SOCIAL PROTECTION DISCUSSION PAPER No. 2513 | MARCH 2025 State of Social Protection Report 2025 The 2-billion-Person Challenge Background Paper #6 Unlocking the Potential of Household Surveys to Measure Women’s Access to Social Protection Claudia P. Rodriguez Alas Ana Verónica López Ingrid Mujica © 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 This paper uses household survey data from 27 countries to assess sex-differentiated access to social protection programs and their impact on mitigating gender gaps in the labor market. The analysis includes indicators of coverage, distribution of social protection recipients, and adequacy of benefits, all disaggregated by sex, to estimate two indices. The first index assesses gender inequalities in the provision of social protection benefits and ranks countries by their level of ‘gender progressivity’. The second index measures the net earnings received by men and women from both labor and social protection transfers, quantifying whether the social protection system reduces or exacerbates labor market gender inequalities. This paper demonstrates the construction and interpretation of these indices and provides practical recommendations for adapting household surveys to collect the data needed to scale them across emerging and developing economies. JEL Codes: J16, J29, I38, O15 Keywords: Gender, gender gaps, social protection, social protection systems, social protection benefits, social assistance, social insurance, coverage, average per capita transfer, incidence, labor markets, labor earnings, labor participation. STATE OF SOCIAL PROTECTION REPORT 2025 | 1 Acknowledgments This paper was developed under the guidance of Iffath Sharif (Global Director, Social Protection), Loli Arribas-Baños (Practice Manager, Social Protection), and Jamele Rigolini (Senior Advisor, Social Protection). It contributes to the State of Social Protection Report 2025: The 2-Billion-Person Challenge (World Bank 2025a). We are grateful for the valuable contributions of Emil Tesliuc, Ana Sofia Martinez Cordova, Afrah Alawi Al-Ahmadi, Kathleen G. Beegle, Anna Tabitha, Georgina Marin, Alessandra Heinemann, Christina Louise Lowe, and Robert Palacios. Special thanks to Zurab Sajaia and Sergiy Radyakin for improving ADePT functionalities to produce the sex-disaggregated indicators featured in this paper. The authors would also like to acknowledge the entire ASPIRE team, particularly Johanna Estefania Andrango Brito, Usama Zafar, Bhavya Paliwal, Muhsine Senart, Leopoldo Tornarolli, Xuejiao Xu, and Vikesh Ramesh Mahboobani Martinez, for their efforts in harmonizing the household surveys used in the report and for their excellent research assistance. Additionally, we extend our thanks to Matthew Naumann and Fiona Mackintosh for their editorial support and to Agnes Nderakindo Mganga for her outstanding administrative assistance. 2 | STATE OF SOCIAL PROTECTION REPORT 2025 Table of Contents Abstract 1 Acknowledgments 2 1. Why Should We Care about Measuring Women’s and Girls’ Access to Social Protection? 5 2. Understanding Differences in Labor Force Participation and Earnings between Men and Women 7 3. Understanding Differences in Access to Social Protection between Men and Women 10 3.1 Coverage 10 3.2 Distribution of beneficiaries 12 3.3 Average per capita transfers 14 4. Net Impact: Countries in Which Social Protection Programs Are More Effective at Offsetting Gender Gaps in Labor Market Earnings 16 4.1 Index 1: Measuring gender inequality in the provision of social protection (SP) benefits 16 4.2 Index 2: Measuring the sex-differentiated net effect on access to labor market and SP systems 18 5. Setting the Data Agenda: Improving Household Survey Design to Generate Sex-Disaggregated Social Protection Indicators 20 Conclusion 24 References 25 Annex 1. Countries and Household Surveys Used in the Analysis 27 Annex 2. Social Protection Performance Analysis 28 STATE OF SOCIAL PROTECTION REPORT 2025 | 3 Table of Figures Figure 1.1 Of the 71 Surveys Analyzed, Nearly One-Third Collect Social Protection Information an the 7 Individual Level Figure 2.1 Women’s Labor Force Participation is Lower than Men’s and this Gap Varies across Countries 8 Figure 2.2 Women’s Earnings in the Labor Market Are on Average Lower Than Men’s Earnings 9 Figure 3.1 Women Are More Likely to Receive Social Assistance Transfers Than Men 11 Figure 3.2 The Distribution of Women and Men among Social Protection Recipients Vary across 12 Countries Figure 3.3 Social Insurance Programs Are the Drivers of the Lower Female-to-Male Ratio of Average per 15 Capita Social Protection Transfers Figure 3.4 Female-to-Male Ratio of Social Protection Coverage by Quintile 16 Figure 4.1. Social Protection Systems Help Reduce the Gender Earnings Gap, but Are Insufficient to 19 Eliminate It Figure A2.1 Distribution of Social Assistance Recipients by Male and Female 29 Figure A2.2 Distribution of Social Insurance Recipients by Male and Female 29 Table of Boxes Box 3.1 Sex Differences in Social Protection Coverage and Transfer Levels Vary Across Income 15 Distribution Box 5.1. Recipients versus Beneficiaries of Social Protection Programs 21 Tables Table 4.1: List of Countries Ranked by the Gender Progressivity of their Social Protection Systems 18 Table 5.1. Four Scenarios to Conduct Sex-Disaggregated Analysis of Social Protection Performance 21 Table A2.1 Female-to-Male Ratios - Coverage and Average per Capita Transfers of Social Protection 28 Programs 4 | STATE OF SOCIAL PROTECTION REPORT 2025 1. Why Should We Care about Measuring Women’s and Girls’ Access to Social Protection? Women and girls face gender-specific risks and vulnerabilities that affect their well-being and increase their likelihood of living in poverty. They encounter structural obstacles that restrict their access to labor markets, asset ownership, education, and health care. Gender vulnerabilities and risks have compounding effects throughout women’s life cycles and various life events—from childhood to old age. Women and girls often bear the burden of care responsibilities and unpaid household work. They are also at higher risk of suffering gender- based violence and exploitation. These vulnerabilities and constraints arise not only from economic conditions but also from discriminatory social norms, regulatory barriers, unequal power dynamics, and entrenched stereotypes. These factors contribute to women’s unequal access to the labor market and human capital acquisition, perpetuating cycles of poverty and exclusion (FAO 2023; Gavrilovic et al. 2022). Social protection (SP) systems can be powerful tools to address gender inequalities if they are designed considering women’s and girls’ risks and vulnerabilities throughout their life cycle. A growing body of evidence shows that social assistance (SA) programs can improve women’s socioeconomic status, psychosocial well-being, agency, health and education outcomes, and service utilization (Alfers et al. 2021; Bastagli et al. 2016; Camilletti 2020; Perera et al. 2022). Cash transfers have been associated with higher investments and savings among women, suggesting a positive effect on their financial resilience (Perera et al. 2022). Meanwhile, labor market programs, have effectively improved women’s labor force participation rates, encouraging greater engagement in wage labor, and contributing to overall employment growth among women (Chinen et al. 2017; Escudero et al. 2017). Enhanced interventions—such as livelihood programs that combine cash transfers with skill-building initiatives—have displayed positive effects on women’s savings, asset ownership, and earning capacity (Dickson and Bangpan 2012). Evidence also shows that the effects of cash transfers in reducing violence against women and children are overwhelmingly positive (Baranov et al. 2021). Despite a robust body of literature documenting the positive impact of SP interventions,1 little is known about sex-differentiated access to SP globally and the size of coverage gaps. This paper addresses the following key questions: How does access to SP programs vary between women and men? Do women receive higher or lower SP transfer amounts than men? Is there a sex difference in the aggregate (countrywide) SP transfers received? How does women’s participation in the labor market compare with their participation in SP programs? How do women’s average labor market earnings compare with the SP transfers 1 The World Bank’s Atlas of Social Protection: Indicators of Resilience and Equity (ASPIRE) harmonizes social protection in three thematic areas and 12 program categories as follows: (1) social assistance programs comprise conditional and unconditional cash transfer programs, social pensions, public works, school feeding, food and in-kind transfers, fee waivers and targeted subsidies, and other social assistance programs; (2) labor market programs comprise active and passive labor market interventions; and (3) social insurance comprises contributory pensions and other social insurance. STATE OF SOCIAL PROTECTION REPORT 2025 | 5 they receive? The answers to these questions will help determine whether SP transfers are reducing or exacerbating the inequalities in the labor market for the countries analyzed. Household survey data provide a unique opportunity to answer these questions identifying the recipient of the SP benefits in the household. The objective of this paper is to leverage household survey data to develop two indices that measure sex-differentiated access to SP systems and their impact on mitigating gender gaps in the labor market.2 Household surveys are crucial because while administrative data and impact evaluations provide important information on specific SP programs, estimating countrywide sex differences requires the disaggregation of coverage and transfer value indicators by sex across all programs in a country.3 Therefore, nationally representative household surveys are the only source to conduct systemwide labor market and SP analysis. For the analysis to be feasible, household surveys need to track information on all large- and medium-size SP programs and include information on program participation and the value of the transfers. This information should be collected at the individual level to account for the share of recipients and direct beneficiaries who are male and female, as well as the benefits allocated to each group. This paper uses data from 27 countries with SP information available at the individual level in nationally representative household surveys. The analysis should ideally be conducted for all emerging and developing economies; however, for most countries, information was either not available or not captured at the individual level to allow for sex disaggregation.4 We analyzed a stock of 71 household surveys from 2016 to 2022 and determined that only 27 comply with the criteria of having all SP data collected at the individual level (see Figure 1.1). Most of these surveys are concentrated in the Latin America and the Caribbean and Europe and Central Asia regions, limiting their ability to represent regional or global conclusions. However, the analysis of the 27 countries still shows the usefulness of these indicators in tracking sex differences in SP indicators and their impact on mitigating gender gaps in the labor market. Therefore, this paper serves as proof of concept for constructing and interpreting two useful indices. It also provides practical recommendations on adapting household surveys to collect the data needed to escalate the formulation of indices to the rest of the emerging and developing economies. 2 This paper builds on the methodology developed in the study “Do Women Have the Same Access as Men to the Social Protection System? The Case of Mongolia” (Tesliuc et al., forthcoming). 3 The indices focus exclusively on social protection programs that provide monetary transfers. 4 Some surveys collect SP data at the household level for programs aimed at households or families. Although this allows the identification of direct beneficiaries, it cannot determine the individual who receives the transfer and controls the resources. 6 | STATE OF SOCIAL PROTECTION REPORT 2025 FIGURE 1.1 Of the 71 Surveys Analyzed, Nearly One-Third Collect Social Protection Information at the Individual Level Level at which social protection variables are captured in household surveys 25 2 20 15 Number of countries 5 10 2 2 20 4 4 5 10 6 6 2 3 2 2 1 0 East Asia and Europe and Latin America Middle East South Asia Sub-Saharan the Pacific Central Asia and the and North Africa Caribbean Africa All individual level variables Mixed levels - household & individual All household level variables Source: Original calculation using Atlas of Social Protection: Indicators of Resilience and Equity (ASPIRE) household survey data. https://www. worldbank.org/aspire Note: n = 71 countries - latest survey only. All variables captured at the individual level = 27; all variables captured at the household level = 6; variables captured using a mix of levels = 38. This paper is organized according to the building blocks needed to construct the indices. First, we analyze gender gaps in the labor market by measuring the differences in labor market participation and monthly earnings between men and women. Second, we quantify the access gap to social protection by measuring the differences in SP coverage, incidence of beneficiaries, and transfer values received by men and women. Third, we construct two indices. The first compares the country-level aggregate of labor earnings and SP transfers to rank countries by their level of gender progressivity. The second examines the net effect of SP transfers in mitigating the gender gap in the labor market. Finally, the paper provides recommendations on how to improve household surveys to make sex-disaggregated analysis feasible at the global level. 2. Understanding Differences in Labor Force Participation and Earnings between Men and Women Based on their gender responsiveness, social protection (SP) systems may exacerbate or mitigate gender gaps in the labor market. Before analyzing sex differences in coverage and SP transfers received by men and women, this section presents an overview of labor market indicators. It quantifies differences in labor force participation and monthly earnings between men and women across 27 countries. The analysis examines aggregate differences in labor earnings at the country level, comparing them with aggregate social spending to measure net effects. For consistency, the labor market indicators used in this section correspond to STATE OF SOCIAL PROTECTION REPORT 2025 | 7 the same years as the household surveys used to measure access to social protection. The indicators are presented as female-to-male ratios, so the statistics for men are normalized to 100 percent. Globally, women’s labor force participation remains, on average, lower than that of men. Currently, roughly half of global working-age women participate in the labor force, compared to 80 percent of working-age men (ILO 2024). Using the sample of 27 countries, Figure 2.1 illustrates female labor force participation as a percentage of the corresponding male rate and the magnitude of the gap. On average, women’s labor force participation is 31 percent lower than that of men across this sample. There is variation in the size of the participation gaps across countries. For example, in Moldova, female labor force participation is nearly equal to that of men; in contrast, in the Republic of the Marshall Islands, Kosovo, and Tunisia it is less than half the rate of men’s participation rate. This disparity is even more pronounced in the Islamic Republic of Iran, where female labor force participation is only 21 percent that of men. FIGURE 2.1. Women’s Labor Force Participation is Lower Than Men’s and this Gap Varies across Countries Female labor force participation as percentage of male participation (100% = male level) 100% 80% 60% 40% 20% 0% Brazil 2021 Bolivia 2021 Tunisia 2022 Uruguay 2021 Peru 2021 Moldova 2018 Chile 2020 Paraguay 2022 Sri Lanka 2019 Colombia 2022 Congo, Dem. Rep. 2020 Kosovo 2022 Maldives 2019 Mexico 2022 Georgia 2021 Mongolia 2020 Vanuatu 2019 Iran, Islamic Rep. 2020 Panama 2021 Türkiye 2019 Lao PDR 2017 RMI 2019 Kiribati 2019 Serbia 2019 Malaysia 2016 Kazakhstan 2021 Ecuador 2021 -20% -40% -60% -80% EAP ECA LAC MNA SAR SSA -100% Gap to gender parity Source: ILO Modelled Estimates (ILOEST) and Projections database. Note: RMI refers to The Republic of the Marshall Islands. The labor force participation rate is the proportion of the population ages 15–64 that is economically active (employed and unemployed). The female-to-male ratio is calculated by dividing the women’s labor force participation rate, by the corresponding men’s rate. EAP= East Asia and the Pacific; ECA= Europe and Central Asia; LAC = Latin America and the Caribbean; MNA= Middle East and North Africa; SAR = South Asia; SSA = Sub-Saharan Africa. 8 | STATE OF SOCIAL PROTECTION REPORT 2025 Gender gaps in labor earnings persist across countries.5 However, great heterogeneity exists across country income groups. The sample of 27 countries with labor earnings data shows that for every dollar earned by men, women make, on average, only 85 cents. Given the sample’s bias toward upper-middle-income countries (UMICs), this figure may be optimistic. An International Labor Organization (ILO) study using 2019 data found that, globally, women earned only 51 cents for every dollar earned by men (ILO 2023). Figure 2.2 shows the disparity across the sample countries. The labor income gap is greater in countries such as the Democratic Republic of Congo, where women earn 70 percent of what men earn. In other words, for every 1 Congolese franc men earn, women make 30 cents less. Conversely, Ecuador and Panama are the only two countries in the sample where women earn more than men (8 and 10 cents more, respectively). FIGURE 2.2. Women’s Earnings in the Labor Market Are, on Average, Lower Than Men’s Earnings Female monthly earnings as percentage of male earnings (100% = male level) 120% 100% 805 60% 40% 20% 0% Panama 2021 Colombia 2022 Tunisia 2022 Bolivia 2021 Chile 2020 Brazil 2021 Georgia 2021 Sri Lanka 2019 Maldives 2019 Paraguay 2022 Ecuador 2021 Congo, Dem. Rep. 2020 Lao PDR 2018 Uruguay 2021 Peru 2021 Türkiye 2019 Mexico 2022 Mongolia 2020 Kazakhstan 2021 Moldova 2018 Malaysia 2016 Serbia 2019 Vanuatu 2019 -20% -40% EAP ECA LAC MNA SAR SSA Gap to gender parity Source: ILOEST and Projections database. Note: The female-to-male ratio in earnings is calculated by dividing women’s average monthly labor earnings by the corresponding monthly men’s earnings. Of the 27 countries in the original sample, earnings data were not available for The Republic of the Marshall Islands, Kiribati, Kosovo, and the Islamic Republic of Iran. 5 The analysis is based on wage employment earnings only and does not include enterprise or farm labor. STATE OF SOCIAL PROTECTION REPORT 2025 | 9 3. Understanding Differences in Access to Social Protection between Men and Women 3.1 Coverage This study uses the coverage indicator to examine gender disparities in access to social protection (SP). The analysis focuses on determining the percentage of women within the total female population and men within the total male population who receive social protection benefits. As with the labor market indicators, a ratio is calculated to normalize the male coverage indicator to 100 percent. In addition, the analysis is based on SP recipients, rather than beneficiaries. Although data on recipients provide information on who controls the SP transfers, they underestimate the number of beneficiaries in household-targeted programs. SP programs cover a higher percentage of women than men: on average, 1.2 women receive an SP transfer for every man across the sample of 27 countries. However, there is significant heterogeneity in coverage rates across countries. The number of women recipients of SP benefits in Kazakhstan and Ecuador is more than double the number of men (2.1 women per 1 male recipient); while in Peru, Kiribati, and Malaysia, the coverage rates between men and women are nearly the same. On the other hand, in a third of the countries in the sample, SP coverage is biased toward men, particularly in the Lao People’s Democratic Republic (PDR), and the Democratic Republic of Congo, where women receive SP programs at less than half the rate of men. These results can be partly explained by the fact that their household surveys collect only limited social protection data.6 Social Assistance (SA) programs7 are the main drivers behind the higher ratio of women recipients in social protection. On average, across the sample of countries, 1.4 women receive a SA transfer for every man. However, this ratio is 1 for social insurance (SI), indicating that an equal number of women and men receive contributory benefits, even though women’s participation in the labor market is lower than men’s. This may be in part due to women’s longer life expectancy, which translates to women receiving old- age contributory benefits for a longer period than men, as well as more women receiving survivors’ pensions. Figure 3.1 shows that a higher proportion of women receive SA programs in 19 of the 25 countries with SA data.8 6 For Lao PDR, this paper uses the 2018 Expenditure and Consumption Survey, while for the Democratic Republic of Congo, it utilizes the 2020 Enquête par Grappes à Indicateurs des Objectifs De Développement (EGI-ODD). 7 Social assistance encompasses noncontributory programs, such as conditional and unconditional cash transfer programs, social pensions, public works, school feeding, food and in-kind transfers, school feeding, fee waivers and targeted subsidies, and other social assistance programs. 8 Lao PDR’s 2018 survey does not collect SA data and the limited SA data collected by Democratic Republic of Congo’s survey has too few observations to generate significant results. 10 | STATE OF SOCIAL PROTECTION REPORT 2025 FIGURE 3.1. Women Are More Likely to Receive Social Assistance Transfers Than Men Female-to-male ratio of social assistance and social insurance coverage 350% 300% 250% 200% 150% 100% 50% 0% Kazakhstan 2021 Panama 2021 Mexico 2022 Vanuatu 2019 RMI 2019 Türkiye 2019 Moldova 2018 Lao PDR 2018 Tunisia 2022 Malaysia 2016 Paraguay 2022 Peru 2021 Congo, Dem. Rep. 2020 Mongolia 2020 Kiribati 2019 Bolivia 2021 Serbia 2019 Ecuador 2021 Brazil 2021 Maldives 2019 Iran, Islamic Rep. 2020 Georgia 2021 Uruguay 2021 Sri Lanka 2019 Chile 2020 Kosovo 2022 Colombia 2022 EAP ECA LAC MNA SAR SSA Social Assistance Social Insurance 100% = male level Source: Original calculation using ASPIRE household survey data. https://www.worldbank.org/aspire. Note: RMI refers to The Republic of the Marshall Islands. Coverage is the percentage of the population receiving social assistance and social insurance transfers. The female-to-male ratio is determined by dividing the women’s coverage rate by the men’s coverage rate. Due to the unavailability of labor market data in most surveys, these programs are not presented in the chart. However, recipiency rates significantly vary across countries and SP areas. In Ecuador, 3.3 women receive social assistance for every man, whereas in Kosovo, the ratio is 0.5:1, with women receiving benefits at half the rate of men. The results in Ecuador are driven by the Bono de Desarrollo Humano; and in Kosovo, by the social welfare payments, the only SA program collected in the survey. Only in six countries in the sample do men receive SA benefits at a higher ratio than women (the Islamic Republic of Iran, Kosovo, Sri Lanka, the Republic of the Marshall Islands, Vanuatu, and Bolivia). In terms of SI, Kazakhstan reports the highest female-to-male ratio of recipients, with 2.2 women receiving old age pension for every male recipient. In contrast in Lao PDR, women receive SI benefits at less than half the rate of men (0.4:1)9. 9 This ratio by the household survey may be underestimated as the national social insurance program covered 94 percent of the population as of 2024 (ILO 2024). STATE OF SOCIAL PROTECTION REPORT 2025 | 11 3.2 Distribution of beneficiaries Results for sex-disaggregated coverage rates are influenced by the distribution of men and women among program recipients. Unlike coverage, which calculates the percentage of recipients within the relevant population groups, this indicator measures the percentage of women and men who receive benefits among the total number of recipients. Figure 3.2 shows that the percentage of women receiving SP benefits is higher than that of men in two-thirds of the countries in our sample (proportion of women greater than 50 percent). The average share of women in SP programs across the sample is 53 percent; however, it varies significantly across countries. FIGURE 3.2. The Distribution of Women and Men among Social Protection Recipients across Countries 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% so 019 Sr ia 19 18 va 2 DR 0 Re 20 De Rep 19 si 2 Co lam RM 21 az 8 o 0 0 Pa es 2 1 rib 16 Mo bia 1 Ma bia 2 Mo gia 1 ma 9 Ma ay 2 1 ia 1 Pa ru 1 Ur ile 2 1 k 1 Co dor 1 Va iye 2 1 Me lia 2 1 at 21 ldi 202 2 or 02 gu 202 nis 2 Pe 202 Ch 202 an 02 ua 02 2 ldo 202 2 lay 02 o P 02 Br 201 xic 02 2 ug 02 na 01 Bo ti 20 20 0 0 lom 20 Ki a 20 Tu il 20 Tü o 20 ng 20 nu 0 m. . 20 Se 20 Ko a 2 I2 u2 Ge ay 2 iL 2 Ec an 2 La p. 2 a v v u v st o i r rk l kh ra i c za Ka s o, I n, ng Ira Female Male Source: Original calculation using ASPIRE household survey data. https://www.worldbank.org/aspire Note: RMI refers to The Republic of the Marshall Islands. Social assistance programs typically have a higher proportion of female recipients compared to social insurance programs. In the sample, women make up the majority of recipients of social assistance programs in Kazakhstan and Moldova. Kazakhstan’s survey shows that women make up 69 percent of social assistance recipients. This is primarily driven by targeted social assistance and state benefits for children programs, where women represent 95 percent and 97 percent of recipients, respectively. Due to differences in longevity in Kazakhstan, women also represent a majority of the contributory and social pension recipients, accounting for about 72 percent and 66 percent, respectively. Similarly, Moldova’s survey10 indicates that women comprise 65 of the SA program recipients, largely driven by child allowance programs where women make up nearly 100 percent of the recipients. 10 Moldova’s 2018 Household Budget Survey. 12 | STATE OF SOCIAL PROTECTION REPORT 2025 Women in Moldova also constitute about 65 percent of contributory pension recipients. Ecuador and Brazil also report the highest share of women in their SA programs. Based on the survey, women comprise 77 percent of recipients of Ecuador’s Bono de Desarrollo Humano and 89.5 percent of recipients of Bono Gallegos Lara.11 In Brazil, women represent 92 percent of Bolsa Familia recipients and 68 percent of Outros Programas Sociais do Governo (other social assistance programs). Women in Brazil also constitute 57 percent of old age pensioners. (See Annex 2 for SA and SI distribution of beneficiaries). However, some countries show SA systems biased toward men. The survey for the Islamic Republic of Iran12 only reports information on two programs: cash subsidies and scholarships, and aid from social and charity institutions. Both programs have a higher proportion of male recipients (83 and 63 percent, respectively). Bolivia’s survey13 also shows that the share of female recipients is lower than 50 percent. The survey collects an array of social assistance programs, including some that are received exclusively by women: the Bono Juana Azurduy for prenatal and postpartum checkups and the Subsidio Universal Prenatal del Estado. However, men make up most of the recipients in the rest of SA programs. For example, approximately 60 percent of those receiving the prenatal or lactation subsidy are men. Additionally, SI programs report a higher share of male recipients. The combination of all programs leads to a higher incidence of men among all SP recipients in Bolivia. The low female-to-male coverage ratio and the low share of women in SP programs in some countries may also stem from the limited SP data collected by their household surveys and may not necessarily reflect the true situation on the ground. For instance, the 2018 survey in Lao PDR collected data solely on the social insurance scheme, revealing that men made up 69 percent of its recipients. However, the survey did not capture information on social assistance (SA) or labor market programs, despite the country offering a variety of non-contributory benefits and services.14 Similarly, the 2019 survey for the Republic of the Marshall Islands15 reports social assistance data primarily for the Four Atolls 177 program, where men make up 51 percent of recipients. However, the survey does not include information on the 2022 expansion of the conditional cash transfer program, which targets pregnant women and mothers with children under 5. By January 2025, the program had covered 2,438 families, with payments being transferred to bank accounts mostly owned by women (World Bank 2025b). The Democratic Republic of Congo’s survey collects only a few positive observations on SA and SI programs for them to be significant. But overall, the data show that men constitute 68 percent of total SP recipients. 11 Ecuador’s 2021 Encuesta Nacional de Empleo Desempleo y Subempleo. 12 The Islamic Republic of Iran’s 2020 Household Expenditure and Income Survey. 13 Bolivia ‘s 2021 Encuesta de Hogares. 14 Lao PDR offers a range of social assistance programs aimed at supporting poor and disadvantaged families. These include scholarships, a school meals program, technical and vocational training, child grants, and special assistance for vulnerable groups such as children, persons with disabilities, and survivors of unexploded ordnance (UXO) (ILO 2024). 15 The Republic of the Marshall Islands’ 2019 Household Income and Expenditure Survey. STATE OF SOCIAL PROTECTION REPORT 2025 | 13 3.3 Average per capita transfers This section analyzes differences in average per capita transfers received by women and men. Transfer values are expressed in 2017 dollar purchasing power parity (PPP) per day, and the average is calculated only for the recipients of the indicated transfer. The female-to-male ratio of average per capita transfers is then estimated to determine sex biases. This indicator can only be calculated if transfer values are collected by the household surveys. Our sample is reduced to 26 countries as Vanuatu’s survey does not collect these values. Additionally, some programs within countries are excluded because of a lack of transfer value information. Across the sample, men receive, on average, higher SP transfers than women: for every 1 dollar in SP transfers men receive, women receive only 81 cents. The ratio of transfer values is higher for men in all countries in the sample except for Malaysia, Georgia, and Mongolia. The programs driving these results in Malaysia’s survey16 are workers’ compensation, transfers from the Welfare Department and Bantuan Rakyat 1Malaysia (BR1M). In Georgia,17 the slightly higher ratio is driven by noncontributory disability pensions. On the other hand, Ecuador and Colombia18 report the smallest female-to-male transfer ratio: for each 1 dollar in transfers men receive, women receive 48 and 50 cents, respectively. The results for both countries are mostly driven by the value of contributory old age pensions, which are higher for men. Gaps in social protection transfers reflect the gender gaps in labor force participation and labor earnings. The female-to-male ratio of the average per capita SA transfers is 97 percent, meaning, for every US$1 men receive on transfers, women receive 97 cents. However, the same female-to-male ratio for SI is 85 percent, bringing the overall ratio of social protection to 81 percent. SI benefits, including contributory labor market programs, depend directly on the amount of time and level of contributions to a social insurance scheme. Therefore, gaps in labor force participation and labor market earnings may translate into gaps in access and level of SI benefits if insurance schemes do not incorporate design features to support women. Figure 3.3 shows that only in Peru,19 women receive a higher SI transfer than men (8 cents higher on average). In the Republic of the Marshall Islands and Malaysia,20 women and men receive nearly the same value transfer, while in the rest of the countries, the amount of the SI transfers is higher for men. The figure also shows that in 9 out of 27 countries, men receive a higher SA transfer than women. 16 Malaysia’s 2016 Household Income Survey. 17 Georgia’s 2021 Household Integrated Survey. 18 Colombia’s 2022 Gran Encuesta Integrada de Hogares (GEIH). 19 Peru’s 2021 Encuesta Nacional de Hogares sobre Condiciones de Vida y Pobreza (ENAHO). 20 Malaysia’s 2016 Household Income Survey. 14 | STATE OF SOCIAL PROTECTION REPORT 2025 FIGURE 3.3. Social Insurance Programs Are the Drivers of the Lower Female-to-Male Ratio of Average per Capita Social Protection Transfers Female-to-male ratio of average per capita transfers 160% 140% 120% 100% 80% 60% 40% 20% 0% Ecuador 2021 Colombia 2022 Uruguay 2021 Maldives 2019 Kosovo 2022 Chile 2020 Lao PDR 2018 Bolivia 2021 Tunisia 2022 Türkiye 2019 Kazakhstan 2021 Paraguay 2022 Serbia 2019 Panama 2021 Moldova 2018 Brazil 2021 Malaysia 2016 Peru 2021 Mongolia 2020 RMI 2019 Kiribati 2019 Iran, Islamic Rep. 2020 Sri Lanka 2019 Mexico 2022 Georgia 2021 EAP ECA LAC MNA SAR Social Assistance Social Insurance 100% = male level Source: Original calculation using ASPIRE household survey data. https://www.worldbank.org/aspire. Note: RMI refers to The Republic of the Marshall Islands. Average per capita social protection transfer expressed in 2017 dollar PPP per day. The female-to-male ratio is calculated by dividing women’s average per capita transfer by men’s corresponding transfer. Due to the unavail- ability of labor market data in most surveys, the programs are not presented in the chart. EAP= East Asia and the Pacific; ECA= Europe and Central Asia; LAC = Latin America and the Caribbean; MNA= Middle East and North Africa; SAR = South Asia; SSA = Sub-Saharan Africa. Box 3.1. Sex Differences in Social Protection Coverage and Transfer Levels Vary Across Income Distribution Depending on the country, sex differences in social protection coverage may vary drastically across welfare groups. Leveraging data from ten countries across Latin America and the Caribbean (LAC), Figure 3.4 shows that the female-to-male coverage ratio is higher for all countries except Bolivia. However, the ratios vary across countries and income quintiles. For example, in Ecuador, 3.6 women in the poorest quintile have social protection coverage for every man, while only 1.2 of women in the richest quintile are covered for every man in the same quintile. The results are influenced by social assistance programs and reflect the income progressivity of these programs by covering a higher percentage of recipients in the first quintile. Brazil, Colombia, and Uruguay show similar progressive distributions. Meanwhile, Bolivia, Chile, and Peru present more even female-to-male ratios across the quintile distribution. STATE OF SOCIAL PROTECTION REPORT 2025 | 15 FIGURE 3.4. Female-to-Male Ratio of Social Protection Coverage by Quintile 400% 350% 300% 250% 200% 150% 100% 50% 0% Bolivia Brazil Chile Colombia Ecuador Mexico Panama Paraguay Peru Uruguay 2021 2021 2020 2022 2021 2022 2021 2022 2021 2021 Poorest quintile Q2 Q3 Q4 Richest quintile 100% = male level Source: Original calculation using ASPIRE household survey data. https://www.worldbank.org/aspire. Note: Coverage is the percentage of the population receiving social protection transfers. The female-to-male ratio is calculated by dividing the women’s coverage rate by the men’s coverage rate. Quintiles are generated using the pretransfer per capita income. 4. Net Impact: Countries in Which Social Protection Programs Are More Effective at Offsetting Gender Gaps in Labor Market Earnings 4.1 Index 1: Measuring gender inequality in the provision of social protection (SP) benefits This section looks at the aggregated, country-wide compounded effects of labor market earnings and SP spending indicators. First, we estimate the female aggregate SP spending relative to the SP spending accrued to men for each country. We estimate aggregate SP spending by multiplying the female-to-male ratio of SP coverage by the female-to-male ratio of average SP per capita transfers received. Similarly, we estimate the female aggregate labor earnings by multiplying the ratio of female-to-male labor participation by the ratio of female-to-male average monthly labor market earnings.21 21 This aggregate labor earnings assumes that wage differences observed in the survey hold, on average, for all types of formal and informal employment. Therefore, we can aggregate the earning gap, or the sum of wages and informal sector earnings for women relative to the labor market earnings accrued to men. 16 | STATE OF SOCIAL PROTECTION REPORT 2025 An index can be constructed to measure the impact of SP programs in closing the gender gap in take-home income (from labor and SP transfers) between women and men.22 With this index, SP programs can be classified as gender regressive or progressive based on their ability to close those gaps. The results are depicted in Table 4.1 as follows: • Strongly regressive, if the proportion of SP transfers received by women is lower than their share of aggregate labor income and also lower than the share of SP transfers received by men (Column C is lower than 100 percent and also lower than Column B). • Mildly regressive, if the proportion of SP transfers received by women is higher than their share of aggregate labor income but lower than the share of SP transfers received by men (Column C is lower than 100 percent but higher than Column B). • Progressive, if the proportion of SP transfers received by women is greater than the share of transfers received by men (Column C is greater than 100 percent but less than 150 percent). • Strongly progressive, if the proportion of SP transfers received by women is greater than the share of transfers received by men by at least 50 percent (Column C is equal to or greater than 150 percent). The index indicates that, despite progress, some SP systems have a limited impact on mitigating the gender gaps in income. Figure 4.1 and Table 4.1 summarize the main results, showing that less than half of the countries in the analysis are progressive toward women. Kazakhstan, Georgia, and Tunisia stand out as strongly progressive systems because the female-to-male ratio of aggregate SP transfers is 150 percent or higher and more than double the aggregate labor earnings taken home by women. Another seven countries, including three in Latin America and the Caribbean, are progressive since the share of the SP spending accrued to women is still higher than the share accrued to men. Based on the household survey data, achieving gender equality in social protection spending remains a pending agenda in half of the countries analyzed.23 Most countries in Latin America and the Caribbean, for example, show moderately gender regressive results as aggregate SP spending for women is lower than for men, but at least higher than the aggregate labor earnings accrued to women. The same applies to Türkiye, Sri Lanka, and Maldives. Lao PDR and the Democratic Republic of Congo show regressive results, as their aggregate social protection spending for women is lower than that for men and also falls below women’s aggregate labor earnings. Given that our sample is not representative of regions, we cannot draw regional or global conclusions. However, the global situation is likely worse than what is reflected in our sample of countries. 22 The construction of this index is conceptually and methodologically similar to the indices used in concentration curves that measure income inequality. 23 When interpreting sex-disaggregated indicators, it is important to consider that household survey data have inherent caveats and limitations. If household surveys do not include data on large- and medium-size social protection programs, the accuracy of the results may be compromised. STATE OF SOCIAL PROTECTION REPORT 2025 | 17 TABLE 4.1. List of Countries Ranked by the Gender Progressivity of their Social Protection Systems Female-to-male ratio of aggregate labor Female-to-male ratio of aggregate Level of gender Country market earnings (100% male base) SP transfers (100% male base) progressivitya (A) (B) (C) (D) Kazakhstan 64% 212% Strongly progressive Georgia 54% 163% Strongly progressive Tunisia 37% 150% Strongly progressive Mongolia 67% 138% Progressive Malaysia 63% 131% Progressive Moldova 85% 127% Progressive Brazil 63% 114% Progressive Uruguay 70% 108% Progressive Panama 73% 103% Progressive Serbia 73% 103% Progressive Ecuador 75% 98% Mildly Regressive Colombia 70% 96% Mildly Regressive Chile 58% 91% Mildly Regressive Paraguay 66% 91% Mildly Regressive Maldives 49% 90% Mildly Regressive Peru 72% 89% Mildly Regressive Bolivia 71% 85% Mildly Regressive Mexico 52% 79% Mildly Regressive Sri Lanka 39% 61% Mildly Regressive Türkiye 45% 57% Mildly Regressive Lao PDR 70% 36% Regressive Congo, Dem. Rep. 64% 28% Regressive Source: Original elaboration. Note: a. The level of progressivity refers to the extent to which the gap in total social protection (SP) aggregate spending between men and women is reduced. 4.2 Index 2: Measuring the sex-differentiated net effect on access to labor market and SP systems Do social protection transfers reduce or exacerbate gender inequalities in labor earnings? This section creates an index that estimates the aggregate net earnings of men and women by combining labor market earnings and social protection transfers. The index measures whether the social protection system reduces or exacerbates labor market inequality between the earnings of women and men. The net index shows that SP systems narrow the gender gap in take-home income (from labor market earnings and SP transfers) but do not completely close it. To estimate the net effect, aggregate labor earnings and SP spending for both women and men, are weighted by 18 | STATE OF SOCIAL PROTECTION REPORT 2025 their share in terms of gross domestic product (GDP). The weighted female labor earnings and SP receipts are divided by the corresponding male earnings to obtain the female-to- male ratio of net effects. The combined effect of these differences is illustrated in Figure 4.1, which shows that women’s aggregate labor market earnings are lower than men’s across all countries, ranging from 85 percent of men’s earnings in Moldova to just 39 percent in Sri Lanka. The figure illustrates that no SP system across the sample countries manages to completely mitigate the net earnings gap in the labor market. On average, across the sample, women’s aggregate earnings are only 68 percent of men’s, with differences observed across countries. Moldova has a smaller earnings gap, with men earning 14 percent more than women, while Mongolia’s gap is 19 percent. Both countries have female-progressive social protection systems, but aggregate social protection spending does not fully mitigate the gaps in aggregate labor earnings. Türkiye and Sri Lanka show the largest gaps, indicating that the SP systems in these countries hardly alleviate gaps in labor earnings. For example, aggregate female earnings (from labor and SP) in Türkiye are 48 percent of aggregate male earnings. FIGURE 4.1. Social Protection Systems Help Reduce the Gender Earnings Gap, but Are Insufficient to Eliminate It Female-to-male ratio of combined income from labor market earnings and social protection receipts 250 200 150 Percentage 100 50 0 Georgia 2021 Kazakhstan 2021 Mexico 2022 Panama 2021 Colombia 2022 Tunisia 2021 Peru 2021 Ecuador 2021 Congo, Dem. Rep. 2020 Serbia 2019 Uruguay 2021 Paraguay 2022 Lao PDR 2018 Brazil 2021 Chile 2020 Maldives 2019 Bolivia 2021 Sri Lanka 2019 Mongolia 2020 Türkiye 2019 Moldova 2018 Malaysia 2016 EAP ECA LAC MNA SAR SSA Agreggate LM Earnings Aggregate SP Spending Net E ect (100% =male level) Source: Original calculation using ASPIRE household survey data. https://www.worldbank.org/aspire. Note: Combined net earnings are estimated by weighting the female and male aggregate labor earnings and aggregate SP receipts by their share in terms of GDP. The female-to-male ratio is estimated by dividing the female combined weighted earnings from labor and SP receipts by the male corresponding earnings. Earnings data were not available for The Republic of the Marshall Islands, Kiribati, Kosovo, and the Islamic Republic of Iran. EAP= East Asia and the Pacific; ECA= Europe and Central Asia; LAC = Latin America and the Caribbean; MNA= Middle East and North Africa; SAR = South Asia; SSA = Sub-Saharan Africa. STATE OF SOCIAL PROTECTION REPORT 2025 | 19 SP programs alone cannot completely mitigate gender inequalities in the labor market. Governments need to implement comprehensive policies that address gender norms and structural barriers to labor market access. However, the effectiveness of SP measures can be enhanced by paying careful attention to the program mix, design features, program size, and transfer levels. The gender-transformative design of SP systems is critical to shifting deep- rooted structural barriers to gender equality in labor markets. Improving women’s access to social insurance programs by embedding female-preference design parameters is one way to counteract some of the gaps that contributory programs inherit from labor markets. Social assistance programs can have a gender redistributive effect when designed to protect women throughout their life cycle and include measures to address limiting social norms and gender- based violence. Ensuring adequate transfer levels is as important as the size of the program to help offset inequalities rooted in labor market participation and earnings. Policy makers also need to be mindful to prevent SP programs from inadvertently reinforcing gender stereotypes about women’s role as caregivers and jeopardizing their ability to participate in paid work due to the assignment of unequal responsibilities (OECD 2019), as well as the lack of quality and affordable care services. 5. Setting the Data Agenda: Improving Household Survey Design to Generate Sex-Disaggregated Social Protection Indicators Improving the availability of sex-disaggregated social protection (SP) data is crucial to measuring the impact of SP programs in reducing gender inequalities. Identifying the individual recipients and beneficiaries of SP programs helps generate sex-disaggregated indicators to monitor gender biases in SP systems. Three gaps can be estimated: (1) in access to labor market and SP programs; (2) in the level of labor earnings and SP benefits received; and (3) in the aggregate volume of SP spending. With the right information from household surveys, this type of analysis can track progress on the gender dimension of social protection programs over time and across countries. Unfortunately, not all household surveys collect SP information in a format that allows for such analysis. This section develops general recommendations for national statistical offices to improve the availability of sex- disaggregated information in their household surveys. 20 | STATE OF SOCIAL PROTECTION REPORT 2025 Box 5.1. Recipients Versus Beneficiaries of Social Protection Programs Although the terms ‘recipients’ and ‘beneficiaries’ are often used interchangeably, they actually refer to different concepts. Beneficiaries are individuals or households targeted by the program based on its assistance unit (AU). For example, contributory pensions, social pensions, school feeding, and public works programs tend to target individuals. These programs may benefit various individuals within the same household. Some unconditional or conditional cash transfers are targeted to the entire household—in this case—all members of the household are considered direct beneficiaries. R ecipients, on the other hand, refer to individuals who collect the benefits. In programs targeted to individuals, the recipient and direct beneficiary are the same, except in the case of minors or people who are incapacitated and have a legal representative collecting the benefits. For household-targeted programs, usually, one individual receives the transfer on behalf of all household members. The distinction of these concepts is important because sex-disaggregated statistics for these groups will be possible depending on the level at which the data are captured. Analysis of program recipients is possible, only if the household survey captures information at the individual level. However, analysis of direct beneficiaries can still be conducted with information captured at the household level, as long as the program’s AU is the household. Therefore, some household surveys capture SP program data based on the program’s AU. However, other surveys collect information at the household level regardless of the program AU. If the data for programs targeted at individuals are collected at the household level, no sex-disaggregation of recipients or direct beneficiaries is possible. The combination of AU and level of data capture leads to the four scenarios illustrated in Table 5.1. TABLE 5.1. Four Scenarios to Conduct Sex-Disaggregated Analysis of Social Protection Performance Program Assistance Unit (AU) Individual level and Household AU Individual level and Individual AU Does [NAME] benefit from the program? Does [NAME] benefit from the program? [Person ID] [Person ID] Indicators can be generated for the recipient and direct beneficiary. Indicators can be generated for recipients and Level of data direct beneficiaries of the program (all household captured by the members). survey Household level and Household AU Household level and Individual AU Does this household benefit from the program? Does any member of the household benefit from the program? [Household ID] [Household ID] Indicators can be generated for direct beneficia- ries (all household members) but not for recipients. No sex-disaggregation is possible (problem!) Source: Original elaboration. STATE OF SOCIAL PROTECTION REPORT 2025 | 21 Therefore, household surveys that collect SP data at the individual level offer greater flexibility for a richer analysis based on recipients and direct beneficiaries. If this is not possible, surveys should at least collect data at the level of the program assistance unit. National Statistical Offices (NSOs) can take the following steps to ensure that household surveys collect social protection data suitable for sex-disaggregated analysis of program performance: (a) Improve SP program representation in the survey Ensuring that the household survey captures representative data on SP systems is crucial to generate accurate statistics on system performance. Some of the results presented in this paper were limited by the availability of data in the survey. If comprehensive information on the three areas of social protection is not collected, the results may be skewed based only on the type of program with available information. Coordination between NSOs and government agencies implementing SP programs can ensure accurate SP program representation in the survey. Program implementers can provide the NSO with a list of large- and medium-size programs that comprise the bulk of social expenditure in the country. In addition, knowledge of program characteristics—including transfer amounts, frequency of disbursement, and target population—can inform the formulation of survey questions, reference periods, and the selection of appropriate modules for question placement. Dialogue between these stakeholders is also important to determine if new questions or modules need to be added, or whether the sample size needs to be adjusted to increase program representativity. Survey questions and the sample can be calibrated using program size information from administrative records. Based on the program size, survey questions can be formulated to collect data on a single program or aggregate several small programs into one question for adequate representation in the sample. Expanded coverage results can then be cross- checked with administrative records to assess the accuracy of the survey in collecting data on the size of the SP system. Oversampling can be considered to increase the number of positive observations for small programs. Administrative data, reflecting the ‘true’ size of the programs, can inform adjustments to the survey sample or questions as needed. (b) Collect data at the individual level SP data collected at the individual level facilitate various approaches to sex-disaggregated indicators. When data are collected at the individual level, only recipients in the household are tagged with a transfer value or a binary variable indicating program participation. When individuals are tagged, it is possible to disaggregate indicators by individual characteristics, such as sex, age, educational qualification, and work status. This also provides greater flexibility in generating statistics for various types of individuals and households: program recipients, direct beneficiaries (regardless of the program AU), and direct and indirect beneficiaries. Meanwhile, when data are collected at the household level, all members of the household are tagged with the same transfer value (or binary value), making it impossible to identify the program recipient or the direct beneficiary (when the program targets individuals). See Box 5.1. 22 | STATE OF SOCIAL PROTECTION REPORT 2025 SP data collected at the household level will not show significant gaps between male and female statistics. Sex-disaggregated indicators for all household members will reflect the sex distribution of the household, which, on average, tends to be 50/50. Therefore, indicators generated for all household members show little or nonexistent differences among females and males, and gaps may be reduced. Data collected at the household level provide less flexibility in the analysis because statistics can only be generated for the household, which will include direct and direct beneficiaries or direct beneficiaries in the case of household-targeted programs. (c) Collect information on transfer values The availability of monetary data allows the generation of monetary-based indicators, including impact. When the survey collects information on transfer values for programs paid in cash or the estimated value of in-kind transfers, indicators such as the transfer level (absolute and relative), the incidence of benefits, benefit-cost ratio, and simulated impact on poverty and inequality reduction can be estimated. Sex-disaggregated data on benefits can help quantify gender gaps and determine whether the SP programs reduce or reinforce gender gaps in labor market earnings. In summary, to determine gender gaps in SP access and benefits received, NSOs need to design survey questions that collect representative SP data at the individual level, including benefit values. The availability of this information, combined with sex-disaggregated administrative records, will enable the quantification of gender gaps and biases. Such analysis is essential for designing gender-smart SP programs that reduce inequalities between men and women. STATE OF SOCIAL PROTECTION REPORT 2025 | 23 Conclusion Social protection systems, when designed with a gender-sensitive approach, serve as effective mechanisms for addressing gender-specific risks and vulnerabilities impacting women and girls. Research indicates that social assistance programs, including cash transfers and labor market interventions, can enhance women’s socioeconomic status, health, education, financial resilience, and contribute to the reduction of gender-based violence. However, there are still gaps in understanding sex-differentiated access to social protection and measuring their efficacy in alleviating gender inequalities. This paper leverages household survey data to assess gender disparities in access to social protection and its impact on mitigating differences in labor market participation and earnings. By analyzing data from 27 countries, our analysis finds that while on average women receive social protection benefits at a higher rate than men, the transfer amounts they receive are generally lower. This discrepancy is largely driven by contributory old-age pensions, which tend to be higher for men due to gender gaps in labor force participation and accumulated contributions. After analyzing the net effect of aggregate take-home income from labor market earnings and social protection transfers, we conclude that while social protection plays a significant role in mitigating the gap in labor market earnings, it cannot fully close the gender gap in take-home income on its own. To achieve lasting change, governments must implement comprehensive policies that address gender norms and remove structural barriers to labor market access. The effectiveness of social protection measures can be significantly enhanced by carefully considering the program mix, design features, program scale, and transfer levels. The analysis in this paper relies on household surveys that collect data on all large- and medium-sized social protection programs at the individual level, including information on program participation and transfer values. This underscores the crucial importance of improving the design of household surveys to collect such data in order to quantify gender gaps and biases in social protection. The availability of this information will not only inform the design of gender-sensitive social protection programs but also enable the tracking of their performance and impact in reducing gender inequalities over time. 24 | STATE OF SOCIAL PROTECTION REPORT 2025 References Alfers, L., R. Holmes, C. McCrum, and L. Quarterman. 2021. “Gender and Social Protection in the COVID-19 Economic Recovery: Opportunities and Challenges.” Social Protection Approaches to COVID-19 Expert Advice Service (SPACE), DAI Global UK Ltd., United Kingdom. Bastagli, F., J. Hagen-Zanker, L. Harman, V. Barca, G. Sturge, T. and Schmidt. 2016. “Cash Transfers: What Does the Evidence Say? A Rigorous Review of Programme Impact and the Role of Design and Implementation Features.” Overseas Development Institute. Baranov, Victoria, Lisa Cameron, Diana Contreras Suarez, and Claire Thibout. 2021. “Theoretical Underpinnings and Meta-analysis of the Effects of Cash Transfers on Intimate Partner Violence in Low- and Middle-Income Countries.” The Journal of Development Studies 57 (1): 1–25. DOI: 10.1080/00220388.2020.1762859. Camilletti, E. 2020. “Social Protection and Its Effects on Gender Equality: A Literature Review.” Innocenti Working Paper 2020-06. Florence: UNICEF Office of Research – Innocenti. Chinen, M., T. de Hoop, L. Alcázar, M. Balarin, and J. Sennett. 2017. “Vocational and Business Training to Improve Women’s Labour Market Outcomes in Low- and Middle-Income Countries: A Systematic Review.” Campbell Systematic Reviews 13: 1–195. https://doi.org/10.4073/csr.2017.16. Dickson, K., and M. Bangpan. 2012. “Providing Access to Economic Assets for Girls and Young Women in Low-and Lower-Middle-Income Countries: A Systematic Review of the Evidence.” London: EPPI-Centre, Social Science Research Unit, Institute of Education, University of London. Escudero, Verónica, Jochen Kluve, Elva López Mourelo, and Clemente Pignatti. 2017. “Active Labour Market Programmes in Latin America and the Caribbean Evidence from a Meta Analysis.” ILO Working Papers 994970193302676, International Labour Organization. FAO (Food and Agriculture Organization). 2023. “The Status of Women in Agrifood Systems.” Rome. https://doi.org/10.4060/cc5343en. Gavrilovic, M., M. Rubio, F. Bastagli, R. Hinton, S. Staab, R. G. Goulder, C. Bilo, R. Khan, et al. 2022. “Gender-Responsive Social Protection Post–COVID-19.” Science 375 (6585): 1111–1113. https://doi.org/10.1126/science.abm5922. International Labour Office. 2024. “Report of the Mid-Term Review of the National Social Protection Strategy in the Lao People’s Democratic Republic.” Bangkok: International Labour Office, © ILO. STATE OF SOCIAL PROTECTION REPORT 2025 | 25 International Labour Organization. 2024. “ILO Modelled Estimates database,” ILOSTAT. Accessed 04-12-2024. https://ilostat.ilo.org/data/. International Labour Organization. 2023. “Spotlight on Work Statistics: New Data Shine Light on Gender Gaps in the Labour Market.” ILO Brief No. 12, March 2023. Geneva: International Labour Organization. Perera, Camila, Shivit Bakrania, Alessandra Ipince, Zahrah Nesbitt-Ahmed, Oluwaseun Obasola, Dominic Richardson, Jorinde Van de Scheur, and Ruichuan Yu. 2022. “Impact of Social Protection on Gender Equality in Low- and Middle-Income Countries: A Systematic Review of Reviews.” Campbell Systematic Reviews 18: e1240. OECD. 2019. “Enabling Women’s Economic Empowerment: New Approaches to Unpaid Care Work in Developing Countries.” OECD Publishing, Paris, https://doi. org/10.1787/ec90d1b1-en. Tesliuc, Emil, Ana Sofia Martinez Cordova, Mario Gronert, and Usama Zafar. Forthcoming. Do Women Have the Same Access as Men to the Social Protection System? The Case of Mongolia.” Washington, DC: World Bank. World Bank. 2025a. State of Social Protection Report 2025: The 2-Billion-Person Challenge. State of Social Protection Report. Washington, DC: World Bank. World Bank. 2025b. RMI Multisectoral Early Childhood Development Project – II”. Washington DC: World Bank data accessed 10 February, 2025. World Bank. 2024. “ASPIRE (Atlas of Social Protection: Indicators of Resilience and Equity) database.” Accessed 04-12-2024. https://www.worldbank.org/aspire. 26 | STATE OF SOCIAL PROTECTION REPORT 2025 ANNEX 1. Countries and Household Surveys Used in the Analysis No. Country Income Classification Year Survey Name 1 Bolivia Lower middle income 2021 Encuesta de Hogares 2 Brazil Upper middle income 2021 Pesquisa Nacional por Amostra de Domicilios 3 Chile High income 2020 Encuesta de Caracterización Socio-Económica Nacional 4 Colombia Upper middle income 2022 Gran Encuesta Integrada de Hogares (GEIH) 5 Congo, Dem. Rep. Low income 2020 Enquête par Grappes à Indicateurs des Objectifs De Developpement (EGI-ODD) 6 Ecuador Upper middle income 2021 Encuesta Nacional de Empleo Desempleo y Subempleo 7 Georgia Upper middle income 2021 Household Integrated Survey 8 Iran, Islamic Rep. Upper middle income 2020 Household Expenditure and Income Survey 9 Kazakhstan Upper middle income 2021 Household Budget Survey 10 Kiribati Lower middle income 2019 Household Income and Expenditure Survey 11 Kosovo Upper middle income 2017 Household Budget Survey 12 Lao PDR Lower middle income 2018 Expenditure and Consumption Survey 13 Malaysia Upper middle income 2016 Household Income Survey 14 Maldives Upper middle income 2019 Household Income and Expenditure Survey 15 The Republic of the Upper middle income 2019 Household Income and Expenditure Survey Marshall Islands 16 Mexico Upper middle income 2022 Encuesta Nacional de Ingresos y Gastos de los Hogares 17 Moldova Upper middle income 2018 Household Budget Survey 18 Mongolia Upper middle income 2020 Household Socio-Economic Survey 19 Panama High income 2021 Encuesta de Mercado Laboral 20 Paraguay Upper middle income 2022 Encuesta Permanente de Hogares 21 Peru Upper middle income 2021 Encuesta Nacional de Hogares sobre Condiciones de Vida y Pobreza (ENAHO) 22 Serbia Upper middle income 2019 Household Budget Survey 23 Sri Lanka Lower middle income 2019 Household Income and Expenditure Survey 24 Tunisia Lower middle income 2021 Enquête nationale sur le budget la consommation et le niveau de vie des ménage 25 Türkiye Upper middle income 2019 Household Income and Consumption Expenditure Survey 26 Uruguay High income 2021 Encuesta Continua de Hogares 27 Vanuatu Lower middle income 2019 Household Income and Expenditure Survey STATE OF SOCIAL PROTECTION REPORT 2025 | 27 ANNEX 2. Social Protection Performance Analysis TABLE A2.1. Female-to-Male Ratios - Coverage and Average per Capita Transfers of Social Protection Programs SP Average SP Coverage Survey PC Transfer Country Region Income Classification (female-to- Year (female-to-male male ratio) ratio) Sri Lanka SAR Lower middle income 2019 75.3% 81.1% Türkiye ECA Upper middle income 2019 76.3% 74.7% Maldives SAR Upper middle income 2019 113.4% 79.3% Tunisia MENA Lower middle income 2021 160.7% 93.7% Mexico LAC Upper middle income 2022 123.6% 69.6% Chile LAC High income 2020 138.0% 66.0% Congo, Dem. Rep. AFR Low income 2020 47.3% 60.3% Paraguay LAC Upper middle income 2022 110.8% 68.6% Lao PDR EAP Lower middle income 2018 44.3% 81.6% Georgia EAP Upper middle income 2021 144.3% 113.0% Malaysia EAP Upper middle income 2016 104.0% 125.5% Brazil LAC Upper middle income 2021 164.8% 69.2% Uruguay LAC High income 2021 151.0% 71.3% Kazakhstan ECA Upper middle income 2021 212.5% 99.9% Peru LAC Upper middle income 2021 104.2% 85.7% Colombia LAC Upper middle income 2022 174.6% 52.2% Panama LAC High income 2021 113.1% 91.0% Ecuador LAC Upper middle income 2021 205.0% 47.7% Bolivia LAC Lower middle income 2021 94.8% 89.5% Serbia ECA Upper middle income 2019 122.0% 84.3% Mongolia ECA Lower middle income 2020 128.5% 107.6% Moldova ECA Upper middle income 2018 150.9% 84.3% Iran, Islamic Rep. MENA Lower middle income 2020 58.0% 58.7% Kiribati EAP Lower middle income 2019 101.8% 67.7% Kosovo ECA Upper middle income 2017 85.5% 71.3% The Republic of the Marshall Islands EAP Upper middle income 2019 70.0% 109.3% Vanuatu EAP Lower middle income 2019 80.3% n.a. Source: World Bank Atlas of Social Protection: Indicators of Resilience and Equity (ASPIRE). Note: EAP= East Asia and the Pacific; ECA= Europe and Central Asia; LAC = Latin America and the Caribbean; MNA= Middle East and North Africa; SAR = South Asia; SSA = Sub-Saharan Africa. 28 | STATE OF SOCIAL PROTECTION REPORT 2025 Ka Ec za 100% 10% 70% 40% 30% 80% 20% 50% 60% 90% ua 0% 100% 10% kh 70% 40% 30% 80% 20% 50% 60% 90% 0% do Mo stan r ng 20 Br 202 o 21 Mo zil 1a Mo lia 2 l 2 ldo 020 Co dova 021 va lom 2 Ch 201 Ka b 01 8 za ia 8 kh 20 Ma ile 2 s lay 02 0 Ur tan 22 ug 20 Ur sia ug 20 ua 21 ua 16 y y Ch 202 Br 202 ile 1 az 1 i Ki 20 rib 2 Se l 20 a 0 Co rbi 21 Tu ti 2 lom a 2 nis 01 b 02 Ge ia 2 9 Ec ia 20 1 or 02 Note: RMI refers to The Republic of the Marshall Islands. Note: RMI refers to The Republic of the Marshall Islands. ua gi 1 d 22 Se a 20 Me or 2 rb 21 xic 02 Me ia 2 Pa o 2 1 0 na 02 Mo xico 21 m 2 Female ng 20 Female Ko a 20 o 2 so 21 Tü lia 2 2 v rk 0 Ki o 20 Ma iye 20 rib 1 7 Male ldi 20 2 Male Va ati 2 Pa ves 1 nu 01 FIGURE A2.1 Distribution of Social Assistance Recipients by Male and Female na 20 FIGURE A2.2. Distribution of Social Insurance Recipients by Male and Female Pa atu 9 ma 19 ra 20 gu 2 ay 21 Pa Per 021 Bo 20 ra u 2 l 2 g 0 Sr ivia 2 Ma uay 21 i L 20 Source: Original calculation using ASPIRE household survey data. https://www.worldbank.org/aspire. Source: Original calculation using ASPIRE household survey data. https://www.worldbank.org/aspire. an 2 1 lay 202 Tü ka 2 si 2 rk 01 Bo a 20 iye 9 liv 16 2 ia Pe 021 2 ru RM 021 2 Va I 2 R 02 Ira S nuat 019 Ma MI 1 n, ri u 2 Co ld 20 Isl La 0 ng L ive 19 am nk 21 o, ao s 2 ic a 2 De PD 01 Re 01 m. R 9 p 9 Re 20 Ko . 20 p. 18 so 20 20 vo 20 20 17 STATE OF SOCIAL PROTECTION REPORT 2025 | 29 Social Protection & Jobs Discussion Paper Series Titles 2025 No. Title 2513 State of Social Protection Report 2025: The 2-Billion-Person Challenge. Background Paper # 6: Unlocking the Potential of Household Surveys to Measure Women’s Access to Social Protection 2512 State of Social Protection Report 2025: The 2-Billion-Person Challenge. Background Paper # 5: Riding the Demographic Wave: Pensions and Retirement Income in an Aging World 2511 State of Social Protection Report 2025: The 2-Billion-Person Challenge. Background Paper #4: Optimizing Labor Market Programs and Strengthening Delivery Systems for Impact and Scale 2510 State of Social Protection Report 2025: The 2-Billion-Person Challenge. Background Paper #3: Wake-Up Call for Social Assistance? The Unfinished Mission to Reach the Poor and Beyond 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 This paper uses household survey data from 27 countries to assess sex-differentiated access to social protection programs and their impact on mitigating gender gaps in the labor market. The analysis includes indicators of coverage, distribution of social protection recipients, and adequacy of benefits, all disaggregated by sex, to estimate two indices. The first index assesses gender inequalities in the provision of social protection benefits and ranks countries by their level of ‘gender progressivity’. The second index measures the net earnings received by men and women from both labor and social protection transfers, quantifying whether the social protection system reduces or exacerbates labor market gender inequalities. This paper demonstrates the construction and interpretation of these indexes and provides practical recommendations for adapting household surveys to collect the data needed to scale them across emerging and developing economies. JEL CODES J16, J29, I38, O15 KEYWORDS Gender, gender gaps, social protection, social protection systems, social protection benefits, social assistance, social insurance, coverage, average per capita transfer, incidence, labor markets, labor earnings, labor participation. 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