under 13 percent countries at the same income level—so this (that is, revenues net of grants), which is often revenue is not resulting in greater pro-poor considered the “tipping point” necessary to spending (Beegle et ­ a l. 2016; de la Brière execute basic state functions and to sustain et ­al. 2017­). 20   Acceler ating Pover t y Reduc tion in Afric a Making Public Spending Go Further for Finally, combining the insights on taxa- the Poor tion and spending practices, it emerges that many in the bottom 40 percent of income are Turning from raising more money toward often net taxpayers instead of net ­recipients. spending more effectively and with a pro- That is, in the aggregate, the total cash ben- poor focus, there is a large unfinished efit transferred to the poorest 40 percent of agenda. A key area to make public spend- ­ the population through subsidies and direct ing more pro-poor is to address high subsidy transfers is smaller in absolute magnitude expenditures (particularly fuel, energy, and than the burden created by direct and indi- fertilizer subsidies), which are often regres- rect tax instruments (de la Fuente, Jellema, sive with little impact on ­ poverty. The lack and Lustig 2018­ ). Although these calculations of impact from agricultural input subsidies refer only to the cash-based financial position gets magnified when they crowd out other purchasing power of individuals—excluding investments in the sector that could raise the value of in-kind benefits like education, p roductivity. Cash transfers seem more ­ health, or infrastructure services—they give effective and efficient than subsidies where cause for ­ pause. evidence exists (Dabalen et ­ a l. 2017­). But To accelerate poverty reduction in Africa, more research is needed to compare their a careful reexamination of its fiscal systems performance relative to other competing from a pro-poor perspective is ­needed. It also needs like spending on education, health, requires a better understanding of the politi- WASH, public goods in agriculture (such cal dynamics of pro-poor policy ­ making. as research and irrigation), rural infrastruc- ture, and ­ security. An Important Role Remains for Official Spending patterns from a “pro-poor” per- Development Assistance spective have a mixed track record—with some sectors generally reaching international Taken together, the low base on which to expenditure targets (like education) but oth- tax, the low capacity to tax more, and the ers falling short for many countries (health, political inability (or lack of will) to channel WASH, and ­ agriculture). Although many revenues from natural resources into pro- countries are close to meeting or exceeding poor social spending result in a large financ- global targets for spending as a share of GDP ing gap for critical s ­ pending. Although or government expenditures, absolute spend- improving revenue and spending perfor- low. ing levels are still very ­ mance is important, even with improve- And within-sector spending is often inef- ments, official development assistance ficient and sometimes regressive (such as (ODA) will remain critical for the poorest spending more on services used dispropor- ­countries. tionately by the nonpoor than the p ­ oor). Aid makes up more than 8 percent of Inefficiency in spending on services mani- gross national income (GNI) for half of low- fests itself in several ways—for example, in income countries in Africa (figure ­ O.4); ODA high rates of absenteeism among teachers supports key sectors for reducing poverty, and supplies not reaching frontline ­ providers. including health, agriculture, and ­ education. As a result of both limited spending on pro- But although global ODA has been increas- poor sectors and inefficiency in the spend- ing and reached an all-time high of US$140 ing, many poor still pay for access to basic billion in 2016 (at ­ c urrent prices), ODA services critical for human development; out- to African countries increased from 2013 to of-pocket expenditures are h ­ igh. Notably, 2017 (from US$45.8 billion to US$46.3 bil- resource-rich countries spend less on educa- lion), after a dip to $42.5 billion in 2016. tion and health than other African countries But in per capita terms, ODA declined of similar income level (Cockx and Francken from US$48.30 to US$42.60 because of the 2014, 2016­). region’s population growth. O v e r v i e w   21 FIGURE O.4  ODA is a large share of GNI in low-income countries 20 18 16 14 12 % of GNI 10 8 6 4 2 0 M uin e T ab a ng anz we De io ia .R a Ch p. ad To ad rk gas o a F ar Be so in Ug nin -B da Co M u m ali r s M rra an i oz L da bi ne ca Mmal Ni ue Ga Re law (24 r bi bli 25 ) LibThe (26%) er (2 ) (3 ) Su %) te , R ia Iv . G ire w a Ca Ke ini e a n M so ia it o é b e a ua d Ve al ria nc e r a ut ng s Bo Af a ts rica Se Ga na he n m es ia i n d’ ep Si Rw und m pu i ( % a, c % ia 7% n a ia ge Bu oro So A itiu l G ip b e m pi Es han m ny m Ca Senani au e h ol to Prí rd Bu a g a ur th Za roo yc bo Co N da o, Eth an Cô ngo ger Le mb Eq an o g ib e in c Na ll a m in at iss a ea an q am eo o 4 w Zi Gu a e M Gu So Co To o fri Sã lA ra nt Ce Upper middle Low income Lower middle income and high income Source: Organisation for Economic Co-operation and Development (OECD) 2017 data, h ­ ttps://data.oecd.org/. Note: GNI = gross national ­income; ODA = official development ­assistance. ODA data do not include aid inflows from international charities, international nongovernmental organizations, and private donations. The proportion of aid going to African frag- several international debt relief initiatives ile and conflict-affected states also continued were implemented—debt has been rising to ­decline. A total of 13 OECD Development more rapidly in Africa than in other regions Assistance Committee (DAC) donors, includ- since 2009. So, while governments could ing the European Union (EU) institutions, borrow domestically and internationally to reduced their contributions to African fragile finance more spending on social sectors and and conflict-affected states between 2014 and WASH, many will find it ­difficult. 2015 (ONE 2017­ ). The overall decline, at least in part, is because the donor countries were spending more in their own countries on refu- Way Forward: Four Primary gees and asylum ­ seekers. Policy Areas The issuing of debt over the past decade in the face of macroeconomic slowdown In conclusion, from the wide range of over the past couple of years, combined with themes and issues discussed across the chap- insufficient revenue and lagging ODA com- ters of this report—focused on raising the mitments, has put country debt concerns incomes of Africa’s poor and accelerating back on the ­ r adar. Although debt levels poverty reduction—four areas for primary remain below those in the late 1990s—when advanced. policy attention are ­ 22   Acceler ating Pover t y Reduc tion in Afric a Accelerate the Fertility Transition productivity and secondary town develop- ment particularly ­ effective. More-integrated Rapid population grow th in A frica—­ approaches—tackling both supply- and averaging 2.7 percent per year—remains a demand-side constraints at once — are defining feature that holds poverty reduc- needed, both to raise agricultural produc- tion back for many countries and households tivity and to increase the return to infor- on the ­ continent. It elevates the fiscal needs mal nonagricultural household enterprises, for social services, which only pay off much where most of the poor will find off-farm ­ later. High fertility has also been an impor- ­employment. tant direct contributor to Africa’s explosive Inclusive value chain development pro- urban growth, making it hard for urban vides a market-based solution to integrate, centers to keep up the infrastructure base to especially for nonstaple ­foods. But comple- remain productive and create ­ employment. mentary public investment (in agricultural And high fertility limits women’s income- research and extension, irrigation, and rural earning ­opportunities. infrastructure) remains key, especially for Accelerating fertility reduction is there- staple crop ­ productivity. fore an important entry point for accelerating Finally, technological leapfrogging and Africa’s poverty ­ reduction. A 1 percent fall new business models bring previously unat- in the dependency rate is associated with a tainable markets and production tech- 0.75 percentage point fall in headcount pov- niques within reach of the poor (such as erty (Cruz and Ahmed 2016­ ). Investments in solar pump irrigation, and mechanization family planning programs can play an impor- in agriculture, and e-commerce household tant cost-effective complementary role, in enterprises). This, too, requires comple- addition to female education, programs offer- mentary public investments in ICT infra- ing life skills for women and girls, addressing structure and ­skills. social norms around gender through social and behavior change communication, and reducing child ­ marriage. Mitigate Fragility Risk and conflict have long permeated Leverage the Food System African ­ livelihoods. This substantially com- Much poverty reduction remains to be plicates Africa’s poverty-reduction ­ efforts. gained from leveraging Africa’s food sys- Shocks are frequent, conflicts often cast tem, on and off the ­ f arm. Raising small- a long shadow, coping capacity is mostly holder agricultural labor productivity inadequate (especially for the poor and increases the income of the poor directly near-poor), and uninsured risks hold and and reduces the price of food for the urban push people back into ­ p overty. Climatic p oor. Urbanization and economic growth ­ change is making weather patterns even are boosting domestic demand for higher- more erratic and extreme, and the upsurge value agricultural products, also creating in terror-related conf lict adds further employment opportunities off the farm ­uncertainty. up and down the value chains, often par- Twenty-nine percent of Africa’s poor live ticularly for ­ women. Rising agricultural in fragile states, a share projected to increase productivity will also increase demand for to 50–80 percent by 2030. This trend puts nonagricultural goods and services, facili- fragile and conflict-affected states at the tating intersectoral and rural-urban labor center of Africa’s fight against p ­ overty. ­reallocation. Climate change and conflict may further However, not all agricultural development interact to increase each other’s occurrence and urbanization models are equally poverty and detrimental effects (Hsiang, Burke, and reducing, with raising smallholder staple crop Miguel 2013­). O v e r v i e w   23 Better risk and conflict management to Finally, not all countries are struggling with address fragility is the third policy entry fiscal deficits, but pro-poor spending and point for accelerating poverty reduction in spending efficiency can be improved in most ­ A frica. Many of the solutions exist, with a of them, and especially in the resource-rich role for both the private and public sectors, ­countries. but the most important hurdle remains incen- tivizing public and private actors to act now, before the shocks and conflict ­occur. A more Notes helps. productive agriculture also ­ 1. Throughout this report, “Africa” refers to Sub-Saharan ­A frica. Address the Poverty Financing Gap 2. This ambition is articulated in SDG 1, Target 1.1 ( ­http://www.un.org/sustainable​ Making progress in these three policy areas development/poverty/). It is tracked by mea- requires public financing focused on the suring progress on the proportion of people poor, including to overcome Africa’s poor living below the $1.90-a-day international initial conditions in human d ­ evelopment. poverty line (in 2011 purchasing power In Africa’s few non-low-income countries, ­parity). the challenge is not so much the amount of 3. Ranking countries from those with the larg- resources required to address poverty, but est number of poor, Nigeria accounts for rather the decision and effort to redirect about one-quarter of Africa’s poor (85.2 resources to policies and programs that ben- million); the next four (the Democratic Republic of Congo, Tanzania, Ethiopia, and efit the ­poor. However, for most countries Madagascar) for another quarter; and the in Africa, which house most of the poor, next five (Mozambique, Uganda, Malawi, current domestic resources are not nearly Kenya, and Zambia) for the following sufficient to address poverty—and insuffi- 25 ­percent. cient domestic revenue mobilization, lagging 4. In Africa, the likelihood of being poor is ODA commitments, and rising debt levels 3 percentage points lower on average when following the macroeconomic slowdown an individual has some primary education; further shrink their fiscal ­ space. 7 percentage points lower given completed In principle, the discovery of natural primary or incomplete secondary education; resources across Africa over the past two 10 percentage points lower given completed decades could ­ help. Yet poverty reduction secondary education; and 12 percentage points lower given tertiary education (con- and multiple human development indicators trolling for the area of residence, household are often worse in resource-rich countries in structure, and demographic characteristics) Africa than in other countries at the same (Castañeda et ­ ). al. 2018­ level of income. 5. The increase in smallholder staple crop pro- In addition to the continued need for ODA ductivity is often referred to as the “Green to address the fiscal gap, Africa’s fiscal sys- Revolution,” in reference to Asia’s rapid tems need to become more effective in rais- increase in smallholder staple crop produc- ing incomes (including through addressing tivity in the 1960s and 1970s, through a domestic tax compliance and international package of modern inputs (seeds, fertilizer, tax avoidance) as well as in making public and pesticides); water control; and reduction spending more pro-poor and more ­ efficient. volatility. in price ­ 6. Side-selling is a practice by which farm- ers divert part or most of their contracted These four primary policy entry points are production to other ­ b uyers. It is greater relevant across countries, albeit to different when limited value addition does not per- degrees. Fertility is, for example, already ­ mit price premiums to make contracts more lower in southern Africa than in western i ncentive-compatible. On the other hand, ­ and eastern ­ A frica. Risks are pervasive the wide availability of undifferentiated everywhere but take on different ­ f orms. staples and the limited opportunity for value 24   Acceler ating Pover t y Reduc tion in Afric a addition also increases the opportunity for and Christopher U ­ dry. 2015. “A Multifaceted buyers to breach the contracts and reduces Program Causes Lasting Progress for the Very their incentives to engage in contracting to Poor: Evidence from Six ­ Countries.” Science begin ­with. 348 (6236): 773–89.   7. The much wider availability of motorcycle and Beegle, Kathleen, Luc Christiaensen, Andrew motorized tricycle taxi services able to navi- Dabalen, and Isis ­ Gaddis. 2016. Poverty in a gate Africa’s rugged rural roads, following the Rising ­Africa. Washington, DC: World ­ Bank. import of much cheaper models from China Berdegué, Julio, and Isidro ­ Soloaga. 2018. “Small and India, is a good example of the impor- and Medium Cities and Development of tance of transport services for c ­ onnectivity. Mexican Rural A ­ reas.” World Development The trend led the World Bank to raise its 107: 277–88. estimated distance of an all-season road pro- ­ Bhalotra, Sonia, and Samantha R ­ awlings. viding rural connectivity from 2 kilometers to 2013. “Gradients of the Intergenerational at least 5 kilometers, in constructing its 2016 Transmission of Health in Developing Rural Access Index ­ (https://datacatalog.world- ­C ountries.” Re view of Economics and bank.org/dataset ​/rural-access-index-rai). Statistics 95 (2): 660–72.   8. Similarly, although there is a positive effect Bhutta, Zulfiqar A ­ . Das, Ajumand Rizvi, ­ ., Jai K of city size and urban concentration on Michelle ­ F. Gaffey, Neff Walker, Susan growth in high-income countries, no such Horton, Patrick Webb, Anna Lartey, and effect has been found so far in low- and Robert ­ E. ­B lack. 2013. “Evidence-Based m iddle-income ­ ­ c ountries. If anything, the Interventions for Improvement of Maternal effect is likely negative (Frick and Rodríguez- and Child Nutrition: What Can Be Done Pose 2016, 2018­ ). and at What Cost?” The Lancet 382 (9890): 452–77. Blattman, Christopher, and Jeannie ­ Annan. 2016. “Can Employment Reduce Lawlessness and References Rebellion? A Field Experiment with High-Risk Adjognon, S ­ ., ­ ­. G L. ­ ­ . Liverpool-Tasie, and S. O Men in a Fragile ­ State.” American Political T. ­ ­ A. ­ R eardon. 2017. “Agricultural Input Science Review 110: 1–17. Credit in Sub-Saharan Africa: Telling Myth Bongaarts, ­ John. 2017. “Africa’s Unique Fertility from ­Facts.” Food Policy 67 (C): 93–105. ­Transition.” Population and Development Allen, Thomas, Philipp Heinrigs, and Inhoi Re vie w 43 (Issue Supplement Fer tilit y H eo. 2018. “Agriculture, Food and Jobs in ­ Transition in Sub-Saharan Africa): 39–58. West ­ A frica.” West African Papers N ­ o. 14, Branisa, Boris, Stephan Klasen, and Maria Organisation for Economic Co-operation and Z iegler. 2009. “Why We Should All Care ­ Development, ­Paris. about Social Institutions Related to Gender Ambler, Kate, Alan de Brauw, and Susan Inequality.” Proceedings of the German ­ Godlonton. 2018. “Agriculture Support Services D e velopm e nt E c o nom i c s C o n f e r e n c e , in Malawi: Direct Effects, Complementarities, Hannover, ­ No. 15, Verein für Socialpolitik, and Time D ­ ynamics.” IFPRI Discussion Paper Ausschuss für Entwicklungsländer, ­ Göttingen. No. 01725, International Food Policy Research ­ ———. 2013. “Gender Inequality in Social Institute, Washington, D ­ C. Institutions and Gendered Development Antoine, Kassia, Raju Jan Singh, and Konstantin ­Outcomes.” World Development 45: 252–68­ . ­ M. W ­ acker. 2017. “Poverty and Shared ———. 2014. “The Institutional Basis of Gender Prosperity: Let’s Move the Discussion beyond Inequality: The Social Institutions and Gender ­G row th.” Forum for Social Economics Index ­(SIGI).” Feminist Economics 20 (2): 46 (20): 192–205. 29–64. Bah, El-hadj, and Lei ­ F ang. 2015. “Impact C a s t a ñe d a , A nd ré s , D u ng D oa n , Dav id of the Business Environment on Output Newhouse, Minh Cong Nguyen, Hiroki and Productivity in A ­ frica.” Journal of Uematsu, and João Pedro ­ A zevedo. 2018. Development Economics 114: 159–71. “A New Profile of the Global P ­ oor.” World Banerjee, Abhijit, Esther Duflo, Nathanael Development 101: 250–67. Goldberg, Dean Karlan, Robert Osei, William Cattaneo, ­ Umberto. 2017. “Poverty Headcount Parienté, Jeremy Shapiro, Bram Thuysbaert, A f r i c a .” P roj e c t io n s i n S ub - S a h a r a n ­ O v e r v i e w   25 Background note prepared for Accelerating Daidone, Silvio, Benjamin Davis, Joshua Dewbre, Poverty Reduction in Africa , World Bank, Borja Miguelez, Ousmane Niang, and Luca Washington, ­DC. Pellerano. 2017. “Linking Agriculture and ­ Chamberlin, Jordan, and Thomas ­ S. ­J ayne. Social Protection for Food Security: The Case 2017. “Does Farm Structure Matter? The of ­Lesotho.” Global Food Security 12 (March): Effects of Farmland Distribution Patterns on 146–54. Rural Household Income in ­ Tanzania.” MSU de la Brière, Bénédicte, Deon Filmer, Dena International Development Working Paper Ringold, Dominic Rohner, Karelle Samuda, 157, Michigan State University, East ­ Lansing. and Anastasiya ­ Denisova. 2017. From Mines Christiaensen, Luc, Lionel Demery, and Stefano and Wells to Well-Built Minds: Turning Sub- P aternostro. 20 03. “Macro and M icro ­ Saharan Africa’s Natural Resource Wealth into Perspectives of Growth and Poverty in ­ Africa.” Human ­C apital. Directions in Development World Bank Economic Review 17 (3): 317–47. Series. Washington, DC: World ­ ­ Bank. Christiaensen, Luc, Joachim De Weerdt, and de la Fuente, A lejandro, Jon Jellema, and Ravi ­ K anbur. 2019. “Decomposing the Nora ­ Lustig. 2018. “Fiscal Policy in Africa: Cont ribution of M ig ration to Pover t y Welfare Impacts and Policy ­ E ffectiveness.” Reduction: Methodology and Application Background paper prepared for Accelerating to ­Tanzania.” Applied Economics Letters Poverty Reduction in Africa , World ­ B ank, 26 (12): 978–82. Washington, DC. Christiaensen, Luc, and Ravi ­ K anbur. 2017. de Silva, Tiloka, and Silvana T ­ enreyro. 2017. “Secondary Towns and Poverty Reduction: “Population Control Policies and Fertility Refocusing the Urbanization A ­ genda.” Annual ­C o nve r g e n c e .” Jo u r n a l o f E c o n o m i c Review of Resource Economics 9: 405–19. Perspectives 31 (4): 205–28. Christiaensen, Luc, and Will ­ M artin. 2018. D ei n i nger, K lau s , a nd Fa ng ­ X ia. 2016 . “Agriculture, Structural Transformation and “Quantifying Spillover Effects from Large Land- Poverty Reduction: Eight New ­ I nsights.” based Investment: The Case of ­ Mozambique.” Wo rl d De velop m e nt 109 (S eptember): World Development 87: 227–41. ­ 413–16. ———. 2018. “A ssessi ng t he L ong-Ter m Christiaensen, Luc, and Yasuyuki ­ Todo. 2014. Performance of Large-Scale Land Transfers: “Poverty Reduction during the Rural–Urban Challenges and Opportunities in Malawi’s Estate Transformation: The Role of the Missing ­Sector.” World Development 104: 281–96. ­Middle.” World Development 63 (C): 43–58. Diao, X i n shen , Ja me s T hu rlow, S a muel Cockx, Lara, and Nathalie ­ F rancken. 2014. Benin, and Shenggen F ­ an. 2012. Strategies “Extending the Concept of the Resource and Priorities for Afric an Agriculture: Curse: Natural Resources and Public Spending Economywide Perspectives from Country on ­H ealth.” Ecological Economics 108: ­Studies. Washington, DC: International Food 136–49­. Policy Research Institute ­ (IFPRI). ———. 2016. “Natural Resources: A Curse on Dixit, Siddharth, Indermit Gill, and Chinmoy Education ­S pending.” Energy Policy 92: ­Kumar. 2018. “Are Economic Relations with 394–408. India Helping Africa? Trade, Investment Collier, Paul, and Anke ­ Hoffler. 2004. “Greed and Development in the Middle-Income and Grievance in African Civil ­ Wars.” Oxford S outh.” Research paper, Duke Center for ­ Economic Papers 56: 563–95. International Development, Duke University, Cruz, Marcio, and ­ S . Amer ­ A hmed. 2016. Durham, ­NC. “On the Impact of Demographic Change Dollar, David, and Aart ­ K raay. 2002. “Growth on Growth, Savings and P ­ overty.” Policy Is Good for the ­ Poor.” Journal of Economic Research Working Paper 7805, World Bank, Growth 7 (3): 195–225. Washington, ­DC. Dorosh, Paul, and James ­ T hurlow. 2018. Dabalen, Andrew, Alejandro de la Fuente, “Beyond Agriculture versus Non-Agriculture: Aparajita Goyal, Wendy Karamba, Nga Thi Decomposing Sectoral Grow th-Pover t y Viet Nguyen, and Tomomi T ­ anaka. 2017. Linkages in Five African ­ C ountries.” World Pathways to Prosperity in Rural M ­ alawi . Development 109: 440–51. Directions in Development ­Series. Washington, Eberhard-Ruiz, Andreas, and Alexander M ­ oradi. DC: World ­ Bank. 2018. “Regional Market Integration and City 26   Acceler ating Pover t y Reduc tion in Afric a Growth in East Africa: Local but No Regional Haushofer, Johannes, and Ernst F ­ ehr. 2014. Ef fec ts?” C SA E Work i ng Paper S eries “On the Psychology of P ­ overty.” Science 344 2018– 09, Centre for the Study of African (6186): 862–67. Economies, University of O ­ xford. Hazell, Peter, Colin Poulton, Steve Wiggins, F e r r a nt , G a ë l l e , a n d A l e x a n d r e ­K ol e v. and Andrew D ­ orward. 2010. “The Future 2016. “Does Gender Discrimination in of Small Farms: Trajectories and Policy Social Institutions Matter for Long-Term ­P r i o r i t i e s .” Wo r l d D e v e l o p m e n t 3 8 : Growth? Cross-Country ­ Evidence.” OECD 1349–61. Development Centre Working Paper N ­ o. 330, Hsiang, Solomon, Marshall Burke, and Edward Organisation for Economic Co-operation and M iguel. 2013. “Quantifying the Influence of ­ Development (OECD), ­ Paris. Climate on Human ­ C onflict.” Science 341 Fisker, Peter, and Ruth H ­ ill. 2018. “Mapping (6151): 1235367. the Nature of Risk in Sub-Saharan ­ A frica.” IMF (International Monetary F ­ und). 2015. Background paper prepared for Accelerating Regional Economic Outlook: Sub-Saharan Poverty Reduction in Africa , World Bank, ­A frica. Dealing with the Gathering ­ C louds. Washington, ­DC. Washington, DC: ­ I MF. Frick, Susanne, and Andrés ­ R odríguez-Pose. Ingelaere, Bert, Luc Christiaensen, Joachim 2016. “Average City Size and Economic De Weerdt, and Ravi K ­ anbur. 2018. “Why ­G rowth.” Cambridge Journal of Regions, Secondary Towns Can Be Important for Economy and Society 9 (2): 301–18­ . Poverty Reduction: A Migrant P ­ erspective.” ———. 2018. “Change in Urban Concentration World Development 105: 273–82. and Economic ­ Growth.” World Development Jalloh, Abdulai, Gerald ­ C . Nelson, Timothy 105: 156–70. S. Thomas, Robert Zougmoré, and Harold Roy- ­ Gaspar, Vitor, Laura Jaramillo, and Philippe Macauley, ­ eds. 2013. West African Agriculture ­ Wingender. 2016. “Tax Capacity and Growth: and Climate Change: A Comprehensive Is There a Tipping ­ Point?” IMF Working Paper ­Analysis. Washington, DC: International Food WP/16/234, International Monetary Fund, Policy Research Institute ­ (IFPRI). Washington, ­DC. James, ­ J effrey. 2016. The Impact of Mobile Gibson, John, Gaurav Datt, Rinku Murgai, Phon es on Pove r t y an d In equ alit y in and Martin ­ R avallion. 2017. “For India’s Developing ­C ountries. Cham, Switzerland: Rural Poor, Growing Towns Matter More ­Springer. Than Growing ­ Cities.” World Development Jayne, Thomas ­ S., Jordan Chamberlin, and Derek 87: 413–29. Headey. 2014. “Land Pressures, the Evolution ­ Gill, Indermit, and Kenan ­ Karakülah. 2018. “Is of Farming Systems and Development Strategies China Helping Africa? Growth and Public in Africa: A ­ Synthesis.” Food Policy 48: 1–17. Debt Effects of the Subcontinent’s Biggest Jedwab, Remi, Luc Christiaensen, and Marina Investor.” Global Working Paper ­ ­ No. 3, Center ­ Gindelsky. 2017. “Demography, Urbanization for International and Global Studies, Duke and Development: Rural Push, Urban Pull University, Durham, N ­ C. a nd…Urba n P ush?” Jour n al of Urban Grimm, Michael, Anicet Munyehirwe, Jörg Economics 98 (C): 6–16. Peters, and Maximiliane ­ S ievert. 2017. Jones, V ­ an. 2018. “How Hello Tractor’s Digital “A First Step Up the Energy Ladder? Low Cost Platform Is Enabling the Mechanization of Solar Kits and Household’s Welfare in Rural African ­Farming.” AgFunder News, July 4. ­Rwanda.” World Bank Economic Review Klasen, ­ Stephan. 2006. “Pro-Poor Growth and 31 (3): 631–49. Gender ­ I nequality.” In Pro-Poor Growth: Günther, Isabel, and Kenneth ­ H arttgen. 2016. Polic y and Evide nce , edited by Lukas “Desired Fertility and Number of Children ­ Menkhoff. Berlin: Duncker and H ­ umblot. Born across Time and ­ Space.” Demography Lall, Somik, ­ J. Vernon Henderson, and Anthony 53: 55–83. ­J . ­Venables. 2017. Africa’s Cities: Opening Hallward-Driemeier, Mary, Tazeen Hasan, and Doors to the W ­ orld. Washington, DC: World Anca Bogdana ­ Rusu. 2013. “Women’s Legal ­Bank. Rights over 50 Years: What Is the Impact of Le Goff, Maëlan, and Raju Jan ­ S ingh. 2014. Reform?” Policy Research Working Paper “Does Trade Reduce Poverty? A View from 6617, World Bank, Washington, ­ DC. ­Africa.” Journal of African Trade 1: 5–14. O v e r v i e w   27 Luo, ­ X ., ­ Y. Wang, and ­ X. ­ Z hang. 2019. I n Afric an De velopme nt in Historic al “E-Commerce Development and Household Perspective , edited by E . A kyeampong, Consumption Growth in C ­ hina.” Policy R. Bates, N. Nunn, and J. Robinson, 153–207. Research Working Paper 8810, World Bank, Cambridge, U.K.: Cambridge University Press. Washington, ­DC. Resnick, D ­ anielle. 2017. “Governance: Informal Martin-Shields, Charles ­ P., and Wolfgang ­ Stojetz. Food Markets in Africa’s ­ C ities.” In IFPRI 2019. “Food Security and Conflict: Empirical Global Food Policy Report, 50–57. Washington, Challenges and Future Opportunities for DC: International Food Policy Research ­ Institute. Research and Policy Making on Food Security Rodrik, ­ Dani. 1998. “Trade Policy and Economic and ­C onflict.” World Development 119: Performance in Sub-Saharan ­ A frica.” NBER 150–64. Working Paper 6562, National Bureau of McCullough, Ellen ­ B . 2017. “Labor Productivity Economic Research, Cambridge, ­ M A. and Employment Gaps in Sub-Saharan ———. 2016. “An African Growth Miracle?” ­Africa.” Food Policy 67: 133–52. Journal of African Economies 27 (1): 1–18. Mellor, John W ­ illiams. 2017. Agricultural Rondinelli, Dennis, and Kenneth R ­ uddle. 1983. Development and Economic ­ Transformation. Urbanization and Rural De velopment: Cham, Switzerland: Springer International A Spatial Policy for Equitable G ­ rowth. ­Publishing. New York: ­ Praeger. Nagler, ­ Paula. 2017. “A Profile of Non-Farm Sakyi, Daniel, José Villaverde, Adolfo Maza, and Household Enterprises in Sub- Saharan Isaac ­ B onuedi. 2017. “The Effects of Trade ­ A frica.” Background note prepared for and Trade Facilitation on Economic Growth in Accelerating Poverty Reduction in Africa , Africa.” African Development Review 29 (2): World Bank, Washington, ­ DC. 350–61. Nagler, Paula, and Wim ­ N audé. 2017. “Non- Sheahan, Megan, and Christopher B B arrett. ­. ­ Farm Entrepreneurship in Rural Sub-Saharan 2014. “Understanding the Agricultural Input Africa: New Empirical ­ Evidence.” Food Policy Landscape in Sub-Saharan Africa: Recent 67: 175–91. Plot, Household, and Community-Level OECD (Organisation for Economic Co-operation Evidence.” Policy Research Working Paper ­ and ­ Development). 2017. Revenue Statistics in 7014, World Bank, Washington, ­ DC. Africa 1990 –2015. Paris: OECD ­ Publishing. Swinnen, Johan, and Rob K ­ uijpers. 2017. ONE. 2017. The 2017 DATA Report: Financing “Inclusive Value Chains to Accelerate Poverty for the African C ­ entury. Annual statistical Reduction in ­ A frica.” Background note pre- report, The ONE Campaign, Washington, pared for Accelerating Poverty Reduction in ­DC. Africa, World Bank, Washington, DC. Otte, ­ J ., ­ A . Costales, ­ J . Dijkman, ­ U . Pica- Swinnen, Johan ­ F. ­M ., Anneleen Vandeplas, Ciamarra, T ­ . Robinson, V ­ . Ly, and ­ . Ahuja, C and Miet ­ M aertens. 2010. “Liberalization, ­ D. Roland-Holst, ­ eds. 2012. Livestock Sector Endogenous Institutions, and Growth: A Development for Poverty Reduction: An Comparative Analysis of Agricultural Reforms Economic and Policy Perspective—Livestock’s in Africa, Asia, and E ­ urope.” World Bank Many ­Virtues. Rome: Food and Agriculture Economic Review 24 (3): 412–45. Organization of the United Nations ­ (FAO). Tschirley, David, Thomas Reardon, Michael Pace, Noemi, Silvio Daidone, Benjamin Davis, Dolislager, and Jason S ­ nyder. 2015. “The Sudhanshu Handa, Marco Knowles, and Rise of a Middle Class in East and Southern Robert ­ Pickmans. 2018. “One Plus One Can A f r ic a: I mpl ic at ions for Food System Be Greater than Two: Evaluating Synergies Transformation: The Middle Class and Food of Development Programmes in ­ M alawi.” System Transformation in ­ E SA.” Journal of Journal of Development Studies 54 (11): International Development 27 (5): 628–46. 2023–60. Victora, Cesar ­ G ., Linda Adair, Caroline Fall, Pauw, K a rl , a nd Ja me s T ­ hu rlow. 2 011. Pedro ­ C . Hallal, Reynaldo Martorell, Linda “Agricultural Growth, Poverty, and Nutrition Richter, Harshpal Singh Sachdev, and Maternal in ­Tanzania.” Food Policy 36: 795–804. and Child Undernutrition Study ­ G roup. Platteau, Jean-Philippe. 2014. “Redistributive 2008. “Maternal and Child Undernutrition: Pressures in Sub-Saharan Africa: Causes, Consequences for Adult Health and Human Consequences, and Coping Strategies.” ­Capital.” The Lancet 371 (9609): 340–57. 28   Acceler ating Pover t y Reduc tion in Afric a Waithaka, M ­ ., G. C. Nelson, T. S. Thomas, ———. 2018­d. World Development Report 2018: and M. Kyotalimye, eds. 2013. East Learning to Realize Education’s P ­ romise. African Agriculture and Climate Change: A Washington, DC: World ­ Bank. Comprehensive ­Analysis. Washington, DC: ———. 2019a. Africa’s Pulse: An Analysis of International Food Policy Research Institute Issues Shaping Africa’s Economic Future , ­(IFPRI). vol. 19 (April). Washington, DC: World Bank. World ­ B a n k. 2015. Wo rl d De velop m e nt ———. 2019b. Women, Business, and the Law Report 2015: Mind, Society, and ­ B ehavior. 2019: A Decade of R ­ eform. Washington, DC: Washington, DC: World ­ Bank. World ­Bank. ———. 2018­a . Africa’s Pulse: An Analysis of ———. 2019c. World Development ­ I ndicators Issues Shaping Africa’s Economic Future , (database). World Bank, Washington, DC. vol. 17 (April). Washington, DC: World Bank. Yoon, Jisu, and Stephan ­ K lasen. 2018. “An ———. 2018­ b. “All Hands on Deck: Reducing Application of Partial Least Squares to the Stunting through Multisectoral Efforts in Construction of the Social Institutions and Sub-Saharan A ­ frica.” Report, World Bank, Gender Index (SIGI) and the Corruption Washington, ­DC. Perception Index ­ (CPI).” Social Indicators ———. 2018­c . Poverty and Shared Prosperity Research 138 (1): 61–88. 2018: Piecing Together the Poverty ­ Puzzle. Washington, DC: World ­ Bank. Introduction A frica’s turnaround over the past And following the collapse in world com- couple of decades has been dra- modity prices, economic progress slowed, as matic.1 After many years in decline, Africa’s economic growth dropped substan- the continent’s economy picked up in the tially. In per capita terms, gross domestic mid-1990s, expanding annually at a robust product (GDP) turned even slightly negative annual average of 4.5 percent into the early during 2016–18. Without Nigeria, South 2010s. People became healthier and better Africa, and Angola—Africa’s three largest nourished, youngsters attended schools in economies, each highly dependent on com- much greater numbers, and both men and modities—the downfall was less severe, drop- women got greater control over their lives. ping to slightly below 2 percent growth per There was also substantial poverty reduc- capita during 2016–18. More recently, along tion, with the share of Africans living in with the overall recovery of the world econ- extreme poverty—defined as living on less omy, Africa’s growth prospects are improv- than US$1.90 per person per day—declin- ing again (World Bank 2019). ing from 54 percent in 1990 to 41 percent by Africa’s turnaround happened in the con- 2015 (World Bank 2018). text of rapid poverty reduction across the These improvements notwithstanding, world, especially in East and South Asia, progress on the nonmonetary dimensions of which found themselves at similarly high well-being was from very low levels. Many poverty levels in the early 1990s. The share remain undernourished, illiterate, and unem- of the world’s extreme poor is now reaching powered, with gender gaps pronounced in all 10 percent (World Bank 2018), and a world three of these dimensions (Beegle et al. 2016). free of extreme poverty has come increas- Exposure to domestic violence remains high, ingly into sight (Ravallion 2013). and exposure to political violence has even With Africa’s poverty rates still high and increased since 2010. As Africa’s popula- progress over the past couple of years stalling, tion continued to expand rapidly (by 2.7 the world’s poor have become increasingly percent annually), the number of extreme concentrated in Africa—from 15 percent of poor in Africa also rose, from an estimated the world’s poor in 1990 to 56 percent in 278 ­million in 1990 to 413 million in 2015 2015 (figure I.1). The ambition to eradicate (World Bank 2018). poverty worldwide has now also formally 29 30   Acceler ating Pover t y Reduc tion in Afric a FIGURE I.1  More than half of the world’s extreme poor live in Africa a. Share of the world’s extreme poor, by region, 1990 b. Share of the world’s extreme poor, by region, 2015 0% 1% 1% 3% 6% 4% 15% 52% 28% 56% 29% 1% 3% 1% East Asia and Pacific Latin America and the Caribbean South Asia Rest of the world Europe and Central Asia Middle East and North Africa Sub-Saharan Africa Source: World Bank 2018. Note: “Extreme poor” refers to the percentage of the population living below US$1.90 per person per day. been adopted as a global goal, and world- poverty-reduction prospects. Its key entry wide attention is increasingly turning toward point is increasing the earnings of the poor, accelerating poverty reduction in Africa. The and so it focuses on their livelihood strate- 2015–16 migration crisis in Europe adds fur- gies and increasing the productivity of their ther political impetus. assets (labor and land). This report examines policy entry points The report proceeds as follows. Chapters 1 for accelerating poverty reduction in the and 2 review the key features of Africa’s pov- region. It is the second of a two-part volume erty and the high-level impediments to accel- on poverty in Africa. The first report, Pov- erating poverty reduction: (a) persistently erty in a Rising Africa (Beegle et al. 2016), high fertility, (b) poor initial conditions, and reviewed Africa’s poverty status in its mon- (c) growth patterns that insufficiently (or etary and nonmonetary dimensions and its unsustainably) benefit the poor. Natural- evolution since the early 1990s. It focused resource richness and fragility and conflict specifically on data considerations. emerge as two further features that challenge This second report focuses on how to poverty-reduction efforts. Chapters 3 and 4 accelerate poverty reduction, with an eye explore how to increase earnings for the poor on the United Nations Sustainable Develop- and near poor on and off the farm, within ment Goal (SDG) of eradicating poverty by and outside the agriculture sector and rural 2030. 2 It draws on global historical expe- areas, respectively. Chapter 5 examines the rience in poverty reduction, as well as on implications of risk and conflict on an agenda recent successes in Africa, and accounts to alleviate poverty. Chapter 6 concludes by for Africa’s specific conditions and over- reviewing the poverty-reducing potential and arching global trends in shaping Africa’s performance of Africa’s current fiscal systems I n t r o d u c t i o n    31 and identifies options to increase poverty more efficient fiscal policy. Some eastern Afri- reduction by raising more resources, allocat- can countries have made some progress along ing them better, and improving the efficiency each of the four paths identified, with some of countries’ spending. success in accelerating poverty reduction (such Chapters 1 and 2 raise the impor- as Ethiopia and Rwanda), though not always tance of four structural inequalities that on fertility reduction (for example, Uganda). hold poverty reduction back in Africa: the human ­ development trap, deep-seated gen- der inequality, lack of infrastructure, and Notes political realities. These are longstanding, 1. Throughout this report, “Africa” refers to entrenched issues, often requiring sustained Sub-Saharan Africa. efforts to address. As such, they are the top- 2. This ambition is articulated in SDG 1, Target ics of four “Fundamentals” sections, with 1.1 (http://www.un.org/sustainabledevelopment​ inserted discussions on these areas between poverty/). It is tracked by measuring progress /­ the chapters. on the proportion of people living below the Notably, four critical areas emerge from $1.90-a-day international poverty line (in 2011 purchasing power parity). the range of themes and issues that arise across the chapters of this report—fragility, fertility, food systems, and fiscal space—as outlined in more detail in the overview. References As an Africa-wide report, this volume does Beegle, Kathleen, Luc Christiaensen, Andrew not aim to provide country-specific strategies Dabalen, and Isis Gaddis. 2016. Poverty in a but rather to identify broad common entry Rising Africa. Washington, DC: World Bank. points to accelerate Africa’s poverty reduc- Ravallion, Martin. 2013. “How Long Will tion. Policy packages will need to be tailored It Take to Lift One Billion People Out of Poverty?” World Bank Research Observer to each setting. For example, conflict and 28  (2): 139–58. high fertility are prominent challenges in the World Bank. 2018. Poverty and Shared Prosperity Sahel but arguably less so in southern Africa. 2018: Piecing Together the Poverty Puzzle. Yet, in both cases, there is substantial scope Washington, DC: World Bank. for increasing agricultural productivity and ———. 2019. Africa’s Pulse: An Analysis of Issues leveraging the food system, for better risk Shaping Africa’s Economic Future, vol. 19 management, and for more pro-poor and (April). Washington, DC: World Bank. Poverty in Africa Luc Christiaensen and Ruth Hill 1 I n 2015, the world embraced the ambition to eradicate poverty by 2030. A review of Africa’s poverty status today and its prospects for tomorrow show that, although the region has made substantial progress since the early 1990s, the number of poor has con- increase. By 2030, Africa’s poverty rate will still be around 20 percent, under most tinued to ­ scenarios, and the world’s poverty will become increasingly concentrated within ­ Africa. How Africa can accelerate its poverty reduction is a global ­ challenge. Most of Africa’s poor live concentrated in a limited number of countries: 5 countries account for more than 50 percent of Africa’s poor; 10 countries account for 75 ­ percent. Poverty rates are particularly high in fragile states, where poverty decline is also particularly ­ slow. Four out of five of the poor live in rural areas, earning their living predominantly in f ­arming. Both chronic and transitory poverty states persist, underscoring the importance of asset building management. Measured gender gaps in poverty are small, though likely under- as well as risk ­ estimating the pernicious consequences of structural gender ­ inequalities. The poor also have weak links with the state—that is, weak access to good-quality public goods (infrastructure) and services as well as a limited voice in public ­policy. These stylized facts provide important entry points for poverty-reducing policy design, though caution remains ­ warranted. They only indicate symptoms of poverty, not necessarily ­ causes. Africa’s recent experience shows further that its poverty rate has not only been higher than in most other low- and middle-income countries; it has also declined more ­ slowly. Three fac- tors contribute to this: •  High population ­ g rowth. Per capita incomes have grown more slowly because a much larger share of gross domestic product (GDP) growth is eroded by faster population ­growth. •  Poor initial ­conditions. Although Africa’s poverty-to-growth elasticity is lower than in other low- and middle-income countries, this is not the case in comparison with other, c ountries. Poverty itself is impeding the conversion of household income equally poor ­ growth (not to be equated with GDP growth) in poverty reduction, just like in other poor ­countries. 33 34   Acceler ating Pover t y Reduc tion in Afric a •  Composition of Africa’s ­ g rowth. Africa has been less efficient at converting (per capita) GDP growth into household income growth. This is plausibly linked to the composition of its growth process over the past couple of decades (more in capital-intensive natural ­ anufacturing). resources, less in labor-intensive agriculture or m The scope and need for more pro-poor growth policies in Africa is l ­arge. Fifty million more people could be lifted out of poverty by 2030 if the incomes of the poor were to grow percentage points faster ­ 2 ­ annually. Combined with lower population growth and better initial conditions, growth processes that foster growth in the places and sectors where the poor live and work, giving them a better chance of raising their incomes directly, could thus go a long way toward accelerating poverty reduction (now and in the f provide the ­ uture). These insights ­ overarching backdrop to the report. Poverty Today and Tomorrow In the rest of the low- and middle-income world, poverty reduction during 1990–2015 Over the past decades, Africa has made was faster—especially in East Asia but also substantial progress in reducing extreme in South Asia—and population growth poverty, with the share of Africans living lower. As a result, world poverty is increas- ­ on less than US$1.90 a day in 2011 pur- ingly concentrating in A ­ frica. About three chasing power parity (PPP) terms declining in five of the world’s poor are now living in by 13 percentage points, from 54 percent Africa—amounting to 57 percent in 2015, in 1990 to 41 percent in 2015 (figure 1.1).1 up from 15 percent in 1990. 2 Accelerating Unfortunately, given high population poverty reduction in Africa is central to the growth (2.7 percent per year), the number of world’s ambition of eradicating poverty by Africans living in poverty nonetheless rose, 2030—as expressed in United Nations (UN) from an estimated 278 million in 1990 to Sustainable Development Goal 1 (SDG 1), 413 million in 2015. Target 1.1, adopted in 2015.3 FIGURE 1.1  The poverty rate in Africa has gone down, but the number of African people living in poverty has increased 1 billion Total population 512 million 413 million Number of poor 278 million Population living on US$1.90/day or less 1990 1993 1996 1999 2002 2005 2008 2011 2013 2015 Source: World Bank PovcalNet data, ­http://iresearch.worldbank.org/PovcalNet. P o v e r t y i n A f r i c a   35 Scenarios for Poverty Reduction in already clear that eradicating Africa’s pov- Africa erty would not be feasible by 2030 and that the world’s poverty would increasingly con- So, what are the prospects for Africa’s A frica. centrate in ­ poverty reduction in the future? In setting Which scenarios could bring Africa’s pov- SDG 1, it was calculated that the world could erty down faster? To address this question, a eradicate poverty by 2030 if everyone’s per- (figure 1.2). new baseline scenario is run first ­ sonal income in all low- and ­ middle-income Each country’s average annual per capita countries continued to expand by around GDP growth rate during 1998–2013 is 4.9 percent per year throughout 2008–30 applied to the country’s 2013 income distri- (Ravallion 2013).4 Under such a scenario, bution until 2030. This assumes distribution- the poverty rate in Africa would decline neutral income growth of 2.8 percent a year to 19.2 ­p ercent. Further simulation stud- on average 6 and would bring Africa’s pov- ies, using different assumptions, all situated erty rate down to 22.8 percent in 2030; the Africa’s poverty rate in 2030 well above the number of poor would decline to 323 million eradication target of 3 percent (Cattaneo s cenario). Only if per (figure 1.2, baseline ­ 2017). 5 So, when adopting SDG 1, it was FIGURE 1.2  Africa cannot eradicate poverty by 2030 but can accelerate poverty reduction 70 450 59.0 57.6 400 60 58.1 56.1 54.4 350 50.3 50 46.9 45.7 Number of poor (millions) 44.2 300 42.6 41.0 Poverty rate (%) 40 250 22.8 (323) 30 22.0 200 (296) 150 20 20.8 100 19.1 19.4 (295) 10 (271) (275) 50 278 328 350 376 396 385 390 401 398 395 390 0 0 1990 1993 1996 1999 2002 2005 2008 2010 2011 2012 2013 2015 2020 2025 2030 Poverty rate: Baseline scenario Scenario 2a (m = 1) Scenario 3 (fragility) Number of poor Scenario 1 (fertility) Scenario 2b (m = 2) Actual Source: World Bank ­calculations. Note: The poverty rate is the percentage of the population living at or below US$1.90 per d­ ay. The baseline scenario assumes average distribution-neutral growth of 2.8 percent per ­year. Scenario 1 assumes a low-fertility population growth scenario instead of the historical population growth r ­ ates. Scenario 2a assumes more pro-poor growth, with average GDP growth the same as in the baseline scenario but with the incomes of the poor growing 1 percentage point faster than the historical a ­ verage. Scenario 2b also holds average GDP growth the same as in the baseline scenario, but assumes annual income growth among the poor that is 2 percentage points faster than the historical a ­ verage. Scenario 3 assumes that increased policy attention would make Africa’s fragile states grow 3 percentage points faster than their historical ­average. 36   Acceler ating Pover t y Reduc tion in Afric a capita incomes grew about three times as fast Outlook from Recent Poverty Trends (by 8 percent per year) would poverty come How has Africa performed in the more down to 3 percent (not ­ shown). This is highly recent past—that is, since the SDG 1 tar- ­unlikely. get was adopted? First, it is worth noting In a first alternative scenario (scenario 1), that it has become increasingly possible the slow downward trend in population to track Africa’s performance on meeting growth is accelerated, using the UN’s low- SDG 1 given the increasing availability fertility population growth scenario during of nationally representative surveys with 2013–30 instead of the historical population which to monitor well-being and poverty growth ­ rates. Doing so would bring pov- (box 1.1). However, estimates of pov- erty rates and numbers down to 22 ­ percent erty for recent years always include some and 296 million, respectively ( ­figure 1.2, GDP growth-based poverty estimates for s cenario 1). Given the multiple channels ­ some countries that are in between sur- through which fertility affects growth vey ­years. Given the assumptions that go and poverty, this likely underestimates into GDP growth-based poverty estimates, the ­poverty-reducing effects of an acceler- these poverty estimates can only be an ated fertility ­ decline. Chapter 2 (“Africa’s indication (Ferreira, Azevedo, and Lakner Demography and Socioeconomic Structure”) 2017). elaborates on this in more ­ depth. Unfortunately, the drop in economic per- In a second set of scenarios (scenarios formance of African economies after 2013 2a and 2b), average income growth stays has not been good for poverty ­ reduction. the same, but the incomes of the poor in The latest 2015 poverty numbers reflect this each country are now growing faster than decline in economic performance, with the the country’s historical income ­ g rowth.7 drop in Africa’s poverty rate slowing to If the incomes of the poor in each coun- 0.72 percentage points per ­ year. This con- try were to grow 1 percentage point faster trasts with the 1 percentage point decline than their historical average, such redistri- projected in the base-case scenario described bution of growth from the nonpoor to the ­ earlier. Hence, based on the latest available poor would bring the poverty rate in 2030 poverty numbers, Africa’s fight against pov- down to 20.8 percent and the number of erty was already off track in 2015, even rela- poor to 295 million (figure 1.2, scenario scenario. tive to the base-case ­ 2a: m = 1). If the incomes of the poor were Si nc e t hen , t he sit u at ion h as not to grow 2 percentage points faster, pov- improved, with annual per capita GDP erty rates and numbers would decline to growth for Africa as a whole even slightly 19.4 percent and 275 million, ­ respectively. negative each year (−0.3 percent in 2018). Almost 50 million more people would The most recent aggregate GDP growth have been lifted out of poverty ­ (figure 1.2, forecasts suggest some recovery, to 2.8 per- scenario 2b: m = 2). cent in 2019 and 3.3 percent in 2020 (World Under scenario 3, increased policy atten- Bank 2019). But this barely makes up for tion to the challenges of Africa’s fragile the decline in Africa’s GDP during 2016–18. states is assumed to pay off in the form of It also remains well below the 1998–2013 faster (distribution-neutral) income growth per capita average of 2.8 percent per year in these ­ states. An income increase of 1, 2, assumed in the base-case s ­ cenario. Over the or 3 percentage points over their historical past five years, progress in poverty reduction growth rates would reduce Africa’s pov- in Africa has in all likelihood been lagging erty rate to 22 percent, 20 percent, and well behind the base-case ­ scenario. 19.1 ­ p ercent, respectively, compared with In addition to accelerating growth across 22.8 percent in the baseline scenario (scenario 3, countries, the simulations also suggest that in figure 1.2). P o v e r t y i n A f r i c a   37 BOX 1.1  Efforts to improve Africa’s poverty data are starting to pay off For a long time, knowledge of Africa’s economic measure poverty has come an expansion of other and social transformation has been compromised surveys such as agriculture production surveys by weaknesses in the underlying d ­ ata. Some of the and censuses, business registries, and population key constraints to better statistical systems and ­censuses. data include infrequent surveys, low access to data Beyond improving data collection, the support to produced, poor coordination and integration of sta- statistical capacity helped reform institutional incen- tistical systems, minimal use of statistical evidence tives, turning statistical agencies into professional for decision making, and insufficient institutional and functionally productive o ­ rganizations. Africa’s capacity and political incentives, culminating in Statistical Capacity Indicator (SCI), which grades inadequate and unreliable ­ fi nancing. country statistical systems on the quality, frequency, These weaknesses were documented in the pre- and timeliness of core economic and social data, cursor report to this one, Poverty in a Rising Africa has now not only caught up with other low-income (Beegle et ­al. 2016). In 2015, and partly in response countries, but even looks already to be edging a bit to the stark findings in that report, the World Bank higher. Sustaining this commitment will be essential ­ committed to put more effort into strengthening to reliably track ­ poverty. the capacity of national statistical systems in low- Building on this experience, the World Bank is income countries, including support for conducting now working on a model that scales up capacity at least one national household survey to measure support for a core package (a “minimum data pack- well-being and poverty every three ­ years. age”) of economic and social statistics in all African There has been substantial progress since that countries. This regional approach aims to harmo- ­ report (figure B1.1.1). More than half of Africa’s nize and benchmark country statistics, to facilitate countries (26) had completed at least one such sur- peer learning, and to scale up ­ i nnovations. These vey by the second quarter of 2018. For the period ongoing and planned efforts complement and build 2018–20, 34 countries (accounting for 76.4 percent on country-developed and owned National Strate- of the population in Africa) have an ongoing or gies for the Development of Statistics ( ­ NSDS). They planned ­ survey. Along with the effort to increase also work with regional and global partners on the the frequency and quality of household surveys to statistics ­agenda. FIGURE B1.1.1  African countries’ poverty status can now be estimated from recent household surveys 30 26 25 24 Number of countries planning surveys completing or 20 17 15 10 9 5 4 1 1 0 ey 10 4 2) an ard 8 19 20 –1 w 01 (Q rv 20 20 20 ) 10 on 2 su 8 re 3 in in in –1 20 No fo (Q d d d 15 ne ne ne Be 20 an an Pl Pl Pl No survey Completed Ongoing or planned Source: World ­Bank. 38   Acceler ating Pover t y Reduc tion in Afric a work on different fronts will be needed ­ qually. This is clearly not always the case e to accelerate Africa’s poverty reduction— (box 1.2). that is, to accelerate the fertility transition With this important caveat in mind, the scenario 1), to make growth more inclusive (­ available data indicate that child poverty (scenario 2), and to improve growth condi- is especially pervasive in ­ A frica. Half of tions in fragile states (scenario 3). Africa’s poor are below the age of 15. This should not ­surprise. At 2.7 percent per year, Africa’s population growth is still high, Africa’s Poverty in Profile and its population is predominantly young Which groups are affected most by pov- (43 percent are below 15 years o ­ ld). Most erty? Where do the poor live, and what do children live in larger households, which they do for a living? How does their status tend to be poorer (Castañeda et ­ a l. 2018; evolve over time? And, how empowered are Newhouse, Suarez-Becerra, and Evans they to change their fate? Answers to these 2016). 8 Childhood poverty affects malnu- questions can provide useful entry points trition, school achievements, and thus the in designing poverty-reducing ­ p olicies. long-term earnings potential for the poor Nonetheless, care must be taken not to over- and the prospects of exiting poverty in findings. They indicate symp- interpret the ­ adulthood (see Fundamentals 1, “Africa’s toms of poverty, not necessarily ­ causes. Human Development ­ Trap”). Greater focus on larger households in fighting poverty is thus called for, both to reduce poverty Which Individuals Are More Likely today and to build the capacity of poor chil- Poor? dren to exit poverty as adults (Watkins and Poverty, as currently assessed, is measured Quattri 2016). at the household level, because this is how Contrary to expectations, the household- consumption data are c ­ ollected. It assumes based consumption data further indicate that members in the same household share that gender gaps in poverty are modest: BOX 1.2  To measure gender and age gaps in poverty, you need to get into the household Our current approach to measuring poverty falls far van de Walle (2017), in this case using nutritional short when it comes to informing on age and gen- poverty. status as a proxy for individual ­ der gaps in p­ overty. Several recent papers use more- Attempting to account for differences in con- detailed consumption data from Africa to move sumption within households greatly increased pov- away from assuming equal intrahousehold sharing erty for women in Malawi (Dunbar, Lewbel, and and to better inform on gender and child ­ poverty. Pendakur 2013) but not in Côte d’Ivoire (Bargain, For example, the pover t y headcou nt rate Donni, and Kwenda 2014). In Senegal, there was increased by 6–7 percentage points in Burundi when inequality in nonfood consumption between men using declared food shares for each household mem- and women but not so much in food consump- ber to correct for intrahousehold allocation pat- tion (Lambert, Ravallion, and van de Walle 2014). terns (Mercier and Verwimp 2017). This was mainly Similarly, no appreciable gender difference in nutri- because a significant additional number of children ent intake inadequacy was found when comparing were counted as poor who had been considered non- household -based adult male equivalent estimates poor under the standard calculations because they with those obtained from individual 24-hour recall were living in nonpoor ­ households. The substantial measures in Ethiopia (Coates et ­ a l. 2017). When presence of poor children in nonpoor households in there were differences, it mainly concerned children Africa is also underscored by Brown, Ravallion, and under three years ­old. P o v e r t y i n A f r i c a   39 Africa’s female share of the poor is about a disadvantage, because it can be correlated the same as the male share (50.2 percent and with wealth and urban status (when deriv- 49.8 ­ percent, respectively) (Munoz Boudet ing from the human immunodeficiency a l. 2018). When looking at the gender et ­ virus and the acquired immune deficiency dimension of poverty by household type, ­ s yndrome [ H I V/A I DS], for ­ e xample). female-headed households are not systemati- Similarly, the internally displaced are not cally poorer, either, and many have seen their al. necessarily always the poorest (Beegle et ­ poverty falling even faster than male-headed 2016). Despite these caveats, systematic, households (Castañeda et ­ al. 2018; Milazzo Africa-wide data about the size and pov- and van de Walle 2017).9 erty status of these groups are hard to come Yet evidence also shows that multiple by. But they often also live geographically ­ structural inequalities confront women rela- concentrated in certain regions (for example, ­ tive to men (such as lower education levels, ethnic groups or pastoralists) or at the out- lower ownership and control over assets, skirts of settlements, as in Ethiopia (World less labor market engagement [linked to Bank, ­ forthcoming). More specific studies gendered time-use patterns], and lower are needed. social indicators), as further discussed in Fundamentals 2 (“The Nexus of Gender Where Do the Poor Live, and What Inequality and ­ Poverty”). A gendered lens Do They Do for a Living? in policy design aimed at reducing poverty is also n­ eeded. More broadly, an enhanced Most of Africa’s poor live in a limited num- focus on intrahousehold allocation pro- ber of countries: 10 out of 48 countries house cesses and on age- and gender-differentiated three-quarters of Africa’s ­ poor.10 These are individual data collection is called for (Doss large countries in terms of overall popula- 2013) (box 1.2). tion, but they are not always the poorest A third demographic trait of poverty, countries in terms of poverty ­ rates. in addition to age and gender, is the edu- Poverty rates are highest in the Sahel cation profile: poor people are consider- countries and the northern regions of the ably less ­ e ducated. Among poor adults in coastal West African countries, extending Africa, two in five have no formal educa- east into Ethiopia and southeast into the tion, reflecting a legacy of poor educational Congo Basin and its eastern surrounding outcomes (Castañeda et ­ a l. 2018). Gross regions in Burundi, Rwanda, Tanzania, and primary school enrollment rates in Africa Uganda (map 1.1, panel a ­ ). These are mostly have increased substantially in the past also landlocked r ­ egions. In some of these two decades (from 73.4 percent in 1996 countries and regions, poverty numbers are to 98.4 percent in 2014). Unfortunately, also high (map 1.1, panel ­ b). Poverty rates learning remains low (World Bank 2018) and numbers are also high in Madagascar and secondary school enrollment limited and ­ Mozambique. Rates and numbers are (gross enrollment of 42.7 percent in 2014). much lower in the higher-income coun- Education, and human development more tries of southern Africa, except for Lesotho, generally, are critical factors for the agenda Eswatini, and Zambia, where poverty rates to reduce poverty in Africa (as discussed high. The low poverty rate in the north- are ­ further in Fundamentals 1). ern regions of the Sahel countries is likely Finally, other demographic and socio- linked to their high urbanization and the cultural traits frequently associated with poor representation of pastoralists in house- higher poverty incidence include orphan- hold ­surveys.11 hood, disability, displacement (internally or Beyond this broad-brush picture, the cor- internationally), and ­ethnicity. The available respondence between poverty rates and the evidence suggests, however, that orphan- number of the poor is ­ l imited. This poses hood, for example, does not always confer a policy c ­ hallenge. The Central African 40   Acceler ating Pover t y Reduc tion in Afric a MAP 1.1  Africa’s poverty and poor are concentrated in a limited number of (often landlocked) countries and regions within these countries a. Poverty rate b. Number of extreme poor 2011 PPP, % of population at