SOCIAL PROTECTION DISCUSSION PAPER No. 2509 | MARCH 2025 State of Social Protection Report 2025 The 2-Billion-Person Challenge Background Paper #2 Adaptive Social Protection Agenda: Lessons from Responses to COVID-19 Shock Emil Daniel Tesliuc Maria Belen Fonteñez © 2025 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: +1 (202) 473 1000; Internet: www.worldbank.org. This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. RIGHTS AND PERMISSIONS The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: +1 (202) 522 2625; e-mail: pubrights@worldbank.org. Abstract The paper examines the social protection response to the COVID-19 pandemic across 76 emerging and developing economies (EDEs) to identify lessons on how to make these systems more resilient against risks, shocks, and crises at the individual, household, or national level. The COVID-19 pandemic triggered significant expansions in social protection systems across EDEs, with responses varying based on countries’ existing infrastructure and income levels. The analysis of 76 EDEs revealed that countries used approximately 37 percent of their social protection programs to respond to COVID-19, with social assistance programs being the most frequently used response (73 percent of total programs). EDEs increased their real per capita social protection spending by an average of 28 percent, with low-income countries (LICs) and high-income countries (HICs) showing the largest increases at around 40 percent and 32 percent, respectively. The effectiveness of responses was strongly correlated with preexisting social protection systems, economic conditions, labor market factors, and digital infrastructure. Countries with more developed social protection systems, formal labor markets, and digital payment infrastructure before the pandemic were better positioned to rapidly scale up their responses, highlighting the importance of maintaining robust routine social protection programs and delivery systems to enable effective crisis response. JEL codes: H53, I38, J28, O15 Keywords: social protection system, COVID-19 pandemic, evolution of social protection spending, shock response 2 Acknowledgments This paper was prepared by the Social Protection Global Unit at the World Bank as a part of its flagship report ‘State of Social Protection Report 2025: The 2-Billion-Person Challenge.’ (World Bank, 2025). The paper was authored by Emil Daniel Tesliuc and Maria Belen Fonteñez, with contributions from Claudia Patricia Rodríguez Alas, Robert Palacios, Johanna Estefania Andrango Brito, and Ana Sofia Martinez Cordova. The authors are grateful for the valuable collaboration with the peer authors who led other background papers on social protection, social assistance, social insurance, labor market, and gender. This paper was made possible thanks to the Atlas of Social Protection: Indicators of Resilience and Equity (ASPIRE), which served as the primary data source for the analyses. The team is also deeply grateful for the guidance provided by the peer reviewers: Aline Coudouel, David Coady (IMF), Carolina Diaz-Bonilla, Phillippe Leite, Anita Schwarz, Joana Silva, and Ruslan Yemtsov, at the concept stage, and Brooks Evans (IMF), Ugo Gentilini, Ruth Hill, Harry Edmund Moroz, and Joana Silva, at the decision stage. The report was produced under the guidance of Iffath Sharif (Global Director, Social Protection), Michal Rutkowski (former Global Director, Social Protection), and Loli Arribas- Baños (Practice Manager, Social Protection). Lastly, the team benefited from valuable support from Matthew Naumann (editor), Fiona Mackintosh (acquisitions editor), Agnes Nderakindo Mganga (Program Assistant), and Alexandra Humme (Senior External Affairs Officer). Table of Contents Abstract .................................................................................................................. 2 Acknowledgments ................................................................................................... 3 I. Introduction .......................................................................................................... 7 II. Macroeconomic Context 2017–2022 ..................................................................... 9 III. The Evolution of Social Protection Spending before, during, and after the COVID-19 Crisis (2017–2022) ................................................................................................. 12 IV. Zooming into 2020–2021: The Social Protection System Responses of EDEs to the COVID-19 Shock .................................................................................................... 17 V. What Economy or Program Characteristics Were Associated with a Stronger or Faster Response to COVID-19? ......................................................................................... 26 VI. What Factors Were Associated with a Stronger Response to the COVID 19 Shock? Suggestions for Strengthening ASP in EDEs .............................................................. 33 Statistical Annex .................................................................................................... 37 References ............................................................................................................ 46 Figures Figure 2.1. Evolution of Per Capita GDP in Emerging and Development Economies, 2017–2022 .............................................................................................................. 9 Figure 2.2. GDP Growth Rate in EDEs 2017–2022, by Income Group .......................... 10 Figure 2.3. Evolution of Inflation, Consumer Prices (Annual Percentage Rise) in EDEs 2017–2022, by Income Group ................................................................................. 11 Figure 2.4. Evolution of Extreme Poverty Headcount across EDEs, 2017–2022, by Income Group ....................................................................................................... 12 Figure 3.1. Evolution of Real Per Capita Social Protection Expenditure, by Social Protection Area, 2017–2022 in Constant US$ PPP 2017 ............................................ 14 Figure 3.2. Social Protection Expenditure as Percentage of GDP in EDEs, by Social Protection Area, 2017–2022 .................................................................................... 15 Figure 3.3. Evolution of Social Protection Spending as Percentage of GDP, by Income Group, 2017–2022, Base 2017 = 100, Unweighted .................................................... 16 Figure 3.4. Social Protection Expenditure as Percentage of GDP, 2017–2022 .............. 17 4 Figure 4.1. Number and Share of Programs Used During COVID-19, by Total, Income Group, and Region ................................................................................................. 19 Figure 4.2. Expenditure in US$ PPP 2017 Per Capita, 2019, and Increase during COVID- 19, by Region, Income Group, and Total EDEs .......................................................... 22 Figure 4.3. Expenditure as Percentage of GDP, 2019, and Increase during COVID-19, by Region, Income Group, and Total EDEs .................................................................... 23 Figure 4.4. Expenditure as Percentage of GDP, 2019, and Increase during COVID-19, by Social Protection Area, Region, Income Group, and Total EDEs ................................. 25 Figure 5.1. Social Protection Expenditure, 2019, and during COVID-19, by Social Protection Area and Income Group ......................................................................... 27 Figure 5.2. Increase in Expenditure in Per Capita Constant US$ PPP 2017 during the COVID-19 Pandemic and Gross National Income per Capita, by Economy and Income Group ................................................................................................................... 28 Figure 5.3. Increase in Expenditure Per Capita Constant US$ PPP 2017 during the COVID- 19 Pandemic and Change in 2020 GDP, by Economy and Income Group .................... 29 Figure 5.4. Increase in Expenditure in Per Capita Constant US$ PPP 2017 during COVID- 19 and Rural Population, by Economy and Income Group ......................................... 30 Figure 5.5. Increase in Expenditure in Per Capita Constant US$ PPP 2017 during COVID- 19 and Wage Employment, by Economy and Income Group ...................................... 31 Figure 5.6. Increase in Expenditures in Per Capita Constant US$ PPP 2017 during COVID- 19 and Account Ownership, by Economy and Income Group..................................... 32 Figure 5.7. Increase in Expenditures in Per Capita Constant US$ PPP 2017 during COVID- 19 and Mobile Ownership, by Economy and Income Group ....................................... 33 Figure A1.1. GDP Growth Rate in EDEs, 2017–2022, by Economies’ Income Group ..... 37 Figure A1.2. Evolution of Social Protection Expenditure as a Percentage of GDP, 2017– 2022 ..................................................................................................................... 37 Figure A1.3. Evolution of Real Per Capita Social Protection Expenditure, by Income Group and Social Protection Area, 2017–2022, in US$ PPP 2017 ......................................... 40 Figure A1.4. Evolution of Real Per Capita Social Protection Expenditure, by Region and Social Protection Area, 2017–2022, in US$ PPP 2017 ................................................ 41 Figure A1.5. Evolution of Social Protection Expenditure as Percentage of GDP, by Region and Social Protection Area, 2017–2022 ................................................................... 42 5 Figure A1.6. Expenditure in Constant US$ PPP 2017 during COVID-19, by Social Protection Area, Region, Income Group, and Total EDEs ........................................... 43 Figure A1.7. Evolution of Social Protection Spending as Percentage of GDP, by Region, 2017–2022, ........................................................................................................... 45 Tables Table 4.1. Programs Used by Economies to Respond to COVID-19, by Social Protection Area ...................................................................................................................... 20 Table 6.1. Factors Associated with Stronger Social Assistance and Labor Market Responses to COVID-19 ......................................................................................... 33 Table A.1. Country-Level Expansion during COVID-19 ............................................... 38 6 I. Introduction The COVID-19 pandemic was an unprecedented covariate shock that triggered an unparalleled global social protection response, thereby offering a unique range and depth of lessons for the future design and implementation of adaptive social protection (ASP) systems. Due to the public health threat as well as government measures such as lockdowns and mobility restrictions to check the virus’ spread, the effects of the COVID-19 shock reverberated rapidly around the globe, heavily disrupting economic activity in most economies as well as global trade flows, and damaging the livelihoods of billions of people (ILO 2022; World Bank 2022b). To cope with the economic and social crisis that resulted from this shock, social protection systems were deployed at an unmatched scale to protect the livelihoods of those affected, a group much larger in size than the typical social protection caseload in normal times (Bastagli and Lowe 2021; Beazley, Marzi, and Steller 2021; Gentilini 2022; ILO 2021). This background paper to the 2025 State of Social Protection report offers new data and insights on social protection responses to COVID-19 in emerging and developing economies (EDEs) to draw lessons for the development of ASP systems. The lessons are primarily based on detailed program-level administrative data from 1,774 programs in 76 EDEs using World Bank ASPIRE data. The background paper explores EDEs’ responses across all types of social protection programs and for the population as a whole, thereby complementing past studies exploring the pandemic response for particular social protection instruments (for example, Beazley, Bischler, and Doyle 2021 and Gentilini 2022, on cash transfers; Kamran et al. 2023, on labor market programs; World Bank 2022b, on the impact of the fiscal policy, of which social protection policies, in protecting households) or population groups (for example, Alfers, Chen, and Plagerson 2022 and Alfers, Ismail, and Valdivia 2020, on informal workers; Gavrilovic et al. 2022 on women and girls; and WFP/IPC-IG/UNICEF 2021, on certain categories of refugees and migrants). 7 To understand the extent and nature of EDEs’ system adaptation to face the COVID- 19 crisis, the paper tracks the evolution of social protection spending between 2017 and 2022 and holistically examines how EDEs leveraged their social protection systems to respond to this shock. It examines which types of social protection programs EDEs used to expand their caseloads and/or spending and identifies factors associated with effective ASP systems. The focus and main contribution of the paper is to track and analyze the evolution of social protection spending, at the program level, by categories of programs (labor market, social insurance, and social assistance programs) and the aggregate social protection spending. Among the multiple indicators that could be used to track the changes in spending, this background paper focuses on the change in real spending per capita (expressed as spending on social protection in US$ 2017 PPP 1) and the change in relative spending (social protection spending as a percentage of GDP 2). While program-level coverage is examined, creating comprehensive coverage measures for program categories or overall social protection proved challenging due to inevitable double-counting of recipients at individual or household levels. Although household survey data typically serves as the primary source for tracking social protection coverage across program groups, limited availability of such data before and after the COVID-19 shock means that the results, presented in Okamura et. al (2025), remain partial and primarily reflect middle-income countries. The background paper is organized as follows: Section II reviews the macroeconomic context before, during, and after the onset of the COVID-19 shock (2017–2022). Section III presents the evolution of social protection spending during the same period. Section IV presents the types of programs used by EDEs to respond to the COVID-19 shock and quantifies—for the first time in the literature—the expansion of social protection system spending for EDEs as a whole and at the country level. Section V identifies the economy or program characteristics that are correlated with a stronger or faster response to 1 PPP = Purchasing power parity. 2 GDP = Gross domestic product. 8 COVID-19. Finally, Section VI distills the lessons learned from the COVID-19 social protection expansion by reviewing the factors associated with a stronger response to the COVID-19 shock and offers suggestions for strengthening ASP in EDEs. II. Macroeconomic Context 2017–2022 The evolution of EDEs’ social protection spending before, during, and after the onset of COVID-19 can be better understood when contextualized within the main macroeconomic developments in the same period. The macroeconomic environment of the EDEs between 2017 and 2022 underwent two distinct periods: a period of sustained growth and stable macroeconomic environment until 2020, followed by a volatile macroeconomic environment between 2020 and 2022. The stability was interrupted by the COVID-19 shock in 2020–2021 and continued to be affected by other large covariate shocks in 2022. 3 This section analyzes the evolution of economic growth, inflation, and poverty between 2017 and 2022, as they comprise the backdrop to a holistic examination of trends in social protection expenditure during that period. Figure 2.1. Evolution of Per Capita GDP in Emerging and Development Economies, 2017–2022 A. Average Year-to-Year Change, B. Average Growth in GDP per Capita US$ PPP GDP Per Capita US$ PPP 2017, Unweighted 2017, (Base 2017 = 100), Unweighted 106 105.1 110 102.7 103.3 107.7 104 102.1 108 102 106 104.9 104.2 100 104 102.7 98 102 96 94.5 100.0 100 99.2 94 92 98 90 96 88 94 2018 2019 2020 2021 2022 2017 2018 2019 2020 2021 2022 Source: Original elaboration for this publication using World Bank, World Development Indicators. Note: The graph uses simple averages. Graph based on 76 observations, including 63 low- and middle- income countries (LMICs) and 13 high-income countries (HICs) monitored by ASPIRE. GDP = Gross domestic product; PPP = Purchasing power parity. 3 Inflation in 2022 was at the highest levels in more than 25 years, as well as food insecurity, partly due to the global fallout of the Russian Federation’s invasion of Ukraine. 9 After years of roughly stable growth at approximately 2–3 percent per year just before the onset of COVID-19, EDEs’ per capita GDP fell by 5.5 percent in 2020 then rebounded in 2021 (Figure 2.1A). Other indicators, such as the evolution of GDP in constant PPP, show impacts of similar magnitude. 4 Economic growth resumed in 2021, from a lower base, across most EDEs; the average growth rate of GDP per capita in 2021 was 5.1 percent (Figure 2.1B). Compared to pre-COVID-19 levels, EDEs’ GDP per capita recovered in 2021 and continued to grow in 2022 (Figure 2.1B). As described by Alon et al. (2023), emerging economies registered a steep short- term contraction in GDP as the pandemic took hold in 2020, resulting in a U-shaped pattern (Figure 2.2). Across countries grouped by their level of income, the higher the income group, the higher the average contraction. Upper-middle income economies (UMICs) and HICs registered the deepest economic decline (−7.1 percent and −5.3 percent, respectively), LMICs experienced the least decline (−2.1 percent), while low- income countries (LICs) were not affected significantly (+0.7 percent). Figure 2.2. GDP Growth Rate in EDEs 2017–2022, by Income Group 5 10.0 Percentage grrowth in GDP 5.0 0.0 2017 2018 2019 2020 2021 2022 -5.0 -10.0 Low-income (N=11) Lower-middle-income (N=21) Upper-middle-income (N=31) High-income (N=13) Total (N=76) Source: Original elaboration for this publication using World Bank, World Development Indicators. 4 Sensitivity analysis was conducted by estimating the evolution of GDP in US$ 2017 PPP. 5 Figure 2.2 reflects the growth pattern of 76 EDEs for which we had detailed program-level social protection data; the same pattern is observed for all EDEs (150), see Statistical annex. 10 Note: The graph uses simple averages. Graph based on 76 observations, including 63 low- and middle- income countries and 13 high-income countries monitored by ASPIRE. EDEs = Emerging and developing economies; GDP = Gross domestic product. Inflation also spiked, from an average of around 4–7 percent before COVID-19 to over 14 percent in 2022. Consumer price inflation increased from a global EDEs’ average of about 6 percent per year during 2017–2019, to about 10 percent during 2020 and 2021, up to 14.1 percent in 2022 (Figure 2.3). This increase played out differently in LICs and HICs. Before the COVID-19 shock, inflation was generally low for HICs and UMICs and higher and more volatile in LMICs and LICs. LICs and LMICs experienced the sharpest rise in consumer price inflation in 2020 (the 2020 spike in LMICs’ inflation was driven by Zimbabwe), while UMICs and HICs did not experience it until 2021. In other words, high inflation has been a feature since the beginning of the period only for LICs and LMICs; however, by 2022, inflation was high, on aggregate, in all groups of countries. Figure 2.3. Evolution of Inflation, Consumer Prices (Annual Percentage Rise) in EDEs 2017–2022, by Income Group 35 30 Inflation, annual (percentage) 25 20 14.1 15 11.5 10 7.2 6.8 4.3 4.2 5 0 2017 (N=73) 2018 (N=73) 2019 (N=72) 2020 (N=71) 2021 (N=70) 2022 (N=70) Low-income Lower-middle-income Upper-middle-income High-income Total Source: Original elaboration for this publication using World Bank, World Development Indicators. Note: The graph uses simple averages. Graph based on 76 observations, including 63 low- and middle- income countries and 13 high-income countries monitored by ASPIRE. EDEs = Emerging and developing economies. 11 The COVID-19 shock also reversed the fall in extreme poverty across all EDEs, with the largest increases in poverty experienced by LICs and LMICs (World Bank 2022a; 2022b). During COVID-19, EDEs suffered the greatest rise in poverty since the Asian financial crisis in 1997–1998. Based on the most recent data from the Poverty and Inequality Platform of the World Bank as of November 21, 2024, the average extreme poverty headcount across the 150 EDEs had fallen from 14.2 percent in 2017 to 13.7 percent in 2019. These gains were reversed by the COVID-19 pandemic, with the extreme poverty headcount increasing to 14 percent in 2020. While extreme poverty dropped to 13.8 percent in 2022, it was still above the pre-COVID-19 level. Figure 2.4. Evolution of Extreme Poverty Headcount across EDEs, 2017–2022, by Income Group Poverty Headcount Ratio (percentage) 50 43.4 45 41.1 39.8 40.8 40.1 38.9 40 35 30 25 20 15 10.6 9.7 9.6 9.6 9.4 9.2 10 10.7 9.5 9.1 9.5 8.9 8.5 5 2.2 1.9 1.8 2.1 1.9 1.7 0 1.1 0.5 0.6 0.5 0.5 0.4 2017 (N=69) 2018 (N=69) 2019 (N=67) 2020 (N=70) 2021 (N=69) 2022 (N=67) Low-income Lower-middle-income Upper-middle-income High-income Total Source: Original elaboration for this publication using World Bank, World Development Indicators. Note: The graph uses simple averages. Graph based on 76 observations of countries with available data, including 60 low- and middle-income countries and 13 high-income countries monitored by ASPIRE. EDEs = Emerging and developing economies. III. The Evolution of Social Protection Spending before, during, and after the COVID-19 Crisis (2017–2022) What was the trajectory of social protection expenditure in the years before, during, and after the COVID-19 crisis? This section presents the evolution of social protection spending over 2017–2022, in both real and relative terms. It examines the change in absolute real per capita spending (in constant purchasing power parity: US$ 2017 PPP) 12 and in relative terms, as social protection spending as a share of GDP. The main source of data used in this section is the detailed ASPIRE administrative dataset on spending on social protection programs between 2017 and 2022 in 76 EDEs. Looking first at real social protection expenditure, constant increases were already evident in the years before COVID-19, then jumped substantially with the crisis, and remained higher in 2022 than in 2019 (Figure 3.1). Social protection spending expressed in constant 2017 PPP increased substantially during the COVID-19 years, as an economic stabilizer reacting to the fall in GDP. The average increase in real terms of social protection spending from 2019 to the peak expenditure in 2020 or 2021 was, on average, about 22.2 percent in per capita terms and about 19.2 percent for overall real spending. On average, this increase amounted to US$7.7 billion (in constant US$ PPP 2017) at the peak of COVID-19. The response to the COVID-19 shock combined an asymmetric parametric adjustment: the purchasing power of social insurance programs was kept relatively constant (amid a fall in GDP), while per capita spending on labor market 6 and social assistance programs increased by 207 percent and 52 percent, respectively. The latter sustained, in most economies, a significant horizontal expansion. 7 During 2021 and 2022, real social protection spending was more or less maintained at this higher plateau, and it did not increase in line with GDP growth. 6 This background paper only quantifies expenditure on supply-side labor market programs. Expenditure on demand-side labor market interventions, such as firm liquidity support programs (tax reliefs, credit facilities, and loan payment facilities), are out of the scope of this background paper. However, Kamran et. al (2023) estimated that firm liquidity support policies accounted for most of the labor market spending, amounting to a global average of US$32.1 billion in 2020 and covering about 217,000 firms, in the sample of 61 economies with data on measures toward firms. 7 Horizontal expansions in social protection systems refers to permanent or temporary increases in the number of program beneficiaries. 13 Figure 3.1. Evolution of Real Per Capita Social Protection Expenditure, by Social Protection Area, 2017–2022 in Constant US$ PPP 2017 1,500 107.7 115 102.7 104.9 104.2 100.0 99.2 1,250 in constant USD PPP $2017 1,100.1 100 change year over year 1,062.6 1,003.4 1,000 850.3 887.7 835.1 85 750 682.4 749.5 723.5 70 612.6 623.2 648.2 500 55 250 279.1 284.9 190.1 194.8 204.7 244.0 32.3 32.3 34.8 101.0 65.7 35.8 0 40 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) LM SA SI GDP pc YoY - base 2017 = 100 Source: Original elaboration for this publication using Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire. Note: This graph uses simple averages. Graph based on a total number of observations of 76 countries, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. The social protection spending for 2022 was estimated for the following countries and economies: Benin, Cameroon, Côte d’Ivoire, Guinea-Bissau, Niger, Togo, Zimbabwe, Türkiye, Colombia, Dominican Republic, Ecuador, and West Bank and Gaza. LM = labor market programs; SA = social assistance; SI = social insurance; GDP pc YoY = Gross domestic product per capita, year-over-year; PPP = Purchasing power parity. When assessed in relative terms, social protection spending as a percentage of GDP was stable in the years before the pandemic hit and saw a major spike in 2020, which was partly reversed in the subsequent years (outpaced by GDP growth). Between 2017 and 2019, EDEs’ social protection spending was quite stable, mirroring the overall macroeconomic stability (and constant GDP growth) during those years. Countries spent on average around 5 percent of GDP on social protection during the pre-COVID period (Figure 3.2). By keeping the same share of social protection spending as a percentage of GDP over 2017–2019, economies passed some of the growth dividend to the beneficiaries of social protection programs, and the increase was relatively even across social protection areas (social assistance, social insurance, and labor market programs). However, in 2020, social protection spending as a percentage of GDP jumped to 6.4 14 percent, due to both the COVID-19-induced decline in GDP and the significant social protection response to the crisis, which varied by instrument (discussed further in section IV). After 2020, however, the relative spending on social protection fell in both 2021 (to 5.9 percent of GDP, −0.5 percentage points) and 2022 (to 5.3 percent of GDP, −0.6 percentage points), once the peak responses to the COVID-19 crisis had ended, resulting in an inverted U-trend in social protection spending. Figure 3.2. Social Protection Expenditure as Percentage of GDP in EDEs, by Social Protection Area, 2017–2022 10 115.7 119.9 120 109.0 108.0 9 104.8 110 8 100.0 Change year over year 100 Percentage of GDP 7 6.4 6 5.9 90 3.9 5.3 5.1 5.0 5.1 3.8 5 80 3.5 3.5 3.7 3.5 4 70 3 60 2 2.0 1.7 50 1 1.3 1.3 1.3 1.5 0 0.2 0.2 0.2 0.5 0.3 0.2 40 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) LM SA SI GDP YoY - base 2017 = 100 Source: Original elaboration for this publication using Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire. Note: This graph uses simple averages. Graph based on a total number of observations of 76 countries, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. The social protection spending for 2022 was estimated for the following countries and economies: Benin, Cameroon, Côte d’Ivoire, Guinea-Bissau, Niger, Togo, Zimbabwe, Türkiye, Colombia, Dominican Republic, Ecuador, and West Bank and Gaza. LM = labor market programs; SA = social assistance; SI = social insurance. GDP YoY = Gross domestic product, year-over-year; EDEs = Emerging and developing economies. The inverted U-shape trend of social protection spending—a sudden spike during the pandemic response and then a retraction in the percentage of GDP spent on social protection—is evident across all income groups (Figure 3.3). This is partly influenced by variation in GDP performance across economies at different income levels, but it also reflects differences in the extent of their social protection system 15 expansion during the pandemic (discussed in section IV.). The reduction in relative spending occurred in 2022 when the COVID-19 emergency was most pronounced in LICs, UMICs, and HICs. After the peak pandemic response, social protection spending as a percentage of GDP was still slightly higher in 2022 compared to pre-pandemic levels, across most income groups except LMICs. The inverted U shape of social protection spending can also be observed in most regions, though the size and timing of their pandemic responses varied extensively (see Statistical annex). East Asia and Pacific and Latin America and the Caribbean were the regions with the largest increases: on average, compared to 2019, the growth was 52 percent in 2020 and 64 percent in 2021 in East Asia and Pacific and 44 percent in 2020 and 39 percent in 2021 in Latin America and the Caribbean. At the income group level, there was also extensive variation across EDEs (as illustrated by the widely varying gray trend lines in Figure 3.4). Figure 3.3. Evolution of Social Protection Spending as Percentage of GDP, by Income Group, 2017– 2022, Base 2017 = 100, Unweighted 140 120 100 80 60 40 20 0 Low-income Lower-middle- Upper-middle- High-income Total income income 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) Source: Original elaboration for this publication using Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire. Note: This graph uses simple averages. Graph based on a total number of observations of 76 countries, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. The social protection spending for 2022 was estimated for the following countries and economies: Benin, Cameroon, Côte d’Ivoire, Guinea-Bissau, Niger, Togo, Zimbabwe, Türkiye, Colombia, Dominican Republic, Ecuador, and West Bank and Gaza. GDP = Gross domestic product. 16 Figure 3.4. Social Protection Expenditure as Percentage of GDP, 2017–2022 18 16 14 Percentage of GDP 12 10 8 6.4 5.9 5.1 5.0 5.1 5.3 6 4 2 0 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) Source: Original elaboration for this publication using Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire. Note: This graph uses simple averages. Graph based on a total number of observations of 76 countries, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. The social protection spending for 2022 was estimated for the following countries and economies: Benin, Cameroon, Côte d’Ivoire, Guinea-Bissau, Niger, Togo, Zimbabwe, Türkiye, Colombia, Dominican Republic, Ecuador, and West Bank and Gaza. GDP = Gross domestic product. IV. Zooming into 2020–2021: The Social Protection System Responses of EDEs to the COVID-19 Shock As an unprecedented global shock, the COVID-19 pandemic triggered a significant, though varied, response from social protection systems across EDEs. Countries achieved social protection system expansions by increasing their financing, caseloads (horizontal expansions) and the value of benefits provided (vertical expansions), with the specific mix depending on economy contexts and income levels. Those entering the pandemic with more developed social protection systems, formal labor markets, and digital payment infrastructure were better positioned to rapidly scale up their social protection responses by expanding existing programs and/or introducing new programs. By understanding how social protection systems responded to the COVID-19 shock and which economy, program and wider contextual characteristics were associated with a stronger response, EDEs can draw lessons and take actions to be better prepared in a time of permacrisis and recurrent and continued global and regional shocks. 17 Using ex post data on social protection spending and beneficiaries in large and medium-size social protection programs, this section examines the extent to which and ways in which EDEs expanded their social protection systems in 2020 and 2021 to respond to the COVID-19 pandemic. The analysis tracks the size of the pandemic response and breaks it down for 76 EDEs, drawing from data on large and medium-size social assistance, labor market, and social insurance programs in each economy. For each social protection program, this section compares the ex post spending based on the figures recorded during the peak year of the economy’s pandemic response (either 2020 or 2021). The reported increases in spending during the peak pandemic response measure the magnitude of the increase in both indicators, relative to 2019 levels. IV.1. Which Social Protection Programs Were Used by EDEs to Respond to COVID- 19? EDEs used about 37 percent of their social protection programs to respond to COVID-19, although this varied across income groups and regions (Figure 4.1). Analysis of 1,774 programs across 76 EDEs captured in ASPIRE 8 sheds light on how many programs were used as a crisis response tool. LICs and UMICs on average used a higher proportion of their existing programs (41 and 42 percent, respectively), while LMICs and HICs used 36 and 24 percent. Regionally, East Asia and Pacific used more programs than any other region—more than half of their programs were implemented to respond to COVID-19, followed by South Asia which used 42 percent of its programs. On the other hand, Sub-Saharan Africa, Europe and Central Asia and Latin America and the Caribbean used about one-third of their social protection programs for the crisis response, with Middle East and North Africa using only 22 percent, the lowest percentage of all regions. 8 From the full set of social protection programs, those used to respond to COVID-19 have been identified from previous studies (from social protection trackers used for De La Flor Giuffra, 2021; Gentilini et al. 2020, 2021, 2022; Kamran et al. 2023) and added to ASPIRE. In addition, existing programs in ASPIRE were tagged as COVID-19 responses if they had an increase of more than 25 percent in expenditure and/or beneficiaries in 2020 or 2021 compared to 2019. 18 Figure 4.1. Number and Share of Programs Used During COVID-19, by Total, Income Group, and Region A. Total EDEs B. By Income Group and Region 2,500 37% 40% 70 60% 52% 35% 60 50% 2,000 1,774 42% 42% 30% 41% Percentage of programs Percentage of programs 50 Number of programs Number of programs 36% 34% 35%35% 40% 1,500 25% 649 40 36 20% 30% 2626 27 22% 30 13 1,000 1,125 15% 22 24% 6 20 20% 11 17 16 9 20 15 23 15 10% 8 20 500 6 8 6 8 18 3 10% 5% 10 14 15 11 11 11 9 8 0 0% 0 0% Upper-middle-income (N=31) East Asia and Pacific (N=8) Europe and Central Asia (N=23) Middle East and North Africa (N=8) Latin America and the Caribbean (N=12) South Asia (N=6) Low-income (N=11) High-income (N=13) Lower-middle-income (N=21) Sub-Saharan Africa (N=19) Total (N=76) Source: Original elaboration for this publication using Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire Note: The graph uses simple averages. Graph based on a total number of observations of 76 countries, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. The number of observations for each income group and region is based on data availability. EDEs = Emerging and developing economies. The response to the COVID-19 shock differed across social assistance, labor market, and social insurance programs. Table 4.1 shows the great variability in social protection instruments used across regions and income groups. Globally across EDEs, social assistance programs were the most frequent response to COVID-19 (73 percent of 19 the total number of programs used). This is not surprising as programs 9 represented 37 percent of total social protection programs before COVID-19, making the scaling up of cash programming during the pandemic more feasible (particularly when combined with the operational challenges of in-kind provision in the context of lockdowns and social distancing) (Gentilini 2022; Lowe, McCord, and Beazley 2021). Social assistance programs comprised a higher proportion of the response in LICs (88 percent) and a lower proportion in HICs (54 percent); the latter group also leveraged a higher proportion of labor market programs (36 percent). LMICs used similar proportions of programs as UMICs (approximately three-quarters of which were social assistance programs and a little under one-quarter labor market programs). By region, Sub-Saharan Africa and South Asia used a larger percentage of social assistance programs to ameliorate the effects of the pandemic (83 and 78 percent, respectively) compared to other regions with available data. Europe and Central Asia on the other hand reports the smallest proportion of social assistance programs used (67 percent) and instead has the highest share of labor market programs (27 percent) compared to other regions. Table 4.1. Programs Used by Economies to Respond to COVID-19, by Social Protection Area Percentage of Programs Composition of Social Used with Respect to Total Protection Programs Used Number of Programs LM SA SI SP LM SA SI SP Low-income (N = 11) 25 50 6 41 10 88 1 100 Lower-middle-income (N = 21) 22 42 28 36 15 77 8 100 Upper-middle-income (N = 31) 40 42 45 42 21 74 6 100 High-income (N = 13) 32 19 40 24 36 54 10 100 Sub-Saharan Africa (N = 19) 19 43 13 34 14 83 4 100 East Asia and Pacific (N = 8) 54 52 45 52 21 72 7 100 Europe and Central Asia (N = 23) 40 32 50 35 27 67 6 100 Latin America and the Caribbean (N = 12) 27 38 36 35 18 76 5 100 Middle East and North Africa (N = 8) 36 41 27 38 17 70 13 100 South Asia (N = 4) 24 48 42 42 12 78 10 100 Total (N = 76) 33 38 32 37 20 73 6 100 9 This includes unconditional and conditional cash transfers, public works and non-contributory social pensions, among other programs with cash payments. 20 Source: Original elaboration for this publication using Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire Note: The table uses simple averages. Graph based on a total number of observations of 76 countries, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. Number of observations for each income level group and region is based on data availability. LM = Labor market; SA = Social assistance; SI = Social insurance; SP = Social protection. IV.2. What Happened with Social Protection Program Expenditure during 2020– 2021? To respond to COVID-19, EDEs increased their real per capita social protection spending, on average, by about 28 percent. Per capita spending increased the most at the extreme ends of the income spectrum, increasing by about 40 percent in LICs and 32 percent in HICs (although from very different base levels), compared to 11 percent in LMICs and 29 percent in UMICs (Figure 4.2). However, sharp differences arise when analyzing the data at income level. For example, Liberia, Chile, and Indonesia almost tripled their real per capita spending, while in Sri Lanka, budget per capita decreased by about 2 percent, respectively. By region, East Asia and Pacific saw the largest growth in expenditure in the sample, with an 89 percent increase relative to real social protection per capita spending in 2019 (though these base levels varied greatly, being far higher in Europe and Central Asia and Latin America and the Caribbean than in other regions in the sample). By contrast, Middle East and North Africa and South Asia spending barely showed an increase—on average, only 16 and 8 percent, respectively. Similar trends are observed when social protection spending is assessed in relative terms, as a percentage of GDP, though in this case South Asia has a relatively larger budget increase than Europe and Central Asia (explained by its relatively larger decline in GDP and its larger base level) (Figure 4.3). 21 Figure 4.2. Expenditure in US$ PPP 2017 Per Capita, 2019, and Increase during COVID-19, by Region, Income Group, and Total EDEs A. Total Social Protection Expenditure in US$ PPP B. Increase in Social Protection Expenditure, 2017, Per Capita Percentage 3,205 3,500 100 Percentage change during COVID-19 89 90 3,000 773 in constant USD PPP $2017 80 2,500 2,144 70 2,432 58 2,000 60 361 49 50 1,500 1,784 1,279 40 1,172 1,156 1,140 40 906 29 32 28 1,000 420 88 30 265 1,068 253 20 393 907 427 858 888 20 16 500 225 204 11 8 39 39 83 479 10 354 142 29 176 0 0 11 28 South Asia (N= 4) Total (N= 76) Low-income (N= 11) Lower-middle-income (N= 21) Sub-Saharan Africa (N= 19) Upper-middle-income (N= 31) High-income (N=13) East Asia and Pacific (N= 8) Europe and Central Asia (N= 23) Middle East and North Africa (N= 8) Latin America and the Caribbean (N= 12) South Asia (N= 4) Total (N= 76) Low-income (N= 11) Lower-middle-income (N= 21) Sub-Saharan Africa (N= 19) Upper-middle-income (N= 31) High-income (N=13) East Asia and Pacific (N= 8) Europe and Central Asia (N= 23) Middle East and North Africa (N= 8) Latin America and the Caribbean (N= 12) 2019 Increase in per capita real spending Source: Original elaboration for this publication using Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire Note: Comparison peak COVID-19 spending. The graph uses simple averages and is based on a total number of observations of 76 countries, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. The number of observations for each income group and region is based on data availability. Aggregated indicators are calculated using simple averages of country-level expenditure in US$ PPP 2017, by income group and region. PPP = Purchasing power parity; EDEs = Emerging and developing economies. 22 Figure 4.3. Expenditure as Percentage of GDP, 2019, and Increase during COVID-19, by Region, Income Group, and Total EDEs A. Total Social Protection Expenditure, Percentage of B. Increase in Social Protection Expenditure, GDP Percentage 12 10.9 90 Percentage change during COVID-19 81 10.1 80 10 2.8 8.0 1.6 7.9 8.3 70 Percentage of GDP 8 60 52 8.5 1.5 6.6 2.0 8.1 5.8 2.7 50 6 6.8 1.6 39 37 4.4 6.0 37 2.6 40 33 35 31 4 5.2 3.0 5.1 0.7 30 2.4 22 2.1 3.7 18 18 0.8 20 2 0.6 3.2 0.6 1.8 2.2 10 1.5 0 0 South Asia (N= 4) Total (N= 76) Low-income (N= 11) Lower-middle-income (N= 21) Sub-Saharan Africa (N= 19) Upper-middle-income (N= 31) High-income (N=13) East Asia and Pacific (N= 8) Europe and Central Asia (N= 23) Middle East and North Africa (N= 8) Latin America and the Caribbean (N= 12) South Asia (N= 4) Total (N= 76) Low-income (N= 11) Lower-middle-income (N= 21) Sub-Saharan Africa (N= 19) Upper-middle-income (N= 31) High-income (N=13) East Asia and Pacific (N= 8) Europe and Central Asia (N= 23) Middle East and North Africa (N= 8) Latin America and the Caribbean (N= 12) 2019 Increase in spending as percentage of GDP Source: Original elaboration for this publication using Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire Note: Comparison peak COVID-19 spending. The graph uses simple averages and is based on a total number of observations of 76 countries, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. The number of observations for each income group and region is based on data availability. Aggregated indicators are calculated using simple averages of country-level expenditure as percentage of GDP by income group and region. GDP = Gross domestic product; EDEs = Emerging and developing economies. When comparing social protection areas, labor market programs saw the largest overall increase in real per capita spending in the sample, but this was driven by higher-income economies, whereas increases in social assistance budgets were more widespread across economy income groups. During peak COVID-19, real per capita spending on labor market programs increased by 207 percent (albeit from a low 23 base), while social assistance budgets increased by 52 percent and social insurance programs saw little spending change, increasing by only 11 percent in the sample (Figure 4.4). These averages mask substantial variation across economy income groups and regions. The expansion in real per capita social protection expenditure in LICs is mostly explained by social assistance, while in HICs it was primarily due to the increase in labor market program budgets. The increase in LMICs’ and UMICs’ spending showed almost similar proportions of increases in social assistance, labor market, and social insurance program. Regionally, the increase in per capita spending in Sub-Saharan Africa 10 and Europe and Central Asia was driven by labor market program budget increases, whereas in Latin America and the Caribbean and, to a lesser extent, in South Asia, it was driven by social assistance. The increase in per capita spending in the East Asia and Pacific and Middle East and North Africa sample was mainly due to higher social insurance program spending. When looking at the duration of pandemic responses through different instruments, evidence from system-wide administrative data up until 2022 suggests that the duration of the social assistance response tended to be longer. Many economies maintained a larger fiscal envelope in 2020 and 2021 and this slightly decreased by 2022, while the sharp labor market program increase was largely confined mainly to 2020 (see Statistical annex). 10 For Sub-Saharan Africa, this was mainly due to the spending in Seychelles, which implemented a measure to guarantee job retention in businesses that were economically affected by the pandemic. 24 Percentage of GDP 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Low-income (N= 11) Lower-middle-income (N= 21) 0.1 0.2 Upper-middle-income (N= 31) 0.5 0.3 0.2 2019 High-income (N=13) 1.4 1.1 0.3 Sub-Saharan Africa (N= 19) Increase East Asia and Pacific (N= 8) 0.4 0.4 0.2 0.3 0.1 0.1 A. Labor Market Europe and Central Asia (N= 23) 0.9 0.6 Latin America and the Caribbean (N= 12) 0.5 0.3 0.3 0.2 Middle East and North Africa (N= 8) 0.2 0.0 South Asia (N= 4) 0.3 0.1 0.1 0.2 Total (N= 76) 0.5 0.2 0.3 25 Protection Area, Region, Income Group, and Total EDEs Percentage of GDP 0.0 1.5 2.0 2.5 3.0 3.5 4.0 Low-income (N= 11) Lower-middle-income (N= 21) 1.3 1.2 1.0 0.5 0.3 0.5 0.8 0.9 Upper-middle-income (N= 31) 2.8 2019 High-income (N=13) 2.6 1.1 1.0 1.7 1.6 Sub-Saharan Africa (N= 19) 0.3 East Asia and Pacific (N= 8) 1.2 1.3 0.8 0.8 Increase Europe and Central Asia (N= 23) 0.4 Latin America and the Caribbean (N= 12) 3.6 2.1 2.2 2.0 1.8 1.5 B. Social Assistance Middle East and North Africa (N= 8) 0.6 1.1 Figure 4.4. Expenditure as Percentage of GDP, 2019, and Increase during COVID-19, by Social South Asia (N= 4) 2.2 1.7 0.5 1.7 Total (N= 76) 2.1 0.8 1.3 C. Social Insurance 8 6.8 7.0 7 0.6 0.7 5.8 6 6.4 Percentage of GDP 6.1 4.6 0.8 5 3.8 4.9 4.0 4 0.6 3.3 3.0 0.5 0.5 3 1.0 3.3 3.5 0.3 4.0 2 2.7 1.2 0.7 0.9 2.3 1 0.1 0.6 0.1 1.1 0 0.8 South Asia (N= 4) Latin America and the Caribbean (N=12) Middle East and North Africa (N= 8) Total (N= 76) Low-income (N= 11) Lower-middle-income (N= 21) Sub-Saharan Africa (N= 19) Upper-middle-income (N= 31) High-income (N=13) East Asia and Pacific (N= 8) Europe and Central Asia (N= 23) 2019 Increase Source: Original elaboration for this publication using Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire Note: Comparison peak COVID-19 spending. The graph uses simple averages and is based on a total number of observations of 76 countries, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. The number of observations for each income group and region is based on data availability. Aggregated indicators are calculated using simple averages of country-level expenditure as percentage of GDP, by income group and region. GDP = Gross domestic product; EDEs = Emerging and developing economies. V. What Economy or Program Characteristics Were Associated with a Stronger or Faster Response to COVID-19? The previous sections quantified how countries adapted their spending in response to the COVID-19 crisis. Going beyond this analysis, this section seeks to unwrap the factors and drivers that facilitated the response to COVID-19, focusing mainly on factors and drivers influencing increased social protection spending during the pandemic. Given that the larger COVID-19 responses were implemented through social assistance and labor market programs, this section primarily analyzes the factors and drivers associated with increased spending on these types of programs. 26 In most countries, there is a strong and positive correlation between the size of the real per capita social protection spending before the crisis (in 2019) and their expansion during 2020-21, although this varies across instrument. As shown in Figure 5.1A, having a social protection system with high social assistance spending in 2019 is associated with larger expansion during COVID-19. This is almost a perfect and positive correlation, indicating that having a larger baseline—that is, a strong social assistance system before COVID-19—helped economies react and protect the most vulnerable during the crisis. Labor market spending levels (before and during COVID-19) are also positive correlated, although the effect is relatively lower (Figure 5.1B). Figure 5.1. Social Protection Expenditure, 2019, and during COVID-19, by Social Protection Area and Income Group A. Social Assistance B. Labor Market 5.0 as percentage of GDP during COVID-19 as percentage of GDP during COVID-19 10.0 4.0 SA Expenditure, LM Expenditure, 3.0 5.0 2.0 1.0 0.0 0.0 0.0 2.0 4.0 6.0 8.0 0.0 0.5 1.0 1.5 2.0 SA Expenditure, LM Expenditure, as percentage of GDP 2019 as percentage of GDP 2019 Source: Original elaboration for this publication using Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire Notes: SA = Social assistance; LM = labor market programs; GDP = Gross domestic product; CI = Confidence interval. Relatedly, the data shows a strong and significant relationship between increases in per capita spending and the economy’s level of economic development. This is true for both social assistance programs (Figure 5.2A) and labor market programs (Figure 5.2B), though the relationship is stronger for the latter. As expected, HICs and UMICs, given their better initial fiscal conditions, were able to expand their spending more 27 strongly beyond 2019 levels, which in turn also implies that these economies were better prepared to cope with the negative effects of the COVID-19 pandemic. Figure 5.2. Increase in Expenditure in Per Capita Constant US$ PPP 2017 during the COVID-19 Pandemic and Gross National Income per Capita, by Economy and Income Group A. Social Assistance B. Labor Market 2000 1500 in USD PPP $2017, per capita in USD PPP $2017, per capita Delta SA Expenditure COVID Delta LM Expenditure COVID 1500 1000 1000 500 500 0 0 -500 6 7 8 9 10 6 7 8 9 10 GNI per capita 2019, in natural logarithm GNI per capita 2019, in natural logarithm Source: Original elaboration for this publication using World Bank World Development Indicators data and Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire Notes: SA = Social assistance; LM = labor market programs; GNI per capita = Gross national income per capita; CI = Confidence interval; PPP = Purchasing power parity. To capture the effect of the economic downturn during 2020 due to COVID-19, Figure 5.3 explores the relationship between the absolute change in real per capita expenditure and 2020 GDP growth. The 2020 change in GDP had an inverse relationship with the expansion of real per capita spending in both social assistance and labor market programs, which indicates that economies with a larger GDP decline had a stronger response to the COVID-19 shock. However, these effects were weak and not significant for either social assistance or labor market program spending. 28 Figure 5.3. Increase in Expenditure Per Capita Constant US$ PPP 2017 during the COVID-19 Pandemic and Change in 2020 GDP, by Economy and Income Group A. Social Assistance B. Labor Market 2000 1500 in USD PPP $2017, per capita in USD PPP $2017, per capita Delta SA Expenditure COVID Delta LM Expenditure COVID 1500 1000 1000 500 500 0 0 -30 -20 -10 0 10 -30 -20 -10 0 10 Change in constant GDP 2020 Change in constant GDP 2020 Source: Original elaboration for this publication using World Bank World Development Indicators data and Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire. Notes: SA = Social assistance; LM = labor market programs; GDP = Gross domestic product; PPP = Purchasing power parity; CI = Confidence interval. Having a proportionately smaller rural population also correlated with greater expansions in social assistance and labor market program budgets during the pandemic. Indeed, there is a strong, significant, and negative correlation between the expansion in per capita real spending and the percentage of population that is rural in each EDE, particularly for social assistance budget increases (Figure 5.4). This is also associated with the economy income group. These findings are consistent with previous studies that showed that the epicenter of the pandemic was frequently in urban, rather than rural, areas, resulting in greater efforts to protect urban populations, often through more expensive programming (such as wage subsidies replacing formal workers’ salaries). 29 Figure 5.4. Increase in Expenditure in Per Capita Constant US$ PPP 2017 during COVID-19 and Rural Population, by Economy and Income Group A. Social Assistance B. Labor Market 2000 1500 in USD PPP $2017, per capita in USD PPP $2017, per capita Delta LM Expenditure COVID Delta SA Expenditure COVID 1500 1000 1000 500 500 0 0 0 20 40 60 80 0 20 40 60 80 Rural population, Rural population, as percentage of total population as percentage of total population Source: Original elaboration for this publication using World Bank World Development Indicators data and Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire Notes: SA = Social assistance; LM = labor market programs; PPP = Purchasing power parity; CI = Confidence interval. Similarly, having a greater share of workers in wage employment before the pandemic was linked to larger real per capita spending increases during the crisis (Figure 5.5). Not surprisingly, this relation is stronger in labor market programs than in social assistance programs. This positive and significant relationship is also associated with the economy’s level of economic development. Apart from Azerbaijan, HICs and UMICs tend to have more salaried workers, which, on average, account for 63 percent and 81 percent of the total population of the economies in the sample, respectively, whereas this is 41 percent and 17 percent in LMICs and LICs, respectively. 30 Figure 5.5. Increase in Expenditure in Per Capita Constant US$ PPP 2017 during COVID-19 and Wage Employment, by Economy and Income Group A. Social Assistance B. Labor Market 2000 600 in USD PPP $2017, per capita in USD PPP $2017, per capita Delta SA Expenditure COVID Delta LM Expenditure COVID 1500 400 1000 200 500 0 0 -200 0 20 40 60 80 100 0 20 40 60 80 100 Wage Employment 2019, Wage Employment 2019, as percentage of total population as percentage of total population Source: Original elaboration for this publication using World Bank World Development Indicators data and Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire Notes: SA = Social assistance; LM = labor market programs; PPP = Purchasing power parity; CI = Confidence interval. The share of population with bank or mobile money accounts before COVID-19 was correlated with the increase in real per capita social protection spending (Figure 5.6). Countries where more of the population owned such financial accounts had more possibilities to expand spending, especially in labor market programs. Typically, this corresponds to economies with higher income levels, although there are exceptions in Mongolia and Sri Lanka. The relationship is positive with the expansion in social assistance and labor market spending—and higher for the latter. 31 Figure 5.6. Increase in Expenditures in Per Capita Constant US$ PPP 2017 during COVID-19 and Account Ownership, by Economy and Income Group A. Social Assistance B. Labor Market 2000 600 in USD PPP $2017, per capita in USD PPP $2017, per capita Delta SA Expenditure COVID Delta LM Expenditure COVID 1500 400 1000 200 500 0 0 -200 20 40 60 80 100 20 40 60 80 100 Account Ownership Account Ownership Source: Original elaboration for this publication using World Bank World Development Indicators data and Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire Notes: SA = Social assistance; LM = labor market programs; PPP = Purchasing power parity; CI = Confidence interval. Similarly, and related to access to account ownership, having a mobile phone was fundamental for economies to expand their responses to COVID-19 (Figure 5.7). In effect, having access to a mobile phone expanded the options for delivering benefits, including enabling people to receive benefits without leaving their homes to cash out money at the bank in times of social distancing rules. This association was stronger among labor market programs (mainly due to programs implemented by HICs, such as Slovenia and Lithuania) than in social assistance programs. 32 Figure 5.7. Increase in Expenditures in Per Capita Constant US$ PPP 2017 during COVID-19 and Mobile Ownership, by Economy and Income Group A. Social Assistance B. Labor Market 2000 600 1500 in USD PPP $2017, per capita in USD PPP $2017, per capita Delta SA Expenditure COVID Delta LM Expenditure COVID 400 1000 200 500 0 0 -500 -200 50 60 70 80 90 100 50 60 70 80 90 100 Mobile Ownership Mobile Ownership Source: Original elaboration for this publication using World Bank World Development Indicators data and Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) administrative data. https://www.worldbank.org/aspire Notes: SA = Social assistance; LM = labor market programs; PPP = Purchasing power parity; CI = Confidence interval. VI. What Factors Were Associated with a Stronger Response to the COVID 19 Shock? Suggestions for Strengthening ASP in EDEs The exploratory analysis presented in this background paper suggests that the following factors were associated with a stronger social protection response to COVID-19: Table 6.1. Factors Associated with Stronger Social Assistance and Labor Market Responses to COVID-19 The Strength of • Economies with higher spending levels before the pandemic were able to the Preexisting mount larger expansions during COVID-19. Social Protection • This is correlated with economy income level, as higher-income economies System had more developed social protection systems precrisis. Economic • The 2020 economic decline had a weak but positive effect on the expansion Conditions of social assistance and labor market program real per capita spending. • Economies with a proportionately larger rural population saw less of an increase in social assistance and labor market spending, as both the pandemic economic impact and response was concentrated in urban areas. 33 Labor Market • A higher pre-pandemic share of wage/salaried employment was positively Factors correlated with the increase in labor market program spending per capita during the COVID-19 pandemic. • Similarly, but to a lesser extent, growth in real spending per capita on social assistance was positively associated with a larger proportion of salaried employment. • This likely reflects the ability to more easily deliver benefits to formal sector workers and having more generous programs for them: for example, wage subsidies/unemployment insurance benefits to temporarily replace formal sector wages. Digital • Greater access to bank accounts and mobile phones before the crisis Infrastructure facilitated the expansion of real per capita spending, enabling contactless delivery of benefits. This effect was stronger among labor market programs. • This digital infrastructure was more prevalent in higher-income economies. The variation in social protection responses illustrates a major lesson from the pandemic: the importance of having adequate and large routine social protection programs to enable more effective shock responses. The COVID-19 crisis shone an intense spotlight on the variable state of social protection programming as the pandemic took hold. Countries with better coverage and higher spending on social protection before the pandemic were better able to expand during the crisis. These findings were also evident within each social protection area. As observed earlier, economies (mainly higher-income ones) that had existing labor market programs in place were able to use them in the COVID-19 response and to recover faster. Similarly, economies that had invested more extensively in flagship cash transfer programs before COVID-19 made heavy use of them to scale up and reach those in need. Relatedly, the crisis highlighted the importance of establishing robust delivery systems that can support both routine provision and shock response. Social protection delivery systems played a pivotal role during the crisis, with the quality and shock-responsiveness of economies’ information systems, registration, payment, and case management mechanisms all contributing to the effectiveness of their emergency response (Burattini et al. 2022; Gentilini et al. 2021; Hammad et al. 2021; Lowe, McCord, and Beazley 2021; Rigolini et al. 2023). Other studies identified that cash transfer expansion was positively associated with social registry coverage; cross-economy 34 analysis in World Bank (2022b) found that where governments relied on digital databases and data exchange to identify the population, their emergency social protection programming reached on average 51 percent of their population, while economies that had to undertake new data collection to identify participants typically reached only 16 percent of the population. Beazley, Marzi, and Steller (2021) observed that responses also tended to be faster in economies that could enroll using preexisting data, had social registries with more than 15 percent population coverage, and had already established electronic payments in their routine programming. But in most economies, strong routine systems with advanced adaptive features are still a long way from fruition (Bowen et al. 2020; Lindert et al. 2020). Successful shock responses also depended on the wider enabling infrastructure used by social protection systems. Higher pre-pandemic rates of financial inclusion and mobile phone ownership were associated with increased per capita social protection spending during the response. Earlier literature has identified a large role of digital and financial infrastructure in enabling shock-responsive expansion in coverage (Beazley, Marzi, and Steller 2021; World Bank 2022b). The data in this report find that greater financial and digital inclusion levels were clearly correlated with the scale of cash transfer expansion, but not with the scale of overall social protection expansion. However, even in the latter case, having existing infrastructure available to distribute payments was associated with increased spending on social protection during the COVID-19 crisis, highlighting the importance of wider infrastructure to enable governments to rapidly disburse additional funds during a crisis. These findings are aligned and complement earlier findings based on preliminary fiscal and social protection spending data (World Bank 2022b). While the fiscal measures helped reduce potential poverty increases—by about one-quarter in LICs, one-third in LMICs, and half in UMICs—the effectiveness varied significantly between richer and poorer nations. The crisis revealed three key lessons: the critical importance of countries’ borrowing capacity to finance fiscal responses, the challenges of reaching 35 households and protecting jobs in informal economies, and the need for efficient delivery systems that can quickly identify and support vulnerable populations beyond just the chronically poor. Many countries, particularly low- and middle-income ones, faced constraints due to limited finance, low formalization levels, and weak support delivery systems, often leading them to rely on less efficient measures like subsidies. The experience underscores the need to better prepare fiscal systems for future crises through debt management, contingent financing preparation, and improved delivery systems, while also recognizing that fiscal policy alone may have limitations in protecting poor households, necessitating support for other welfare-protection instruments. The emergence of new global crises following the COVID-19 pandemic makes it difficult to assess the extent to which some of the COVID-19 expansion will be permanent. Preliminary data suggest that some of the COVID-19 expansion lasted beyond the pandemic and may have helped bridge the gap toward universal social protection. Social protection expenditure in 2022 was still higher than pre-pandemic levels, with real per capita social protection expenditure around 13 percent higher in 2022 than 2019 (maintaining two-thirds of the surge between 2019 and 2020). Nevertheless, 2022 was characterized by the highest global levels of inflation in more than 25 years, as well as dramatic spikes in food insecurity, partly due to the global fallout of Russia’s full-scale invasion of Ukraine. The extent to which the expansion will also outlast these crises will only become clearer with time. 36 Statistical Annex Figure A1.1. GDP Growth Rate in EDEs, 2017–2022, by Economies’ Income Group 12 10 Percentage grrowth in GDP 8 6 4 2 0 -2 2017 2018 2019 2020 2021 2022 -4 (N=150) (N=150) (N=149) (N=149) (N=149) (N=148) -6 -8 -10 Low-income Lower-middle-income Upper-middle-income High-income Total Source: Original elaboration for this publication using World Bank, World Development Indicators. Note: The graph uses simple averages. Graph based on 150 observations, including 127 low- and middle- income countries and 23 high-income countries monitored by ASPIRE. GDP = Gross domestic product; EDEs = Emerging and developing economies. Figure A1.2. Evolution of Social Protection Expenditure as a Percentage of GDP, 2017–2022 A. Average Year-to-Year Change, Unweighted B. Cumulative Growth, (Base 2017 = 100), Unweighted 140 125.8 140 125.9 120 102.0 116.5 98.1 120 105.6 100 92.6 90.6 100.0 98.1 100.0 100 80 80 60 60 40 40 20 20 0 0 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) Source: Original elaboration for this publication using ASPIRE administrative data. https://www.worldbank.org/aspire. Note: The graph uses simple averages. Graph based on a total number of observations of 76 countries, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. Sample sizes varies across years. The number of observations for each income group and region 37 is based on data availability. Aggregated indicators are calculated using simple averages of country-level expenditure as percentage of GDP variation across years, by income group and region. GDP = Gross domestic product. Table A.1. Country-Level Expansion during COVID-19 Expenditure Per Capita in Expenditure as % GDP US$ PPP 2017 Delta Delta Income Region Country 2019 COVID [COVID- 2019 COVID [COVID- Group 2019] 2019] AFR Benin LMIC 68.0 76.8 8.8 2.2 2.4 0.26 AFR Burkina Faso LIC 57.8 81.2 23.4 2.5 3.4 0.90 Central African AFR LIC 14.7 15.2 0.5 1.9 2.0 0.09 Republic Congo, Democratic AFR LIC 3.9 6.7 2.8 0.4 0.6 0.25 Republic of AFR Côte d’Ivoire LMIC 70.3 76.7 6.4 1.3 1.4 0.15 AFR Ethiopia LIC 17.4 36.2 18.8 0.8 1.7 0.87 AFR Gabon UMIC 237.5 274.4 36.9 1.8 2.4 0.56 AFR Ghana LMIC 56.6 59.6 3.0 1.0 1.1 0.08 AFR Guinea-Bissau LIC 70.2 74.7 4.5 3.7 3.9 0.18 AFR Liberia LIC 6.7 16.6 10.0 0.5 1.3 0.81 AFR Nigeria LMIC 20.5 29.7 9.2 0.4 0.7 0.25 AFR Seychelles HIC 1,750.7 3,195.1 1,444.4 6.2 13.3 7.07 AFR Togo LIC 44.1 55.0 10.9 2.2 2.8 0.54 EAP China UMIC 1,013.1 1,124.8 111.7 6.3 6.5 0.24 EAP Indonesia UMIC 90.4 198.6 108.2 0.8 1.9 1.08 EAP Lao PDR LMIC 51.2 51.8 0.6 0.8 0.8 0.01 EAP Mongolia UMIC 872.8 1,544.5 671.7 7.5 13.6 6.18 EAP Philippines LMIC 201.7 288.9 87.2 2.4 3.8 1.44 EAP Tonga UMIC 99.0 168.4 69.4 1.6 3.0 1.36 ECA Albania UMIC 905.1 1,090.0 184.9 8.3 9.7 1.32 ECA Azerbaijan UMIC 890.5 1,076.9 186.4 5.6 7.7 2.10 ECA Bulgaria HIC 2,132.6 3,116.0 983.4 9.8 12.9 3.11 ECA Czechia HIC 3,340.7 3,986.1 645.4 9.0 11.2 2.17 ECA Estonia HIC 3,324.2 3,991.0 666.8 10.1 12.3 2.21 ECA Georgia UMIC 1,003.3 1,302.8 299.5 7.5 9.2 1.76 ECA Hungary HIC 3,105.1 3,317.6 212.5 10.3 10.6 0.25 ECA Kazakhstan UMIC 664.9 1,014.0 349.1 2.5 4.0 1.54 ECA Kosovo UMIC 624.8 838.5 213.6 5.9 7.8 1.91 ECA Kyrgyz Republic LMIC 438.1 442.9 4.8 8.6 9.6 1.02 ECA Latvia HIC 2,427.4 2,861.3 433.9 9.0 10.2 1.17 ECA Lithuania HIC 2,552.6 3,527.6 975.0 7.6 10.2 2.60 ECA Moldova UMIC 127.7 185.4 57.7 1.1 1.5 0.40 ECA Montenegro UMIC 1,958.8 2,141.3 182.5 10.8 13.7 2.90 ECA Slovak Republic HIC 2,552.5 3,126.2 573.6 9.6 11.9 2.34 ECA Slovenia HIC 4,128.4 5,038.4 910.1 11.9 14.7 2.82 ECA Tajikistan LMIC 114.9 122.1 7.2 3.7 3.9 0.15 ECA Türkiye UMIC 2,798.7 2,944.5 145.8 12.2 12.2 -0.03 ECA Ukraine UMIC 1,535.5 1,665.4 129.9 14.4 15.0 0.60 ECA Uzbekistan LMIC 454.8 558.9 104.2 5.9 7.1 1.22 38 Expenditure Per Capita in Expenditure as % GDP US$ PPP 2017 Delta Delta Income Region Country 2019 COVID [COVID- 2019 COVID [COVID- Group 2019] 2019] LAC Argentina UMIC 1,893.2 2,342.3 449.1 9.2 12.6 3.38 LAC Brazil UMIC 1,673.1 2,247.5 574.4 12.0 16.3 4.26 LAC Chile HIC 1,345.7 3,248.5 1,902.9 6.4 14.2 7.83 LAC Colombia UMIC 902.7 1,083.6 180.9 6.7 8.3 1.56 LAC Dominican Republic UMIC 506.2 1,020.8 514.6 3.1 6.6 3.53 LAC Ecuador UMIC 590.2 670.1 80.0 5.5 6.6 1.17 LAC Grenada UMIC 729.0 884.2 155.2 5.0 6.8 1.87 LAC Guatemala UMIC 144.8 268.8 124.0 1.9 3.7 1.77 LAC Mexico UMIC 851.3 984.7 133.4 4.7 5.7 0.99 LAC Panama HIC 1,012.8 1,705.9 693.1 3.1 5.9 2.76 LAC Peru UMIC 217.8 425.8 208.0 1.8 4.0 2.13 LAC St. Lucia UMIC 434.6 461.5 26.9 2.9 4.3 1.35 MNA Egypt, Arab Rep. LMIC 739.5 788.6 49.1 5.9 6.1 0.17 MNA Iran, Islamic Rep. UMIC 831.0 950.1 119.2 6.1 6.1 0.06 MNA Iraq UMIC 923.7 938.1 14.3 7.7 10.4 2.62 MNA Tunisia LMIC 960.9 1,107.9 147.0 8.0 10.0 2.04 SAR Afghanistan LIC 31.7 62.8 31.1 1.5 3.0 1.59 SAR Bangladesh LMIC 81.7 87.8 6.1 1.5 1.6 0.10 SAR Maldives UMIC 351.1 430.1 79.0 2.0 3.5 1.53 SAR Pakistan LMIC 68.5 92.1 23.7 1.3 1.8 0.46 SAR Sri Lanka LMIC 369.1 361.8 −7.4 3.0 3.2 0.19 EDEs Average 887.7 1,140.3 252.6 5.1 6.6 1.56 Source: Original elaboration for this publication using ASPIRE administrative data. https://www.worldbank.org/aspire. Note: Aggregated indicators are calculated using simple averages of country-level expenditure as percentage of GDP and in US$ PPP 2017. Average for the total based on 76 observations, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. LIC = low-income countries; LMIC = lower-middle-income countries; UMIC = upper-middle-income countries; HIC= high-income countries; AFR = Sub-Saharan Africa; ECA = Europe and Central Asia; EAP = East Asia and Pacific; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; SAR = South Asia; GDP = Gross domestic product; PPP = Purchasing power parity. 39 Figure A1.3. Evolution of Real Per Capita Social Protection Expenditure, by Income Group and Social Protection Area, 2017–2022, in US$ PPP 2017 A. Social Assistance B. Social Insurance 2,500 800 Annual per capita constant 700 2,000 Annual per capita constant 600 USD PPP $2017 USD PPP $2017 1,500 500 400 1,000 300 200 500 100 0 0 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) C. Labor Markets 450 Annual per capita constant 400 350 USD PPP $2017 300 250 200 150 100 50 0 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) Low-income Lower-middle-income Upper-middle-income High-income Total Source: Original elaboration for this publication using ASPIRE administrative data. https://www.worldbank.org/aspire Note: The graph uses simple averages. Graph based on 76 observations, including 63 low- and middle- income countries and 13 high-income countries monitored by ASPIRE. The number of observations varies per year. For 2022, social protection spending was estimated for the following countries and economies: Benin, Cameroon, Côte d’Ivoire, Guinea-Bissau, Niger, Togo, Zimbabwe, Türkiye, Colombia, the Dominican Republic, Ecuador, and West Bank and Gaza. Aggregated indicators were calculated using simple averages of country-level annual per capita spending by income groups. PPP = Purchasing power parity. 40 Figure A1.4. Evolution of Real Per Capita Social Protection Expenditure, by Region and Social Protection Area, 2017–2022, in US$ PPP 2017 A. Social Assistance B. Social Insurance 700 2,000 600 in constant USD PPP $2017 in constant USD PPP $2017 500 1,500 400 1,000 300 200 500 100 0 0 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) C. Labor Markets 250 in constant USD PPP $2017 200 150 100 50 0 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) Sub-Saharan Africa East Asia and Pacific Europe and Central Asia Latin America and the Caribbean Middle East and North Africa South Asia Total Source: Original elaboration for this publication using ASPIRE administrative data. https://www.worldbank.org/aspire. Note: The graph uses simple averages. Graph based on 76 observations, including 63 low- and middle- income countries and 13 high-income countries monitored by ASPIRE. The number of observations varies per year. For 2022, social protection spending was estimated for the following countries and economies: Benin, Cameroon, Côte d’Ivoire, Guinea-Bissau, Niger, Togo, Zimbabwe, Türkiye, Colombia, the Dominican Republic, Ecuador, and West Bank and Gaza. Aggregated indicators were calculated using simple averages of country-level annual per capita spending by regions. PPP = Purchasing power parity. 41 Figure A1.5. Evolution of Social Protection Expenditure as Percentage of GDP, by Region and Social Protection Area, 2017–2022 A. Social Assistance B. Social Insurance 4 8 7 Percentage of GDP Percentage of GDP 3 6 5 2 4 3 1 2 1 0 0 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) C. Labor Markets 1.0 0.8 Percentage of GDP 0.6 0.4 0.2 0.0 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) Sub-Saharan Africa East Asia and Pacific Europe and Central Asia Latin America and the Caribbean Middle East and North Africa South Asia Total Source: Original elaboration for this publication using ASPIRE administrative data. https://www.worldbank.org/aspire. Note: The graph uses simple averages. Graph based on 76 observations, including 63 low- and middle- income countries and 13 high-income countries monitored by ASPIRE. The number of observations varies per year. For 2022, social protection spending was estimated for the following countries and economies: Benin, Cameroon, Côte d’Ivoire, Guinea-Bissau, Niger, Togo, Zimbabwe, Türkiye, Colombia, the Dominican Republic, Ecuador, and West Bank and Gaza. Aggregated indicators were calculated using simple averages of country-level annual spending as percentage of GDP, by regions. 42 in constant USD PPP $2017 100 150 200 250 300 350 400 450 0 50 Low-income (N= 11) 1 4 Lower-middle-income (N= 21) 77 8 36 Upper-middle-income (N= 31) High-income (N=13) 419 2 12 42 104 315 2019 69 Sub-Saharan Africa (N= 19) 6 9 East Asia and Pacific (N= 8) 53 A. Labor Market Europe and Central Asia (N= 23) 218 146 76 63 44 73 46 Latin America and the Caribbean (N= 12) Region, Income Group, and Total EDEs Increase 30 1 10 Middle East and North Africa (N= 8) 50 9 28 22 South Asia (N= 4) 35 72 Total (N= 76) 107 43 in constant USD PPP $2017 100 200 300 400 500 600 700 800 0 Low-income (N= 11) 87 Lower-middle-income (N= 21) 107 Upper-middle-income (N= 31) 383 26 20 244 139 High-income (N=13) 711 461 250 2019 13 Sub-Saharan Africa (N= 19) East Asia and Pacific (N= 8) 88 106 280 83 Europe and Central Asia (N= 23) 396 101175313 Increase Latin America and the Caribbean (N= 12) 607 252 355 B. Social Assistance 84 19 Middle East and North Africa (N= 8) 103 19 South Asia (N= 4) 310 290 Figure A1.6. Expenditure in Constant US$ PPP 2017 during COVID-19, by Social Protection Area, 205 107 311 Total (N= 76) C. Social Insurance 2,500 2,075 in constant USD PPP $2017 2,000 208 1,530 1,867 1,500 131 1,399 1,000 797 722 711 573 595 47 84 35 74 500 273 628 208 750 648 55 561 91 11 15 365 0 0 258 90 South Asia (N= 4) Total (N= 76) Low-income (N= 11) Lower-middle-income (N= 21) Sub-Saharan Africa (N= 19) Upper-middle-income (N= 31) High-income (N=13) East Asia and Pacific (N= 8) Europe and Central Asia (N= 23) Middle East and North Africa (N= 8) Latin America and the Caribbean (N= 12) 2019 Increase Source: Original elaboration for this publication using ASPIRE administrative data. https://www.worldbank.org/aspire Note: The graph uses simple averages. Graph based on 76 observations, including 63 low- and middle- income countries and 13 high-income countries monitored by ASPIRE. The number of observations varies per year. Aggregated indicators were calculated using simple averages of country-level annual per capita spending by income groups and regions. EDE = Emerging and developing economies; PPP = Purchasing power parity. 44 Figure A1.7. Evolution of Social Protection Spending as Percentage of GDP, by Region, 2017–2022, Base 2017 = 100, Unweighted 180 160 140 120 100 80 60 40 20 0 Sub-Saharan East Asia and Europe and Latin America Middle East South Asia Total Africa Pacific Central Asia and the and North Caribbean Africa 2017 (N = 74) 2018 (N = 76) 2019 (N = 76) 2020 (N = 76) 2021 (N = 72) 2022 (N = 72) Source: Original elaboration for this publication using ASPIRE administrative data. https://www.worldbank.org/aspire. Note: This graph uses simple averages. Graph based on a total number of observations of 76 countries, including 63 countries in low- and middle-income countries and 13 high-income countries monitored by ASPIRE. The social protection spending for 2022 was estimated for the following countries and economies: Benin, Cameroon, Côte d’Ivoire, Guinea-Bissau, Niger, Togo, Zimbabwe, Türkiye, Colombia, Dominican Republic, Ecuador, and West Bank and Gaza. GDP = Gross domestic product. 45 References Alfers, L., M. Chen, and S. Plagerson. 2022. “Informal Workers and Social Contracts in the Global South.” Edward Elgar. Alfers, L., G. Ismail, and M. Valdivia. 2020. “Informal Workers and the Social Protection Response to COVID-19: Who Got Relief? How? And Did It Make a Difference?” WIEGO, Policy Insights 2. Manchester. Alon, T., M. Kim, D. Lagakos, and M. Van Vuren. 2023. “Macroeconomic Effects of COVID- 19 across the World Income Distribution.” IMF Economic Review 71 (1): 99–147. doi: 10.1057/s41308-022-00182-8. Bastagli, Francesca, and Christina Lowe. 2021. “Social Protection Response to Covid-19 and Beyond: Emerging Evidence and Learning for Future Crises.” ODI, London. https://odi.org/en/publications/social-protection-response-to-covid-19-and- beyond-emerging-evidence-and-learning-for-future-crises/. Beazley, R., M. Marzi, and R. Steller. 2021. “Drivers of Timely and Large-Scale Cash Responses to COVID-19: What Does the Data Say?” SPACE. London and Bonn. Beazley, Rodolfo, Jana Bischler, and Alexandra Doyle. 2021. “Towards Shock-responsive Social Protection: Lessons from the COVID-19 Response in Six Countries.” Oxford Policy Management, Oxford. https://www.opml.co.uk/files/Publications/A2241-maintains/maintains- towards-shock-responsive-social-protection-synthesis-report.pdf. Bowen, T., del Ninno, C., Andrews, C. et al. (2020) Adaptive social protection : building resilience to shocks. International Development in Focus. Washington, DC: World Bank Group (https://openknowledge.worldbank.org/handle/10986/33785). De La Flor Giuffra, Luciana, Mujica Canas, Ingrid Veronica, Maria Belen Fontenez, David Locke Newhouse, Claudia P Rodriguez Alas, Gayatri Sabharwal, and Michael Weber. 2021. Taking Stock of COVID-19 Labor Policy Responses in Developing 46 Countries. Washington, DC: World Bank Group. https://documentsinternal.worldbank.org/search/32939420. Gavrilovic, M., M. Rubio, F. Bastagli, R. Hinton, S. Staab, R. G. Goulder, ... and C. Tabbush. 2022. “Gender-responsive Social Protection Post–COVID-19.” Science 375 (6585): 1111–1113. https://www.science.org/doi/10.1126/science.abm5922. Gentilini, Ugo. 2022. “Cash Transfers in Pandemic Times Evidence, Practices, and Implications from the Largest Scale Up in History.” World Bank, Washington, DC. https://documents1.worldbank.org/curated/en/099800007112236655/pdf/P17 658505ca3820930a254018e229a30bf8.pdf. Gentilini, Ugo, Mohamed Bubaker Alsafi Almenfi, Pamela Dale, Robert J. Palacios, Harish Natarajan, Guillermo Alfonso Galicia Rabadan, Yuko Okamura, John D. Blomquist, Miglena Abels, Gustavo C. DeMarco, and Indhira Vanessa Santos. 2020. “Social Protection and Jobs Responses to COVID-19: A Real-Time Review of Country Measures (September 18, 2020) (English).” COVID- 19 Living Paper, Washington, DC: World Bank Group. http://documents.worldbank.org/curated/en/295321600473 897712/Social-Protection-and-Jobs-Responses-to-COVID-19-A-Real-Time- Review-of-Country-Measures-September-18-2020. Gentilini, Ugo, Mohamed Bubaker Alsafi Almenfi, John D. Blomquist, Pamela Dale, Luciana De La Flor Giuffra, Vyjayanti Tharmaratnam Desai, Maria Belen Fontenez, Guillermo Alfonso Galicia Rabadan, Veronica Lopez, Ana Georgina Marin Espinosa, Harish Natarajan, David Locke Newhouse, Robert J. Palacios, Ana Patricia Quiroz, Claudia P Rodriguez Alas, Gayatri Sabharwal, and Michael Weber. 2021. “Social Protection and Jobs Responses to COVID-19: A Real-Time Review of Country Measures (May 14, 2021).” Vol. 1 of 2, Washington, DC: World Bank Group. Gentilini, Ugo, Mohamed Bubaker Alsafi Almenfi, Yuko Okamura, John Austin Downes, Pamela Dale, Michael Weber, David Locke Newhouse, Claudia P Rodriguez Alas, Mareeha Kamran, Ingrid Veronica Mujica Canas, Maria Belen Fontenez, Sandra 47 Asieduah, Vikesh Ramesh Martinez Mahboobani, Gonzalo Javier Reyes Hartley, Gustavo C. Demarco, Miglena Abels, Usama Zafar, Emilio Raul Urteaga, Giorgia Valleriani, Jimmy Vulembera Muhindo, Sheraz Aziz, and Hrishikesh Tirumala Madabushi Matam I. 2022. “Social Protection and Jobs Responses to COVID-19: A Real-Time Review of Country Measures.” Vol. 1 of 2, Washington, DC: World Bank Group. https://documentsinternal.worldbank.org/search/33720967. ILO (International Labour Organization). 2021. World Social Protection Report 2020-22: Social Protection at the Crossroads - In Pursuit of a Better Future. International Labour Organization, Geneva. https://www.ilo.org/global/publications/books/WCMS_817572/lang-- en/index.htm. ILO. 2022. ILO Monitor on the World of Work. 9th edition. https://www.ilo.org/resource/brief/ilo-monitor-world-work-9th-edition. Kamran, Mareeha, Ingrid Veronica Mujica Canas, Maria Belen Fontenez, David Locke Newhouse, Claudia P Rodriguez Alas, and Michael Weber. 2023. Jobs Watch COVID-19: Exploring Two Years of Labor Market Policy Responses to COVID-19 - A Global Effort to Protect Workers and Jobs. Washington, DC: World Bank Group. https://documentsinternal.worldbank.org/search/33978243. Lindert, K., Karippacheril, T.G., Rodriguez Caillava, T. and Nishikawa Chavez, K. (2020) Sourcebook on the foundations of social protection delivery systems. Washington, DC: World Bank Group (https://openknowledge.worldbank.org/handle/10986/34044). Lowe, Christy, Anna McCord, and Rodolfo Beazley. 2021. “National Cash Transfer Responses to Covid-19: Operational Lessons Learned for Social Protection System-Strengthening and Future Shocks.” Working paper, ODI Global. https://odi.org/en/publications/national-cash-transfer-responses-to-covid-19- operational-lessons-learned-for-social-protection-system-strengthening-and- future-shocks/. 48 Okamura, Yuko, Hrishikesh TMM Iyengar, and Colin Andrews, 2025. “Wake-up Call for Social Assistance? Unfinished Mission to Reach the Poor and Beyond” World Bank, Washington, D.C., forthcoming. WFP (World Food Programme), IPC-IG (International Policy Centre for Inclusive Growth), and UNICEF (United Nations Children’s Fund). 2021. Social Protection and Venezuelan Migration in Latin America and the Caribbean in the Context of COVID-19 - 2021. https://www.wfp.org/publications/social-protection-and- venezuelan-migration-latin-america-and-caribbean-context-covid. World Bank. 2022a. Poverty and Shared Prosperity 2022: Correcting Course. Washington, DC: World Bank. http://hdl.handle.net/10986/37739. World Bank. 2022b. World Development Report 2022: Finance for an Equitable Recovery. Washington, DC: World Bank Group. https://www.worldbank.org/en/publication /wdr2022. World Bank. 2025. State of Social Protection Report 2025: The 2-Billion-Person Challenge. State of Social Protection Report. Washington, DC: World Bank. 49 Social Protection & Jobs Discussion Paper Series Titles 2025 No. Title 2509 State of Social Protection Report 2025: The 2-Billion-Person Challenge. Background Paper #2: Adaptive Social Protection Agenda: Lessons from Responses to COVID-19 Shock 2508 State of Social Protection Report 2025: The 2-Billion-Person Challenge. Background Paper #1: Mind the Gap: Coverage, Adequacy and Financing Gaps in Social Protection for the Extreme Poor and the Poorest Quintile 2507 Service Integration and Case Management for People on the Move: A Review of Selected International Practices 2506 Impact of Climate Change and the Green Transition on Human Capital: A Review of the Evidence from Europe and Central Asia 2505 A slippery slope: the opportunities and risks of digital approaches and technology in Social Protection Systems 2504 De Jure and De Facto Coverage of Parental Benefits in Nepal 2503 Awareness, Access, and Perceptions around Parental benefits among Urban Argentinians 2502 Regulating Markets So More People Find Better Jobs 2501 São Tomé and Príncipe Unpacking Migration Dynamics: Critical Issues and Policy Recommendations To view Social Protection & Jobs Discussion Papers published prior to 2021, please visit www.worldbank.org/sp. ABSTRACT The paper examines the social protection response to the COVID-19 pandemic across 76 emerging and developing economies (EDEs) to identify lessons on how to make these systems more resilient against risks, shocks, and crises at the individual, household, or national level. The COVID-19 pandemic triggered significant expansions in social protection systems across EDEs, with responses varying based on countries’ existing infrastructure and income levels. The analysis of 76 EDEs revealed that countries used approximately 37 percent of their social protection programs to respond to COVID-19, with social assistance programs being the most frequently used response (73 percent of total programs). EDEs increased their real per capita social protection spending by an average of 28 percent, with low-income countries (LICs) and high-income countries (HICs) showing the largest increases at around 40 percent and 32 percent, respectively. The effectiveness of responses was strongly correlated with preexisting social protection systems, economic conditions, labor market factors, and digital infrastructure. Countries with more developed social protection systems, formal labor markets, and digital payment infrastructure before the pandemic were better positioned to rapidly scale up their responses, highlighting the importance of maintaining robust routine social protection programs and delivery systems to enable effective crisis response. JEL CODES H53, I38, J28, O15 KEYWORDS Social protection system, COVID-19 pandemic, evolution of social protection spending, shock response ABOUT THIS SERIES Social Protection & Jobs Discussion Papers are published to communicate the results of The World Bank’s work to the development community with the least possible delay. This paper therefore has not been prepared in accordance with the procedures appropriate for formally edited texts. For more information, please visit us online at www.worldbank.org/socialprotection