Overcoming Barriers To Women’s Economic Empowerment in the Philippines December 2021 © 2021 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 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, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions 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: 202- 522-2625; e-mail: pubrights@worldbank.org. Cover photo: The World Bank Overcoming Barriers To Women’s Economic Empowerment in the Philippines December 2021 Contents Acknowledgement 12 Executive Summary 13 Introduction 28 CHAPTER 1 Gender Gaps and the Role of Domestic Obligations, 34 Social Norms, and Childcare CHAPTER 2 Education 43 I. Gender Gaps in Enrollment and Attainment 44 II. Learning Outcomes 49 III. University Graduates’ Fields of Study 51 CHAPTER 3 Women in The Labor Market 53 I. The Tenacity of Gender Gaps in the Labor Force 54 II. What Dissuades Women’s Participation in the Labor Force? 58 III. Social Attitudes and Childcare 66 CHAPTER 4 Women’s Employment 74 I. Gender Gaps in Type of Work, Sectors, and Occupations 75 II. Family Responsibilities in Women’s Employment 81 III. Gender Pay Gaps 85 CHAPTER 5 Gender Inequality and the COVID-19 Pandemic 95 I. Impact of COVID-19 on Women’s Employment 98 II. Effect of the Pandemic on Women’s Income 110 III. Coping Mechanisms and Food Security 113 CHAPTER 6 Women, Care Work, Market Work, and the Pandemic: 117 Some Preliminary Findings from a Rapid Survey I. Care Work and Time Use 119 II. Childcare Arrangements 121 III. Attitudes toward Women’s Work 129 Conclusion 127 Appendix A. The Effect of Increased Female Labor Force Participation  134 on Long-Term Economic Growth in the Philippines Appendix B. Unconditional Quantile Decomposition of the Gender Pay Gap 139 Appendix C. Estimates from High Frequency Survey 148 References 149 Figures Executive Summary FIGURE ES.1 Cross-Country Comparison of Fertility Rates, 1960–2018, Total Births per 16 Woman FIGURE ES.2 Cross-Country Comparison of Youth Dependency Ratios, 1960–2018, Percent 16 FIGURE ES.3 Gross Enrollment Rates, by Educational Level and Gender, 1981–2017, Percent 18 FIGURE ES.4 Cross-Country Comparison of Female Labor Force Participation, 2019 18 FIGURE ES.5 Labor Force Participation, by Gender, 1988–2020 19 FIGURE ES.6 Breakdown of Working-Age Population Ages 15+, 2020, Thousands (Percent of 21 Total) FIGURE ES.7 Share of Male and Female Employment, by Occupation and Level of Skill, 21 2020, Percent FIGURE ES.8 Endowments and Returns Effects on Gender Pay Gap, 2020 23 Introduction FIGURE I.1 Labor Force Participation Rates, by Region and Country, 2019 29 FIGURE I.2 Increment to GDP per capita Level due to Higher Female Labor Force 30 Participation, 2020–50 FIGURE I.3 Cumulative Gains to GDP from Higher Female Labor Force Participation, 30 2020–50 CHAPTER 1 Gender Gaps and the Role of Domestic Obligations, Social Norms, and Childcare FIGURE 1.1 Cross-Country Comparison of GSNI and GSNI2 by Gender, Latest Year, Percent 40 CHAPTER 2 Education FIGURE 2.1 Gross Enrollment Rates, by Educational Level and Gender, 1981–2017, Percent 44 FIGURE 2.2 Educational Attainment of Working-Age Population, by Gender, 2020, Percent 45 FIGURE 2.3 Current Enrollment, by Gender and Age, 2020, Percent 45 FIGURE 2.4 Highest Level of Educational Attainment, by Gender and Age, 2020, Percent 46 FIGURE 2.5 Rural and Urban Female School Enrollment, by Age, 2012–20, Percent 46 FIGURE 2.6 Rural and Urban Labor Force Participation of Out of School Youth, by Age, and 47 Gender, 2020, Percent FIGURE 2.7 Rural and Urban Marriage Rates for Out-of-School Youth, by Age and Gender, 47 2020, Percent FIGURE 2.8 Education Levels of Adult Women Ages 15+, by Region, 2020, Percent 48 FIGURE 2.9 School Enrollment, by Age and Income Quantile, 2017, Percent 49 FIGURE 2.10 Female Expected Years of Schooling vs. Learning-Adjusted Years of Schooling, 50 the Philippines and Comparator Countries, 2020 FIGURE 2.11 Share of Students with Basic Skills, the Philippines and Comparator Countries, 50 2018, Percent Executive Summary FIGURE 2.12 Harmonized Test Scores, by Gender, the Philippines and Comparator Countries, 51 2020 FIGURE 2.13 Gender of University Graduates in the Philippines, by Field of Study, 2017, 51 Percent FIGURE 2.14 Share of Female Graduates, by Field of Study, the Philippines and Comparator 52 Countries, Latest Years, Percent CHAPTER 3 Women in The Labor Market FIGURE 3.1 Labor Force Participation, by Gender, 1988–2020 55 FIGURE 3.2 Labor Force Participation, by Gender and Area, 1988–2020, Percent 56 FIGURE 3.3 Regional Differences in Labor Force Participation, by Gender, 2020 56 FIGURE 3.4 Cross-Country Comparison of Female Labor Force Participation, the Philippines 57 and Comparator Countries, 2019 FIGURE 3.5 Labor Force Participation, by Educational Attainment and Gender, 2020 58 FIGURE 3.6 Labor Force Participation, by Gender and Age Group, 2002–20, Percent 59 FIGURE 3.7 Rural and Urban Labor Force Participation, by Gender, 2020, Percent 59 FIGURE 3.8 Marriage and Labor Force Participation, by Age Group, Location, and Gender, 60 2020, Percent FIGURE 3.9 Factors Influencing Labor Force Participation, by Gender, Marital Status, and 61 Children, 2020, Percent FIGURE 3.10 Female Employment, by Income Quintile and Presence of Extended Family 62 Members in the Household, 2017, Percent FIGURE 3.11 Female Employment, by Presence of Extended Family Members and Preschool- 62 Age Children in the Household, 2017, Percent FIGURE 3.12 Preschools and Women’s Labor Force Participation, by Region 63 FIGURE 3.13 Gender Equality Indicators, the Philippines and Comparator Countries, 2019, 67 Percent FIGURE 3.14 Share of Seats in Parliament Held by Women, 2018, Percent 67 FIGURE 3.15 Agreement with “Men should have more right to a job than women when jobs 68 are scarce,” by Country, 1994–2020, Percent FIGURE 3.16 Agreement by Women with “Men should have more right to a job when jobs 68 are scarce,” Philippines, 2019, Percent FIGURE 3.17 Bias Against Women’s Economic Opportunities, by Country, 2019, Percent 68 FIGURE 3.18 Agreement by Women with “A preschool child suffers with a working mother,” 68 Philippines, 2019, Percent FIGURE 3.19 Attitude toward “Women should work when they have children,” Philippines, 69 2012, Percent FIGURE 3.20 Women’s Work Status when They Have Children, Philippines, 2012, Percent 69 FIGURE 3.21 Attitude towards Who Should Provide Childcare, by Gender, Philippines, 2012, 71 Percent FIGURE 3.22 Best Childcare Option, by Parents’ Employment Type, Philippines, 2012, 71 Percent FIGURE 3.23 Attitude toward Parental Paid Leave, Philippines, 2012, Percent 72 Executive Summary FIGURE 3.24 Opinion on “How should this paid leave period be divided between the mother 72 and the father?” Philippines, 2012, Percent FIGURE 3.25 Mother vs. Daughter Employment Status, Philippines, 2019, Percent 72 FIGURE 3.26 Mother vs. Daughter Education Level, Philippines, 2019, Percent 72 FIGURE 3.27 Mother vs. Daughter Occupation, Philippines, 2019, Percent 72 FIGURE 3.28 Mother vs. Daughter Employment Status, by Age and Income Group, 73 Philippines, 2009, Percent FIGURE 3.29 Women vs. Husband Employment Status, by Age and Income Group, 73 Philippines, 2009, Percent CHAPTER 4 Women’s Employment FIGURE 4.1 Trends in Vulnerable Employment, by Gender, 1988–2020, Percent 75 FIGURE 4.2 Trends in Type of Employment, by Gender, 1988–2020, Percent 76 FIGURE 4.3 Average Total Time Worked, by Age and Gender, 2012–20, Hours per Week 78 FIGURE 4.4 Average Time Spent on Market Work, by Age and Gender, 2012–20, Hours per 78 Week FIGURE 4.5 Sectoral Trends in Employment, 2012–20, Percent 78 FIGURE 4.6 Share of Employment, by Industry and Gender, 2020, Percent 79 FIGURE 4.7 Employment, by Occupation and Gender, 2020, Percent 80 FIGURE 4.8 Educational Attainment of Employed, by Gender, 2020, Percent 80 FIGURE 4.9 Wage Work as a Share of Employment, by Education and Gender, 2020, 81 Percent FIGURE 4.10 Educational Attainment, by Familial Status and Gender, for Employed 82 Population, Ages 25–39, 2020, Percent FIGURE 4.11 Female Employment Type, by Familial Status, Ages 25–39, 2020, Percent 83 FIGURE 4.12 Average Hours Worked, by Familial Status and Gender, Ages 25–39, 2020 83 FIGURE 4.13 Share of Rural Employment in Agriculture, by Familial Status and Gender, 84 2020, Percent FIGURE 4.14 Share of Employed Women in Wage Employment, by Familial Status and Type 84 of Support, Ages 25–39, 2020, Percent FIGURE 4.15 Average Total Hours Worked, by Familial Status, Gender, and Type of Support, 85 Ages 25–39, 2020 FIGURE 4.16 Density of Daily and Hourly Log Wages, by Gender, 2012 and 2020 86 FIGURE 4.17 Ratio of Women’s to Men’s Average Wages, by Educational Attainment, 2012 87 and 2020, Percent FIGURE 4.18 Endowments and Returns Effects on Gender Pay Gap, 2012 89 FIGURE 4.19 Endowments and Returns Effects on Gender Pay Gap, 2020 89 FIGURE 4.20 Average Daily Wage, by Gender and Percentile, 2012 and 2020, PhP 90 FIGURE 4.21 Detailed Endowments and Returns Effects on Gender Pay Gap, 2012 91 FIGURE 4.22 Detailed Endowments and Returns Effects on Gender Pay Gap, 2020 91 FIGURE 4.23 Endowment and Return Effects of Education on Gender Pay Gap, 2012 92 FIGURE 4.24 Endowment and Return Effects of Education on Gender Pay Gap, 2020 92 FIGURE 4.25 Endowment and Return Effects of Employment on Gender Pay Gap, 2012 92 Executive Summary FIGURE 4.26 Endowment and Return Effects of Employment on Gender Pay Gap, 2020 92 FIGURE 4.27 Endowment and Return Effects of Demographic Characteristics on Gender Pay 93 Gap, 2012 FIGURE 4.28 Endowment and Return Effects of Demographic Characteristics on Gender Pay 93 Gap, 2020 FIGURE 4.29 Endowment and Return Effects of Childcare Assistance on Gender Pay Gap, 94 2012 FIGURE 4.30 Endowment and Return Effects of Childcare Assistance on Gender Pay Gap, 94 2020 CHAPTER 5 Gender Inequality and the COVID-19 Pandemic FIGURE 5.1 Employment Trends, by Gender and Area, 2019 to 2021, Percent 98 FIGURE 5.2 Reason for Not Working, by Gender and Area, April 2020, Percent 99 FIGURE 5.3 Reason Not Working for those Working in the Previous Quarter, April 2020, 99 Percent FIGURE 5.4 Change in Employment Status Between March 2020 and December 2020, 100 Percent FIGURE 5.5 Proportion of People who Lost their Jobs, by Gender, Area, and Round, 2020, 101 Percent FIGURE 5.6 Recovery of Jobs Lost in August 2020, Percent 102 FIGURE 5.7 Proportion of Jobs Lost, by Sector, August 2020, Percent 103 FIGURE 5.8 Employment, by Sector, Gender, and Area, August 2020, Percent 103 FIGURE 5.9 Reason for Not Working, by Gender, August 2020, Percent 104 FIGURE 5.10 Reason for Not Working, by Gender, December 2020, Percent 104 FIGURE 5.11 Reason for Women Not Working, by Location, Income Group, and Family, August 105 2020, Percent FIGURE 5.12 Reason for Women Not Working, by Location, Income Group, and Family, 105 December 2020, Percent FIGURE 5.13 Primary Assistant with Distance Learning, by Sex of Household Head, December 106 2020, Percent FIGURE 5.14 Primary Barriers to Children Learning Effectively, December 2020, Percent 107 FIGURE 5.15 Access to Internet in the Home, by Quintile and Round, 2020, Percent 107 FIGURE 5.16 Proportion of Employed People Working from Home, by Gender, 2020, Percent 108 FIGURE 5.17 Loss of Income, by Quintile, August 2020, Percent 110 FIGURE 5.18 Loss of Income, by Gender and Urban/Rural, 2020, Percent 111 FIGURE 5.19 Loss of Income, by Industry and Gender, August 2020, Percent 111 FIGURE 5.20 Income Changes for Non-farm and Farm Businesses compared to Previous 112 Period, 2019 to August 2020, Percent FIGURE 5.21 Coping Mechanisms, by Gender of Household Head, August 2020, Percent 113 FIGURE 5.22 Coping Mechanisms, by Quintile, August 2020, Percent 114 FIGURE 5.23 Inability to Purchase Food, by Gender of Household Head, August 2020, Percent 114 FIGURE 5.24 Food Insecurity, August and December 2020, Percent 115 FIGURE 5.25 Food Insecurity, by Gender, Income, and Number of Children in the Household, 116 August 2020, Percent FIGURE 5.26 Top Three Pressing Problems, 2019 and August 2020, Percent 116 FIGURE 5.27 Groups that Need More Assistance, August 2020, Percent 116 Executive Summary CHAPTER 6 Women, Care Work, Market Work, and the Pandemic: Some Preliminary Findings from a Rapid Survey FIGURE 6.1 Relation to Child of Primary Caregiver, May 2021, Percent 119 FIGURE 6.2 Respondents’ Role in Caregiving to Children, May 2021, Percent 119 FIGURE 6.3 Number of Hours per Week Spent in Different Activities, May 2021 120 FIGURE 6.4 Perceived Increase in Time Use on Caregiving during COVID-19, Percent 120 FIGURE 6.5 Activities Unable to Do because of Domestic Work, Percent 121 FIGURE 6.6 Activities Unable to Do because of Childcare, Percent 121 FIGURE 6.7 Decision-Maker on Childcare, Percent 121 FIGURE 6.8 Amount Willing to Pay per Week for High-Quality Childcare, Percent 121 FIGURE 6.9 Reason Why Children Ages 0-4 Years Are Not Enrolled in Childcare, Percent 122 FIGURE 6.10 Primary Method of Childcare, Percent 122 FIGURE 6.11 Women’s Opinions on Childcare Arrangements and Women’s Work, Percent 123 FIGURE 6.12 Opinion on Whether Women Contribute Better to Society by Having a Job, 124 Percent FIGURE 6.13 Opinion on Whether Women Contribute Better to Family by Having a Job, 124 Percent FIGURE 6.14 Opinion on Whether Women Contribute Better to Society by Staying Home, 125 Percent FIGURE 6.15 Hesitancy about the Contribution of Women’s Work to Society, Percent 125 FIGURE 6.16 Main Obstacles to a Mother’s to Return to Work, Percent 125 FIGURE 6.17 Opinion on Best Way to Organize Family-Work Life When There Is a Child 126 Under School Age, Percent FIGURE 6.18 Respondents’ Parents’ Family-Work Life When They Had a Child Under School 126 Age, Percent Appendix A. The Effect of Increased Female Labor Force Participation on Long-Term Economic Growth in the Philippines FIGURE A.1 Female Labor Force Participation 134 FIGURE A.2 Increment to GDP Growth from Higher FLFP, Percentage Points (Scenario less 134 Baseline) FIGURE A.3 Increment to GDPPC Level from Higher FLFP 135 FIGURE A.4 Increment to Total GDP Level from Higher FLFP 135 FIGURE A.5 Cumulative Gains to GDP from Higher FLFP 136 FIGURE A.6 Robustness 1: Extra GDP Growth from Higher FLFP, Percentage Points 137 FIGURE A.7 Robustness 2: Extra GDP Growth from Higher FLFP, Percentage Points 137 Appendix B. Unconditional Quantile Decomposition of the Gender Pay Gap FIGURE B.1 Aggregate Effects, Linear Model, 2020 142 FIGURE B.2 Aggregate Effects, Probit Reweighting Model, 2020 142 FIGURE B.3 Detailed Effects, Linear Model, 2020 143 FIGURE B.4 Detailed Effects, Probit Reweighting Model, 2020 143 FIGURE B.5 Endowment and Returns Gaps at Top Percentiles, 2020, Percent 144 Executive Summary Appendix C. Estimates from High Frequency Survey FIGURE C.1 Share of Men and Women, by Sector, 2020, Percent 148 FIGURE C.2 Proportion of Jobs Lost by Sector and Gender, 2020, Percent 148 FIGURE C.3 Employment, by Sector, Gender, and Region, 2020, Percent 148 FIGURE C.4 Employment, by Sector, Gender, and Quintiles, 2020, Percent 148  Tables TABLE 3.1 Determinants of Women’s Labor Force Participation, 2020, Average Marginal 65 Effects TABLE 4.1 Estimates of Endowments and Returns Effects on Gender Pay Gap, 2012 and 89 2020 TABLE B.1 Unconditional Quantile Regression, by Gender, 2012 145 TABLE B.2 Unconditional Quantile Regression, by Gender, 2020 146 Acknowledgement The members of the core team that prepared this report are Nadia Belhaj Hassine Belghith (EEAPV, co-TTL), Benjamin Lavin (EEAPV), and Hannah LaPalombara (EEAPV), with the help of Sharon Faye Alariao Piza (EEAPV), Vincent M. Abrigo (EACPF), Francine Claire Fernandez (EEAPV) and guidance of Helle Buchhave (SEAS1, TTL). Additional input and feedback were provided by Claire Nicole Charamnac (SEAS1), and Forest Brach Jarvis (EAPCE). The analysis of the long-term growth impact of women’s labor force participation was prepared by Steven Pennings (DECMG) with support from Kevin Thomas Garcia Cruz (EEAM2). The team worked under the guidance of Ndiamé Diop (Country Director, EACPF), Achim Fock (Manager Operations, EACPF), Rinku Murgai (Practice Manager, EEAPV), and Janmejay Singh (Practice Manager SEAS1, SEAS2). The team gratefully acknowledges advice from Souleymane Coulibaly (Program Leader, EEADR) and Madhu Raghunath (Sector Leader, SEADR). The team is grateful to Maria Loreto (Malu) Padua (SEAS2), Ririn Salwa Purnamasari (EEAPV), Grace Borja (DFAT) and the DFAT team, Clarissa Crisostomo David (ECREA) and the EXT team, and Yoonyoung Cho (HEASP) and the EAPGIL team for their constructive comments during the decision meeting. The team also offers its thanks to Mildred Thomas Gonsalvez (EEAPV) and Kristiana Gizelle Torres Rosario (EACPF) for their precious help in the preparation of the report. The report benefited from consultations and insights from the Philippines National Economic and Development Authority (NEDA) and the Philippines Commission on Women (PCW). The team acknowledges guidance from the Philippines Statistical Authority (PSA) and NEDA as well as technical and financial support from the Australian Government. The report was prepared as part of The World Bank’s Philippines Women’s Economic Empowerment Program funded by the Australia – World Bank Growth and Prosperity Trust Fund. Executive Summary Overcoming Barriers To Women’s Economic Empowerment In The Philippines 13 Executive Summary Executive Summary Gender equality is intrinsically linked to sustainable and inclusive growth and development. No society can afford to lose out on the skills and capabilities of half of its population if it wants to realize the promise of a more prosperous future. The Philippines Country Gender Action Plan (CGAP) for fiscal year (FY) 2020 through FY2024 identifies addressing women’s low labor force participation as a priority area (World Bank 2020a). With that priority in mind, this report examines the constraints for women’s employment by looking at education, labor market participation, and wages with a focus on the role of childcare and social norms. Gender disparities are complex1 because of their multidimensional aspects and changing nature—and are among the most entrenched forms of inequality and the greatest barriers to human development. In the Philippines, women can play a catalytic role in promoting social development in order to achieve the country’s AmBisyon Natin 2040 vision. Expanding women’s opportunities, enhancing their productivity, and promoting their economic participation are among the most promising avenues for fostering inclusive development in the Philippines. In recent years, the Philippines’ legal framework has helped move the country toward gender equality. The passage in 2009 of the comprehensive Magna Carta of Women Act helped to eliminate legal barriers that women face in the workplace.2 More recent legislation, including the Domestic Workers Act in 2013 and the 105-Day Expanded Maternity Leave Law in 2019, has expanded protections for women. In 2021, the Philippines received a score of 78.8 out of 100 on the World Bank’s Women, Business and the Law index, which measures legislation affecting women in the realms of the workplace, parenthood, entrepreneurship, pay, pension, mobility, marriage, and assets. In fact, the Philippines received a perfect score of 100 in the workplace, pay, and entrepreneurship subcomponents; these scores indicate that women face no legal barriers to labor force participation, workplace discrimination and sexual harassment are legally prohibited, wages and access to different sectors are legally mandated as equal, and women have equal access to credit, bank accounts, and business registration (World Bank 2021). The Philippines has embraced gender equality among its current policy priorities. The country has made notable progress toward closing the gender gaps in many human development dimensions, but it has a long way to go yet before realizing the full potential of its female workforce and achieving equality between men and women in economic opportunities. The World Bank Group’s Country Partnership Framework (CPF) FY2020–FY2025 aims to help the Philippines reach its AmBisyon Natin 2040 vision through reducing core constraints to inclusive growth and poverty reduction (World Bank Group 2019a). It aims to accomplish this by promoting investment in human capital, engendering economic opportunity and jobs and job creation, and building peace and resilience. The CPF summarizes the main gender gaps that act as obstacles to this vision, including gaps in labor force participation and the labor market. The Philippines FY2020–FY2024 CGAP serves to guide the World Bank in its efforts to address the obstacles to gender equality outlined in the CPF. It focuses on four pillars: (1) increasing female labor force participation, (2) improving education quality and outcomes, (3) reducing gender-based vulnerabilities related to conflict and disasters, and (4) enhancing gender parity in the public sector. 1  The report considers the supply of women’s labor and does not address factors on the demand side (that is, firms’ demand for female labor). 2  For more on the Magna Carta of Women Act, see https://psa.gov.ph/content/q-magna-carta-women-republic-act-no-9710. 14 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Executive Summary Increasing female labor force participation is instrumental for accelerating economic growth and fulfilling the Philippines’ aspiration to achieve upper-middle-income status. Lower women’s labor force participation represents a missed opportunity for economic growth and increased prosperity in the Philippines; an increase of women’s labor supply by a mere 0.5 percentage points (pp) per year would increase gross domestic product (GDP) per capita by about 6 percent by 2040 and almost 10 percent by 2050, translating into an additional US$500 (real 2010 US dollars) in GDP per capita by 2040 and US$1,325 by 2050, which would provide households with substantial income gains. The Philippines’ GDP would be US$70 billion higher than the baseline projection by 2040 and US$190 billion higher by 2050. Understanding the complexities of gender gaps in the Philippines is a daunting task because several factors on the supply and demand sides are at play and often mutually reinforce each other. Despite an extensive literature analyzing the drivers of—and obstacles to—women’s labor force participation worldwide, and some empirical evidence of correlation for several factors, there is no conclusive consensus on which factors universally matter. This report aims to contribute to a better understanding of gender inequalities in the Philippines’ labor market to help inform strategies to promote women’s economic opportunities. The report is not intended to comprehensively address all factors affecting female labor force participation, but instead focuses on a subset of determinants of female labor force participation: education, childcare, and social norms. The selected determinants are those supply-side factors identified as being of particular interest in the CGAP, as well as in the recent literature, and are not as well understood as other determinants. Although gender parity has been reached in education, female labor force participation remains low, despite demographic and economic changes in the Philippines. The Philippines has seen a consistent slow decline in the fertility rate but has yet to experience the demographic dividend seen in other countries in the East Asia and Pacific region. Since the late 1960s, the Philippines’ fertility rate has fallen at a slower rate than other countries in the East Asia and Pacific (EAP) region and the region as a whole (figure ES.1). By 2018, the fertility rate in the Philippines was 2.6 births per woman, slightly higher than most comparator countries but much lower than the 1960 rate of 7.1 births per woman. Although the Philippines’ youth dependency ratio is still above the regional average, it has continued to trend downward, reaching 48 percent in 2019 (figure ES.2). The Philippines’ youth dependency ratio has not decreased to the same level as the regional average, which fell from 70 percent in 1960 to 28 percent in 2019. However, the Philippines is not facing a rapidly aging population, unlike many of its peers. The United Nations projects that the share of the working-age population will rise slightly from 64 percent in 2020 to 66 percent in 2040, whereas the old-age population will rise slightly from 5 percent to 8 percent in the same period (World Bank Group 2019b). These data indicate that, although the country has not seen a large demographic dividend, it will have a demographic advantage in future years as the population remains fairly young and the share of the population above 65 years old is not expected to increase rapidly. This demographic advantage will allow for the possibility of higher productivity and economic growth; however, in order to harness the potential of this demographic advantage, the Philippines must promote gender equality in human capital and economic opportunities so that women—who currently make up 39 percent of the labor force—can more easily participate. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 15 Executive Summary FIGURE ES.1 Cross-Country Comparison of Fertility Rates, 1960–2018, Total Births per Woman Source: World Development Indicators, 2020. Note: EAP = East Asia and Pacific; PHL = Philippines FIGURE ES.2 Cross-Country Comparison of Youth Dependency Ratios, 1960–2018, Percent Source: World Development Indicators, 2020. Note: The youth dependency ratio is the ratio of younger dependents (under 15 years old) to the working-age population, those ages 15 to 64 years old. EAP = East Asia and Pacific; PHL = Philippines. 16 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Executive Summary Over the past few decades, the Philippines has experienced considerable poverty reduction and broad economic improvements. Since 1985, poverty in the Philippines has steadily decreased from 49.2 percent to 16.7 percent in 2018, with much of this progress realized in the last decade. In line with this trend, the middle class grew by 5 million people from 2006 to 2018, resulting in a middle class of nearly 12 million Filipinos; meanwhile, the economically secure population grew by 20 million to reach 44 million people in the same period. The Philippines’ Human Development Index improved from 0.590 in 1990 to 0.712 in 2018 (UNDP 2019), and basic services such as electricity and drinking water reached over 90 percent of the population. As a reflection of the economic and well-being improvements, 74 percent of the population in 2018 reported satisfaction with their living standard. The Philippines made important advances toward gender equality, but progress has plateaued in recent years. According to the 2021 World Economic Forum Global Gender Gap Report, the Philippines has closed 78.4 percent of its overall gender gap, placing it 17th out of 156 countries. This is higher than the EAP average of 68.8 percent, making the Philippines the second-best performer in the region after New Zealand, but represents a 1.5-pp and nine-rank decrease from 2018 (WEF 2021). In terms of area-specific gender gaps, the Philippines has fully closed the educational and health and survival gaps, but has closed only 79 percent of the economic gap and 36 percent of the political empowerment gap. Although these proportions are above the regional averages, all but the Philippines’ health and survival indicator decreased from 2018. An additional indicator of the country’s challenges in reaching gender equity is reflected in the United Nations Development Programme’s Gender Inequality Index, on which the Philippines performs poorly, ranking 107th out of 189 countries with a score of 0.43 in 2020, worse than the EAP average of 0.324 (UNDP 2020b).3 The large difference in the World Economic Forum and Gender Inequality Index rankings is due to the low levels of female labor force participation in the Philippines and the weights each measure assigns to gender parity in labor force participation. 3  GII scores range from 0 to 1, with 0 as most equal and 1 as most unequal. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 17 Executive Summary Women have historically fared well in terms of access to education, reaching higher enrollment and completion rates than men at both secondary and tertiary levels. By 2017—and at least as early as 1981—both men and women had reached universal primary school enrollment (figure ES.3). In the same year, women had a higher rate of secondary enrollment, at 91 percent, whereas men’s enrollment rate was 82 percent, representing a reverse gender gap of 9 pp. The gender gap in favor of women at the tertiary level was also 9 pp, with women’s enrollment at 40 percent in comparison to men’s enrollment of 31 percent. According to the Philippines’ 2020 Labor Force Survey, 25 percent of women 25 to 39 years had completed university compared to 19 percent of men. These statistics indicate that a larger share of women than men in the Philippines are receiving and completing university education. FIGURE ES.3 Gross Enrollment Rates, by Educational Level and Gender, 1981–2017, Percent Source: United Nations Educational, Scientific and Cultural Organization’s Institute for Statistics, 2020. The Philippines trails its peers in female labor force participation, with one of the highest labor force gender gaps among regional comparators. At 49 percent, the Philippines’ female labor force participation in 2019 was the lowest among comparator countries in the region (figure ES.4, panel a), and its labor force gender gap of 27 pp was the second largest across these countries, after Indonesia (figure ES.4, panel b). With significant economic growth since 1988, female labor force participation has remained below East Asian peer countries but has moved closer to the U-Shaped Female Labor Force Function (figure ES.4, panel c). FIGURE ES.4 Cross-Country Comparison of Female Labor Force Participation, 2019 a. Labor Force Participation of Peer Countries, b. Gender Gaps in Labor Force Participation of by Gender, 2019, Percent Peer Countries, 2019, Percentage Points 18 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Executive Summary c. Female Labor Force Participation, by GDP per capita, 2019 Source: World Development Indicators, 2020. Note: Female labor force participation by GDP per capita is for the year 2019 with East Asian countries marked in orange except for the Philippines, which includes data points for the years 1988, 1997, 2008 and 2019. CHN = China; IDN = Indonesia; KHM = Cambodia; MYS = Malaysia; PHL = Philippines; THA = Thailand. The gender gap in the labor force has persisted across decades. Although the gender gap in the country’s labor force has fallen by about 7 pp since 1988, much of this progress is attributable to a decrease in the labor force participation of men rather than an increase in the labor force participation of women (figure ES.5). Progress toward closing the gap has been minimal in recent years, with the gap shrinking by a mere 4 pp since 2004. FIGURE ES.5 Labor Force Participation, by Gender, 1988–2020 Overseas Men Overseas Women Source: Based on data from the 1988–2020 Labor Force Surveys. Note: The labor force participation (LFP) rate is the proportion of the total labor force to the total household population ages 15 years and over (excluding overseas workers). The proportion of overseas workers is the number of Filipinos working abroad over the total population ages 15 years and over. Estimates represent annual averages except for 2020, for which only January and April quarters are available. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 19 Executive Summary Childcare, social norms, and stereotypes about gender roles in the household are holding back women’s participation in the labor market. Biases in gender social norms against more enhanced forms of women’s economic participation are widespread in the Philippines. Data from the International Social Survey Programme’s World Value and Gender Role surveys show that about 70 percent of male and female respondents favor men over women for a job; over 80 percent agree that the man’s job is to earn money while the woman’s job is to look after the home; and about half believe that a preschool child suffers with a working mother. Biases have intensified in this thinking, and the extent of the backlash was higher among women. Gender attitudes and beliefs about women’s work are mutually reinforcing and act as a barrier to women’s participation in economic life. They are associated with a reduction of about 14–22 pp in the probability of women’s engagement in the labor market. The number of young children also tends to depress women’s employment likelihood. One additional child aged two years or younger is associated with a 7-pp reduction in the probability of women’s participation in the labor market, but the presence of domestic help in the household reduces the negative effect of childcare on women’s employment likelihood. The number of children aged three years and older does not seem to matter significantly. Household income security, in particular the presence of overseas workers and of men with salaried employment in the household as well as higher education levels of the head, seem to penalize women’s participation in the labor force. The negative relationship is higher for women with low education than for highly educated ones, who are probably less influenced by stigmas attached to women working if they are in more secure economic environments. Education helps to advance women’s labor force participation, but high returns to labor force participation would occur only at the college level and above. Relative to women with primary education or less, women with a college degree or above are 27-pp more likely to participate in the labor force. Many women are engaged in high-skill occupations, but a large proportion of women are held back from productive employment opportunities by their family responsibilities. Men and women who are in the labor market exhibit different patterns of employment. According to the 2020 Labor Force Survey, 75 percent of men aged 15 years and older were in the labor force, compared with 48 percent of women, with employment rates for both groups reaching about 95 percent (figure ES.6). Women are concentrated in low-skill positions (because of economic necessity) and high-skill occupations (because of high rates of education). Although the largest share of both employed men and women worked in private establishments (59 percent and 40 percent, respectively) and self-employment (26 percent for both men and women), a larger share of women were employed without pay in family-owned businesses and in private households as domestic workers—occupations that tend to offer narrower avenues for skills development and career growth. Marital status and children also affect employment patterns; single working women of peak reproductive ages of 25 to 39 are more likely to both engage in wage work and work longer hours than married women and those with young children, suggesting that gendered family duties may lead women to choose more flexible employment options. 20 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Executive Summary FIGURE ES.6 Breakdown of Working-Age Population Ages 15+, 2020, Thousands (Percent of Total) Working Age Population 72,841 Male Female 36,571 36,270 (50.2%) (49.8%) Labor Force Out of Labor Force Labor Force Out of Labor Force 27,372 9,199 17,562 18,708 (74.8%) (25.2%) (48.4%) (51.6%) Employed Unemployed Employed Unemployed 25,860 1,512 16,683 879 (94.5%) (5.5%) (95%) (5%) Private Private Gov’t/Gov’t With pay Without pay household establishment corporation Self-employed Employer (FOB) (FOB) 6,732 81 300 15,308 1,749 692 997 (26%) (0.3%) (1.2%) (59.2%) (6.8%) (2.7%) (3.9%) Private Private Gov’t/Gov’t With pay Without pay household establishment Self-employed Employer corporation (FOB) (FOB) 4,403 310 1,550 6,586 2,146 36 1,652 (26.4%) (1.9%) (9.3%) (39.5%) (12.9%) (0.2%) (9.9%) Source: Based on data from the 2020 Labor Force Survey. Note: FOB = family-owned business; Gov’t = government. Women occupy a large share of leadership positions and high-skill occupations. Women occupy 49 percent of managerial roles in the country, while representing 65 percent of professionals and 51 percent of technicians and associated professionals (figure ES.7). The high prevalence of women in these positions likely reflects Filipino women’s higher rates of educational attainment relative to men, particularly at the tertiary level. However, the concentration of women in high-skill positions declines considerably when they have young children. FIGURE ES.7 Share of Male and Female Employment, by Occupation and Level of Skill, 2020, Percent Source: Based on data from the 2020 Labor Force Survey. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 21 Executive Summary Women earn more on average than men, but the gender earnings gap shows large variations across the pay distribution. Women earn about 5 percent more in mean daily wages and 7 percent more in mean hourly wages relative to men. However, the distribution of women’s wages is wider than men’s wages, indicating that women tend to cluster at two ends of the wage distribution— in low-paying jobs and in high-paying jobs—as opposed to medium-paying positions, where men are more clustered. This clustering suggests that women at the low end of the distribution must work to avoid falling further into poverty, whereas women at the high end of the distribution may be more likely to remain in the labor force because of high earnings potential. At the bottom of the pay distribution, the daily wage is over 50 percent higher for men than for women; at the top of the distribution, the daily wage is about 20 percent higher for women than for men. The pay differential in favor of men in low-wage groups is explained by the fact that women have a lower skill level than men; however, even when women have the same qualifications and attributes as men, they still receive lower pay or returns for those attributes (figure ES.8). This gap in returns does not seem to be related to obvious discriminatory policies or practices in the labor market, but rather to women’s inability to take up some jobs that offer more adequate returns to their skills. In part, women’s earning disadvantage at the lower end of the distribution may be due to their desire to work in jobs that offer more flexible hours and work arrangements, as well as to social stigmas that deem certain types of jobs to be men’s work. The situation is different at the upper end of the wage distribution, where women have higher pay than men because they have higher qualifications and characteristics, essentially higher tertiary education and higher concentration in high-skill occupations. The overall wage gap in favor of men at the bottom of the distribution has increased over time, whereas the gap in favor of women at the top has narrowed over time. The gender pay gap is smallest at the median of the distribution. As shown in figure ES.8, the pay differential between men and women is lowest at the median of the distribution, where men’s wages were only 4 percent higher relative to women in 2020. This represents a considerable improvement from 2012, when the daily wage was around 18 percent higher for men at the median. The decline in the gender pay difference for this group was due to a reduction in the gender gap in the returns to the number of hours worked as well as the returns to employment in agriculture and low-end services (such as shop and market sales and other unskilled services). However, it is worth noting that despite the reduction of the gender pay differential for the median group in 2020, women continue to be disadvantaged by the higher burden exerted by children on their returns. 22 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Executive Summary FIGURE ES.8 Endowments and Returns Effects on Gender Pay Gap, 2020 .4 Difference in log daily wage .2 0 -.2 -.4 -.6 .1 .2 .3 .4 .5 .6 .7 .8 .9 Quantiles Total wage gap Pure Endowments Effect Pure Returns Effect Source: Based on data from the 2020 Labor Force Survey. Note: The decomposition estimates the difference between men and women’s log wages. Negative values mean that the gap is at the advantage of women. The COVID-19 crisis has had severe effects on women’s and men’s employment, but emerging opportunities for women to work from home and to engage in e-commerce offer reasons for optimism. Job losses have been widespread in Philippines as a result of COVID-19. Results from the first round of the 2020 High Frequency Survey show that, on average, about 30 percent of both men and women experienced job losses between March and August 2020. Data from the April 2020 Labor Force Survey show similar results of high unemployment and reduced labor force participation for both men and women, but the share of women who have left the labor force, at least in the short term, appears to be higher. Although the second round of the High Frequency Survey shows substantial job recovery by December 2020 after the initial shock of COVID-19’s onset, women’s job recovery has lagged behind men’s overall, across income quintiles, and in both urban and rural areas. A positive change has been in work from home (WFH). More than a third of women (36 percent) who remained employed were able to WFH—10 pp more than the share of men who did so—and this reverse gender gap increased to 16 pp by December 2020 (35 percent of women versus 19 percent of men). The pandemic has also opened some new WFH opportunities with industries such as business process outsourcing, increasing opportunities for WFH in order to adapt to COVID-19 measures. Furthermore, the rise of e-commerce in the country—which allows for WFH and flexible schedules—represents a promising area of growth for women entrepreneurs. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 23 Executive Summary E-commerce offers new opportunities for women’s economic empowerment. A 2021 report by the International Finance Corporation (IFC), Women and E-commerce in Southeast Asia, analyzes gender-disaggregated data from Indonesian and Filipino sellers that utilize Lazada, an online platform, as well as interviews with sellers in both countries and e-commerce experts (IFC 2021). The report notes that closing gender gaps in e-commerce could add more than $280 billion to the value of the sector in Southeast Asia from 2025 to 2030. The report also finds that two- thirds of the Filipino sellers in Lazada were women-owned businesses, and 70 percent of these were microenterprises, compared with 60 percent of male-owned Filipino businesses in Lazada. Judging from the prevalence of women entrepreneurs—and women-owned microenterprises—on a platform like Lazada, e-commerce represents an area for potential growth in women’s economic participation in the Philippines. Given women’s prevalence in service sectors in the Philippines, job increases and growing digitalization of this sector could especially benefit them. COVID-19 has had disproportionately negative impacts on the merchandise value of women’s business in the e-commerce sector. Women and E-commerce in Southeast Asia’s analysis of Lazada shows that, in the last two quarters of 2019 (before the onset of the pandemic), women sellers were substantially outselling men in the Philippines, with an average gross merchandise value (GMV) that was 106 percent of the average GMV of men-owned businesses. However, by the last two quarters of 2020, women’s average GMV was 79 percent of men’s, representing a 27-pp decrease in average GMV. This change was mostly due to a decline in women-owned businesses’ GMV in the electronics and general merchandise sectors on Lazada, which are higher value than other sectors on the platform. Despite a decrease in GMV during the pandemic, Lazada has seen an increase in the share of businesses that are women-owned, from 60 to 66 percent. Both men and women sellers have also branched out into new selling categories on Lazada, with the shares of men- owned and women-owned businesses that sell in more than one product category increasing by 10 pp and 8 pp, respectively, from the end of 2019 to the end of 2020. This finding may indicate that entrepreneurs have turned to product diversification as a coping mechanism during COVID-19, which is perhaps better facilitated by e-commerce platforms. Although COVID-19 appears to have caused a setback for women-owned businesses in selling online, pre-pandemic trends show that the Filipino women can be quite successful in the e-commerce sector. Although lockdown and quarantine measures were the main reasons for not working for both women and men early in the pandemic, women were disproportionately affected by caregiving and family duties, especially as the initial shock subsided. About 10 percent of women who stopped working in August 2020 did so because they had to take care of children or of a sick family member (compared with only 3 percent of men); these proportions increased to 33 percent of women and 8 percent of men in December 2020. Much of this may be an indirect effect of COVID-19—as more members of the household stayed home, the domestic labor burden on women increased. Among those who have continued working, a larger share of men than women experienced declines in their incomes early on during the crisis—especially among poor households— although women reported slightly higher rates of decreased income by December 2020. The shock of the pandemic has increased food insecurity, suggesting that many households have struggled to cope with income and job losses. Nearly 90 percent of women and men were worried about their financial situation in August 2020, demonstrating that COVID-19 presents a large challenge moving forward for the Philippines. 24 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Executive Summary Agreement with social norms in the Philippines that position women as the primary caregivers for children and prioritize women’s domestic role over their economic activity appears prevalent among both men and women, discouraging women’s labor force participation. Preliminary findings from a rapid survey looking at perceptions of care work, childcare, and time use suggest that social norms and beliefs about women’s role in the household may act as barriers to women’s participation in the labor market. The survey shows that women spend significantly more time on domestic and care activities and are seen as the primary caregivers, but fathers are seen as playing an important role in childcare as well. Although a relatively important proportion of women indicated that they would spend more time on working and skills development in the absence of childcare duties, respondents—and women in particular— demonstrated a reluctance to use childcare services outside of the home, seemingly because of a strong belief that childcare duties should fall to mothers and other family members. This finding suggests that, if social norms are not addressed, the availability of more formal childcare services may not increase women’s labor force participation. Other barriers to mothers’ participation in the labor market appear to be a belief that children’s sociopsychological development and school performance may suffer when mothers work, mixed sentiments about whether and how a woman should return to work after having a child, and an uncertain opinion on whether women’s contribution to society is to stay home and take care of children or work outside the home if there is a choice to be made between the two. Finally, the survey suggests that adults replicate their parents’ model of gendered division of labor in their own households as adults—with the mother staying home and the father working full time when they have a child under school age—showing that social norms may persist along generational lines. Policy Considerations. Although a more complete policy paper is forthcoming, the following are several suggested areas for shaping policies to support women’s labor force participation and economic activity. 1. Decrease time spent on domestic activities. Policies that aim to reduce the amount of time spent on domestic labor could free up time for women to increase participation in the labor market. Several examples of policies and country experiences show that the expansion of access to more modern services and household appliances contributes to the reduction in time women spend on domestic work and to the increase in their labor force participation. 2. Provide alternatives to childcare in the home. Much of the literature suggests that increased access to childcare allows mothers to participate in the labor market. However, findings in this report show that many households in the Philippines are reluctant to use childcare outside of the home, and their willingness to pay for childcare services appears to be low. As such, increasing childcare options alone is likely insufficient to raise female labor force participation in the country. 3. Address gendered social norms that affect women’s participation in the labor market. Social norms that appear to contribute to depressed female labor force participation in the Philippines include a preference for mothers and family members to care for children (as opposed to formal childcare outside the home) and the belief in women’s role as caregiver and men’s role as breadwinner. Although addressing social norms themselves with policies is challenging, the literature points to interventions such as media campaigns, behavior Overcoming Barriers To Women’s Economic Empowerment In The Philippines 25 Executive Summary change communication, and attitude change interventions as possible solutions to changing opinions about masculinity, gender in the workplace, and gender roles. Examples include (1) an intervention in India in which families were shown informational videos aimed at improving familial support for women’s work outside of the home; (2) a study in Saudi Arabia that showed that men tended to overestimate other men’s opposition to women’s work and female students tended to underestimate their peers’ intentions to work, but both groups showed changes in opinion and behavior when presented with the correct statistics; and (3) a project in India in which a nongovernmental organization held regular classroom discussions on gender roles and discrimination in schools, leading to increased support for gender equality and more gender-progressive behavior. In the Philippines, mixed results on the perceived contribution of women to society suggest that a significant portion of the population currently believes that women play important roles in both the home and the workforce; this group could perhaps be convinced to support women’s labor force participation even when there are young children in the home. 4. Support more flexible work arrangements, particularly remote work models. The COVID-19 pandemic severely disrupted established work patterns, accelerating the adoption of remote work across various sectors of the economy. Findings from this report suggest that arrangements which allow women to work from home show promise in the Philippines, as it provides flexibility that is compatible with care work. In addition, remote work arrangements also reduce the amount of time workers spend on the road, freeing up more time for productive use. The Philippine House of Representatives’ proposed amendment of the Telecommuting Act (Republic Act 11165) to expand work from home capabilities and support would be an important step forward in this regard. 5. Strengthen support for women entrepreneurs, particularly in e-commerce. Along with the adoption of remote working arrangements, findings from this report also suggest that e-commerce could be a valuable tool in increasing labor force participation among women. Engaging in e-commerce provides women with flexibility, allowing them to grow their business while engaging in domestic activities and childcare. As such, e-commerce provides a window of opportunity for the emergence of more female entrepreneurs. Data from the sample of women entrepreneurs on the e-commerce platform Lazada suggest that providing support in the form of basic skills training and expanding access to credit could particularly be impactful in supporting women owned enterprises (IFC 2021). 6. Support networking among working women and exposure to role models. Programs that enable women to network with other women, and potentially with men, could help erode the restrictive gendered norms and create a virtuous cycle to promote women’s economic activity. The literature shows that networking among women can notably be valuable for female entrepreneurs. Similarly, exposure to counter-stereotypical role models can influence girls’ and women’s gender stereotypes and shift cultural norms about women’s work. Furthermore, e-commerce platforms could serve as important forums for such networking opportunities. 7. Scale efforts in reskilling and upskilling women, particularly in fields that have high growth potential. The disruptions caused by the pandemic has shifted how businesses operate, accelerating the adoption and raising the share of businesses which offer remote working arrangements. However, women who stand to benefit greatly from the changing landscape of work may not be able to take advantages of this opportunity if they lack the necessary skills 26 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Executive Summary or knowledge. Scaling efforts to provide women the opportunity to acquire new skills in fields such as ICT and other STEM-related fields could help them secure more productive work in the new normal. This could also aid in a more inclusive recovery as the country recovers from the effects of the pandemic. 8. Encourage firms to expand opportunities for women who wish to re-enter the labor force. Findings from the report reveal that less than half of women agree that it is easy for a mother to return to work after leaving the labor market, with 14 percent of women viewing the gap in work experience as the main obstacle in returning to the workforce. Firms could be encouraged to adopt “returnship” programs, which typically provide women who wish to re-enter the labor force with skills training, mentorship as well as the potential to get hired after the training period (PwC 2016). These efforts can be complemented by services which help match women with opportunities suited to their strengths, skills and previous work experience. Helping women re-enter the labor force could break the persistence of gendered social norms and could act as a mechanism to improve women’s labor force participation for future generations by reinforcing the belief that taking a break from work should not become a major hindrance to women’s future career prospects. 9. Address gender wage gaps at the bottom of the income distribution. As shown in this report, there are key differences in men’s and women’s wages at the bottom of the pay distribution that largely result from the inability of women, notably those married and those with young children, to engage in jobs that offer higher returns. Policies aiming at supporting women in these groups to engage in higher-paying jobs (through encouraging hiring of women in certain industries, enhancing employers to offer flexible work arrangements, addressing the social stigma associated with certain types of “male-dominated” jobs, and so forth) could help reduce gender inequalities in the labor market. 10. Address the need for greater data on childcare services. Data on childcare service enrollment and quality particularly for community-based and home-based facilities, day care facilities run by nongovernmental organizations, and private nursery schools are limited, making it difficult to connect childcare access and quality to labor market outcomes. 11. Although this report focuses on the interaction between women’s economic opportunities and specific areas such as education, employment, and childcare, there are additional factors pertinent to women’s work in the Philippines that merit research and policy focus. For example, the gender dynamics of migration and remittances from overseas foreign workers may affect the domestic burden and labor force participation of remaining household members. Additionally, recent legislation improving wages and rights of domestic helpers (kasambahay) needs to be monitored for potential unintended consequences on female labor force participation, because demand for such help in lower-income households may decrease. Finally, analyses of the gender distribution across industries with the highest shares of employment—including high-skill jobs in business process outsourcing and call centers—and of the gender distribution within manufacturing as compared with other countries in Southeast Asia merit further research. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 27 Introduction 28 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Introduction Despite decades of economic development and growth, increased women’s education and skills, more gender-inclusive labor regulations, decreases in fertility, and improvements in health, progress towards gender equity in labor force participation remains disappointing worldwide. Women’s labor force participation lags behind men’s participation across the globe, with gaps reaching over 50 percentage points (pp) in some regions (figure I.1, panel a). Women’s participation is particularly low in the Middle East and North Africa and the South Asia regions, with an average of under 25 percent of women in the labor force and a resulting gender gap of 53 pp for both regions. Although the East Asia and Pacific region boasts significantly higher female labor force participation, a gender gap of 17 pp in labor force participation remained in 2019. Within the East Asia and Pacific region, the Philippines has the lowest female labor force participation rate, at 49 percent in 2019 (figure I.1, panel b), and the second-highest labor force participation gender gap after Indonesia. FIGURE I.1 Labor Force Participation Rates, by Region and Country, 2019 a. Labor Force Participation, by Region, 2019, b. Labor Force Participation of Philippines Peer Percent Countries in East Asia and Pacific, by Gender, 2019, Percent Source: World Development Indicators, 2021. Note: CEB = Central Europe and the Baltics; CHN = China; EAP = East Asia and Pacific; ECA = Europe and Central Asia; IDN = Indonesia; KHM = Cambodia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; MYS = Malaysia; NA = North America; PHL = Philippines; SAR = South Asia; SSA = Sub-Saharan Africa; THA = Thailand; VNM = Vietnam. Increasing female labor force participation is instrumental for accelerating economic growth in the Philippines. A World Bank–generated model of the effect of increased female labor force participation on long-term economic growth in the Philippines finds that, if women’s participation increases by 0.5 pp per year from a baseline of 45 percent in 2020, gross domestic product (GDP) and GDP per capita would each increase, on average, by 0.3 pp per year over 2021–50 (see appendix A for more details).4 Although the incremental annual gains in GDP seem modest, the cumulative effects for the economy as a whole and for individual households would be meaningful. Based on the model’s projections, GDP per capita would be about 6 percent higher than the baseline by 2040, and almost 10 percent higher by 2050 (figure I.2). This would translate into an additional US$500 in GDP per capita by 2040 and US$1,325 by 2050, compared to the baseline, which would provide households with substantial income gains. The Philippines’ GDP would be US$70 billion higher than the baseline projection by 2040 and US$190 billion higher 4  At 0.5-pp increase per year, female labor force participation would reach 60 percent by 2050. The model assumes that male labor force participation remains constant at its 2020 rate of 72 percent. This increase would boost GDP growth by 0.22 pp initially, rising to 0.35 pp by 2050. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 29 Introduction by 2050, amounting to over US$1.8 trillion in cumulative gains by 2050 (figure I.3).5 A 2018 report by McKinsey Global Institute also shows important economic gains in increasing women’s participation in the labor force (Woetzel et al. 2018). The report indicates that increasing women’s equality in the East Asia and Pacific region as a whole could add a total of US$4.5 trillion to the region’s collective GDP by 2025, or a 12 percent increase over the baseline projection at the time. According to the McKinsey model, 58 percent of this change would result from raising the female labor force participation rate, 17 percent would result from increasing women’s working hours, and 25 percent would derive from employing women in sectors of higher productivity.6 Evidentially, low female labor force participation in the Philippines and the East Asia and Pacific Region as a whole represents a missed opportunity for economic growth and increased prosperity. FIGURE I.2 Increment to GDP per capita Level FIGURE I.3 Cumulative Gains to GDP from Higher due to Higher Female Labor Force Participation, Female Labor Force Participation, 2020–50 2020–50 Source: World Bank. Source: World Bank. Note: Figure shows scenario less baseline in indicated year. Note: Figure shows scenario less baseline since 2020. A wealth of literature has examined the determinants of women’s labor participation, which fall under demand- and supply-side factors. Demand-side factors are those on the employer side that encourage or “pull” women to find work and are thus related to labor markets, macroeconomic trends, and legal frameworks. Taşseven, Altaş, and Ün (2016) note that typical demand-side factors correlated with increased female labor force participation in the literature include urbanization, increased production, growing service and public sectors, increased availability of part-time work, and enabling legislation that supports women’s economic activities; their research further shows that country-level unemployment rates and GDP per capita are positively correlated with female labor force participation rates. Supply-side factors are those personal and household characteristics that “push” women to find work outside of the household, such as female educational attainment, fertility postponement and falling fertility rates, and changes in beliefs and attitudes toward female employment (Taşseven, Altaş, and Ün 2016). 5  Amounts in this model are in real 2010 US dollars. 6  The World Bank model adopts a more conservative projection than the McKinsey model, which assumes a faster annual increase in female labor force participation and thus predicts growth in GDP over the baseline by 2025 that is roughly equivalent to the growth the World Bank model predicts by 2040. More details about differences in methodology and underlying assumptions between the two models are discussed in appendix A. 30 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Introduction On the supply side, empirical studies point to women’s education and decreased fertility as influential in increasing female labor force participation. In a study of urban married women in Bolivia, Brazil, India, Indonesia, Jordan, South Africa, Tanzania, and Vietnam, Klasen et al. (2021) find that women’s education was positively correlated with female labor force participation, whereas the number of children and the presence of young children (from birth to two years old) in the household were negatively correlated with female labor force participation. Using micro-level data from a region in Ethiopia, Mulugeta (2021) shows that women’s educational attainment and age at first marriage were positively correlated with female labor force participation, whereas the presence of children under the age of five and family size were negatively correlated with participation. Similarly, a study of women’s labor force participation in Jakarta finds a strong correlation between women’s education and their participation in the labor force, but a negative correlation between the presence of children under five in the home and female labor force participation (Widarti 1998).7 Some demand-side factors may also be related to fertility rates and family size. In a study using Turkish firm-level data from 2005 to 2015, Karamollaoglu and Soybilgen (2020) find that larger firms, exporting firms, foreign-owned firms, and firms with higher part-time worker ratios had higher rates of female employment, whereas younger firms and those with higher worker productivity and longer working hours had lower female employment. The authors suggest that some of these effects may be related to balancing work and domestic responsibilities, because larger Turkish firms provide more benefits for mothers and their children, and full-time work and long hours may be incompatible with childcare and domestic work. Apart from female education and fertility rate, another key supply-side determinant of female labor force participation in the literature is household income. Using data from the late nineteenth- century French rural census, Grantham (2012) shows that women in poorer households—proxied by landlessness and lower-skill occupations of husbands—were more likely to enter the workforce. Similarly, Widarti (1998) shows a positive correlation between female labor force participation and poor-quality housing as a proxy for low-income status. Additional studies find a negative relationship between household income and female labor force participation, suggesting that many women may work out of necessity but may leave the workforce as household income increases (Klasen et al. 2021; Mulugeta 2021; Widarti 1998).8 Other factors that have been shown to have positive correlation with female labor force participation include women’s skills training, exposure to mass media, and access to credit (Mulugeta 2021). Despite the empirically positive relationship between women’s labor force participation and their educational attainment, many researchers note that social and cultural norms may dampen these factors’ effect on female labor force participation. In their cross-country study, Klasen et al. (2021) argue that cultural, social, and religious norms affect female labor force participation and may counteract gains from increased women’s education and fewer children; they also conclude that these unobserved characteristics account for the largest portion of between-country differences in female labor force participation trends and rates in their sample. In discussing the positive relationship between women’s education and female labor force participation, Widarti (1998) and Taşseven, Altaş, and Ün (2016) caution that norms surrounding family responsibilities and gender roles may suppress this positive effect in their respective contexts, and Mulugeta (2021) notes that social and cultural norms in Ethiopia may act as barriers 7  In further support of these determinants, an analysis of a late nineteenth-century French rural census by Grantham (2012) shows negative correlations between female labor force participation and (1) presence of children under the age of three and (2) family size. 8  Mulugeta (2021) uses husband’s income and husband’s education as proxies for household income/economic status, and Widarti (1998) uses husband’s education. In addition to a direct measure of household income, Klasen et al. (2021) use household head education as a proxy. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 31 Introduction to women’s labor force participation. Taşseven, Altaş, and Ün (2016) also posit that childcare facilities and cost may act as barriers to labor force participation for mothers. Widarti (1998) attempts to examine the effect of social norms empirically, using women’s place of birth as proxy for more gender-restrictive social, cultural, and religious values and ultimately showing a negative relationship between restrictive norms and female labor force participation. A growing literature suggests that housework, childcare, social norms, and cultural barriers are key factors in determining female labor force participation. In recent years, research has examined how domestic and childcare duties may depress female labor force participation or shape the types of work in which women engage, both globally and in the Philippines specifically (Abrigo and Francisco-Abrigo 2019; Bayudan-Dacuycuy 2019; Clark et al. 2019; Chowdhury et al. 2018; Du and Dong 2013; Epetia 2019; Halim, Johnson, and Perova 2017, 2019; Lavy, Lotti, and Yan 2016; UNDP 2020a; World Bank 2012). Additionally, there has been a shift in interest toward social and cultural norms surrounding gender, and how they affect female labor force participation (Bayudan-Dacuycuy 2019; Oxfam International 2019; UNDP 2020b). Jayachandran (2021) presents a comprehensive discussion of social norms as barriers to female labor force participation, arguing that norms around public harassment and violence toward women, women’s social interactions and freedom of movement, women’s control over household finances, and gender roles regarding breadwinning and domestic duties affect female labor force participation differently depending on how the norms present in a given society. The author further suggests that the norm of women as primarily in charge of house and care work is the most common and the most entrenched across societies. Additionally, Jayachandran describes and assesses examples of policy interventions in different country contexts that have endeavored to address these norms and the resulting effects on female labor force participation. This report focuses on a subset of determinants of female labor force participation: education, childcare, and social norms. As presented previously, a host of factors affects female labor force participation in the literature on both the demand and the supply sides, and, despite some empirical evidence of correlation for several factors, there is no conclusive consensus on which factors—if any—universally matter. This report is not intended to comprehensively address all factors affecting female labor force participation.9 The selected determinants are those supply- side factors identified as being of particular interest in the Philippines Country Gender Action Plan (CGAP) for fiscal year (FY) 2020 through FY2024. Furthermore, the effects of childcare and social norms on female labor force participation are not as well understood or studied in the literature as other determinants. Finally, this report can contribute to a growing literature on the effect of childcare on mothers’ labor force participation. The report’s findings have the potential to inform government policies and programs to promote women’s economic empowerment and inclusive development, particularly as COVID-19 persists. The structure of the report is as follows. Chapter 1 discusses the recent literature on barriers to women’s empowerment in economic opportunities and female labor force participation, with a focus on disproportionate domestic workload and gendered social norms. Chapter 2 presents a gendered analysis of education, focusing on gaps and overall levels of enrollment, achievement, learning outcomes and field of study, paying particular attention to geographic and urban/ rural differences. This chapter aims to help understand gender gaps throughout the life cycle and examine whether women in the Philippines are at a disadvantage when entering the labor market. Chapter 3 examines the gender gaps in labor force participation and how factors such as 9  It is worth noting that, although some factors—such as education and family size—may be more plausibly causally associated with female labor force participation, the impact of social norms is difficult to empirically measure and is most convincingly presented as correlational. 32 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Introduction region, education level, marital status, children, and employment status of family members affect female labor force participation, with a focus on the role of social norms and childcare. Chapter 4 covers trends in women’s employment, the types of employment women hold, the influence of educational differences on employment, and the effects of marriage and young children on women’s employment. The chapter also examines the extent and determinants of the gender pay gap. Chapter 5 looks at the effect of COVID-19 on women’s employment and income, and the coping mechanisms used to manage the challenges caused by the pandemic. Chapter 6 presents the preliminary findings from a recent survey on women’s labor force participation and care work in the Philippines. The concluding chapter provides a policy discussion and suggestions for areas of intervention. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 33 CHAPTER 1 Gender Gaps and the Role of Domestic Obligations, Social Norms, and Childcare 34 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 1 | Gender Gaps and the Role of Domestic Obligations, Social Norms, and Childcare Although the Philippines has achieved a relatively high level of gender equality, labor force participation among women remains low, resulting in the underutilization of the population’s human capital. By many dimensions, the Philippines is a leader in gender equality. It ranks 16th out of 153 countries in the World Economic Forum’s Gender Gap Index for 2020, the country’s legal framework for gender equality is recognized for its comprehensive nature, and the educational attainment of women surpasses that of men. Despite this progress, labor force participation among women remains disproportionately low, with under half of the working-age population of women in the labor force. The government of the Philippines is committed to gender equality and understanding the challenges that hinder women from joining the labor force. Priority Area 1 of the fiscal year (FY) 2020 to FY2024 Philippines Country Gender Action Plan (CGAP) aims to address women’s low labor force participation in order to reduce gender inequality as both an essential development objective and an opportunity for economic growth. This report, funded by the Australian–World Bank Trust Fund for Growth and Prosperity (AGAP) as part of the Women, Work, and Care project, seeks to fill the gender knowledge gap around women’s constraints to work and understand the role of childcare and social norms. This chapter looks at the emerging literature on the drivers of gender gaps in the labor force, with a focus on social norms, housework, and childcare. This review of the literature covers both global and Philippines-specific contexts in order to guide analysis of complex obstacles that women continue to face with an eye toward increasing female labor force participation. Is a large domestic burden holding women back from participating in the labor market? Despite positive economic, educational, and demographic conditions in recent years, female labor force participation has not increased commensurately in most regions. Since the 1990s, most developing countries have seen decreases in fertility rates, increases in female primary and secondary education, and high economic growth rates. Despite these favorable trends, female labor force participation has fallen in East and South Asia in recent years and has only moderately increased in the Middle East and North Africa. This trend contradicts the feminization U hypothesis, in which female labor force participation falls as an economy transitions from a primarily agrarian base to industry and then increases again as education rates rise, fertility rates drop, and the service sector grows. There is also evidence that female labor force participation is countercyclical, suggesting that many women enter the workforce when low levels of household income require it and then leave once income increases (Klasen 2018). Rigid gender and societal norms, which place the burden of domestic work on women, are among the factors impeding women’s labor force participation and affecting the type of jobs they take on. One explanation for low rates of female labor force participation may be that home production activities fall disproportionately on women, who find many labor market opportunities to be incompatible with household work. Indeed, Klasen (2018) suggests that female labor force participation rates in Sub-Saharan Africa remain higher than in other regions, despite higher fertility rates, because of a higher share of employment in agriculture, which is more conducive to childcare. The 2019 Human Development Report lends support to this finding, revealing that low female labor force participation is related to household work (UNDP 2020). Overcoming Barriers To Women’s Economic Empowerment In The Philippines 35 CHAPTER 1 | Gender Gaps and the Role of Domestic Obligations, Social Norms, and Childcare In the Philippines, the persistence of the labor force participation gap may be due to the disproportionate domestic burden placed on women. Bayudan-Dacuycuy (2019) finds that Filipino women who spend more time on housework are less likely to participate in the labor market, but they are more likely to work when their husbands spend more time on home production. Specifically, women’s likelihood of being in the workforce is 42 percentage points (pp) greater for women who do not participate in housework in comparison with those who do, whereas the difference for men only 25 pp. Each week, women 20 to 39 years old spend an average of 20.0 hours on domestic work and 7.5 hours on childcare, compared to men who spend an average of only 6.5 hours on domestic work and 2.0 hours on childcare (Abrigo and Francisco-Abrigo 2019). In 2017, nearly 60 percent of women and only 10 percent of men cited household and family work as the reason for their nonparticipation in the labor force. Furthermore, women with a young child were 7.4 percent less likely to be in the labor force than other women, and married women were 12.5 percent less likely to participate in the labor market than single women (Epetia 2019). Although men spend more hours per week than women on market activities, once home production is accounted for, on average women work more hours than men. Labor force data show that men devote more hours to market activities than women; however, labor force statistics do not account for home production—comprising housework, childcare, and elder care. Analysis of the Philippines Pilot Time Use Survey and Labor Force Survey in 2000 shows that, when home production is factored in so that the total working hours reflect combined market activities and nonmarket home activities, women work a greater average number of hours per week than men (Abrigo and Francisco-Abrigo 2019). This trend holds within each age group; when paid market work and unpaid care and housework are combined in the conception of “work,” women spend about 30 percent more time working than men at ages 15 to 19, about 5 percent more at ages 20 to 39, about 14 percent more at ages 40 to 59, and a striking 65 percent more at ages 60 and above. Women older than 60 years work an average of 40 hours per week in comparison to 24 hours for men in the same age range, when accounting for home production due to a second wave of childcare in multigenerational households. In 2015, the value of unpaid home activities was estimated at PhP2.5 trillion, with about 75 percent contributed by women. The authors also find that, when the value of home production is included, the value of women’s labor contributions is on par with the value of men’s labor. Bayudan-Dacuycuy (2019) finds that the amount of time women spend on housework in the Philippines is positively related to having a toddler in the house. Given the disproportionate domestic burden placed on women, many choose to work in industries or with schedules that are more compatible with home activities. In Vietnam, a large gender earnings gap has remained constant from 2011 to 2014, despite higher educational attainment by women than men (Chowdhury et al. 2018). Much of the earnings gap can be attributed to Vietnamese women working in lower-paying industries; women seem to self-select into jobs that offer better benefits including paid leave, fewer required work hours, and health and social insurance at the expense of better pay. This finding may indicate that Vietnamese women prioritize employment opportunities that better allow them to balance work with childcare and housework, sacrificing higher wages as a result. Additionally, Vietnamese women work an average of two fewer hours per week than their male peers, which is likely related to the need to balance market and household work. The 2012 World Development Report finds that, because of disproportionate care burdens, women tend to prefer occupations with more flexible schedules, as well as to work fewer hours in market labor (World Bank 2012). Occupations with more flexible schedules tend to be lower-paid and lower-quality, and many are in the informal sector. 36 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 1 | Gender Gaps and the Role of Domestic Obligations, Social Norms, and Childcare The presence of childcare may alleviate some of the pressure placed on women, allowing them to enter the workforce. A lack of childcare acts as a barrier to female labor force participation for many mothers. In Indonesia, low female labor force participation is correlated with a lack of childcare. An analysis by Halim, Johnson, and Perova (2019) finds that an additional public preschool per 1,000 children increases the work participation of mothers of preschool-age eligible children by 11–16 percent. The availability of preschools induces mothers to informal sector occupations that do not require full-time commitments. A previous analysis by the same authors shows that mothers in households without elderly family members—who often act as informal childcare providers—are less likely to participate in market work (Halim, Johnson, and Perova 2017). In the peak fertility ages of 26 to 28 years, Indonesian women who live with elderly household members are 10 to 19 pp more likely to work than women who do not. From 2000 to 2014, as the percentage of Indonesian households that listed the grandmother as primary caretaker increased by 4.9 pp, female labor force participation increased by 3.9 pp. For urban Indonesian mothers who do return to the labor market after childbirth, the absence of elderly household members delays this return by an additional two years. Not only are Indonesian women without informal childcare less likely to work, but they are also more likely to transition to unpaid family business within the year of giving birth, with urban mothers remaining in this work for an average of eight years and rural mothers for an average of six years. Furthermore, Indonesian mothers without access to childcare from extended family are more likely to transition out of higher-paying manufacturing jobs to lower- paying jobs in agriculture or sales. In this case of transitioning to lower-paying sectors, mothers without informal childcare are never again as likely to work in manufacturing jobs as they were before giving birth. Decreases in the availability of formal childcare and increases in formal childcare prices have been shown to negatively affect female labor force participation. Following changes in China’s economic system in the late 1970s and early 1980s, publicly funded childcare programs decreased, while costs increased (Du and Dong 2013). As public childcare programs became less available during the period 1996–2006, urban mothers’ labor force participation dropped by 22.1 pp, and their labor hours decreased by an average of 11.5 hours per week. Mothers of young children were most affected; those with children from birth to two years old experienced a 33.4-pp decrease in labor force participation, and the labor force participation rate of mothers with children aged three to six years old fell by 16.0 pp. The authors estimate that the addition of a formal day care center in a community that did not previously have one would increase mothers’ labor force participation by 10.5 percent and time devoted to wage work by 5.8 to 7.1 hours per week. Increased access to formal early childhood care can improve labor market outcomes for mothers, including labor force participation rates and type of employment. A randomized control trial finds that low-income mothers in Kenya who were offered vouchers for childcare were 8.5 pp more likely to be employed than mothers who did not receive vouchers (Clark et al. 2019). Married women who received the voucher worked an average of 4.25 more hours per week. Single mothers who received the voucher were able to switch jobs in the service sector, which provided more regular hours and better wages compensating for working 8.8 fewer hours per week. In a study of a home preschool and care program in Ecuador, Lavy, Lotti, and Yan (2016) find that mothers who participated were 17.6 percent more likely to be employed, 20.7 percent more likely to be working full-time, and 20.4 percent more likely to be employed in the formal sector in comparison to mothers who did not participate in the program. Mothers in the program also earned on average Overcoming Barriers To Women’s Economic Empowerment In The Philippines 37 CHAPTER 1 | Gender Gaps and the Role of Domestic Obligations, Social Norms, and Childcare US$57 more per month than mothers who did not participate. Although Dang, Hiraga, and Nguyen (2019) do not find that childcare use significantly affected mothers’ labor force participation in Vietnam, they do find that it increased mothers’ probability of having a wage-earning job (by 41 percent) and having a formal job (by 26 percent). Even more important, childcare use was also negatively correlated with mothers’ probability of being in the self-employed farm work sector and with increased household income. The effect of available formal childcare on female labor force participation is also dependent on other factors such as household income and existing levels of labor force participation and childcare availability in the community. In other contexts, the argument for childcare increasing female labor force participation is less clear. Lundin, Mörk, and Öckert (2008) find that increasing the supply of childcare did not increase female labor force participation in Sweden, and Fitzpatrick’s (2008) study of public preschool programs in the United States shows that the availability of the program did not increase female labor force participation for any group except rural mothers, who demonstrated a 20 percent increase in probability of being employed. These two studies suggest that additional childcare and preschool programs may have a negligible effect on mothers’ labor force participation in areas that already have high female labor force participation rates, existing subsidized public childcare, and higher household incomes. Alternatively, in these contexts, market demand for additional labor may already be limited. In addition to positive effects on mothers’ labor force outcomes, early childhood care can benefit children’s development. The study by Lavy, Lotti, and Yan (2016) in Ecuador shows that the home preschool program decreased children’s school dropout and grade repetition rates, and increased school attendance. The program also improved children’s math and reading test scores. With regard to mothers’ involvement in their children’s development, women who participated were more likely to report being involved in household decisions surrounding children’s education (9.9 percent more likely) and discipline (8.7 percent more likely). Furthermore, participant mothers were 10 percent more likely than nonparticipants to engage with their children on current learning materials and 6.6 percent more likely to engage in educational recreation with their children. This finding provides evidence that, apart from positive effects on female labor force participation and earnings, early childhood care has intrinsic value in improving children’s development both directly and indirectly via increases in mothers’ decision-making power and investment in children’s learning. A study by Brinkman et al. (2016) finds that, in rural Indonesia, preschool attendance positively affects children’s development. Public childcare services are not widely used in the Philippines despite policy efforts to formalize a national childcare system. In the Philippines, children’s enrollment in early childhood care and development (ECCD) programs remains low. In 2018, under 10 percent of children from birth to four years old were enrolled in 45,295 public ECCD facilities.10 Although a legal and policy framework exists to form a nationwide childcare system (the Republic Act 6972, the Kindergarten Education Act, the Early Years Act and the Enhanced Basic Education Act), the requirement for day care to be devolved to the local (barangay) level poses challenges of implementation, availability, access, and quality.11 Instead, it is thought that a large share of childcare services are provided by the private sector and are generally not affordable for mothers from low-income households (Epetia 2019). Data are limited on the childcare services provided through community-based and 10  Data from the ECCD Council of the Philippines (https://eccdcouncil.gov.ph). 11  The national government is pursuing ECCD strategies to strengthen the childcare system, such as putting into place and enforc- ing quality standards and upgrading day care facilities to national child development centers. These National Child Development Centers are envisioned to be a repository of relevant data in the community on ECCD and could also be used to address local data gaps on childcare. 38 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 1 | Gender Gaps and the Role of Domestic Obligations, Social Norms, and Childcare home-based facilities, on day cares run by nongovernmental organizations, and on private nursery schools; as such, figures on the overall usage of childcare services in the Philippines and how it affects female labor force participation are unclear. Social norms surrounding gender, family responsibilities, and labor act as barriers to women’s economic opportunities. Globally, gender norms typically place women in the role of caregiver and homemaker, and men in the role of breadwinner and provider. Social norms are the beliefs, values, and practices that maintain societal structures (UNDP 2020b). Social norms help determine the range of available and acceptable options and behaviors for individuals based on membership in social categories such as gender, race, ethnicity, and age. They also uphold existing systems and power dynamics in a given society by implicit or explicit rewards and sanctions. Social norms are said to be “sticky” when the stakes for transgressing or maintaining them are particularly high, resulting in intractability in changing said norms. When women in a society move from parity in basic capabilities such as health to enhanced capabilities such as social and political power, social norms surrounding gender often become stickier. For example, the gender wage gap widens as the type of employment moves from basic to enhanced, with the smallest gender pay gap for contributing family workers and the largest gaps for employers, 100 top entertainers, and 500 top billionaires (UNDP 2020b). Furthermore, Klasen (2018) notes that, in some countries, stigma against working women is stronger at higher levels of education; for example, women’s labor force participation in India is negatively correlated with their husbands’ level of education. Ortiz Rodriguez and Kumara Pillai (2019) find that participation in the labor market increases women’s support for gender equality 12 in Mexico, but women’s individual and household characteristics unrelated to their employment— such as education level and age—significantly affect working and nonworking women’s different levels of support for gender equality. According to the United Nations Development Programme’s gender social norms index (GSNI), 91 percent of men and 86 percent of women exhibit at least one gender bias. The GSNI comprises four dimensions—political, educational, economic, and physical integrity—constructed from questions in the World Values Survey waves 5 (2005–09) and 6 (2010–14).13 Globally, GSNI levels are correlated with the Gender Inequality Index, as well as with the ratio of women’s to men’s time spent on unpaid home activities (UNDP 2020b). Within the political and economic dimensions of the GSNI, about 50 percent of the global population, irrespective of gender, indicated that they believe men make better political leaders than women and over 40 percent believe that men make better executives. Compared to countries in the East Asia and Pacific region, the Philippines has one of the highest rates of gender biases. A comparison with peer countries shows that the Philippines has the highest overall GSNI at 99 percent and the second-highest GSNI2 (exhibiting at least two gender biases) at 87 percent, exceeded only by Malaysia. Both men and women express gender biases at high rates. Almost all women (99 percent) in the Philippines exhibit at least one gender bias, and 83.3 percent exhibit at least two (figure 1.1); almost all Filipino men exhibit at least one gender bias, and 90.3 percent exhibit at least two. These figures further support the notion that gender norms are entrenched and maintained in Filipino society by both men and women. 12  The authors define gender quality as a composite score of agree/disagree statements related to women’s status in relation to a husband, women’s role in work and caregiving, women’s rights, and violence against women, among others. 13  For more on how the GSNI is calculated see “Frequently Asked Question–Gender Social Norms Index (GSNI)” at http://hdr.undp. org/sites/default/files/frequently_asked_questions_gsni.pdf. For more on the World Values Survey, see https://www.worldvalues- survey.org/WVSDocumentationWV6.jsp. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 39 CHAPTER 1 | Gender Gaps and the Role of Domestic Obligations, Social Norms, and Childcare FIGURE 1.1 Cross-Country Comparison of GSNI and GSNI2 by Gender, Latest Year, Percent Source: Based on the World Values Survey waves 5 to 7. Note: Data from Indonesia and Vietnam are from the World Values Survey wave 5 (2005–09); data from Malaysia, the Philippines, and Thailand are from World Values Survey wave 6 (2010–14). GSNI = gender social norms index (exhibiting at least one gender bias); GSNI2 = exhibiting at least two gender biases. Gender norms in the Philippines delineate different roles for men and women in economic and household realms. In line with global trends, traditional Filipino gender norms designate men as breadwinners and women as caretakers and homemakers (Bayudan-Dacuycuy 2019). Women in the Philippines are generally seen as having three areas of work: (1) reproductive work, including childbearing and rearing; (2) productive, income-generating activities; and (3) community work that is generally related to reproductive and care roles and is often unpaid (Eviota 1992; Moser 1993). Filipino society’s view of men as the head of household and primary breadwinner is still quite widespread, and thus women’s participation in productive work is often seen as secondary and particularly helpful in the case where a husband’s income is not sufficient to support the family. However, Filipino women are still expected to act in their reproductive and care roles, regardless of whether they work; what’s more, the societal expectation is that women will prioritize reproductive work over productive work if the two are not compatible. Recent qualitative information supports the assertion that men and women are societally assigned different roles in the Philippines. A study by Oxfam International (2019) using data from 27 focus groups in five regions in the Philippines in 2017 reveals that a majority of Filipino participants believe that women should be in charge of domestic and care work. Although women are expected to contribute to household income where possible, participants believe that it is the husband’s responsibility to economically provide for his family. The consensus of the focus groups was that this division of labor would result in a balanced, happy home and familial cohesion. It was stressed that it is generally not socially acceptable for women to forgo domestic work— especially childcare. Men’s contribution to care work is moderately accepted but is often viewed to be temporary and dependent on familial circumstances. Children are expected to contribute to home activities, but these tasks are also gendered; girls are expected to support care and housework, whereas boys are generally assigned tasks that are perceived to require more strength or skills. Women-dominated tasks such as care work are generally viewed as less skills-based than male-dominated employment sectors, although some home activities are assigned high social 40 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 1 | Gender Gaps and the Role of Domestic Obligations, Social Norms, and Childcare value. However, the focus groups did express some potential shifts in gender norms, in that many participants indicated a desire for a less gendered division of labor for younger generations. Gender norms that relegate women to reproductive and care roles and men to income- generating roles affect labor force participation for both men and women in the Philippines. Using data from the 2012 International Social Survey Program, Bayudan-Dacuycuy (2019) finds a bigger increase in women’s market work participation in the Philippines when they do not engage in housework and a larger decrease when their spouses do not share in the household production. Qualitative data from the National Economic and Development Authority (NEDA) indicate that husbands often influence their wives’ retreat from the workforce in order to devote time to care work (NEDA 2019). Furthermore, NEDA creates a patriarchy index based on women’s self-reported decision-making power across a number of household domains such as use of money and health care. Among married women, the labor force participation rate of those with low levels of the patriarchy index was 8 pp higher than of women with moderate levels of the index and 13 pp higher than of women with the highest level of the index. Gender norms also influence men’s labor force participation in the Philippines. Having a young child in the home increases men’s labor force participation by 3.8 percent (Epetia 2019), and married men are 11 pp more likely to be in the labor force than men who have never been married and 5 pp more likely to be in the labor force than men who are separated, annulled, or widowed (NEDA 2019). These data suggest that married men and fathers view their societal role as breadwinners and are more likely to be in the labor force if they occupy those positions. Discrimination in the workplace may also be a contributor to low female labor force participation. Women in the Philippines face no legal barriers and have strong protections against discrimination such as banning the dismissal of pregnant women (World Bank 2021). Despite this, pregnancy is correlated with an overall 13-pp decrease in the likelihood of a woman’s employment and a 37- pp decrease for a married woman (NEDA 2019). Qualitative focus groups reveal that employers believe pregnant women will be less productive, absent or sick more frequently, and more likely to leave their jobs after childbirth. Cho, Doan et al. (2021) argue that because employers incur some maternity benefit costs—as well as the cost of hiring interim workers during pregnant women’s maternity leave—employers may discriminate against women of child-bearing age and may be less likely to hire women with young children because they may believe these women to be less career-oriented and therefore less worthy of institutional investment. More generally, the NEDA study finds that Muslim women face additional discrimination in the workplace because of dress and perceived cultural differences. The global COVID-19 pandemic has further exacerbated gender gaps in labor force participation and economic opportunities. Since the onset of the pandemic, women have lost employment and left the labor force at higher rates than men. The UN Women’s From Insight to Action report (2020) finds that, globally, women’s employment during the pandemic is 19 percent more at risk than men’s. Women are more likely to work in the sectors that have experienced the highest rates of layoffs, including accommodation and food services and as domestic workers. Informal workers—of which women make up a large share—have lost an average of 60 percent of their wages globally, and 22 percent in the East Asia and Pacific region. The report also notes that women have recounted larger decreases in working hours due to the pandemic; in the East Asia and Pacific region, 50 percent of women in the formal sector have reported decreases in working time, in comparison with 35 Overcoming Barriers To Women’s Economic Empowerment In The Philippines 41 CHAPTER 1 | Gender Gaps and the Role of Domestic Obligations, Social Norms, and Childcare percent of men. With the closures of schools and childcare centers, families face increases in unpaid childcare and schooling supervision of remote learning, and the majority of this burden falls on women. Although more global data are needed to further analyze the impacts of COVID-19 on gender gaps in the labor force and economic opportunities, the report cautions that many women may be unable to reenter the workforce, especially without government relief policies that specifically address childcare. Results from the first round of the High Frequency Survey in the Philippines reveal similar job losses between men and women but uneven job recovery for men and women by the end of 2020; whether the job losses will be lasting remains uncertain. Expanded work from home (WFH) and greater adoption of digital technology have emerged as vital to overcome the impacts of the pandemic (World Bank 2020b). Business process outsourcing, such as call centers, has moved to WFH models, albeit with connectivity difficulties (Navarro 2020). Expanded WFH offering greater flexibility could be beneficial for women who face a large domestic burden, making it difficult to work fixed hours outside of the house. However, digitalization in the Philippines remains constrained by low high-speed broadband connections and a reliance on cash (World Bank 2020b). 42 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 2 Education Overcoming Barriers To Women’s Economic Empowerment In The Philippines 43 CHAPTER 2 | Education As shown in the previous chapter, female labor force participation is lagging globally and especially in the Philippines within the East Asia and Pacific region, which has only been exacerbated by the COVID-19 pandemic. Some of the relevant literature indicates that a disproportionate domestic burden may discourage Filipino women from entering the workforce. The literature also suggests that access to early childhood care is beneficial for both mothers and children, and that a lack of available care may decrease the probability of mothers’ employment. Women who do work may be more likely to choose sectors and types of jobs that are more compatible with balancing domestic activities, including childcare. However, public childcare is not widely used the Philippines, and private sector childcare is unaffordable for most low-income mothers. Chapter 1 also delved into the literature on social norms, suggesting that gender norms surrounding family responsibilities, women’s employment, and acceptable types of work for women are particularly strong in the Philippines and may discourage female labor force participation. In assessing women’s economic activity in the Philippines, it is critical to first examine gender gaps in education, because educational attainment is posited to be a key determinant of female labor force participation. Enrollment rates in the Philippines have historically been high, comparing favorably with peer countries. However, disparities in provision and access still exist in the country. This chapter examines how school enrollment and attainment rates vary by gender, geographical location, and household income. The chapter also takes a closer look at how the Philippines compares with its peers with regard to learning outcomes, and the effect poor performance could have on the country’s growth. I. Gender Gaps in Enrollment and Attainment Girls perform better than boys in terms of educational enrollment and attainment. Enrollment rates in all levels of education are higher among girls than among boys. For both boys and girls, enrollment at the primary level has been nearly universal for over three decades. Women have also not been disfavored in educational participation at higher levels. In 1981, the earliest year in which data at every level are available from the United Nations Educational, Scientific and Cultural Organization’s Institute of Statistics, enrollments in secondary and tertiary education were already higher for girls than boys; gross enrollment rates at secondary and tertiary levels were, respectively, 11 and 5 percentage points (pp), in favor of girls (figure 2.1).14 Although enrollment in higher education has become more common in recent years, the gap in favor of girls has persisted. As of 2017, gross enrollment at the primary level was slightly higher for boys than for girls, but the opposite was true for both secondary and tertiary levels, and the gender gap has even widened in favor of girls at the tertiary level.15 FIGURE 2.1 Gross Enrollment Rates, by Educational Level and Gender, 1981–2017, Percent Source: United Nations Educational, Scientific and Cultural Organization’s Institute for Statistics, 2020. 14  The ratio of total enrollment to the population of the age group that officially corresponds to the level of education, regardless of age. For Institute of Statistics data, see http://data.uis.unesco.org/. 15  Higher gross rates of enrollment at the primary level could be indicative of grade repetition or late entry into school. 44 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 2 | Education Higher enrollment in education among girls has materialized into higher education levels for women relative to men. The 2020 Labor Force Survey shows that, among the working-age population, 63 percent of women have completed at least high school whereas only 56 percent of males have; at higher levels of education, 18 percent of women have completed college or have postgraduate degrees, compared to 14 percent of men (figure 2.2). Looking at the current enrollment rates for youth aged 5 to 21 years old, women seem to remain in school longer than men, suggesting that the gap in educational attainment in favor of women will persist (figure 2.3). FIGURE 2.2 Educational Attainment of Working- FIGURE 2.3 Current Enrollment, by Gender and Age Population, by Gender, 2020, Percent Age, 2020, Percent Source: Based on data from the 2020 Labor Force Survey. Note : TVET = technical and vocational education and training. Younger Filipinos are more educated than previous generations. Leaving school with low levels of education has become more uncommon among younger Filipinos. Comparing education levels between those who reached the age of 65 in 2020 and those who were 25 years old in 2020 shows that, whereas 28 percent of men and 22 percent of women had not completed elementary education among those aged 65, only 11 percent of men and 6 percent of women had not finished primary education among those aged 25 (figure 2.4, panels a and b). Completing tertiary education is also much more common among younger Filipinos, with the proportion of those finishing tertiary education substantially higher among younger women (12 percent among those aged 65 compared to 32 percent among those aged 25) and three times higher among younger men (8 percent among those aged 65 compared to 24 percent among those aged 25) (figure 2.4, panel d). Moreover, although younger Filipinos receive higher levels of education, the gender gap in favor of women has persisted across time. A student entering primary school in 2020 is likely to reach at least the end of secondary education, with the chances being higher for girls than boys. In the Philippines, the expected years of schooling has reached 13.2 years for girls and 12.8 years for boys.16 Girls’ school enrollment in the Philippines compares favorably with the country’s peers. Only China and Vietnam performed slightly better than the Philippines, with a mere 0.04 additional years of expected schooling for girls. The pattern of women becoming more educated than men is common to all peer countries with the exception of Thailand, but the reverse gender gap in expected years of schooling in the Philippines is particularly large, surpassed only by Malaysia. The COVID-19 pandemic will likely reduce enrollment rates and therefore the expected years of schooling, but by how much and how it will affect the gender composition of enrollment are ongoing questions. 16  This measure considers only 14 possible years that a student can be enrolled in preprimary, primary, and secondary education. Tertiary education is not counted toward the total number of expected years of schooling. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 45 CHAPTER 2 | Education FIGURE 2.4 Highest Level of Educational Attainment, by Gender and Age, 2020, Percent a. Less than elementary b. Elementary c. Secondary d. Tertiary Source: Based on data from the 2020 Labor Force Survey. Geographic location and income affect enrollment and attainment in education. The gap between urban and rural female enrollment has narrowed significantly over the last decade. School enrollment of teenagers 14 to 18 years old has increased with the passage of the Enhanced Basic Education Act of 2013, which expanded compulsory education beyond the elementary level to the end of secondary schooling, in addition to adding two more years to high school education.17 Between 2012 and 2020, enrollment among 16-year-old girls increased by 14 pp in urban areas, while increasing 24 pp in rural areas (figure 2.5). These increases resulted in a reversal in the urban-rural gap in enrollment among 16-year-olds, with girls’ enrollment in rural areas 4 pp higher than in urban areas. The passage of the Enhanced Basic Education Act also had a similar effect on boys’ enrollment, which increased 22 pp in rural areas (from 62 percent to 84 percent) and 15 pp in urban areas (from 71 percent to 86 percent). FIGURE 2.5 Rural and Urban Female School Enrollment, by Age, 2012–20, Percent Source: Based on data from the 2012 and 2020 Labor Force Survey. 17  For more on the act, see “Implementing Rules and Regulations of the Enhanced Basic Education Act of 2013.” Official Gazette of the Republic of the Philippines at https://www.officialgazette.gov.ph/2013/09/04/irr-republic-act-no-10533/. 46 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 2 | Education School dropout remains a challenge for rural boys, who leave school to join the labor market. In 2020, the gender gap in enrollment for 17-year-olds was similar in rural areas, where enrollment was 9 pp greater for girls, and in urban areas, where the gender gap was 8 pp in favor of girls. However, among the boys of secondary school age who dropped out, a large proportion joined the labor market, particularly in rural areas. Among 17-year-olds out of school, 73 percent of boys in rural areas were in the labor force compared to 32 percent of girls; in urban areas, 58 percent of boys were in the labor force compared to 31 percent of girls (figure 2.6). Women appear frequently to drop out for family reasons. A much larger proportion of 17-year-old girls who have dropped out are married, especially compared to boys of the same age: in rural areas, 37 percent of out-of- school girls aged 17 report being married compared to 2 percent of boys; in urban areas, 21 percent of out-of-school girls report being married compared to less than 1 percent of boys (figure 2.7). FIGURE 2.6 Rural and Urban Labor Force FIGURE 2.7 Rural and Urban Marriage Rates for Participation of Out of School Youth, by Age, and Out-of-School Youth, by Age and Gender, 2020, Gender, 2020, Percent Percent Source: Based on data from the 2020 Labor Force Survey. Educational attainment in the conflict-affected Bangsamoro Autonomous Region in Muslim Mindanao (BARMM) lags far behind other regions. Educational attainment for both men and women is significantly lower in BARMM compared to the rest of the country. In BARMM, 44 percent of women and 48 percent of men have not gone beyond elementary education, and another 20 percent of women and men have not completed high school. In the highest-performing region, National Capital Region, the share of women who have not gone beyond elementary education drops to 9 percent, and only 12 percent have not completed high school (figure 2.8). Overall, the proportion of women who have not completed above elementary education is 15 pp greater in the BARMM than in any other region.18 Leaving education early hurts women’s ability to enter wage work and high-skill occupations. This could, at least partly, explain the high concentration of women in low-productivity sectors and low-profile occupations. 18  For males, the proportion who have not completed above elementary education is 10 pp greater in the BARMM than in any other region. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 47 CHAPTER 2 | Education FIGURE 2.8 Education Levels of Adult Women Ages 15+, by Region, 2020, Percent Source: Based on data from the 2020 Labor Force Survey. Note: BARMM = Bangsamoro Autonomous Region in Muslim Mindanao; CAR =Cordillera Administrative Region ; NCR = National Capital Region; TVET = technical and vocational education and training. Poverty significantly constrains educational attainment. Students in the lowest quintile of household income both enter schooling at a later age and are enrolled at lower rates than their peers in the top income quintile (figure 2.9). Among six-year-old children, only 71 percent of those at the bottom quintile are in the age-appropriate level, compared to 90 percent in the top quintile. This later start drags on the educational trajectory of poor students; among 13-year-olds, 97 percent of those in the top quintile are enrolled in high school, compared to only 71 percent of those in the bottom quintile. This pattern continues years later: at 19 years old, 70 percent of the wealthiest students are in tertiary education, whereas 58 percent of the poorest are no longer in school. Financial constraints appear to limit both men and women from staying in school to gain the skills that they need to escape poverty and achieve their full potential in the labor market. 48 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 2 | Education FIGURE 2.9 School Enrollment, by Age and Income Quantile, 2017, Percent Source: Based on data from the 2017 Annual Poverty Indicators Survey. Note: TVET = technical and vocational education and training. II. Learning Outcomes Despite relatively high education attainment levels, learning outcomes remain low. Although students attend school at high rates and attain relatively high levels of education, these successes are not reflected in student learning outcomes. The expected years of schooling for women is 13.2 years, higher than in most comparator countries (figure 2.10). However, the Philippines ranks last among its peers when using learning-adjusted years of schooling, which the World Bank’s Human Capital Index uses to adjust the average years of schooling with how much students have learned while in school (World Bank 2020c).19 Women are estimated to only have 7.8 learning-adjusted years of schooling and men to have only 7.2 learning-adjusted years of schooling. In Malaysia, the expected years of schooling for women is 0.5 years less than in the Philippines, but the learning-adjusted years of schooling is 1.4 years higher. Although the nonpecuniary returns to education—such as delayed marriage and pregnancy—improve when students remain in school, poor learning outcomes suggest that students might not have the skills they need for productive work once they leave school. 19  Based on results from international assessments, including Trends in International Mathematics and Science Study (TIMSS)/ Progress in International Reading Literacy Study (PIRLS), Programme for International Student Assessment (PISA), PISA+TIMSS/ PIRLS, Southern and Eastern Africa Consortium for Monitoring Educational Quality (SACMEQ), Programme for the Analysis of Education Systems (PASEC), Latin-American Laboratory for Assessment of the Quality of Education (LLECE), Pacific Islands Literacy and Numeracy Assessment (PILNA), Early Grade Reading Assessment (EGRA), and Non-nationally-representative Early Grade Reading Assessment (EGRANR). Overcoming Barriers To Women’s Economic Empowerment In The Philippines 49 CHAPTER 2 | Education FIGURE 2.10 Female Expected Years of Schooling vs. Learning-Adjusted Years of Schooling, the Philippines and Comparator Countries, 2020 Source: Based on data from the 2020 Human Capital Index, World Bank. Poor learning outcomes are reflected across all subjects, with only a small share of students having top level skills. In the 2018 Programme for International Student Assessment (PISA), students in the Philippines ranked last in reading and second to last in both mathematics and science out of 79 participating countries. The country’s performance in reading is much lower than peer countries, with only 19 percent of students reaching level 2, which represents the baseline level of skill needed in the current labor market (figure 2.11) (Hanushek and Woessman 2015). In Indonesia, which has the second-lowest performance, the share of students reaching level 2 is 55 percent greater than in the Philippines; and in Vietnam, the best-performing comparator country, this share is over three times greater. The PISA results for the Philippines in math and science, which represent important foundational skills for STEM (science, technology, engineering, and mathematics) subjects, are slightly better but still far lower than comparator countries. Results from the 2019 Trends in International Mathematics and Science Study (TIMSS) for fourth-graders add further evidence of difficulties in mathematics and science.20 The average scores for the Philippines in both mathematics (297) and science (249) were lowest among all 58 countries. Only 19 percent of students reached the low international benchmark of 400 in mathematics, and only 13 percent achieved the same benchmark in science. In addition, assessment results suggest that the educational system may not be producing enough high performers, which could negatively affect the country’s growth prospects. Only a small share of performers (under 0.1 percent) in each subject on the PISA assessment achieved scores that placed them at level 5 and above, a level in which students demonstrate the ability to solve complex problems. On the TIMSS, the high benchmark was reached by only 1 percent of students in mathematics and science. FIGURE 2.11 Share of Students with Basic Skills, the Philippines and Comparator Countries, 2018, Percent Source: Based on data from the 2015 and 2018 Programme for International Student Assessment. Note: Because 2018 data for Vietnam have not yet been fully validated, 2015 results are used instead. 20  Results available at https://timss2019.org/reports/. 50 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 2 | Education Learning outcomes are higher for women than for men in the Philippines, although their performance is still below that of their peers from comparator countries. In 2017, 2018, and 2020, women in the Philippines had higher harmonized test scores than men did.21 The 14-point reverse gender gap in 2020, measured in TIMSS-equivalent units, is almost identical to the reverse gender gaps in Indonesia and Malaysia (Figure 2.12).22 Women outperform men in all comparator countries except China, showing that the reverse gender gap in learning outcomes is common in Southeast Asia. Despite their relative advantage over their male peers, women in the Philippines have harmonized test scores that show they are learning less than women in peer countries. FIGURE 2.12 Harmonized Test Scores, by Gender, the Philippines and Comparator Countries, 2020 Source: Based on data from the 2020 Human Capital Index, World Bank. Note: Test scores are in harmonized test scores from major international student achievement testing programs, measured in TIMSS-equivalent units. TIMSS = Trends in International Mathematics and Science Study. III. University Graduates’ Fields of Study Women are underrepresented in STEM and other related fields in the Philippines, but are more represented in these fields relative to women in comparator countries. At the tertiary level, the share of women in academic fields mirrors traditional gender roles. Despite performing better than men in standardized mathematics and science tests, women graduates are underrepresented in traditionally male-dominated fields that rely on these skills. Only 24 percent of engineering graduates and 36 percent of STEM graduates are women (figure 2.13). In contrast, 77 percent of education graduates and 72 percent of health and welfare graduates are women. The natural sciences appear to be a positive exception, with women representing 62 percent of graduates in the field. FIGURE 2.13 Gender of University Graduates in the Philippines, by Field of Study, 2017, Percent Source: United Nations Educational, Scientific and Cultural Organization’s Institute for Statistics, 2020. Note: Select fields of study only. ICT = information and communication technology; STEM = science, technology, engineering, and mathematics. 21  Harmonized test scores standardize results from major international student achievement testing programs into comparable scores. 22  In TIMSS-equivalent units, 300 is minimal attainment and 625 is advanced attainment. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 51 CHAPTER 2 | Education Although the share of women graduates in STEM and similar fields is low when compared to men, the percentage of women graduating from these fields compares favorably with peer countries. The percentage of information and communication technology graduates who are female is at least 4 pp greater in the Philippines than in any comparator country (figure 2.14). Additionally, the female share of graduates who complete STEM and engineering degrees is higher in the Philippines than in Indonesia and Thailand. Only Malaysia, which heavily emphasizes these fields of study, surpasses the Philippines, with women representing 26 percent of graduates from STEM fields and 15 percent of graduates from engineering fields. FIGURE 2.14 Share of Female Graduates, by Field of Study, the Philippines and Comparator Countries, Latest Years, Percent Source: Based on data from the United Nations Educational, Scientific and Cultural Organization’s Institute for Statistics. Note: Select fields of study only. These fields overlap and can sum to greater than 100 percent. Data for PHL are from 2017; IDN and MYS from 2018 and THA from 2016. Data for other peer countries are unavailable. ICT = information and communication technology. 52 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 3 Women in The Labor Market Overcoming Barriers To Women’s Economic Empowerment In The Philippines 53 CHAPTER 3 | Women in The Labor Market Chapter 2 demonstrated that, for both men and women, younger generations of Filipinos are more highly educated than older generations. There is a reverse gender gap for enrollment and educational attainment, although levels vary somewhat by geographic location, region, and household income. Learning outcomes and learning-adjusted years of schooling in the Philippines are low among peer countries, but girls consistently achieve better outcomes than boys. Although Filipino women are more likely than men to attend university and are overrepresented in most educational subjects, they are underrepresented in science, technology, engineering, and mathematics, and in information and communication technology. However, rates of female representation in these educational fields in the Philippines are higher than in most peer countries. Overall, chapter 2 showed that women in the Philippines are at an advantage in terms of education. Despite outperforming men in educational attainment and achievement, women continue to lag behind men in economic participation and opportunity. Progress in their participation in the labor market is particularly disappointing. This chapter explores the extent and trends of gender gaps in labor force participation and examines the influence of gender norms and family responsibilities in the persistence of gender inequalities. I. The Tenacity of Gender Gaps in the Labor Force Only marginal improvements have been achieved in women’s labor force participation over the past three decades. A large proportion of women in the Philippines remains out of the labor force, with limited progress achieved in expanding participation rates over the past three decades. In 2020, only 45 percent of working-age women (population 15 years and older excluding overseas workers) reported active participation in the labor market, compared with 72 percent of men. Female labor force participation peaked at 51 percent in 2002 but started declining in 2016, averaging 46 percent over the past four years. 54 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 3 | Women in The Labor Market The gender gap in labor force participation has narrowed slightly since 1988 but remains persistently high. The gap between male and female labor force participation was estimated at 27 percentage points (pp) in 2020 (figure 3.1). This represents a 7.2-pp decrease from 1988, but the reduction was essentially due to a decline in labor force participation among men rather than an increase in labor force participation among women. Overall, labor force participation rates declined for both men and women since 2016—partly because of a change in the educational system, which added two years of schooling in order to receive a high school diploma—but the decline was faster for men than for women. FIGURE 3.1 Labor Force Participation, by Gender, 1988–2020 Overseas Men Overseas Women Source: Based on data from the 1988–2020 Labor Force Surveys. Note: Labor force participation (LFP) rate is the proportion of total labor force to the total household population aged 15 years and over (excluding overseas workers). The proportion of overseas workers is the number of Filipinos working abroad over total population aged 15 years and over. Estimates represent annual averages except in 2020 for which only January and April quarters are available. Overseas employment is increasing faster for women than for men. In 2003, about 2 percent of men and 1.8 percent of women aged 15 and over were working abroad, and by 2020 the shares for each increased to over 3 percent. Overseas employment increased faster for men than for women until 2015, but since then the trends have reversed in favor of women (figure 3.1) Women’s labor force participation is higher and increasing faster in urban areas than in rural zones, where the gender gap is more pronounced but narrowing faster. Between 1988 and 2020, women’s participation in the labor force increased by 0.2 pp in urban areas and declined by 2.8 pp in rural areas.23 At the same time the participation rate for men declined by 3 pp in urban areas and by 11 pp in rural zones (figure 3.2). This led to a faster decline of the gender gap in rural zones—which decreased from 39 pp to 31 pp—than in urban areas, where the gap reduced from 28 pp to 24 pp. The narrowing of the rural gender gap is therefore happening for the wrong reasons and cannot be considered a positive shift. 23 In 1988–2019, women’s labor force participation increased by 3.7 pp in urban areas and declined by 0.5 pp in rural zones. For men, the declines were by 0.1 pp in urban areas and 9 pp in rural zones. The situation deteriorated in 2020, due to the pandemic (the 2020 data cover January and April quarters). Overcoming Barriers To Women’s Economic Empowerment In The Philippines 55 CHAPTER 3 | Women in The Labor Market FIGURE 3.2 Labor Force Participation, by Gender and Area, 1988–2020, Percent Source: Based on data from the 1988–2020 Labor Force Surveys. There are wide regional variations in women’s participation rates, revealing the struggle of women in less developed and fragile settings. Across all regions in Luzon and Visayas, female labor participation rates resemble the national figure, with participation rates between 43 and 53 percent (figure 3.3).24 In contrast, participation rates in Mindanao vary widely: whereas Northern Mindanao has the country’s highest rates of labor force participation for both men and women—as well as the lowest gender gap in labor force participation—the three worst-performing regions in the country are also in this region, including the conflict-affected Bangsamoro Autonomous Region in Muslim Mindanao (BARMM), where the female labor force participation rate is only 29 percent, and the gender gap in participation is estimated to be about 43 pp. This suggests that conflict could keep women out of the labor force despite lower incomes in the region, which generally would push participation in the labor market out of necessity. FIGURE 3.3 Regional Differences in Labor Force Participation, by Gender, 2020 Source: Based on data from the 2020 Labor Force Survey. 24  Estimates are based on the 2020 Labor Force Survey, January quarter. 56 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 3 | Women in The Labor Market The gender gap remains higher in the Philippines than in most of its East Asian peers, but the expansion of higher education for girls augurs well for closing the gap. Female labor force participation rates in the country lag behind all its peers. As of 2019, the Philippines had the lowest female participation rate across all comparator countries (figure 3.4, panel a), with the gender gap in labor force participation exceeded only by Indonesia (figure 3.4, panel b). Among all East Asian countries,25 female labor participation in the Philippines remains one of the lowest, second only to Myanmar. The country actually has moved closer to fitting into a U-shaped relationship between development and women’s labor force participation, where the highest rates of female labor force participation are found at low and high levels of economic development (figure 3.4, panel c).26 This, as well as the large dispersion in female labor force participation at similar income levels, suggests that, in addition to economic factors, noneconomic reasons play an important role in influencing female participation in the labor market. FIGURE 3.4 Cross-Country Comparison of Female Labor Force Participation, the Philippines and Comparator Countries, 2019 a. Labor Force Participation of Peer Countries, b. Gender Gaps in Labor Force Participation of Peer by Gender, 2019, Percent Countries, 2019, Percentage Points c. Female Labor Force Participation, by GDP per capita, 2019 Source: World Development Indicators, 2020. Note: Female labor force participation by GDP per capita is for the year 2019 with East Asian countries marked in orange except for the Philippines, which is in red and includes data points for the years 1988, 1997, 2008 and 2019. CHN = China; IDN = Indonesia; KHM = Cambodia; MYS = Malaysia; PHL = Philippines; THA = Thailand; VNM = Vietnam. 25 Following World Bank regional groupings. 26 The U-shaped hypothesis is the relationship between the female labor force participation rate with economic development where the labor force participation rate of women initially declines with the movement away from agriculture to industry and then rises as countries develop and both women’s fertility falls and education increases.  Overcoming Barriers To Women’s Economic Empowerment In The Philippines 57 CHAPTER 3 | Women in The Labor Market Higher levels of educational attainment generally increase the rate of female labor force participation and influence the type of employment that women have. Women who completed tertiary education have the highest labor force participation rate (74 percent) and the smallest gap with men who completed the same education level (a difference of 10 pp) (figure 3.5). It is worth noting that, although higher levels of education are correlated with higher labor force participation, this may reflect, in part, that women who are more likely to succeed in the labor market are also more likely to pursue higher levels of education. FIGURE 3.5 Labor Force Participation, by Educational Attainment and Gender, 2020 Source: Based on data from the 2020 Labor Force Survey. II. What Dissuades Women’s Participation in the Labor Force? Young and female—a double strike? Women make a late entry to the labor market, which lowers their economic opportunities. The age-participation profiles of men and women in the Philippines differ widely, with high rates of male labor force participation beginning in early adulthood and the later entry of women into the labor force. Labor force participation rises steeply among men from the 20–24 age group to the 25–29 age group, after which participation remains consistently high, reaching its peak in the 40–44 age group and starting to decline among those aged 50 and older (figure 3.6, panel b). In sharp contrast, participation rates among women increase incrementally before reaching their peak in the 45–49 age group (figure 3.6, panel a). The age-participation profiles have seen slight positive changes from 2002 to 2020, with lower rates of labor force participation from ages 15 to 19 following the expansion of high school by two years, and the slight increase in early female labor force participation in the 25–29 age range. 58 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 3 | Women in The Labor Market FIGURE 3.6 Labor Force Participation, by Gender and Age Group, 2002–20, Percent a. Female Labor Force Participation b. Male Labor Force Participation Source: Based on data from the 2002–20 Labor Force Surveys. The pattern of women’s participation in the labor force follows a different trajectory in rural areas than in urban zones. Whereas labor force participation rates among men across different age groups remain relatively similar in both areas, participation rates among women in younger age cohorts are much higher in urban areas. Among urban women, participation rates peak twice and are highest in the 25–29 age group and the 40–44 age group, but participation rates among rural women rise incrementally beginning in the 20–24 age group, with the peak occurring much later, in the 45–49 age group (figure 3.7). In both areas, the gender gaps in labor force participation are highest among the 30–34 age group and lowest in the 15–19 age group, but gaps in rural areas are significantly larger than in urban areas. It is worth noting that the large disparity in labor force participation between men and women, especially in the critical 20–24 and 25–29 age groups, could be a factor in the low overall participation rates of women in the country given that the delay in labor market entry could create an absence of initial work experience, which may then serve as a barrier to women joining the labor market later in life, and could also negatively affect their earning potential by limiting the scope of professional roles they can take. Overall, the late entry of women in the labor market results in both social and individual costs in the form of lower earnings, a shortened working life, and fewer years of experience. FIGURE 3.7 Rural and Urban Labor Force Participation, by Gender, 2020, Percent Source: Based on data from the 2020 Labor Force Survey. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 59 CHAPTER 3 | Women in The Labor Market The proportion of women who marry at younger ages and do not work is much higher among rural women than urban women, who remain single longer. Across all age groups, the proportions of married women are higher in rural areas than in urban areas. Notably, 36 percent of women in rural areas in the 20–24 age group report being married, compared to 29 percent in urban areas, and the highest urban-rural difference in marital status occurs in the 25–29 and 30–34 age groups, where it reaches 10 pp (figure 3.8, panel a). Results further show that labor force participation among married women is the lowest across all age groups compared to women who remain single and to men, whether they are single or married (figure 3.8, panel b). Among married women in the labor force, participation remains low for young age groups and peaks much later— between the ages of 40 and 54—before declining again, which suggests that many married women wait until their children are sufficiently grown before seeking work or reentering the labor market. In stark contrast, married men consistently have the highest rates of labor force participation across all groups, hovering above 90 percent for most of their adult lives. FIGURE 3.8 Marriage and Labor Force Participation, by Age Group, Location, and Gender, 2020, Percent a. Share of Married Women, by Age Group and Location b. Labor Force Participation, by Gender and Marital Status Source: Based on data from the 2020 Labor Force Survey. Child and household care duties may cause women to miss labor market opportunities during the early stages of their careers. Childcare, family duties, and overall cultural norms, which dictate the role of women in the household, seem to hold back women’s participation in the labor force. Among Filipino women aged 15 and above who are out of the labor force, the reasons cited for not seeking employment vary drastically, depending on their current marital status. In 2020, the vast majority of married women (88 percent) cited household and family duties as the main reason they did not look for work, compared to only 11 percent of single women who cite this as the main reason for not seeking employment (figure 3.9, panel a). This trend is consistent across areas, with 87 percent of married women in rural areas and 89 percent of married women in urban areas citing household and family duties as the main reason they did not look for work. Regionally, the rates of married women citing this reason vary from 82 percent in Ilocos to 96 percent in the BARMM. In contrast, housework and family duties do not seem to be a major impediment for men’s participation in the labor market, with 5 percent of single men and 15 percent of married men citing this as a reason for not seeking employment. The possible influence of gender norms, which have historically placed the responsibility of childcare and housework on women while assigning the responsibility of providing for the family on men, can also be seen when looking at the participation rates of 60 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 3 | Women in The Labor Market working-age Filipinos in households that have young children (birth to four years old). In 2020, the labor force participation rate of men with young children in their household was 9 pp higher than those without young children (81 percent and 72 percent, respectively); for women, participation rates fall 7 pp among those with young children at home compared to those without—43 and 51 percent, respectively (figure 3.9, panel b). FIGURE 3.9 Factors Influencing Labor Force Participation, by Gender, Marital Status, and Children, 2020, Percent a. Reasons for Not Seeking Employment, by Marital Status and b. Labor Force Participation for Working- Gender Age Population with Young Children Source: Based on data from the 2020 Labor Force Survey. Extended family members appear to play a role in augmenting participation rates among women in households with younger children, particularly in lower-income households. Living in extended family households is a fairly common practice in the Philippines.27 Data from the Annual Poverty Indicators Survey (APIS) show that 45 percent of women aged 15 and above lived in extended family households in 2017.28 Although data from APIS similarly show a decline in employment among women who live in households with younger children, they also reveal the potential role that extended family members play in assisting with housework or childcare, helping augment participation rates among women. Despite a steep decline in the proportion of women in households with young children and without extended family members who are employed, rates of employment among women with young children rise substantially for those living in extended family households (figure 3.10). This trend holds for households in the bottom, second, and third income quintiles, but is reversed for households in the highest income quintile. For more well-off households, the proportion of women with young children living in nonextended family households who are employed is higher relative to those with extended family members, which could partly reflect the ability of women in these income groups to hire outside help. 27  The analysis here draws on APIS data because the information on household income and other characteristics is not available in the Labor Force Survey. Although labor participation rates in APIS are slightly different from Labor Force Survey rates, patterns are similar in both surveys. 28  Extended family households are defined in this report as households that include family members outside of the nuclear family, which comprises the household head, the spouse of the household head, and their children Overcoming Barriers To Women’s Economic Empowerment In The Philippines 61 CHAPTER 3 | Women in The Labor Market FIGURE 3.10 Female Employment, by Income Quintile and Presence of Extended Family Members in the Household, 2017, Percent Source: Based on data from the 2017 Annual Poverty Indicators Survey. Women who live in households with extended family members and with young children enrolled in preschool have higher rates of employment. When extended family is present and regardless of whether a three- to four-year-old child in the household is in school, women’s employment rate is higher than when a three- to four-year-old is enrolled in school with no extended family. The combination of extended family and at least one three- to four-year-old enrolled in school increases employment rates by 5 pp beyond the presence of extended family alone (figure 3.11). One reason for this could be that having extended family members is complementary to having a young child in preschool, given that hours spent in preschool are not a full school day. As such, having a child enrolled without any familial assistance could make it even more difficult for mothers to seek employment or continue working. FIGURE 3.11 Female Employment, by Presence of Extended Family Members and Preschool-Age Children in the Household, 2017, Percent Source: Based on data from the 2017 Annual Poverty Indicators Survey. The number and characteristics of preschools in the Philippines are mostly similar by region, but data are lacking to draw conclusions about preschools’ effects on women’s participation in the labor force. Data from the Philippine Statistical Yearbook show that, in the 2012–13 school year, CALABARZON—the most populous region in the country—had the highest number of preschools whereas the Cordillera Administrative Region —the smallest region in the Philippines—had the lowest number (figure 3.12, panel a).29 The National Capital Region was an outlier, having the least number of preschools on a per capita basis—less than half the rate of any other region—and with 75 percent of the region’s preschools privately run, compared to an average of 16 percent in other regions. The total number of preschools available does not appear to be predictive of women’s participation in the labor market: the National Capital Region, which has the lowest number of 29  Although the total number of enrolled preschool students could be a more accurate measure of the effect of preschools on women’s labor force participation, region-level enrollment data for preschools are unavailable. Indicators of the quality of the preschools, an important determinant of enrollment, are also unavailable. 62 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 3 | Women in The Labor Market preschools per capita, has labor force participation among women that is above the national average; however, the Autonomous Region in Muslim Mindanao, with an above-average number of preschools per capita, has the lowest female participation rate in the country (figure 3.12, panel b). However, given the absence of recent and spatially disaggregated data on the availability of preschools, it is difficult to infer any causal relationship between the provision of preschools and women’s participation in the labor market. More empirical analysis is needed. FIGURE 3.12 Preschools and Women’s Labor Force Participation, by Region a. Number of Preschools by Region, 2012–13 School Year, b. Number of Preschools per 10,000 Population Total and Female Labor Force Participation, 2013, Percent Source: Based on data from the Philippine Statistical Yearbook, 2013. Note: Data for the number of preschools are for the 2012–13 school year (the most recent available), the female labor force participation rate is for 2013, and the regional population data are based on the 2015 census. ARMM = Autonomous Region in Muslim Mindanao; CAR = Cordillera Administrative Region ; NCR = National Capital Region. An empirical analysis of the potential drivers of women participation in the labor market helps to better understand how women’s own characteristics and household conditions affect their labor supply. The analysis uses a probit model following Klasen and Pieters (2015) and Klasen et al. (2021) to examine the key factors that may explain the low participation of Filipino women in the labor market. The model is estimated for all working-age women, and separately for six subsamples: low-skilled women (with incomplete secondary education and less), high-skilled women (with complete secondary education and above), urban women, rural women, women aged 20–34, and women aged 35–50. The covariates capture two groups of supply-side factors. The first group is women’s own characteristics—education, age, and marital status. The second group is family conditions—household income excluding the woman’s own earnings; education of the household head; presence of a man with salaried employment; presence of overseas working members (to capture income security); number of children in the household children aged 0–2, 3–5, 6–14 (boys), and 6–14 (girls); and presence of extended family members and domestic help.30 Although the model does not include an extensive evaluation of labor demand conditions, some demand effects are captured in two ways: (1) via regional fixed effects dummies and (2) via regional employment shares the shares of male workers in agriculture, manufacturing, other industries (construction, 30  Only extended family members aged 15–50 are included to account for members who can help with childcare. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 63 CHAPTER 3 | Women in The Labor Market mining, utilities), traditional services (accommodation and food, transport, other household services), and modern services (public administration, education, information and communication technology, financial services, and so on) and the share of the region’s working-age population with upper-secondary education and above, to control for the local supply of skilled labor. The model estimates should be understood as reduced-form correlates. Marriage, the presence of young children, and income security seem to penalize women’s participation in the labor force. Results in table 3.1 show that the number of young children (aged 0–2 years) correlates negatively with women’s labor force participation: one additional young child is associated with a 7-pp reduction in the probability of women’s participation. The effect is highest among women aged 35–50. However, the number of children aged 3 years and older does not seem to matter significantly. Being married also appears to reduce women’s likelihood of joining the labor market, supporting the above findings. Although household income does not seem to have an effect on women’s work, other proxies of household income security indicate a negative association between the economic situation of the household and women’s work. In particular, the presence of overseas workers and of men with salaried employment would reduce women’s likelihood of being in the labor force by 4 pp to 10 pp. The negative relationship is higher for women with low education than for highly educated ones, suggesting that women with less education are boxed in by the necessity to work if their household economic conditions are insecure but are possibly more influenced by stigmas attached to women working if they are in more secure economic environments, whereas highly educated women are less constrained by their family circumstances and potentially by gendered stigmas in their labor participation decision. These results are further supported by the negative correlation of household head education with the probability of women’s labor force participation, with the effect increasing at higher education levels of the household head among less educated women. Women’s education and the presence of domestic help seem to enhance women’s labor supply. Increasing women’s education seems to increase their probability of participating in the labor market. However, the relationship appears to follow the U shape meaning that, relative to the reference group of women with primary education and less, the average marginal effects are negative and larger in magnitude with each additional level of educational attainment up to postsecondary schooling and negative again at incomplete college level. For women with a college degree and above, the positive marginal effect is very large and significant. The presence of extended family and domestic help in the household—having someone to help fulfill the obligations of childcare and household chores—is positively and significantly associated with the probability of women’s labor force participation, with the effect of domestic help having a significantly higher magnitude than the effect of extended family members.31 Regional variables do not indicate any crowding-out effect on women’s participation from the relative supply of graduates in the region. However, the local structure of employment seems to depress women’s labor supply probability; regions where manufacturing and modern services make up a high share of employment seem to generate lower employment opportunities for women, suggesting that jobs in these sectors continue to be male dominated. 31 Domestic helpers, known as kasambahay, are a common fixture in middle- and high-income Philippine households, and are often extended family members, meaning that they may be picked up in surveys as either domestic workers or simply household members. Kasambahay frequently work in poor conditions and for very low wages, but recent laws have set minimum wages for domestic workers and detailed their rights and minimum working conditions. This may depress women’s labor force participation, because middle-class women may be unable to pay a kasambahay the minimum wage and instead choose to stay out of the labor force (Institute for Labor Studies 2020). 64 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 3 | Women in The Labor Market TABLE 3.1 Determinants of Women’s Labor Force Participation, 2020, Average Marginal Effects Pr(Labor All Low-skilled High- Rural Urban Women Women Force) working- women skilled women women aged aged age women women 20–34 35–50 Own education (Ref. = Primary & less) Secondary -0.11*** -0.09*** -0.10*** -0.05 -0.03 incomplete Secondary -0.02 -0.02 0 0.02 0.08 complete Postsecond- 0.09*** 0.03 0.15*** 0.13 0.07 ary/ TVET College in- -0.04** -0.05* -0.01 -0.18* 0.08 complete College comp. 0.27*** 0.25*** 0.30*** 0.32*** 0.21** & above Age 0.06*** 0.06*** 0.07*** 0.06*** 0.07*** 0.019*** Married -0.14*** -0.04*** -0.19*** -0.10*** -0.18*** -0.09* -0.18*** Children 0–2 -0.07*** -0.06*** -0.07*** -0.08*** -0.05*** -0.06* -0.14*** Children 3–5 -0.02** -0.01 -0.02* -0.01 -0.02* -0.05 -0.06 Boys 6–14 -0.01** 0 -0.02** 0 -0.02** 0.04* -0.02 Girls 6–14 -0.01 0.01 -0.01 -0.01 0 0.01 0.03 Extended 0.02*** 0 0.03*** 0.01 0.04*** 0.02 0.02 family Domestic 0.35*** 0.63*** 0.26*** 0.31*** 0.36*** 0.45*** 0.56*** help Education HH head (Ref. = Primary & less) Primary -0.01 -0.04** 0.05** 0 -0.01 -0.03 -0.05 Secondary 0.01 0 0.03 0 0.02 0 -0.05 incomplete Secondary -0.03** -0.08*** 0.04** -0.04** -0.02 -0.03 -0.08 complete Postsecond- -0.06*** -0.06 0 -0.06* -0.062* -0.17* -0.16* ary/ TVET College in- -0.02 -0.07** 0.05** 0.01 -0.03 0.01 -0.07 complete College comp. -0.14*** -0.16*** -0.07** -0.11*** -0.14*** -0.13* -0.06 & above Log HH in- -0.01 -0.01 -0.01 0 -0.01 0 0 come Male salaried -0.053*** -0.07*** -0.04** -0.06*** -0.05*** -0.03 -0.101* emp. Overseas -0.10*** -0.10*** -0.10*** -0.08*** -0.12*** -0.06 -0.11* members Urban loca- 0 -0.02 0.01 0.03 0 tion Region employment shares (Ref=other industry) Share agr. -0.87*** -0.93*** -0.93*** -0.72** -0.82** -1.42** -1.13* emp. Share manuf. -1.08*** 0.14 -0.75** -1.51*** -0.54 -2.34* -1.18 emp. Share emp. -1.33*** -1.80*** -1.06*** -0.21 -1.15*** -1.84** -1.76* modern serv. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 65 CHAPTER 3 | Women in The Labor Market Pr(Labor All Low-skilled High- Rural Urban Women Women Force) working- women skilled women women aged aged age women women 20–34 35–50 Share emp. -0.88** -0.45 -0.99** -0.45 -0.42 -0.83 -1.18 trad. serv. Share edu- 0.03 0.14 0.012 0.13 -0.24 0.24 0.35 cated work. No observa- 61098 24576 36522 31116 29982 2503 2190 tions Source: Based on data from the 2020 Labor Force Survey. Note: Using region fixed effects instead of region-level variables produce similar results. HH = household; TVET = technical and vocational education and training. Significance level: * = 10 percent, ** = 5 percent, *** = 1 percent. III. Social Attitudes and Childcare Gender inequality is relatively high in the Philippines, influenced by traditional social norms. The Philippines has made some progress toward closing the gender gap, yet the gaps in economic participation and opportunity remain large. The 2021 Global Gender Gap Report shows that the Philippines has closed 78.4 percent of its gender gap, ranking 17th out of 156 countries in the Global Gender Gap Index (WEF 2021).32 With an EAP average of 68.8 percent, the Philippines ranks favorably relative to peers in the region, placing second only to New Zealand. However, it is worth noting that while the Philippines’ score increased 3.2 pp since 2006, its global ranking fell 11 ranks since then, indicating slower progress in gender equality relative to other countries. In addition, while the Philippines has closed most of the gender gap in educational attainment as well as in health and survival, gaps in economic participation and opportunity as well as political empowerment remain.33 In particular, the country’s progress in economic opportunity and participation remain mixed: while it is one of only 18 countries that have closed at least 79.5 percent of the gender gap in economic participation and opportunity, due, in large part, to the country closing the gender gap in the share of women in managerial positions as well as in professional and technical positions, it ranks 118th in terms of gender parity in labor force participation with a score of 65.3 percent. The extent of gender disparities can also be seen through the Gender Inequality Index (GII).34 With a score of 0.43 in 2020, the Philippines performs worse than the East Asia and the Pacific average of 0.324, ranking 107th out of 189 countries for which data are provided (figure 3.13) (UNDP 2020b). The Philippines’ GII improved from 0.48 in 1995 to 0.43 in 2013, but progress has slowed since then, in particular because of persistent gender gaps in the labor market. Encouragingly, women’s representation in parliament has more than doubled since 1995, reaching a level higher than in all comparators, offering prospects of increased participation of women in decision-making (figure 3.14). 32 Global Gender Gap Index measures gender-based gaps in four key dimensions: economic participation and opportunity, educational attainment, health and survival, and political empowerment. The economic participation and opportunity subindex captures the gaps in three dimensions: labor force participation, remuneration, and advancement. 33 Gender gaps in political empowerment are due to low representation of women in ministerial positions. 34 GII measures gender inequalities in reproductive health, estimated by maternal mortality ratio and adolescent birth rates; empowerment, measured by proportion of parliamentary seats occupied by women and proportion of adult women and men aged 25 years and older with at least some secondary education; and economic status, expressed as labor market participation and estimated by labor force participation rates of women and men aged 15 years and older. The GII assesses the human development costs of gender inequality—higher GII values indicate more disparities between women and men and thus more loss to human development. 66 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 3 | Women in The Labor Market FIGURE 3.13 Gender Equality Indicators, the FIGURE 3.14 Share of Seats in Parliament Held by Philippines and Comparator Countries, 2019, Women, 2018, Percent Percent Source: Based on data from World Values Survey wave 7; Source: Inter-Parliamentary Union, 1997–2018. UNDP 2020. Widespread biases in gender social norms contribute to the persistence of gender inequalities. According to the 2021 Global Gender Gap Report, it would take the world about 267 years to close the gender gap in economic participation and opportunity. The Philippines is no exception, and the time it would take to close the gender gap in economic opportunity in the country might be even longer. Social norms embedded deep within the country’s culture and traditions and the trade- offs women face between participating in the labor market and family care work are mutually reinforcing and seem to obstruct women’s empowerment and gender equality in the labor market (Mukhopadhyay, Rivera, and Tapia 2019; UNDP 2019). World Values Survey (WVS) data show widespread and persistent biases in gender social norms in the Philippines, particularly against more enhanced forms of women’s economic participation; about 70 percent of Filipinos favor men over women for a job (figure 3.15).35 Moreover, 43 percent favor men for economic leadership positions. Women tend to have less bias against gender equality in economic dimensions, yet about 67 percent favor men for a job and 37 percent favor men for economic leadership positions. Biases have intensified in thinking “Men should have more right to a job than women when jobs are scarce” and “Men make better business executives than women do”, with a greater increase among women. The bias against women’s economic empowerment is higher among women in poorer economic classes, those out of the labor force, and those with more children; yet it remains high in the other groups too (figure 3.1). 35  World Values Survey data available at https://www.worldvaluessurvey.org/WVSDocumentationWV6.jsp. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 67 CHAPTER 3 | Women in The Labor Market FIGURE 3.15 Agreement with “Men should have FIGURE 3.16 Agreement by Women with “Men more right to a job than women when jobs are should have more right to a job when jobs are scarce,” by Country, 1994–2020, Percent scarce,” Philippines, 2019, Percent Source: Based on data from World Values Survey, waves Source: Based on data from World Values Survey, waves 3–7. 3–7. Note: OLF = out of labor force. Public opinion about women’s role in the job market and the impact women’s work has on the family suggests that the Philippines is still a patriarchal society and social attitudes remain traditional, although relatively less than in some peer countries. Over 80 percent of Filipinos agree with the statement “A man’s job is to earn money while a woman’s job is to look after home and family,” and over 70 percent agree that being a housewife is as fulfilling as working for pay. About 46 percent of Filipino women believe that preschool children suffer when a mother works, and 27 percent think that a woman having more income than her husband is problematic (figure 3.17). Interestingly, the proportion of women agreeing with these statements is higher than that of men, but lower than in some peer countries in the region. Comparatively, in Vietnam, 78 percent and 45 percent of women, respectively, share these views.36 There is little variations in the proportion of Filipino women agreeing with these views across income, education, and employment groups (figure 3.18). FIGURE 3.17 Bias Against Women’s Economic FIGURE 3.18. Agreement by Women with “A Opportunities, by Country, 2019, Percent preschool child suffers with a working mother,” Philippines, 2019, Percent Source: Based on data from World Values Survey, wave 7, 2017–20. Note: Figures represent the proportion of population that agrees with the indicated statement. 36 The high rates of female labor force participation in Vietnam despite social norms, which would seemingly depress women’s work otherwise, require further investigation but are suggestive of high levels of demand for female labor in some growing industries such as textiles and garments. 68 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 3 | Women in The Labor Market The persistence of stereotypes and social norms that childcare and other housework are the responsibility of women appears to be holding back women’s participation to the labor market. Over half of men and women in the Philippines think that women should stay home when they have a child below school age. This proportion increased by more than 10 pp between 1994 and 2012 for both men and women (figure 3.19). In 2012, only 10 percent of Filipinos considered that women can work full-time outside home in this situation. However, there was a positive shift in the attitude toward women’s work once the youngest child starts school, particularly among women in 2012, 32 percent of women considered they should work full time and another 45 percent thought they should work part-time in this situation compared to 23 and 41 percent, respectively, in 1994. Yet, in 2012, about 55 percent of women stayed home when they had a child under school age and 51 percent also stayed home after the youngest child started school, and these proportions are similar to those in 1994 (figure 3.20). This suggests two things: first, the persistence of the effect of stereotypes about childcare on women’s decision to work and, second, the difficulty women face in rejoining the labor market after staying home. In general, women with upper-secondary- level education and above are more likely than those with lower education to work when they have children; however, almost half (48 percent) of women with higher educational attainment stayed home when they had children who were under the school age (figure 3.20). FIGURE 3.19 Attitude toward “Women should FIGURE 3.20 Women’s Work Status when They work when they have children,” Philippines, 2012, Have Children, Philippines, 2012, Percent Percent Source: Based on data from the International Social Survey Programme, Family and Changing Gender Roles, 1994–2012. Women’s work-family trade-offs and their decision to work may be largely influenced by their family environment. About 59 percent of Filipinos declared that their mother did not work when they were 14 years old, and the situation has not changed over the past 20 years. The proportion of men and women who think that women should stay home when there is a child under school age is larger among those whose mothers did not work. Although higher education levels tend to moderate the attitude of men and women toward whether a woman should work when she has children below school age, the negative influence of their mother’s work status on this opinion remains strong across educational levels. In particular, men of all educational levels whose mothers did not work are likely to believe that women should stay home when there is a child under school age. A large proportion of women (over 60 percent) do not perceive children as restricting their career chances, 70 percent consider that a working mother can establish just as warm and secure a relationship with her children as a mother who does not work, and another 64 percent consider that having a job is the best way for women to be independent. The reality, however, is that women continue to give up on their opportunities to join the labor market once they get married and particularly when they have children. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 69 CHAPTER 3 | Women in The Labor Market Biases and social stereotypes are mutually reinforcing, potentially negatively affecting the inclusion of women in the labor market. Gender attitudes and beliefs about women’s work are highly correlated and seem to be negatively associated with women’s employment likelihood. Data from the 2012 Family and Changing Gender Roles International Social Survey Programme (ISSP) survey show that the belief that “women should stay home when they have children under school age” seems to be associated with a 14-pp reduction in the probability of women’s engagement in paid employment and with an increase in the likelihood of being a housewife.37 Having three children or more below school age is associated with a 23-pp increase in the probability of a woman being a housewife. Estimates from the 2012 and 2019 WVS reveal that the belief that “a preschool child suffers with a working mother” is associated with a 22-pp reduction in the probability of women’s employment. In general, number of children is negatively correlated with women’s employment likelihood.38 Education level, income group, geographic location (urban versus rural and regional), and age group are also related to the probability of women’s employment, but their effect is less important when controlling for social attitudes and number of children. This suggests that the provision of childcare may be essential for women’s economic empowerment but that it will need to be coupled with policies to shift social norms and stereotypes about gender roles in the household. In the absence of childcare options for young children, a large share of women feel compelled to stay out of the labor market—but is there a real demand for childcare assistance? On average, women spend 50 percent more time on caring for family members than men do. Women spend about 25 hours a week looking after family members (for example, children, elderly, ill or disabled family members) whereas men spend only about 17 hours on these tasks. The number of hours increases to 34 for women with children under school age, compared to 17 hours for those who do not have children, but it remains the same for men whether they have children or not. This suggests that women spend a large proportion of their time on childcare. Moreover, the number of hours women with young children spend on family members does not change with their employment status or income group, indicating that women continue to spend much of their time on childcare whether or not they work outside the household and even if they are in wealthier groups. Men tend to underestimate the number of hours their spouse spends on childcare, estimating they spend only 24 hours a week on family members. Likewise, most domestic work falls on women’s shoulders, with over 70 percent declaring responsibility for laundry, cooking, and cleaning. However, men feel more overwhelmed than women: about 16 percent of men declare finding difficulties fulfilling family responsibilities compared to only 8 percent of women. Although most of the findings point to the importance of expanding childcare provision to alleviate women’s burden and enhance their participation in the economy, the perception toward childcare still seems negative. The majority of the population (over 95 percent of both men and women) believe that childcare should be provided by the family and that childcare costs should be covered by the family itself (figure 3.21). Likewise, over 60 percent believe that the best childcare option is for the mother to stay home and for the father to work full-time (figure 37  Data available at https://dbk.gesis.org/dbksearch/sdesc2.asp?no=5900&db=e&doi=10.4232/1.12661. 38  The estimation approach follows the method of Fortin (2015). The model was estimated using linear probability, probit, and logit models. Estimates from the first model and marginal effects from the other two models show very close results. The 2012 ISSP survey includes more variables about gender norms, social attitudes, and mother’s characteristics; the WVS includes more recent data. The WVS includes four rounds for the Philippines (1996, 2001, 2012, and 2019), but only the last two rounds include relatively detailed variables capturing traditional gender role attitudes. The estimation of the model using the four rounds’ data and using proxies such as “men should have more right to a job than women when jobs are scarce” or “men make better business executives than women” to capture traditional views did not show a significant impact on women’s employment probability. The effect of vari- ables related to the attitude toward working mothers remains large and significant even after controlling for other variables such as education, demographic characteristics, and income groups. The latter shows that, although women in poorer groups cannot really afford to stay out of the labor market, gender norms still seem to negatively affect their employment decision. 70 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 3 | Women in The Labor Market 3.22). The perception that the mother staying home is the best childcare option does not change across income groups or men’s education levels, but it is slightly lower among women with higher education levels about 44 percent of women with tertiary education believe that the mother staying home is the best childcare option and 17 percent believe that both parents should work full-time, compared with 69 percent and 8 percent, respectively, among those with only primary education. This suggests that the expansion of education among women may bring positive prospects for changing stereotypes and social norms. FIGURE 3.21 Attitude towards Who Should FIGURE 3.22 Best Childcare Option, by Parents’ Provide Childcare, by Gender, Philippines, 2012, Employment Type, Philippines, 2012, Percent Percent Source: Based on data from the International Social Survey Programme, Family and Changing Gender Roles, 2012. However, attitudes toward paid parental leave are largely positive. In 2020, the Philippines extended paid maternity leave from 60 to 105 days. The government administers a share of the maternity leave benefits, but employers are responsible for the difference between benefits issued through the Social Security System and an employee’s wages (World Bank 2021). Paid leave for fathers has been available since 1998 for a duration of 7 working days (9 calendar days). Enthusiasm among the population for the implementation of parental paid leave remains moderate. About 87 percent of the population believes that parents who work full-time and who have a newborn child should benefit from paid leave benefits if one of them stops working for some time to care for their child (figure 3.23).39 The proportion of those who oppose this idea is the same for men and women, but those who oppose most are young men below 30 years old (21 percent), followed by young women (14 percent). Among those who agree to paid parental leave, over 50 percent think that the employer should pay for the leave and nearly 25 percent think it should be covered by both the government and the employer. Also, among those who agree on parental leave, around 32 percent believe that, when both parents are in a similar work situation and are eligible for paid leave, the mother should benefit entirely from the paid leave period; 31 percent think it should be equally divided between the father and mother (figure 3.24). More women than men favor the allocation of more leave time to the father, particularly among higher education groups, seemingly indicating more willingness among educated women to involve their husbands in childcare work. 39  The data are from 2012, and it is possible that beliefs may have shifted over the past eight years. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 71 CHAPTER 3 | Women in The Labor Market FIGURE 3.23 Attitude toward Parental Paid FIGURE 3.24 Opinion on “How should this paid Leave, Philippines, 2012, Percent leave period be divided between the mother and the father?” Philippines, 2012, Percent Source: Based on data from the International Social Survey Programme, Family and Changing Gender Roles, 2012. Slow intergenerational mobility contributes to the persistence of gender social biases and limited economic opportunities for women. The employment status of women is highly correlated with that of their mother and their husband. Mother’s education level and employment status seem to affect the education and employment of their daughters (figures 3.25 and 3.26). Women of less-educated mothers and those whose mothers did not work or stopped working seem more likely to replicate the mother’s model. Intergenerational mobility across occupation sectors also seems limited, contributing to the persistence of economic gender gaps (figures 3.27 and 3.28). Women seem also to be influenced by the employment and occupation status of their husband (figure 3.29). When men are employed, their wives often stay out of the labor market and remain housewives. This is particularly true among poorer groups and, surprisingly, younger generations. Overall, controlling for other individual characteristics, having a mother who worked is associated with about a 9-pp increase in the probability of women’s participation in paid employment and an 11-pp decrease in the probability of being a housewife. The influence of the husband is also important: having a husband who is employed is associated with about a 17-pp increase in the probability of his wife being a housewife. FIGURE 3.25 Mother vs. Daughter FIGURE 3.26 Mother vs. FIGURE 3.27 Mother vs. Employment Status, Philippines, Daughter Education Level, Daughter Occupation, 2019, Percent Philippines, 2019, Percent Philippines, 2019, Percent 72 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 3 | Women in The Labor Market FIGURE 3.28 Mother vs. Daughter Employment FIGURE 3.29 Women vs. Husband Employment Status, by Age and Income Group, Philippines, Status, by Age and Income Group, Philippines, 2009, 2009, Percent Percent Sources: Figures 3.25, 3.27, 3.28, and 3.29 based on data from the International Social Survey Programme, Social Inequality 2009 (https://dbk.gesis.org/dbksearch/sdesc2.asp?no=5400&db=e&doi=10.4232/1.12777). Figure 3.26 based on data from the World Values Survey, wave 7. Note: N/A = not available; OLF = out of labor force Overcoming Barriers To Women’s Economic Empowerment In The Philippines 73 CHAPTER 4 Women’s Employment 74 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 4 | Women’s Employment The previous chapter revealed persistent large gender gaps in labor force participation, especially in relation to comparator countries. Within the Philippines, there is some regional variation in female labor force participation, suggesting that less developed and fragile settings are less conducive to women’s participation. Chapter 3 also showed that women’s higher educational attainment increases female labor force participation, but the positive effect of education is somewhat diminished by other factors. Among the factors holding back women’s involvement in the labor market, childcare and gender norms seem critical, with some indication that attitudes and female labor force participation patterns persist within families across generations. Having established that there is a persistent gender gap in labor force participation, this chapter dives further into an analysis of women in the labor market, examining the differences between employed men and women. It investigates whether women have the same opportunities as men to access productive jobs and, if not, what might be causing the discrepancy. The chapter further examines the extent and trends of gender pay gaps and investigates their determinants across the entire pay distribution. I. Gender Gaps in Type of Work, Sectors, and Occupations Women’s participation in productive jobs has steadily increased over the past decades. Vulnerable employment in the Philippines has fallen considerably over the past three decades for both men and women, although the shift out of vulnerable employment has been faster for men. Between 1988 and 2020, the Philippines saw large improvements in human development, poverty, and economic growth. During this period, the share of workers in vulnerable employment, defined as those employed in own-account work or unpaid family businesses, declined from 52 percent to 32 percent. This shift has generally been faster for men than women, with vulnerable employment decreasing from 51 percent to 30 percent for men and from 54 percent to 36 percent for women, which has resulted in a larger gap between men and women in 2020 (figure 4.1). Although vulnerable employment remains high in rural areas relative to urban areas, rural areas saw the largest declines in vulnerable employment from 1988 to 2020, decreasing from 61 percent to 39 percent for men, and from 64 percent to 44 percent for women. FIGURE 4.1 Trends in Vulnerable Employment, by Gender, 1988–2020, Percent Source: Based on data from the 1988–2020 Labor Force Surveys. The decline in vulnerable employment has been driven by the shift away from unpaid family business and self-employment to employment in private enterprises and, more generally, to wage work. From 1988 to 2020, the share of women working in unpaid family business fell by 15 percentage points (pp) and self-employment decreased by 3 pp as employment in private Overcoming Barriers To Women’s Economic Empowerment In The Philippines 75 CHAPTER 4 | Women’s Employment establishments grew by 15 pp (figure 4.2, panel a).40 For men, the decline in vulnerable employment was driven equally by the shift away from self-employment, which fell 11 pp, and the decline in the share engaged in unpaid family business, which fell 10 pp (figure 4.2, panel b). Following a similar pattern, employment in wage work increased for both men and women. During 2002–20, the proportion of women in wage work grew from 43 percent to 55 percent and the proportion of men in wage work rose from 43 percent to 56 percent.41 Since 2012, the Philippines has operated on a two-tier minimum wage rate system, wherein separate minimum wages exist for agricultural and nonagricultural sectors that vary by region;42 although the system allows for regional contextualization, its intricacies may hinder enforcement and encourage informal labor (Cho, Doan, et al. 2021). Although minimum wage rates in the Philippines are relatively high in comparison to other countries, it is worth noting that the Philippine Institute for Development Studies cautions that high legal minimum wages can decrease the probability of being employed for young workers, women, and those with lower educational attainment (Cho, Doan, et al. 2021, citing Paqueo et al. 2016). Despite the rise in the share of women working in the private sector, an important proportion of women is still employed in low-wage work that offers limited avenues for career growth. A much larger share of women than men continues to be employed in work that provides less potential for skill advancement; in 2020, 9 percent of women were employed in private households, compared to only 1 percent of men.43 In sharp contrast, a much larger proportion of men (59 percent, compared to 39 percent for women) was employed by establishments in the private sector, which are more likely to offer more opportunities for knowledge transfer and career advancement. Aside from the stark differences in the opportunities such work provides, those employed in private households also received lower pay; in 2020, whereas those employed by private firms earned PhP464 per day on average, those employed by private households earned approximately PHP256 per day on average—less than 60 percent the amount earned in wages from private firms. FIGURE 4.2 Trends in Type of Employment, by Gender, 1988–2020, Percent a. Trends among Women 40 Employment in private establishments refers to persons working for pay, in cash or in kind, in private industry as well as working for a religious group, missionary, union, nonprofit organization, or international organization. 41 Wage work includes work in private establishments, in government corporations and public administration, and in family-owned businesses and private households when the payment basis excludes commission, pakyaw, and other nonsalaried forms of pay- ment. 42 Minimum wage rates for each region are based on regional labor market conditions. 43 Estimates are based on the 2020 Labor Force Survey, January quarter. 76 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 4 | Women’s Employment b. Trends among Men Source: Based on data from the 1988–2020 Labor Force Surveys. Employed women work slightly fewer hours than men but have lower rates of underemployment, a pattern that persists across geographic location. Nationally, employed men work 42 hours per week and employed women work 41 hours per week.44 This small difference is consistent in both urban and rural areas, and varies only slightly between regions. However, the average employee works three more hours a week in urban areas relative to rural areas. Although men work longer hours than women, the underemployment rate (defined as being employed but seeking more hours of work) is higher among men. In urban areas, 12 percent of men would prefer more hours of work compared to 10 percent of women; in rural areas, 20 percent of men and 16 percent of women would prefer to have additional hours of work. The gap between men’s and women’s preferences for additional work varies across regions, but in every region apart from Ilocos (where the gap is close to zero) more men than women prefer to work more hours, likely because of the additional domestic obligations that women face. The gender gap in working hours varies considerably across the life cycle. Among the employed population, the number of hours worked per week is lowest for the youngest age cohort (15–19 years old) and is declining over time because of increased school participation and expansion of the number of years of schooling. The number of hours worked by women in the 20–40 age group is on average three hours lower than that of men. This coincides with peak reproductive ages as well as ages to take care of young children for women (figure 4.3). When considering the total working-age population, women appear to spend nearly half the time men spend in market work. In 2012, the gender gap was highest among the 25–39 age group, and this shifted to the 30–44 age cohort by 2020, probably because of the increase in marriage age over time (figure 4.4). The time spent by women aged 20–34 years on work outside the home seems to have increased by about four hours per week between 2012 and 2020. These changes may be explained on the one hand by the expansion of education among women and on the other hand by delays in getting married and having children. 44 Total of all occupations. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 77 CHAPTER 4 | Women’s Employment FIGURE 4.3 Average Total Time Worked, by Age FIGURE 4.4 Average Time Spent on Market Work, by and Gender, 2012–20, Hours per Week Age and Gender, 2012–20, Hours per Week Source: Based on data from the 2012 and 2020 Labor Force Surveys. Note: Figure 4.3 shows the average number of hours worked in all jobs per week for employed population only. Figure 4.4 shows the average number of hours worked per week for total working-age population. Women have shifted progressively from agriculture to services. The Philippines has undergone structural transformation over the past few decades, with women moving out of agriculture and into services across all regions, and men moving out of agriculture and into both industry and services. At the national level, the share of workers in agriculture fell 11 pp for men and 5 pp for women from 2012 to 2020 (figure 4.5, panel a). For men, the decline in the share of agricultural workers, from 39 percent in 2012 to 28 percent in 2020, was accompanied by a shift toward industry and services, in which shares of male workers grew by 6 pp and 5 pp, respectively, during this period. For women, the decline in the share of workers in agriculture was accompanied by a very small decrease in the share of workers in industry (0.7 pp), and a 6-pp increase in the share of workers in the services sector. For both men and women, this shift was almost entirely driven by changes in rural areas. At the regional level, women’s work in services increased in all regions except SOCCCKSARGEN and CARAGA, which both experienced conflict during 2012-20 period and saw almost no movement away from agriculture (figure 4.5, panel b). FIGURE 4.5 Sectoral Trends in Employment, 2012–20, Percent a. Sectoral Trends in b. Sectoral Trends among Women, by Region Employment, by Gender Source: Based on data from the 2012 and 2020 Labor Force Surveys. 78 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 4 | Women’s Employment Women continue to be absent from employment in industries that tend to require physical strength. In the industrial sector, men occupy a much larger share of employment in fields that have traditionally been male dominated and physically taxing. In 2020, 97 percent of workers in construction and 93 percent of workers in mining and quarrying were men. Within the industrial sector, the share of female employment does not reach beyond 16 percent in any industry except in manufacturing, where women represent 40 percent of the workforce (figure 4.6). In stark contrast, women occupy much larger shares across different subsectors within services. Notably, traditionally female-dominated activities such as education and human health and social work employ a much larger share of women relative to men (74 percent and 67 percent, respectively). The only subsector that employs a disproportionate share of men within the services sector is transportation and storage (96 percent). Other service industries—including professional, scientific, and technical activities and financial and insurance activities—employ either an equal proportion of men and women or a slightly higher proportion of women (48 percent and 61 percent, respectively). FIGURE 4.6 Share of Employment, by Industry and Gender, 2020, Percent Source: Based on data from the 2020 Labor Force Survey. Although many women are engaged in high-skilled occupations, a substantial proportion are still engaged in nonwage work. Women occupy at least half of employment in leadership positions and occupations that require a higher level of knowledge or skills. Women occupy 49 percent of managerial positions in the country, 51 percent of technicians and associate professional positions, and 65 percent of professional positions (figure 4.7, panel a). The Women in Business 2020 report finds similar results for the Philippines, with the report stating that 43 percent of senior management positions in the country were filled by women, ranking the country first in this metric among all 32 countries surveyed (Grant Thornton 2020). The pattern of employment across medium- skilled occupations largely follows the pattern of employment in industry, where traditionally male-dominated occupations continue to employ a larger proportion of men; these include skilled agricultural occupations, craft and trade workers, and plant and machine operators. Overall, a larger proportion of women is employed in high-skilled occupations relative to men; among employed women, 27 percent work in high-skilled occupations compared to 14 percent of men, whereas a larger proportion of men is employed in both medium-skilled occupations (55 percent of men compared to 50 percent of women) and low-skilled occupations (30 percent of men and 23 percent of women) (figure 4.7, panel b). Overcoming Barriers To Women’s Economic Empowerment In The Philippines 79 CHAPTER 4 | Women’s Employment FIGURE 4.7 Employment, by Occupation and Gender, 2020, Percent a. Share of Male and Female Employment, by Occupation and Level of Skill b. Share of Workers, by Occupation and Gender Source: Based on data from the 2020 Labor Force Survey. The gender occupational differences are related to the fact that women hold higher levels of education relative to men. In 2020, 43 percent of employed men had not completed high school, compared to 31 percent of employed women (figure 4.8). In contrast, over 40 percent of employed women had completed postsecondary education and above, compared to only 26 percent of employed men. These figures indicate that women who are more highly educated are disproportionately represented in the employment pool, which could be because the opportunity cost of choosing not to work is much higher for educated women. This mirrors the findings in chapter 3 that women with a college degree and above are 27 pp more likely to participate in the labor market compared to women with primary education or less. The disparity in educational attainment also accounts for, at least in part, the relatively larger proportion of women in managerial and professional roles compared to their male peers. FIGURE 4.8 Educational Attainment of Employed, by Gender, 2020, Percent Source: Based on data from the 2020 Labor Force Survey. Note: TVET = technical and vocational education and training. 80 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 4 | Women’s Employment At low and intermediate education levels, men are more likely to be engaged in wage work. Overall, the proportion of men engaged in wage work is only slightly higher than that of women (56 percent compared to 55 percent). However, differences in the type of work men and women are engaged in become more pronounced when their levels of education are factored in. At every level of education, with the exception of university graduates and those with no education, men are more likely to be engaged in wage work compared to women (figure 4.9). In stark contrast, women are more likely to be in unpaid family business at every level of education and are also more likely to be self-employed at higher levels of education. FIGURE 4.9 Wage Work as a Share of Employment, by Education and Gender, 2020, Percent Source: Based on data from the 2020 Labor Force Survey. Domestic obligations could have an influential role in shaping women’s decision to engage in nonwage work. Among the factors that women are likely to consider when deciding the type of work to engage in is the allocation of their time between housework or care work and market work, especially for women with heavier domestic obligations. As formal wage jobs are typically less flexible than nonwage work because of the set schedules and minimum hours of work required (World Bank 2012), women who shoulder heavier domestic obligations may instead choose to engage in nonwage work despite the benefits that wage work can confer, which include greater stability, more opportunities to build professional networks, and skills development. Women face few legal barriers in the labor force and have equal rights to men. Legislation prohibits discrimination in employment based on gender, and dismissing pregnant women is illegal (World Bank 2020a). Equal compensation for the same work is required, and women have no restrictions on any profession. Maternity leave in the Philippines is generous, with 15 weeks paid for by the government. Paternity leave is less generous, with only seven paid days. Women are also able to sign contracts, register businesses, and open bank accounts in the same way as men. It does not appear that legal constraints are a challenge for women in their labor force choices. II. Family Responsibilities in Women’s Employment Among employed women, those married and with young children have profiles that are significantly different from those of single women.45 Single women, on average, tend to be younger than married women and women with young children, with a mean age of 29, compared to mean ages of 42 for married women and 36 for women with young children. Single women also tend to have higher educational attainment, with 44 percent in this group having completed tertiary education compared to 22 percent among married women and 24 percent among women with young children. Narrowing the analysis to women in peak reproductive ages (25–39 years 45 For the purpose of this analysis, “women with young children” refers to women with at least one young child (from birth to four years old) within the household. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 81 CHAPTER 4 | Women’s Employment old), which also corresponds with heavier domestic duties (Abrigo and Francisco-Abrigo 2019), shows an even larger gap, with 53 percent of single women having completed tertiary education, compared to only 27 percent of married women and 32 percent of women with young children (figure 4.10, panel a). In contrast, the educational differences between men within the same age range are smaller, with 26 percent of single men having completed tertiary education, compared to 13 percent of married men and 14 percent of men with young children (figure 4.10, panel b). Because men are expected to participate in the labor force regardless of skill level or educational attainment, the larger educational gap among the different groups of employed women likely reflects the opportunity cost of choosing to stay at home for domestic obligations, especially for women with higher levels of educational attainment. FIGURE 4.10 Educational Attainment, by Familial Status and Gender, for Employed Population, Ages 25–39, 2020, Percent a. Female Educational Attainment b. Male Educational Attainment Source: Based on data from the 2020 Labor Force Survey. At peak reproductive ages, married women and women with children are more frequently engaged in nonwage work compared to single women. Among employed women aged 25– 39, those who are single are much more likely to engage in wage work (87 percent) relative to married women and women with young children (58 percent and 63 percent, respectively). These differences are largely driven by the share of women working in private establishments (68 percent, compared to 44 percent of married women) and the substantial proportion of women engaged in self-employment, which, compared to their single peers, is 19 pp higher among married women and 16 pp higher among women with young children (figure 4.11). High rates of self-employment and nonwage work, which remain even when comparing women with similar levels of education, suggest that, when married women and women with children do work, they more frequently choose employment that typically has more flexible working arrangements. It is also important to note that a substantial proportion of married women and women with children does not even engage in market work. When women outside the labor force are included, only 26 percent of married women and 27 percent of women with children are in wage work, compared to 58 percent of single women, providing further evidence that the decision to work and the type of work women choose to engage in are influenced by their familial status and the domestic obligations that they bear. 82 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 4 | Women’s Employment FIGURE 4.11 Female Employment Type, by Familial Status, Ages 25–39, 2020, Percent Source: Based on data from the 2020 Labor Force Survey. Family obligations appear to lead women to work fewer hours. Single men aged 25–39 work 2.6 fewer hours, on average, than single women in the same age group (figure 4.12). However, this relationship is reversed when comparing married women or those with children with their male counterparts. In addition, married women and women with children are also more likely to cite personal and family reasons as the main reason why they work less than 40 hours per week (15 percent of married women and 16 percent of women with children, compared to 7 percent of single women). Although these are small differences in hours worked, they do match a common pattern in which family obligations lead women to work fewer hours outside the home to compensate for time spent on house or care work, and men work longer hours to provide financial support to their family. FIGURE 4.12 Average Hours Worked, by Familial Status and Gender, Ages 25–39, 2020 Source: Based on data from the 2020 Labor Force Survey. Married women and women with young children are more likely than single women to work in agriculture, especially in rural areas. At the national level, larger shares of married women and women with young children (17 percent and 14 percent, respectively) are employed in the agricultural sector compared to single women (6 percent). The gap is larger in rural areas, where most agricultural activities are located (figure 4.13). Women employed in agriculture tend to work substantially fewer hours than men, suggesting that women’s work in this sector is often not full- time; whereas the difference in hours worked between men and women in industry and services is 1 hour per week, the difference is starker in agriculture, estimated at 10 hours per week. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 83 CHAPTER 4 | Women’s Employment FIGURE 4.13 Share of Rural Employment in Agriculture, by Familial Status and Gender, 2020, Percent Source: Based on data from the 2020 Labor Force Survey. Extended family and hired help alleviate domestic obligations, allowing women to pursue opportunities in the labor market. Additional support in domestic obligations could allow more women to engage in wage work. One factor that could be driving lower rates of wage workers among women is the difficulty involved in balancing housework with work outside the home, especially when that outside work entails fixed hours and a set schedule, which are common with wage work. Receiving support with domestic obligations, whether through help from extended family members or through hired help, allows women to reallocate responsibilities within the household, enabling them to pursue jobs that may be more advantageous for their career despite being less flexible. Among women aged 25–39 who are married or have children, larger proportions of those who have either extended family or hired help in the household are employed in wage work (figure 4.14). It is also worth noting that having additional support does not seem to have as large an effect on wage work among single women, who are less likely to have heavy domestic obligations. FIGURE 4.14 Share of Employed Women in Wage Employment, by Familial Status and Type of Support, Ages 25–39, 2020, Percent Source: Based on data from the 2020 Labor Force Survey. Additional support in domestic obligations could also allow women to work longer hours. Married women and women with children on average work longer hours when there is extended family or hired help present. Among married women who receive support from extended family members, the average hours worked per week increase by two hours compared to those with no extended family support; for women with children, the additional hours worked increase by four hours per week relative to those with no extended family support (figure 4.15). It is worth noting that having extended family support does not seem to increase the number of hours worked by men, and having hired help adds only slightly to the number of hours worked, which could be because men have less of the household responsibility compared to women. 84 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 4 | Women’s Employment FIGURE 4.15 Average Total Hours Worked, by Familial Status, Gender, and Type of Support, Ages 25–39, 2020 Source: Based on data from the 2020 Labor Force Survey. Empirical analysis further supports findings that employed women with family obligations are less likely to engage in wage work. Estimates of a probit regression show that marriage and presence of young children in the household correlate, respectively, with a 16-pp and 4-pp decline in the probability of engaging in wage work for employed women aged 25–39 years old but that education is significantly positively associated with women’s wage work; relative to the reference group of women with primary education and less, postsecondary and technical and vocational education and training are associated with a 14-pp increase in wage work probability and having a college degree and above is associated with a 31-pp increase. The positive effect of hired domestic and extended family help on the likelihood of women’s wage work vanishes when controlling for other household and individual characteristics (such as education, geographic location, and household income security). III. Gender Pay Gaps Women earn more than men on average, but there are variations across the wage distribution. Women’s wages are on average higher than men’s, but this advantage hides important gaps against women. Using mean daily wages and mean hourly wages, respectively, women appear to earn about 5 percent and 7 percent more than men in the Philippines. However, the distribution of women’s wages is more dispersed and polarized than the distribution of men’s wages, indicating that earnings inequalities are significantly higher among women than among men and that women tend to cluster more than men in low-skilled/low-paying jobs as well as in high-skilled/ high-paying jobs and less in medium-skilled occupations.46 Although inequality and polarization in wage distribution declined for both men and women between 2012 and 2020, the disparity remained significantly higher among women in 2020. The Gini coefficient in daily wages declined from 37.7 to 30.8 for men and from 46.9 to 46.0 for women, and the Foster-Wolfson polarization index declined from 0.114 to 0.086 for men and from 0.247 to 0.165 for women. These effects can be seen in the density of wages in figure 4.16 panels a–d. Women’s wage distributions are characterized by high irregularities, with peaks and troughs, suggesting that women’s earnings are less smoothly or regularly distributed across the range of daily (and hourly) wages than men’s earnings are. A higher proportion of women wage employees than of men is either located at the low end of the wage distribution or receives wages at the upper end of the wage distribution, whereas a larger proportion of men wage employees is located in the middle of the wage distribution. In part because of these irregularities, the transformation of wages into logarithms shows a slightly 46 Polarization is a process in which concentration in high and low wage groups increases while concentration in the middle wage group declines. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 85 CHAPTER 4 | Women’s Employment higher mean for men than for women and equal medians. Although the wage distributions became smoother in 2020 compared with 2012, irregularities remained more marked for women than for men. This indicates that women continue to concentrate in specific ranges of wages reflecting their selective labor market participation.47 In a wage distribution characterized by such irregularities, gender pay gaps based on mean or median estimates are distorted by these clustering effects and provide a biased picture of the extent and importance of the gender earnings gaps. FIGURE 4.16 Density of Daily and Hourly Log Wages, by Gender, 2012 and 2020 a. Density of Daily Log Wages, by Gender, 2012 b. Density of Daily Log Wages, by Gender, 2020 1 1 .8 .8 .6 .6 Density Density Density .4 .4 .2 .2 0 0 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 Range of daily wages Range of daily wages Men wage Women wage Men wage Women wage c. Density of Hourly Log Wages, by Gender, d. Density of Hourly Log Wages, by Gender, 2020 2012 1 1 .8 .8 .6 Density Density Density Density .6 .4 .4 .2 .2 0 0 .5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 .5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 Range of daily wages Range of daily wages Men wage Women wage Men wage Women wage Source: Based on data from the 2012 and 2020 Labor Force Surveys. Note: Wages are transformed into natural logarithms and densities are estimated using the Epanechnikov kernel and bandwidths of 0.1 for men and 0.12 for women. The broken vertical lines show median values for women and men, respectively; the solid lines show the mean values for each gender. Median log (daily and hourly) wages were equal for men and women in 2020, but were slightly higher for men in 2012. Employed women are both more educated and in high-skilled employment at higher rates than men, but women earn less than men at the same level of educational attainment. Across all skill levels and all occupations,48 men receive higher wages than women. Women also earn less than men across all levels of education; however, the wage gap becomes narrower with higher levels of education (figure 4.17). There has been some improvement in terms of narrowing pay differentials; 47 Similar patterns were found in the 2018/19 Global Wage Report, particularly in developing economies where women’s labor force participation is limited. 48 Excluding clerical support work. 86 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 4 | Women’s Employment from 2012 to 2020, women’s wages, expressed as a percentage of men’s wages, have gone up for all education levels, with the exception of those without schooling, which have remained constant, and those with a college degree, which have fallen slightly. FIGURE 4.17 Ratio of Women’s to Men’s Average Wages, by Educational Attainment, 2012 and 2020, Percent Source: Based on data from the 2012 and 2020 Labor Force Surveys. A more complete understanding of the extent and drivers of the gender pay differentials requires an analysis across the entire wage distribution.49 The overrepresentation of women at the lower and upper ends of the wage distribution requires understanding gender pay gaps at different points (quantiles) across the wage distribution to determine whether the gender pay differentials are in favor of men or women and if they are higher at the bottom of the wage distribution the sticky floor effect or at the top of the distribution, when women’s wages stop rising much earlier than men’s (the glass ceiling effect). It is also important to understand the part of the observed pay differential between men and women that is due to differences in labor market attributes and personal characteristics (human capital endowments, occupation levels, sectors of employment, family and demographic characteristics, and so forth), called the wage composition or endowments effect, and the part due to gender differences in the returns to those labor market and personal characteristics, called the wage structure or returns effect, which indicates a gender discrimination effect. The decomposition of the gender pay gap into endowments and returns effects proceeds in three steps. The approach follows the method of Firpo, Fortin, and Lemieux (2009, 2018), using a three- step methodology.50 The first step is to estimate the unconditional quantiles using the recentered influence function (RIF). The second step involves estimating a counterfactual distribution of wages for women, which represents the wages that women would have earned if they had received the same returns for their labor market attributes and characteristics as men. The third step consists in using the counterfactual distribution to decompose the difference in wage distributions between men and women into a part that is entirely explained by differences in attributes and characteristics between men and women composition or endowments effects and a part that is due only to differences in returns to those characteristics (for example, when men and women with similar education levels or occupational status receive different returns to these attributes). The latter is referred to as a wage structure or returns effect and reflects discriminations in gender pay gaps.51 The endowments and returns effects can be further decomposed into the contribution of individual specific characteristics or groups of some characteristics. The main advantage of the 49 Because daily and hourly pay are available only for wage employment, the analysis is constrained when trying to understand the full extent of gender income gaps. 50 The method is similar to the Oaxaca-Blinder decomposition approach but can be applied to various distributional measures, such as quantiles, Gini, interquantiles, and so forth.  51 The discrimination is not necessarily against women and can be against men as well. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 87 CHAPTER 4 | Women’s Employment unconditional quantile decomposition using the RIF-regression method is that it provides a linear approximation of nonlinear functionals of the distribution, such as the quantiles, Gini coefficient, and so forth. In its simplest form, the approach assumes that the conditional expectation of the quantiles can be modeled as a linear function of the explanatory variables, and thus estimated using the ordinary least squares regression. The problem is that, when the linearity does not hold, this affects the regression coefficients and thus the wage structure and composition effects (Firpo, Fortin, and Lemieux 2018; Fortin, Lemieux, and Firpo 2010; Rothe 2015). The issue can be addressed by adjusting the wage distributions by a reweighting factor, which can be estimated using a logit or probit model. Another problem is related to the use of categorical explanatory variables such as education groups, sectors of employment, and so forth. The contribution of these variables to the wage structure or returns effect is sensitive to the choice of the base group. The issue can be addressed using a normalization approach to the base group (Firpo, Fortin, and Lemieux 2007, 2018; Rios-Avila 2019). More technical details about the methodology and the variables used in the regressions can be found in appendix B. Women’s pay disadvantage is hidden by their educational attainment. Men have significantly higher pay than women at the bottom of the wage distribution, mainly driven by higher returns to men’s attributes, but women have higher pay than men at the top of the distribution because they have higher endowments. Figures 4.18 and 4.19 show the decomposition of the gender pay gap for 2012 and 2020 at different (unconditional) quantiles of the daily wage distribution.52 Each chart shows the overall pay gap (the green line), the pure endowments effect (the blue line), the pure returns effect (the orange line), and the specification and reweighting errors (dotted lines).53 In 2020, the daily wage was on average 56 percent higher for men than for women at the lowest decile. The gender pay differential declines as we move up in the wage distribution and favors women starting from the sixth decile. At the top decile, women’s daily wage is on average 20 percent higher than that of men. Women at the bottom of the wage distribution have lower qualifications and labor market attributes than men, which significantly contributes to the pay differential; however, most of the gap is driven by the fact that women receive lower returns to their qualifications then men. At the first decile, differences in endowments between men and women contribute 19 percent to the total pay differential, whereas differences in returns contribute 77 percent. This means that at the lower end of the wage distribution, even when women have the same qualifications and attributes as men, they still receive lower pay or returns for those attributes than men do. This reflects potential discrimination effects against women at this part of the distribution, which contribute to the sticky floor effects. The situation is different at the upper end of the wage distribution, where women have higher pay than men because they have higher qualifications and characteristics. At the ninth decile, women have almost three times greater endowments than men, but they still receive lower returns to these endowments than men. Overall, the returns effects dominate the endowments effects—and are positive—across the entire wage distribution, indicating that at each quantile, women with the same pay-determining characteristics as men receive lower pay or returns to those characteristics. Although the gender gap in returns declines at upper deciles, it still affects women’s pay in a statistically significant negative way—women’s pay would have been on average 16 percent higher if they had received the same returns for their qualifications as men. 52 The results are based on the reweighted-regression decomposition, where the reweighting factor is estimated using the logit model (the results for the linear model and the reweighted decomposition using the probit model are in figures B.1–B.5 in appendix B). 53 The specification error reflects the importance of departures from the linearity assumption of the RIF-regressions and the fact that, except for the mean, the RIF depends on the distribution of log wages (and thus from the explanatory variables X through their effect on log wages). The reweighting allows for assessing the quality of the reweighting (see appendix B and Firpo, Fortin, and Lemieux 2018 for more technical details). 88 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 4 | Women’s Employment FIGURE 4.18 Endowments and Returns Effects FIGURE 4.19 Endowments and Returns Effects on on Gender Pay Gap, 2012 Gender Pay Gap, 2020 .4 .4 Difference in log daily wage Difference in log daily wage .2 .2 0 0 -.2 -.2 -.4 -.4 -.6 -.6 .1 .2 .3 .4 .5 .6 .7 .8 .9 .1 .2 .3 .4 .5 .6 .7 .8 .9 Quantiles Quantiles Total wage gap Pure Endowments Effect Total wage gap Pure Endowments Effect Pure returns effect Total Endowments Effect Pure returns effect Total Endowments Effect Total returns effect Total returns effect TABLE 4.1 Estimates of Endowments and Returns Effects on Gender Pay Gap, 2012 and 2020 First decile Median Ninth decile 2012 2020 2012 2020 2012 2020 Total 0.365*** 0.443*** 0.163*** 0.037*** -0.347*** -0.226*** Pure endowments effect 0.162*** 0.083*** -0.080*** -0.064*** -0.575*** -0.571*** Specification error -0.089*** -0.007 -0.047*** -0.047*** 0.207*** 0.317*** Total endowments 0.073*** 0.076*** -0.127*** -0.111*** -0.368*** -0.254*** Pure returns effect 0.268*** 0.343*** 0.260*** 0.118*** 0.001 0.032** Specification error 0.024** 0.024** 0.029*** 0.029*** 0.021 -0.005 Total returns 0.292*** 0.367*** 0.289*** 0.148*** 0.022 0.028 Source: Based on data from the 2012 and 2020 Labor Force Surveys. Note: The decomposition estimates the difference between men’s and women’s wages. Men are used as the reference group used to compute the wage composition (endowment) and structure (returns) effects, meaning that the counterfactual estimates the wage that would prevail if wage-working women were paid under the wage structure of men. Negative values mean that the gap is at the advantage of women. The unconditional quantiles are estimated using the Epanechnikov kernel and bandwidths of 0.07. The reweighting factors are estimated using the logit model. Significance level: * = 10 percent, ** = 5 percent, *** = 1 percent, based on bootstrap standard deviations using 100 replications. Qualifications and endowments for women at the bottom of the distribution increased faster than for men during 2012–18, but the wage gap increased because of widening inequalities in returns. Between 2012 and 2020, the average daily wage for men increased slightly faster than for women at the lower and upper ends of the wage distribution (figure 4.20). This resulted in an increase of the wage gap in favor of men at the lower quantiles and a decline of the wage gap in favor of women at the upper quantiles. During 2012-20, women at the bottom of the distribution saw a faster increase in their labor market attributes and endowments then men did, which helped to significantly reduce the gender gap in endowments (table 4.1). However, the returns women received for those characteristics seem to have increased at a slower pace than for men, resulting in a widening of the gender gap in returns and therefore in the overall daily wage gap. At the top decile, the endowments of men increased more quickly, helping them to partly close the gap with women. This, coupled with an increased gap in returns in favor of men, helped to reduce the gender Overcoming Barriers To Women’s Economic Empowerment In The Philippines 89 CHAPTER 4 | Women’s Employment differential between men and women though the total gap remains in favor of women.54 In the middle of the wage distribution—second through eighth deciles—women’s pay increased faster than men’s, resulting in a significant reduction of the gender differential (which is in favor of men) at the median and below, and an increase of the gap in favor of women in the sixth through eighth deciles. The changes result from two factors: a decline in the gap in endowments between men and women and, to a greater extent, a reduction of the gender differences in returns. FIGURE 4.20 Average Daily Wage, by Gender and Percentile, 2012 and 2020, PhP Source: Based on data from the 2012 and 2020 Labor Force Surveys. Note: PhP = Philippine peso; Q = percentile. Well-designed policies should consider the endowments and returns effects to address gender pay gaps. The results point to the importance of designing policies to promote equal pay for men and women. The evaluation of the part of the gender pay gap explained by labor market characteristics and endowments can inform policy makers in designing policies that can empower women in poorer groups and increase their abilities and skills to close the pay gap with men. The importance (and increase) in the gap in returns between men and women, particularly among poorer groups, suggests that reducing the gender pay gap requires additional measures that are more targeted to eliminate pay discrimination and promote legal frameworks, and policies that are conducive to further efforts for equal pay for work between women and men. Dividing the endowments and returns effects into the contribution of the different sets of labor market attributes and other characteristics would allow for better informing policy design that targets differences in endowments between women and men such as reducing differences in educational attainment, enhancing diversification across occupations or sectors, improving access to childcare and family support services, and so forth and policies that ensure equal returns to those characteristics for men and women, such as equal returns to education for men and women. Women are favored by better levels of education than men, but at the lower deciles they are disadvantaged by their sector and class of employment. The main two factors that contribute to higher endowments among men than among women, particularly at the quantiles below the median, are the sector of employment and the class of employment (figures 4.21 and 4.22). The gaps in returns between men and women at the bottom of the distribution are essentially due to 54 At the 99th percentile the wage gap is in favor of men even though women have higher endowments. See appendix B for more discussion of this effect. 90 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 4 | Women’s Employment the fact that women tend to occupy jobs that have lower returns than those men hold, probably because of the nature and requirements of these jobs. Gender differentials in returns are also due to demographic characteristics and geographic location. The number of children in the household has a negative impact on the returns of both men and women; however, the impact is larger on women than on men. Similarly, returns to age are positive for men but negative for women, suggesting that men may have better rewards for their experience than women have.55 FIGURE 4.21 Detailed Endowments and Returns FIGURE 4.22 Detailed Endowments and Returns Effects on Gender Pay Gap, 2012 Effects on Gender Pay Gap, 2020 Source: Based on data from the 2012 and 2020 Labor Force Surveys. Note: Q = quantile. The education advantage of women over men is apparent at every point of the wage distribution, though it has declined over time. At each decile, employed women have higher education endowments than men. The education gap in favor of women is significantly higher at the upper deciles than among poorer groups (figures 4.23 and 4.24). This is driven by the fact that proportion of employed women with postsecondary education, and to a greater extent with a college degree, is significantly higher than the proportion of employed men. In contrast, the proportion of employed men who only completed high school is higher than that of employed women. This is true for all deciles, but particularly higher ones. At the median and the top decile, women and men receive equal pay or returns for their education levels meaning that similar education levels for men and women are rewarded the same way. However, at the other deciles, women tend to receive higher returns for their education even when they have same education qualification as men. In 2012, women at the bottom of the distribution were receiving lower returns for their education than men; however, the situation changed in 2020 with the returns gap favoring women. 55 The effect of geographic location on the differential in returns by gender in the median and upper quantiles seems to be driven by the fact that men in higher wage groups in Luzon appear to receive higher returns to their endowments than women do. In particular, men in medium- and low-skilled occupations seem to receive higher returns than women who are in the same occupa- tions. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 91 CHAPTER 4 | Women’s Employment FIGURE 4.23 Endowment and Return Effects of FIGURE 4.24 Endowment and Return Effects of Education on Gender Pay Gap, 2012 Education on Gender Pay Gap, 2020 Source: Based on data from the 2012 and 2020 Labor Force Surveys. Note: Q = quantile. Men tend more than women to work in the private sector and to diversify across sectors, resulting in gender endowment gaps in favor of men. Women tend to work more in government and public corporations whereas men tend to work more in the private sector. However, the gaps in favor of men in the private sector are larger than the gaps in favor of women in government and public corporations. This is true for all deciles except the top one, where employment in government and public corporations is much larger for women than for men, and the gender difference in employment in the private sector is trivial. Similarly, among the poorer groups, there is a much larger concentration of men than of women in the construction and, to a lesser extent, manufacturing sectors. Women’s overrepresentation in wholesale and retail trade and in public administration and human development services, as well as in other services, does not seem to compensate for the gaps in construction and manufacturing. At lower quantiles, women tend to work more hours than men, but at upper quantiles men work more hours. This results in relatively large gaps in endowments in favor of men in employment. However, women working in the private sector and to a lesser extent in government corporations, as well as women working in services, tend to receive higher returns than men (figures 4.25 and 4.26 and tables B.1 and B.2).56 FIGURE 4.25 Endowment and Return Effects of FIGURE 4.26 Endowment and Return Effects of Employment on Gender Pay Gap, 2012 Employment on Gender Pay Gap, 2020 Source: Based on data from the 2012 and 2020 Labor Force Surveys. Note: Q = quantile. 56 The 2018 Global Wage Report found that, in many countries, women’s work tends to be undervalued in highly feminized occupa- tions and sectors, because women working in these sectors seem to receive lower rewards to their skills and experience than men (ILO 2018); however, evidence does not seem to support this claim in the Philippines.  92 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 4 | Women’s Employment Most of the gender gaps in returns seem to be driven by demographic differentials between men and women. First, it is important to note that the number of children below the age of eight years in the household is significantly lower among wage-employed women than among wage- employed men. This is because a large proportion of women with young children are out of the labor force. Among both men and women, the number of children is higher at lower quantiles. The number of children negatively affects the returns of men and women, but the impact is much larger on women. It is also much larger among poorer groups (tables B.1 and B.2).57 The result is large gaps in returns between men and women that seem to be driving the overall gender returns gaps (figures 4.27 and 4.28). This suggests that women have higher endowments than men and would have received higher pay if they received the same returns and rewards to their characteristics as men. However, this inequality in returns does not seem to be driven by discriminatory policies or practices in the labor market, but rather by the higher burden exerted by children on women’s returns than on men’s. FIGURE 4.27 Endowment and Return Effects of FIGURE 4.28 Endowment and Return Effects of Demographic Characteristics on Gender Pay Gap, Demographic Characteristics on Gender Pay Gap, 2012 2020 Source: Based on data from the 2012 and 2020 Labor Force Surveys. Note: Endowments differences are approximately equal to zero across all quantiles in both figures. Q = quantile. Childcare support and family support may help women have higher returns. Wage-working women benefit from higher domestic help and assistance from extended family members than men do. However, the proportion of wage-working women with a young child in school is lower than that of men because women with children aged four years and below tend to stay out of the labor force even when the child is in school. When they benefit from domestic help, however, the proportion of women with young children that joins the labor market and wage employment increases significantly, and it increases even more when young children are in school. Support from extended family members and having young children in school help increase women’s returns at the lower quantiles, and domestic help positively affects women’s returns at higher quantiles (tables B.1 and B.2). These effects are higher for women than for men, resulting in returns gaps in favor of women (figures 4.29 and 4.30), which suggests that expanding childcare and family support services to alleviate the burden on women could not only increase their participation in economic life but also increase their productivity. Doing so would help empower women and reduce the gender gaps in economic opportunities. 57 Surprisingly, most of the negative impact comes from children aged 12–14 years rather than from younger children. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 93 CHAPTER 4 | Women’s Employment FIGURE 4.29 Endowment and Return Effects of FIGURE 4.30 Endowment and Return Effects of Childcare Assistance on Gender Pay Gap, 2012 Childcare Assistance on Gender Pay Gap, 2020 Source: Based on data from the 2012 and 2020 Labor Force Surveys. Note: Q = quantile. 94 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 5 Gender Inequality and the COVID-19 Pandemic Overcoming Barriers To Women’s Economic Empowerment In The Philippines 95 CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic Chapter 4 examined the state of employed women, showing that, although vulnerable employment has fallen and wage work has increased for Filipino women, a large share is still employed in low-wage work. There has been an overall growth in employment in the service sector for both men and women, but women are employed at lower rates in the industry sector. Because of women’s higher educational attainment, they make up at least half of leadership and high-skilled positions. However, a substantial proportion of women are still engaging in nonwage work, with marriage and children increasing the likelihood of women’s participation in this type of work. Although the descriptive analysis suggests that assistance with domestic obligations—such as by the presence of extended family members in the household—may allow women to engage in wage work and work more hours, the effect vanishes when controlling for other characteristics. Although women’s wages are on average higher than men’s, the distribution of wages shows that women earn less than men at the bottom of the pay distribution but have higher pay then men at the top of the distribution because of higher educational endowments. Chapter 4’s analysis of the state of women’s employment indicates that there is still much work to be done to address gender equity in labor markets. However, these data was collected before the onset of COVID-19, which has negatively affected employment patterns for both men and women, globally and in the Philippines. COVID-19 continues to have devastating health and economic impacts. As in most crises, the effects of the pandemic are not gender neutral. Recent reports by the United Nations Development Programme and UN Women reveal the exposure of women and girls to disproportionate economic and social risks from the pandemic due to entrenched inequalities, social norms, and unequal power relations. Women generally have less capacity than men to absorb economic shocks and can be especially hurt by the resulting economic and social fallout because they have weaker coping strategies and are overrepresented in hard-hit economic sectors. The effects are not just economic. The lockdowns and quarantine measures are expected to increase food security risks for women, who tend to eat less to provide for children and other family members. More people at home will mean increased unpaid workloads for women, especially in caring for children and relatives and in managing household tasks, and heightened risks for gender- based violence. The gradual reopening will pose additional challenges for women because it could push them out of the labor force or into part-time jobs or unpaid care work. All this puts women’s capabilities at risk and could undermine hard-won gains in gender equality. Understanding the gender-differentiated impacts of the COVID-19 crisis is key for informing policy responses that reduce women’s vulnerability and sustain efforts in rectifying longstanding gender inequalities. Recent data from High-Frequency Surveys (HFSs) help to provide greater understanding of the gendered impact of the pandemic. This chapter uses information from the Philippines HFS and the Low-Income Household Panel Economic Survey (HOPE survey), as well as recent Labor Force Surveys (LFSs), to examine the effects of the COVID-19 crisis on women’s socioeconomic situation (see box 5.1). 96 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic BOX 5.1 Philippines COVID-19 High-Frequency Survey and Low-Income Household Panel Economic Survey High-Frequency Survey The High Frequency Survey (HFS) in the Philippines is part of global set of surveys implemented by the World Bank and partners to monitor the evolving impacts of COVID-19. The HFS surveys provide a real-time, ground-level picture of what has been happening in certain countries as the COVID-19 pandemic unfolds. In collaboration with the Department of Finance and the National Economic and Development Authority, and with support from the Australian Government, the World Bank implemented two quarterly surveys (household and firm surveys) and community surveys to monitor welfare changes, identify negatively affected groups and track the impacts of the pandemic on employment and firms’ operations in the Philippines. This chapter relies on the first round of the household, firm, and community surveys conducted in July–September 2020 and the second round of the household survey conducted in December 2020. Two additional rounds are planned by the end of April 2022. In the first round, 9,448 households were interviewed through a combination of web and phone; in the second round, 1,805 households were reinterviewed by phone. The survey is weighted to be nationally representative to understand the impacts of the COVID-19 pandemic. Although not gender-focused, the HFS provides valuable information that contributes to assessing gender-specific impacts and coping mechanisms, albeit with limitations. The survey in particular provides a rich set of information on gender impacts, allowing identification of gendered job losses, income impact, and household coping mechanisms, as well as changes in childcare. Although information on the head of household is collected, only one adult member per household was surveyed. This limits the analysis on the differing effects on men and women within the same household. Low-Income Household Panel Economic Survey The Low-Income Household Panel Economic Survey (HOPE survey) is a panel survey that follows individuals in low-income households who are enrolled in the country’s flagship social protection program, Pantawid Pamilyang Pilipino Program (4Ps), and other comparable low-income households; it is funded with financial support from the government of Australia and the World Bank. Four rounds of the HOPE survey were carried out from December 2019 to August 2020. The sample is composed of 580 low-income households (and 1,614 adult individuals) based on an existing set of 4Ps and non-4Ps households that were a part of previous impact evaluation studies, with the first wave conducted in person and subsequent rounds by phone. The survey provides detailed insights on the impacts of COVID-19 on poor households with a focus on social protection mechanisms, as well as education and health. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 97 CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic I. Impact of COVID-19 on Women’s Employment With the shock of COVID-19 and enhanced community quarantine (ECQ), LFS figures show that unemployment quickly increased and has remained well above previous levels for both men and women. Unemployment more than tripled at the start of the COVID-19 pandemic in the Philippines, jumping from 5.3 percent in January 2020 to 17.7 percent in April 2020 (figure 5.1, panel a). About 20 percent of men and 19 percent of women who were working in January 2020 were either unemployed or out of the labor force (OLF) in April 2020, and the gender gap in labor force participation increased by 1.8 percentage points (pp). Urban men and women left the labor force at higher rates than in rural areas (figure 5.1, panel b), but unemployment increased more for both urban and rural men than for women (figure 5.1, panel c). This may be because men were more concentrated in highly affected sectors, such as construction. Accompanying this, there has been marked movement across sectors, particularly for men, from construction, manufacturing and traditional services to agriculture. In the year following the initial shock, labor force participation rates have returned to prepandemic levels, but unemployment for both men and women persists at higher rates. Oddly, April 2021 LFS figures show both the highest rate of female labor force participation in available data and the smallest gender gap. These advances come with the important caveat that the overall percentage of women employed was actually higher in January 2020 than it was in April 2021. FIGURE 5.1 Employment Trends, by Gender and Area, 2019 to 2021, Percent a. Labor Force Participation and Unemployment, by Gender b. Labor Force Participation, by Area and Gender c. Unemployment, by Area and Gender Source: Based on data from the 2019, 2020, and 2021 Labor Force Surveys and Philippine Statistics Authority (PSA) reports. July 2020 to April 2021 are based on information published by PSA. Information on labor force indicators by gender and area are not available in PSA publications. Note: LFP = labor force participation. 98 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic In the initial ECQ period of the COVID-19 pandemic, men and women had very different reasons for not working. Almost half of women not working in April 2020 were doing so because of housekeeping obligations, compared to only 4 percent of men (figure 5.2). Among married women, this figure was much higher at 77 percent, equal in both urban and rural areas. It also appears that women absorbed much of the initial increase in domestic burden. Although 89 percent of men who were newly not working (because they either became unemployed or left the labor market) in April 2020 cited ECQ/COVID-19 as the reason, only 1 percent left for household obligations (figure 5.3). Among women no longer working, 68 percent left because of ECQ/COVID-19 and 24 percent because of domestic obligations. Married women (36 percent) and women with young children (30 percent) had higher rates of leaving work for housekeeping than single women, further suggesting that much of the additional work of caring for family members and children at home was absorbed by women at the beginning of the pandemic. This finding is more pronounced in rural areas, where 42 percent of rural women with children up to two years old and 38 percent with children three to five years old left work for domestic reasons, compared to 18 percent of urban women with children up to 2 years old and 20 percent of urban women with children aged three to five years old. FIGURE 5.2 Reason for Not Working, by Gender FIGURE 5.3 Reason Not Working for those and Area, April 2020, Percent Working in the Previous Quarter, April 2020, Percent Source: Based on data from the 2020 Labor Force Survey. Note: Figure 5.2 shows reason for not working for all unemployed and OLF working-age population; Figure 5.3 shows reason for those who were employed in January 2020 and became unemployed/OLF in April 2020. ECQ = enhanced community quarantine; OLF = out of labor force. Data from the HOPE survey indicate that, among low-income households, the initial COVID-19 shock accompanied a large drop in employment, but there has been significant recovery. The employment ratio, defined as the share of adults 18 years and over who reported working at least one day in the previous seven days, declined from 56 percent to 31 percent from December 2019 to April 2020, a sharper drop than the 13-pp change in employment ratio seen in the LFS from October 2019 to April 2020.58 By August 2020, the employment ratio in low-income households had returned to 84 percent of its December 2019 level. Employment losses in April were particularly large in Luzon and in urban areas. As of August 2020, these areas were still experiencing slower recovery. The employment ratio in Luzon was only 75 percent of that in December 2019, but it was 87 percent of the December 2019 ratio in Visayas and 93 percent of the December 2019 ratio in Mindanao. Likewise, employment in urban areas (79 percent) lagged behind that in rural areas in August 2020 (87 percent). 58 LFS uses a slightly different definition of unemployment, considering adults aged 15 and above. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 99 CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic The HOPE survey did not find significant gendered impacts on the labor market for low-income households, with the slight changes in the gender composition of the workforce affecting men more than women. The short-term impact of COVID-19 and the ECQ appears to have disproportionally affected men, and later waves of the HOPE survey indicate results similar to the initial round. In December 2019, 64 percent of working adults in low-income households were men. This proportion declined to 57 percent in April 2020 as men lost employment at a higher rate, but the proportion returned to 62 percent in June 2020. The share of working women in households with young children (ages five and under) has slightly increased since the onset of the crisis, from 13 percent to 17 percent. As these surveys were conducted before the start of public school classes in October 2020, the potential gendered impact of child remote schooling supervision cannot be seen here. The HFS shows that, on average, women and men were equally affected by job loss in the immediate aftermath of the pandemic, but women in poorer groups continued to lag behind as the situation improved through the end of 2020. Results from the first round of the HFS show that 77 percent of men and 62 percent of women reported having worked in March 2020, but only 58 percent of men and 47 percent of women declared continuing to work in August 2020 (figure 5.4, panel a). On average, about 30 percent of both men and women experienced job losses between March and August. These figures are in line with estimates from LFS data for April 2020. However, job losses have not affected all income groups equally, with women losing more jobs than men at each quintile apart from the middle (figure 5.4, panel b).59 According to the second round of the HFS, 74 percent of men and 64 percent of women report having worked in December 2020, resulting in around a 16-pp increase in employment for both men and women since August. About 63 percent of men and 48 percent of women who were not working in August gained a job in December.60 Although these changes suggest that the employment situation has significantly improved through the end of 2020, some job loss persisted 16 percent of men and 18 percent of women who were working in both March and August were no longer working in December. Women in poorer groups were deeply affected by the persistence of job loss, as 27 percent of women in the poorest quintile who were working in March and August lost their jobs in December, compared to 15 percent of men in the same income group and to 15 percent of women in the richest quintile (figure 5.4, panel c). FIGURE 5.4 Change in Employment Status Between March 2020 and December 2020, Percent a. Share of Respondents Reporting to Work, by Round, Gender, and Quintile 59 The larger increase of job losses among women in poorer quintiles contrasts with results from HOPE surveys. This may be due to the prominence of women in these surveys, who are those most benefiting from government and other support programs for low-income families. 60 Changes in employment between August and December 2020 are regardless of whether or not respondents were employed in March 2020. Among those who were not working in both March and August of 2020, 47 percent of men and 42 percent of wom- en gained jobs by December 2020. 100 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic b. Proportion of People Who Lost their Jobs in c. Proportion of People Who Lost their Jobs in August 2020, by Gender and Quintile December 2020, by Gender and Quintile Source: Based on data from the 2020 Household High-Frequency Survey, Rounds 1 and 2. Note: Q = quintile. Urban areas were more affected by employment loss early in the pandemic but also recovered faster than rural areas. The first round of the HFS shows that about 60 percent of rural women and 64 percent of urban ones were working in March 2020, and the rates dropped to, respectively, 46 percent and 48 percent in August. During the same period, the proportion of working men fell from 80 percent to 63 percent in rural areas and from 75 percent to 53 percent in urban zones. Overall, urban men and rural women were the most affected by job losses in August (figure 5.5). By December, women in both urban and rural areas were more affected by job loss, as 15 percent of urban women and 21 percent of rural women reported no longer working after being employed from March to August, compared with 13 percent of urban men and 19 percent of rural men (figure 5.5). At the regional level, job losses in August were highest in Luzon (outside the National Capital Region), where 37 percent of women and 34 percent of men lost their jobs, and lowest in Mindanao, where 24 percent of women and 23 percent of men did so. The second HFS round suggests that job loss in Mindanao may have been more delayed than in other regions, because it reported the highest level of job loss in December among those who reported working in March and August; 33 percent of women and 35 percent of men in Mindanao reported not working in December, whereas in Visayas, which had the next highest job loss, the rates were 20 percent for women and 8 percent for men. FIGURE 5.5 Proportion of People who Lost their Jobs, by Gender, Area, and Round, 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey, Rounds 1 and 2. Large proportions of the population that lost work in August 2020 recovered jobs by December 2020, although women’s recovery has lagged behind that of men. To assess the recovery of employment lost in the initial shock of COVID-19’s onset, it is prudent to focus on those who were working in March 2020 but lost work by August 2020. Within this group, 74 percent of men and 56 percent of women reported working again in December 2020 (figure 5.6, panel a). Men reported regaining work at higher rates than women in all quintiles, with the middle quintile Overcoming Barriers To Women’s Economic Empowerment In The Philippines 101 CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic showing the highest gains for men and the richest quintile showing the highest gains for women (figure 5.6, panel a). A higher proportion of urban men and women recovered jobs in December than their rural counterparts (figure 5.6 panel b), although the gender gap in recovery was smaller in rural areas (18 pp) than in urban areas (21 pp). Women in households with no young children aged four or younger are more likely to have recovered work than women in households with children; among women who recovered jobs in December, 60 percent have no young children and 10 percent have two or more young children in the household (figure 5.6, panel c). Interestingly, a larger proportion of women with children in the household reported job recovery than did men with children, probably reflecting a need for women to work if men’s income is no longer sufficient to support a large family and if the child is not too young. However, women in households with two or more children aged four or younger are the least likely to have recovered jobs, which may speak to the larger childcare duties that women experience for young children. FIGURE 5.6 Recovery of Jobs Lost in August 2020, Percent a. Proportion of People Who Recovered Jobs Lost b. Proportion of People Who Recovered Jobs Lost in August, by Gender and Quintile, December in August, by Gender and Area, December 2020 2020 c. Proportion of People Who Recovered Jobs Lost in August, by Gender and Number of Children in the Household, December 2020 Source: Based on data from the 2020 Household High-Frequency Survey, Rounds 1 and 2. Note: Round 1 of the High Frequency Survey counts only children aged 5–18. Round 2 counts children aged 5–18 and 0–4. Both variables used here are from Round 2. Q = quintile. Many of the sectors hardest hit by the COVID-19 crisis are feminized, which explains the high job loss among women in some groups. Overall, construction, followed by transport, domestic services, and hospitality (accommodation and food services) are sectors that shed the most jobs during the initial shock of the pandemic. In contrast, information and communication technology (ICT) and financial services, public administration, and agriculture were less affected (figure 5.7).61 Among the industries hit hardest by the crisis, men are overrepresented in construction and transport 61 It is important to note that it remains unknown whether job losses will be temporary or permanent.  102 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic while women are overrepresented in domestic services and hospitality (figure C.1 in appendix C). Even though, on average, women and men lost jobs in August 2020 in similar proportions, there were large variations across industries, income groups, and geographic locations. In many industries, women lost their jobs at a disproportionately higher rate than men. For instance, in hospitality and trade services, where women account for the largest share of the workforce, they were also the most affected by the job cuts (figure C.2).62 At the geographic level, women in rural areas and in the rest of Luzon were more exposed than men to job losses in August 2020 because of women’s higher concentration in sectors cutting employment (for example, trade, hospitality, domestic and other services) (figures 5.8 and C.3). In urban areas, women were more protected by their relatively high employment in public administration and ICT services. Similarly, in poorer groups, most of women’s job losses occurred in hospitality, trade, domestic and other services, and, to a lesser extent, education. The overrepresentation of women in these sectors exposed them to employment cuts at a disproportionately higher rate than men (figure C.4). FIGURE 5.7 Proportion of Jobs Lost, by Sector, FIGURE 5.8 Employment, by Sector, Gender, and August 2020, Percent Area, August 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey, Round 1. Note: Public administration includes government and public services, administrative assistance, and education; ICT & finance includes ICT, financial insurance, banking and real estate; domestic includes employers of domestic personnel (such as maids, drivers), production of goods, and services; other industry includes mining and public utilities; other services includes unions, international organizations, art, and other services. ICT = information and communication technology. Women face elevated economic and health risks from the pandemic. Women made up about 70 percent of domestic workers in August 2020. Although the need for caregiving and cleaning services has increased, quarantine measures and lockdowns have made it difficult to maintain precrisis work arrangements, resulting in a loss of employment among this vulnerable female workforce. In urban areas, over 36 percent of women in domestic work had lost their job by August 2020, and this share increases to nearly 50 percent in the poorest income quintile. Women also made up about 72 percent of the health and social care workforce, and they are more likely to be front-line care providers, especially nurses, and community health workers, which raises their exposure to risk of infection. Although data on the number of people infected by sector are not available, the HFS Round 1 showed that 9 percent of women lost their jobs in health care and social work by August 2020 because they were infected or had to quarantine—compared to less than 1 percent of men. Additionally, the Philippines Overseas Employment Administration 62 HFS shows that even in some industries where women are underrepresented such as manufacturing, mining & public utilities and transport, they fell more victim to the crisis than men. However, LFS data does not support this finding.  Overcoming Barriers To Women’s Economic Empowerment In The Philippines 103 CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic suspended the deployment of overseas health care workers in early April 2020,63 leaving many such workers—a majority of whom were women—unemployed, although an increased demand for health services may have absorbed some of these women (Cho, Johnson, et al. 2021). Given the increased demand for health services and associated risk for health care workers during COVID-19, the Filipino Congress has proposed a minimum wage for nurses in private hospitals, who currently receive salaries well below the average rates of nurses in the public sector. Under the assumption that wage levels for nurses would subsequently rise in the private health sector, this change would overwhelmingly benefit women, who make up the majority of nurses (Cho, Doan, et al. 2021). Childcare and caregiving workloads have a stronger negative impact on women’s employment than on men’s. Although lockdown and quarantine measures were the main reason for not working for both women and men in August 2020, women seemed, disproportionately, to be not working because of caregiving and family duties. Much of this may be an indirect effect of COVID-19 leading more members of the household to staying home, increasing the domestic labor burden. Overall, 10 percent of women who stopped working in August did so because they had to take care of children (7 percent) or of a sick family member (3 percent). In comparison, only 3 percent of men who stopped working in August did so for this reason (figure 5.9). By December of 2020, lockdown/quarantine was a much less prevalent reason for not working, although it was still the primary reason cited by men (figure 5.10). In contrast, the primary reason for women not working in December was caring for children or an ill relative, at 33 percent (32 for childcare, 1 percent for caring for the ill). Although men were much less likely to cite this reason than women in December 2020, at 8 percent, they were slightly more likely to not work for this reason than in August 2020. More men than women indicated that time spent on childcare increased in December 2020 as compared with March 2020, at 29 and 26 percent, respectively. However, a higher percentage of women (28 percent) than men (23 percent) indicated that time spent on housework increased in the same period. FIGURE 5.9 Reason for Not Working, by Gender, FIGURE 5.10 Reason for Not Working, by Gender, August 2020, Percent December 2020, Percent Source: Based on data from the 2020 Household High- Source: Based on data from the 2020 Household High- Frequency Survey, Round 1. Frequency Survey, Round 2. Note: Business closed is unrelated to COVID-19; other Note: Student option was included in Round 2 but not in includes vacation, retired, and other unspecified reasons. Round 1. 63 The suspension was partially lifted in early December 2020. 104 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic Women in lower-income households and urban areas seem to be more negatively affected by the effect of COVID-19 on childcare. In August 2020, childcare and overall caregiving affected women in poorer groups (11 percent)—who cannot afford domestic help and likely have more dependents—more than those in better-off income groups (6 percent) (figure 5.11). Urban women were also more negatively affected by childcare (9 percent) than rural ones (5 percent). However, rural women were more affected than urban women by caregiving for ill relatives (4 percent versus 1 percent). This suggests that, although family ties and community support in rural areas help relieve some of women’s workload in terms of taking care of children, those same factors put more pressure in terms of other caregiving. Additionally, women in large families and in households with three or more children stopped working to take care of their children at a disproportionately higher rate than women in small families. For instance, 15 percent of women in households with three or more children and the same proportion in households with five or more members stopped working to take care of children or a sick relative.64 In comparison, only 3 percent of women in households without children stopped working for caregiving reasons. In December 2020, care for children and sick household members was the most-cited reason for not working among women in the poorest and richest quintiles, urban and rural women, and women in households with children65 (figure 5.12). As in August 2020, poorer women were more likely to cite this reason than richer women (40 percent versus 27 percent), whereas urban women were more likely than rural women to not work for this reason (37 percent versus 29 percent). Women in households with three or more children were particularly likely to not work for care reasons, at 44 percent. Overall, a higher proportion of urban women than rural women cited increases in time spent on housework (32 percent versus 24 percent, respectively) and childcare (32 percent versus 19 percent, respectively) in December 2020, as compared with March 2020 (pre-pandemic).66 FIGURE 5.11 Reason for Women Not Working, FIGURE 5.12 Reason for Women Not Working, by by Location, Income Group, and Family, August Location, Income Group, and Family, December 2020, Percent 2020, Percent Source: Based on data from the 2020 Household High- Source: Based on data from the 2020 Household High- Frequency Survey, Round 1. Frequency Survey, Round 2. Note: Business closed is unrelated to COVID-19; other Note: Student option was included in Round 2 but not includes vacation, retired, and other reasons. Round 1. 64 Round 1 of the survey asks for the number of total members and number of members aged 5–18 (considered children here). This limitation was addressed in the second round of HFS, where respondents were also asked about the number of members aged 0–4. 65 Aged 0–18, given additional questions in Round 2.  66 This urban/rural pattern is the same for men, with 26 percent of urban men reporting increases in housework in comparison with 21 percent or rural men, and 34 percent of urban men reporting an increase in childcare in comparison with 24 percent of rural men.  Overcoming Barriers To Women’s Economic Empowerment In The Philippines 105 CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic As school instruction has adapted to the limitations imposed by COVID-19, increased assistance with children’s education has compounded domestic burdens. In addition to regular childcare activities, household members have added overseeing distance learning to their childcare duties, in particular for young children; results from the second round of the HFS in December 2020 indicated that parents are the primary assistants in distance learning, at about 40 percent, followed by grandparents at 30 percent (World Bank Group 2021). Furthermore, women are much more likely to assist with distance learning, which may affect their economic activity. Female heads of household are 15 pp more likely than male heads of household to be the primary person assisting with distance learning, whereas male heads of household are 26 pp more likely to indicate that their wives usually assist with distance learning than female heads of household are to indicate the same of their husbands67 (figure 5.13). About 25 percent of both male and female heads of household indicate that their parent(s) usually help children with distanced learning, suggesting that having extended family members in the household is a huge benefit. Furthermore, women in poor households may experience an additional burden due to reliance on paper modules in households without internet. The HOPE survey finds that only about 10 percent of students ages six to eight years old were able to use paper modules on their own, requiring an average three to four hours of supervision per day. FIGURE 5.13 Primary Assistant with Distance Learning, by Sex of Household Head, December 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey, Round 2. Although long-term effects of COVID-19 on children’s education are impossible to determine given the pandemic’s ongoing nature, it is likely that school closures and learning loss during the pandemic will have a notable negative impact on educational outcomes of the current cohort of schoolchildren. Children who are enrolled face many challenges to effective learning under the current distance learning modality. There is no indication, however, of a gendered impact on learning losses from the closure of schools in Philippines, nor is there a reason to expect a disproportionately negative long-term impact on girls, given that female educational attainment in the Philippines is higher than male educational attainment and girls typically have better learning outcomes (see chapter 2). Rather, the barriers to effective learning reflect challenges of inequality in access to gadgets and family member support for remote learning. According to the second round of the HFS, 22 percent of households listed a lack of access to gadgets as the primary barrier to children’s learning, with an inability to focus without adult supervision and insufficient internet as the next most prevalent primary barriers (figure 5.14). In addition to these access and modality concerns, other barriers cited frequently were family members’ inability to assist with learning because they lack knowledge about online classes and subject material. Differences in learning loss will most likely be related to income levels, because distance schooling poses concerns in particular for students from low-income households who may have disproportionate access problems (World Bank Group 2021). Although access to internet at home is a concern for many households—only 46 percent of households reported having access to internet at home 67 Given that same-sex marriage is not legal in the Philippines, we can assume that spouses of male heads of household are female, and vice versa.  106 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic in August 2020 and 55 percent in December 2020—there are stark differences across income groups. Despite increased internet access for all income groups from August to December, poorer quintiles reported significantly lower levels of access than richer quintiles (figure 5.15). All told, the extent of the pandemic’s impact on educational outcomes will depend on learning recovery efforts, and those outcomes require monitoring moving forward. FIGURE 5.14 Primary Barriers to Children Learning Effectively, December 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey, Round 2. FIGURE 5.15 Access to Internet in the Home, by Quintile and Round, 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey, Rounds 1-2. Despite the negative impacts of the pandemic on women’s economic activities, there is reason for optimism because of emerging opportunities to work from home. The first round of the HFS found that a larger share of women who remained employed in August 2020—36 percent, compared to 26 percent of men—were able to work from home. By December 2020, 35 percent of women who were employed were able to work from home in comparison with 19 percent of men. However, it is worth noting that, by December 2020, 42 percent of women and 59 percent of men were able to work as usual from the office, an uneven increase from August, when only 33 percent of both men and women could work from the office as usual. Women are employed in services at higher rates than men and in particular make up a majority (54 percent) of business process outsourcing (BPO) workers. BPO companies, which employed 1.3 million people as of 2019, Overcoming Barriers To Women’s Economic Empowerment In The Philippines 107 CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic are an example of an industry that has adopted new work-from-home (WFH) arrangements during the lockdown (Navarro 2020). This increased flexibility and elimination of long commutes not only supports existing working women but also may benefit women who remain out of the labor force because of family obligations. Their higher levels of education may allow women to benefit significantly from both new and growing WFH arrangements. In August of 2020, women with high school degrees or above reported higher rates of WFH than women without high school degrees and men of all educational attainment levels; however, although women were still able to WFH at higher rates than men in December 2020, the educational trend had reversed for both sexes, with those without high school degrees working from home at higher rates than those with high school degrees (figure 5.16). In further support of WFH, there is a proposed bill in the Philippine House of Representatives to amend the Telecommuting Act (Republic Act 11165) in order to (1) cover public and government sector workers, and (2) require employers to provide a monthly stipend for telecommuting employees to cover related expenses (Cho, Doan, et al. 2021). Additional legislation has been proposed that would encourage more flexible work schedules, such as a 35-hour work week. Although these legislative proposals would benefit all employees engaged in WFH, they would be particularly crucial for women, given disproportionate domestic duties. FIGURE 5.16 Proportion of Employed People Working from Home, by Gender, 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey, Rounds 1 and 2. In addition to WFH, the rise of e-commerce is a promising area of growth and women’s economic participation in the Philippines. A 2021 report by the International Finance Corporation, Women and E-commerce in Southeast Asia, analyzes gender-disaggregated data from Indonesian and Filipino sellers that utilize Lazada, an online platform, as well as interviews with sellers in both countries and e-commerce experts (IFC 2021). The report notes the massive growth in the e-commerce sector in Southeast Asia, with the value tripling to $105 billion in gross merchandise value (GMV) between 2015 and 2020. However, the Philippines’ digital economy has the lowest value in the region at $7.5 billion in 2020, in part because of slow and expensive connectivity.68 Overall, the report notes that closing gender gaps in e-commerce could add more than $280 billion to the value of the sector in Southeast Asia from 2025 to 2030. The report also finds that two-thirds of the Filipino sellers in Lazada were women-owned businesses, and 70 percent of these were microenterprises, compared with 60 percent of male-owned Filipino businesses in Lazada. Judging from the prevalence of women entrepreneurs—and women-owned microenterprises—on a platform like Lazada, e-commerce represents an area for potential growth in women’s economic participation in the Philippines. Furthermore, the analysis by Beylis et al. (2020) of job transformations in Latin America during COVID-19 argues that ICT advances in recent years have allowed more service sectors to become digitalized, while service jobs across the region are increasing in all sectors, and in manufacturing in particular. Given women’s prevalence in service sectors in the Philippines, job increases and growing digitalization of this sector could especially benefit them. 68 Compared to Indonesia’s $44 billion in the same year. 108 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic Women-owned businesses in the Philippines seem to enter into the e-commerce setting on unequal footing in comparison to men-owned businesses. The Women and E-commerce in Southeast Asia report notes that women-owned businesses generally face numerous disadvantages: they tend to have fewer employees on average; they are concentrated in lower-value sectors including agriculture, food processing, catering, and beauty; and they are underresourced and may need to rely on multiple selling platforms (IFC 2021). The report’s analysis of Filipino businesses using Lazada shows that women-owned enterprises are more likely than male-owned ones to have been started with personal savings as opposed to a formal loan or private investments, and women- owned businesses are more likely than men-owned businesses to use other social media platforms (Facebook, WhatsApp, Instagram) to sell in addition to Lazada (72 percent of women-owned businesses versus 65 percent of men-owned businesses). Despite the disadvantages facing women-owned businesses in the Philippines, e-commerce presents unique economic opportunities for these entrepreneurs. IFC (2021) report finds that e-commerce is particularly beneficial for female entrepreneurs in the Philippines and Southeast Asia for several reasons. First, e-commerce provides women-owned businesses with access to new and growing online markets. Forty-four percent of women-owned Filipino businesses on Lazada indicated that the reason for joining the platform was to sell online and grow their business— compared to 29 percent of male-owned businesses; 47 percent of Filipino men-owned businesses joined to start a new business, in comparison to 40 percent of Filipino women-owned businesses. Second, online platforms provide women-owned businesses with additional resources for growth; women-owned businesses in Lazada were more likely than men-owned ones to benefit from platform financing and to indicate that they valued Lazada’s training and businesses support. Third, e-commerce may allow women-owned businesses to break into traditionally male sectors. Since 2019, women-owned businesses have dominated sales in the electronics category on Lazada, representing the platform’s largest share of total sales and indicating that e-commerce may facilitate sector switching with more ease. Finally, and perhaps most crucially in the context of COVID-19, selling on e-commerce platforms provides women entrepreneurs with flexibility in work hours, location, and activities. Among Lazada’s sellers, a larger share of women-owned enterprises (52 percent) than men-owned businesses (41 percent) reported that a key benefit of selling online was a flexible work schedule. Women sellers were also more likely than men sellers to list fulfilling personal goals, meeting basic needs, and enjoying time with family and friends as benefits of e-commerce. This suggests that e-commerce is compatible with care work, which would largely benefit women entrepreneurs in the Philippines. Overall, these beneficial aspects of e-commerce indicate that the sector is a promising area for increasing women’s economic participation and future work in the Philippines. COVID-19 has had disproportionately negative impacts on the merchandise value of women’s business in the e-commerce sector. Women and E-commerce in Southeast Asia’s analysis of Lazada shows that, in the last two quarters of 2019 (before the onset of the pandemic), women sellers were substantially outselling men in the Philippines, with an average GMV that was 106 percent of the average GMV of men-owned businesses. However, by the last two quarters of 2020, women’s average GMV was 79 percent of men’s, representing a 27-pp decrease in average GMV. This change was mostly due to a decline in women-owned businesses’ GMV in the electronics and general merchandise sectors on Lazada, which are higher value than other sectors on the platform. Despite a decrease in GMV during the pandemic, Lazada has seen an increase in the share of businesses that are women-owned, from 60 to 66 percent. Both men and women sellers have also branched out into new selling categories on Lazada, with the shares of men-owned and women-owned businesses that sell in more than one product category increasing by 10 pp and 8 pp, respectively, from the end of 2019 to the end of 2020. This finding may indicate that entrepreneurs have Overcoming Barriers To Women’s Economic Empowerment In The Philippines 109 CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic turned to product diversification as a coping mechanism during COVID-19, which is perhaps better facilitated by e-commerce platforms. Although COVID-19 appears to have caused a setback for women-owned businesses in selling online, prepandemic trends show that the Filipino women can be quite successful in the e-commerce sector. II. Effect of the Pandemic on Women’s Income Men who continued working have seen declines in their incomes at higher rates than women for the most part, especially among those in poor households. Income losses have been widespread, with a majority of men and women experiencing either reduced income or loss of income in August 2020: 44 percent of men reported having a decline in income and 14 percent reported receiving no income, compared to 42 percent and 11 percent of women, respectively. Although losses in income were less drastic in December 2020, 29 percent of men experienced a loss in income and 8 percent received no income, whereas 32 percent of women experienced a loss of income and 6 percent received no income. Men in the lowest quintile were most likely to experience income losses in August 2020, with 26 percent earning no income and 40 percent seeing a decline in their income (figure 5.17, panel a). Women in the first and second quintiles similarly had the highest rates of income losses in August 2020, though at a slightly lower rate than men (figure 5.17, panel b). Although men and women in the highest earning quintiles experienced high levels of income losses in the immediate shock of the pandemic, they did so at lower rates than those in lower income groups. Among the low-income households in the HOPE survey, incomes have not recovered to prepandemic levels. The immediate shock in April 2020 saw median weekly household earnings drop from just below PhP350 to under Php50. Earnings increased in the second and third waves, but the August 2020 median weekly household earnings of roughly Php175 represent half of prior income levels. FIGURE 5.17 Loss of Income, by Quintile, August 2020, Percent a. Male b. Female Source: Based on data from the 2020 Household High-Frequency Survey, Round 1. Note: Q = quintile. Rural areas experienced higher rates of income losses in the initial shock of the pandemic, whereas urban areas experienced higher rates later on. Within urban areas, men and women saw very similar rates of income losses in August 2020. In contrast, men fared worse than women in rural areas in the same period, with 17 percent reporting having worked but receiving no income, and another 46 percent having worked but earning less income (compared to 10 percent and 45 percent of women, respectively) (figure 5.18, panel a). Rural women also fared worse than urban women in August 2020, with a 6-pp difference in respondents earning less income. Filipinos in all areas of the country were affected by the pandemic early on, completely losing their income or earning less in August 2020, but at varying levels, ranging from 49 percent in the National Capital Region and its surrounding regions to 64 percent in the rest of Luzon. Men were more affected 110 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic than women in all areas but the rest of Luzon. By December 2020, however, urban areas reported higher levels of income loss among both men and women (figure 5.18, panel b). Urban women reported the highest rate of decreased income at 34 percent, while rural men reported the highest rate of no income, at 9 percent. FIGURE 5.18 Loss of Income, by Gender and Urban/Rural, 2020, Percent a. August 2020 b. December 2020 Source: Based on data from the 2020 Household High-Frequency Survey, Rounds 1 and 2. No industry has been spared, but the income losses of workers vary considerably both across and within industries. More than 75 percent of men and 72 percent of women working in agriculture experienced a decrease or loss in income in August 2020, which helps explain why income losses were higher in rural areas (figure 5.19). Some industries were also more affected than others; over 66 percent of those employed in construction, hospitality, transport, and trade services reported income declines or losses, whereas, among those employed in health and social services, public administration, and ICT and finance, under 35 percent reported income losses. Some industries also saw a large gender gap in income losses; in manufacturing, 58 percent of women reporting having income losses relative to 27 percent of men, but in public administration, the share of individuals who had income losses was 15 pp lower for women. The sample size of December 2020 is too small for a detailed analysis by sector. FIGURE 5.19 Loss of Income, by Industry and Gender, August 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey, Round 1. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 111 CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic The COVID-19 pandemic has also had a substantial and negative impact on family-run businesses. About 64 percent of non-farm family businesses reported experiencing a decline in revenues in August 2020 compared to February of the same year, and about 65 percent of farm businesses saw a decrease in revenue from January to July 2020 compared to the previous year (figure 5.20, panel a). Data from the HOPE survey show that family businesses in low-income households were also greatly affected by the initial wave of COVID-19; the share of households with family businesses declined by 41 percent; among the businesses that remained open, incomes fell by 65 percent. Farm businesses in female-headed households also had wide variation in incomes in August 2020, with 31 percent reporting having no income and 12 percent reporting increased income (figure 5.20, panel b). Data from the HFS community survey provides further evidence that farming households are struggling, with 27 percent of community representatives stating in August 2020 that farmers need more support. It is also worth noting that these surveys took place before multiple natural disasters that occurred in the fourth quarter of 2020, which have affected Filipino farmers even more. Although the proportions of households that reported lower income from family-run businesses were lower in December than in August, they were still substantial, with 53 percent of non-farm family businesses and 42 percent of farm businesses reporting lower income in comparison to revenues in August 2020. FIGURE 5.20 Income Changes for Non-farm and Farm Businesses compared to Previous Period, 2019 to August 2020, Percent a. All Businesses b. By Gender of Head of Household Source: Based on data from the 2020 Household High-Frequency Survey, Round 1. Note: Non-farm businesses are compared to income in February 2020, and farm businesses are compared to income in January to July 2019. Remittance income has not increased substantially in response to the economic challenges of the pandemic and has instead fallen, particularly for male-headed households. A larger share of female-headed households typically receives remittances relative to male-headed households, with 25 percent of female-headed households and 18 percent of male-headed households receiving remittances in August 2020. By December 2020, the proportion of households receiving remittances increased for both sexes of household head; 29 percent of female-headed households and 21 percent of male-headed households reported receiving remittances. The source of these remittances has remained similar between male and female-headed households, with about half coming from other cities in the Philippines and a bit less than half from other countries. Despite changes in the proportion of households receiving remittances, the pandemic has not led to a countercyclical increase in the amount of remittances to support struggling households; only 2 percent of female-headed households and 1 percent of male-headed households reported an increase in remittances in August 2020, with no meaningful difference between international and domestic remittances. Instead, a large percentage of households saw reduced remittances, with almost two-thirds (65 percent) of male-headed households and 57 percent of female-headed 112 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic households receiving either no remittances or lower-than-usual remittances. By December 2020, 8 percent of male-headed households and 2 percent of female-headed households reported an increase in remittances, whereas 49 percent of male-headed households and 41 percent of female- headed households reported decreases in remittances compared to the usual level. III. Coping Mechanisms and Food Security Reducing food consumption has been a common mechanism to manage the shock from the COVID-19 pandemic, but men and women have otherwise adopted different coping strategies. Over 80 percent of households coped with COVID-19 in August 2020 by reducing food consumption or shifting to cheaper food items (figure 5.21). Female- and male-headed households, however, exhibited slightly different patterns. Female-headed households reduced non-food consumption at a greater rate in August 2020 (a 12-pp difference), whereas male-headed households were substantially more likely to engage in additional income-generating activities (a 10-pp difference). This pattern may be because males were expected to earn additional income as breadwinners while many women have taken on additional domestic obligations. It is important to note that, although more women in poorer quintiles and those with children in the household tended to reduce their food and non-food consumption than women in better-off groups and those with no children in the household, there were no significant gender differences across these groups meaning that poorer women or those with children in the household did not seem to have reduced their consumption more than their male counterparts. Data from the second round of the HFS shows that, although the incidence of decreasing food consumption or shifting to cheaper food items lowered slightly in December 2020 as compared with August 2020, it was still quite high, at 70 percent of households. FIGURE 5.21 Coping Mechanisms, by Gender of Household Head, August 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey, Round 1. In August 2020, poorer households were more likely to rely on government assistance to cope with the pandemic, whereas richer households were more likely to rely on savings. Although households across all income quintiles relied heavily on reducing food consumption, government assistance was the first and second most common coping mechanism among households in the two poorest quintiles (figure 5.22). Borrowing from family and friends was much more common in the poorest quintile compared to the richest quintile (64 percent compared to 38 percent, respectively). The use of savings increased in higher income quintiles and was most common in the top quintile, where 64 percent of households used this as a coping mechanism. Although these differences across quintiles speak to varying levels of wealth and the need for social protection across income groups, it is worth noting that the majority of households in all quintiles have Overcoming Barriers To Women’s Economic Empowerment In The Philippines 113 CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic reduced non-food and food consumption, delayed payments, and benefited from government assistance during the pandemic. It is also worth noting, however, that data from the HOPE survey show that the use of coping mechanisms among low-income households such as taking out loans and reducing food and non-food consumption has increased significantly as restrictions have been lifted, but the incidence of emergency social assistance as a coping mechanism has decreased. FIGURE 5.22 Coping Mechanisms, by Quintile, August 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey, Round 1. Purchasing staple foods has been difficult for poorer households. Households in the lowest two income quintiles experienced the greatest difficulty purchasing staple food in August 2020, with 70 percent of all households citing the lack of financial resources as the primary reason for being unable to purchase essential food.69 The pattern of difficulties of purchasing food differs when examining households by gender. A larger share of female-headed households in the second quintile than in the poorest quintile was unable to purchase food (69 percent compared to 54 percent), whereas 63 percent and 64 percent of male-headed households in the poorest two quintiles were unable to purchase these goods (figure 5.23). This was consistent in December 2020, with 60 percent of female-headed households and 54 percent of male-headed households in the second quintile reporting an inability to purchase staple foods, compared with 62 percent of female-headed households and 66 percent of male-headed households in the poorest quintile. This matches the pattern of job and income losses in which women in the second quintile have suffered higher rates of income and job losses than those in the poorest quintile. FIGURE 5.23 Inability to Purchase Food, by Gender of Household Head, August 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey, Round 1. Note: A household was considered unable to purchase food if it could not purchase either (1) rice or (2) meat, fish, or eggs or (3) fruits or vegetables. A substantial proportion of Filipinos experienced food insecurity due to COVID-19, with the poorest households being the most affected. Male- and female-headed households have experienced similar levels of food insecurity due to the pandemic. Data from the HFS show that 69 Staple foods include rice, fruits or vegetables, and meat, fish, and eggs. 114 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic 74 percent of male-headed households and 73 percent of female-headed households had at least one member worried about not having enough food to eat in August 2020 (figure 5.24, panel a). Moreover, 17 percent of male-headed households and 13 percent of female-headed households reported having a household member who went without eating for a whole day because of a lack of resources in August. Although indicators of food security decreased in December 2020, female- headed households reported higher rates of concern about not having enough to eat than male- headed households, at 62 percent and 58 percent, respectively (figure 5.24, panel a). In August, food insecurity was particularly high among those in households in the two poorest quintiles, with 37 percent and 36 percent of households in these quintiles having at least one member who experienced hunger but did not eat (Figure 5.24B). The rate of food insecurity decreases in richer quintiles, but even in the richest quintile, 21 percent of households reported having at least one family member who did not eat despite being hungry. According to the HOPE survey, poorer households experienced a large food security shock in April 2020, with 56 percent of low- income households reporting at least one household member reducing meals in the past seven days. The presence of food insecurity among low-income households was greater for those with three or more children. In later waves of the HOPE survey, indicators of food security improved. Even with overall improvement in food security as of August 2020, some low-income households— particularly in Luzon and at the lower end of the earnings distribution—were more likely to report worsening food security. FIGURE 5.24 Food Insecurity, August and December 2020, Percent a. By Gender of Household Head and Round b. By Quintile, August 2020 Source: Based on data from the 2020 Household High-Frequency Survey, Rounds 1 and 2. Note: Q = quintile. Women with children were more affected by food insecurity and financial anxiety than men were. Women and men in the two poorest quintiles were similarly affected by food insecurity in August 2020; however, the proportion of women with children in the household who felt food Overcoming Barriers To Women’s Economic Empowerment In The Philippines 115 CHAPTER 5 | Gender Inequality and the COVID-19 Pandemic insecure was significantly larger than that of men (figure 5.25). Interestingly, we note that the proportion of respondents who reported being hungry and not having eaten and those who went without eating a whole day was higher among men than women, whether they have children in the household or not. This may be due to the fact that women have less food needs than men or because they feel less the sacrifices they make for their children. About 87 percent of men and 89 percent of women were worried about their financial situation in August 2020. The rates increase to 93 and 94 percent, respectively, in the two poorest quintiles. However, among those who have children, 95 of women were worried, compared to 89 percent of men. FIGURE 5.25 Food Insecurity, by Gender, Income, and Number of Children in the Household, August 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey, Round 1. Community surveys support the finding that food security is a significant issue due to the pandemic. According to community representatives surveyed in August 2020, 30 percent believed food supply is a major concern and 20 percent noted the lack of access to markets, food, and essentials as an important problem, in comparison with pre-COVID-19 levels of 17 percent and 6 percent, respectively (figure 5.26). Other prominent concerns were economic and income opportunities (60 percent) and lack of medical supplies and personal protective equipment (24 percent). Community representatives also noted that vulnerable groups are particularly in need of assistance. A majority (59 percent) cited the elderly as needing more support during this time and 36 percent indicated that persons with disabilities need more assistance (figure 5.27). FIGURE 5.26 Top Three Pressing Problems, 2019 FIGURE 5.27 Groups that Need More Assistance, and August 2020, Percent August 2020, Percent Source: Based on data from the 2020 Community High-Frequency Survey, Round 1. Note: 4Ps = Pantawid Pamilyang Pilipino Program; PPE = personal protective equipment. 116 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 6 Women, Care Work, Market Work, and the Pandemic: Some Preliminary Findings from a Rapid Survey Overcoming Barriers To Women’s Economic Empowerment In The Philippines 117 CHAPTER 6 | Women, Care Work, Market Work, and the Pandemic: Some Preliminary Findings from a Rapid Survey The previous chapters show that, although women fare better than men in education, and despite progress in the Philippines on economic and human development, women’s participation in the labor market remains very low. Among factors that seem to constrain women’s labor supply, marriage, childcare, income security, and social norms seem to have a significant effect; and educational advancement is positively associated with women’s participation in the labor force. When they work, women tend to cluster in low- skilled/low-paying and in high-skilled/high-paying jobs—as opposed to medium positions, where men are more clustered—but concentration in high-skilled positions tends to be much lower for women with young children. Early findings suggest that the COVID-19 crisis led to significant job and income losses for both men and women, but married women and those with children in their household were more affected by the indirect effects of the pandemic through increased pressure for engaging in unpaid work, such as childcare and housework, than by the economic fallout of the crisis. This chapter attempts to understand people’s perception about care work, childcare services, and time use in the Philippines while the COVID-19 crisis continues unfolding. It presents preliminary findings from a rapid survey conducted in May 2021 (box 6.1). BOX 6.1 About the Survey The survey covers a sample of 1,200 adults aged 18 years and older (50 percent women and 50 percent men, 53 percent urban and 47 percent rural) and is representative at the national and urban-rural levels. Fieldwork was conducted from April 25 to May 8, 2021 using face-to-face (tablet-assisted) interviews (83 percent) and telephone (computer-assisted) interviews (17 percent). The questionnaire includes five modules on the socio-demographic characteristics of the respondent and the respondent’s household, employment and income, domestic obligations and time use, perception about childcare, and attitudes toward women’s work. It was prepared in consultation with the National Economic and Development Authority. Profile of respondents The average number of household members of respondents is 4.9, varying from 2.0 to 14.0 (on average, 5.0 in urban areas and 4.8 in rural areas). About 75 percent of respondents (national, urban/rural, and by gender) have at least one member aged four years and below or five to nine years in their household. The education profile is quite similar among men and women (about 70 percent of each have a high school diploma and above), but more men than women have postsecondary technical education (7 percent versus 3 percent). About 63 percent of men and 26 percent of women were working during the week preceding the interview, and 64 and 23 percent, respectively, were working before the first quarantine. The proportion of men currently working is higher in rural areas (66 percent versus 61 percent in urban areas), but the opposite is true for women (30 percent in urban versus 23 percent in rural areas). More women in low-income households (27 percent) and with higher education (29 percent) are working than women in better-off households (25 percent) and with lower education (22 percent). The proportion of women who are not currently working is significantly higher among women who have young children in their household; 31 percent of women who have no young children in their household are working, compared with 21 percent of those who 118 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 6 | Women, Care Work, Market Work, and the Pandemic: Some Preliminary Findings from a Rapid Survey BOX 6.1 About the Survey have at least a child aged four or younger and 24 percent of those who have at least one child aged five to nine. Over 80 percent of men stopped working because of business closure or quarantine measures compared with 51 percent of women; about 30 percent of women (and 35 percent of those who have children in their household) stopped working to take care of children. Among respondents currently working, about 82 percent of women work in the services sector (56 percent in traditional services and 26 percent in modern services) and 60 percent of men work in agriculture and industry (27 percent in agriculture, 6 percent in manufacturing, and 27 percent in construction and other industries); 32 percent work of men in traditional services and 8 percent in modern ones.a About 25 percent of women and 17 percent of men are self-employed, and about 11 percent of both men and women are in unpaid family business. About 43 percent of men and 37 percent of women are wage employees in government or private establishments. a. Traditional services include wholesale & retail trade, food & accommodation, transportation, and employers of domestic personnel. Modern services include public administration, ICT, finance, education and human development. I. Care Work and Time Use Mothers are the primary caregivers to children in the household, but fathers do play an important role. The number of primary caregivers for children in the household varies from one to four and is similar in urban and rural households. About 31 percent of respondents name the mother as the primary caregiver, and 28 percent indicate that both the mother and the father hold the role (figure 6.1). Primary care is provided by family members, with a marginal role given to babysitters and domestic helpers. More female respondents (59 percent) indicate that they play the primary role in caregiving than male respondents (38 percent) (figure 6.2). Notably, the proportion of working women indicating they are primary caregivers is higher than the proportion of nonworking ones— 62 percent versus 57 percent. FIGURE 6.1 Relation to Child of Primary FIGURE 6.2 Respondents’ Role in Caregiving to Caregiver, May 2021, Percent Children, May 2021, Percent Source: Women work and childcare survey, 2021. Note: Female relatives include sisters, grandmothers, and aunts. Male relatives include brothers, grandfathers, and uncles. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 119 CHAPTER 6 | Women, Care Work, Market Work, and the Pandemic: Some Preliminary Findings from a Rapid Survey Women spend much more time on domestic and caregiving activities than men and much less time in economic activities. Among working respondents, men report spending an average of 33 hours per week in their job while women report 26 hours; the number of hours increases to 34 for men with young children aged four years and below and to 28 hours for women with children in this age range. This suggests that young children may constrain women’s decision to work but not necessarily the number of hours they spend in their jobs. Women spend on average 30 percent more hours per week than men on domestic activities and childcare and about two times more time taking care of elderly, sick, and disabled household members (figure 6.3). Working women also spend much more time on domestic and caregiving activities than working men, though they spend less time on these activities than nonworking women. Although men recognize that their partners spend more time than they do on domestic activities, they tend to underestimate the time spent by their partners on caregiving for children and elderly, ill, and disabled members. More women than men consider that the number of hours they spend on domestic work and childcare has increased since the beginning of the quarantine, and the gender gap in this perceived increase is higher among those living in households with young children; about 54 percent of women indicate they are spending more time on domestic activities and 64 percent indicate they are spending more time on childcare compared with, respectively, 49 and 56 percent of men. However, a much larger proportion of men than women reports that time spent on caregiving for elderly and ill members has increased since the quarantine (figure 6.4). FIGURE 6.3 Number of Hours per Week Spent in FIGURE 6.4 Perceived Increase in Time Use on Different Activities, May 2021 Caregiving during COVID-19, Percent Source: Women work and childcare survey, 2021. Women would spend more time in work and on skills development if they could devote less time to domestic activities and childcare, whereas men would spend more time socializing, although there are important variations across education and income groups. On average, about 33 percent of women indicate there are activities they are unable to do because of domestic work, and 28 percent report they are unable to do other activities because of childcare; the proportions among men are, respectively, 31 and 26 percent. On average, more women than men would spend time working (34 percent versus 26 percent) and developing their skills (10 percent versus 7 percent); more men than women would spend time socializing (46 percent versus 23 percent) if they had to work less on domestic activities. However, there are large variations among women depending on their geographic location, education, and income group. Women in rural areas, lower income groups, and lower education levels would spend more time working and developing their skills, whereas urban women and those in higher income groups would spend more time on their own wellness or socializing (figure 6.5). Similarly, rural, low-education, and low-income women would spend more time working and developing their skills than urban, better-educated, 120 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 6 | Women, Care Work, Market Work, and the Pandemic: Some Preliminary Findings from a Rapid Survey and higher-income women if they could spend less time on childcare. However, although a higher proportion of men than women would spend more time socializing if they did not have to care for children, more men than women would also spend more time working (figure 6.6). This supports earlier findings that men are involved in caregiving for children and that childcare may not be a major obstacle to women’s work. FIGURE 6.5 Activities Unable to Do because of FIGURE 6.6 Activities Unable to Do because of Domestic Work, Percent Childcare, Percent Source: Women work and childcare survey, 2021. Note: Low-income groups are respondents whose household monthly income is PhP10,000 or less, and low-education groups are high school undergraduates and below. Groups with children are respondents living in households where there are children aged four years and below. II. Childcare Arrangements Willingness to use childcare services is limited, and reluctance appears to come primarily from women, particularly in low-income households. Over 95 percent of both men and women believe that childcare should be provided by family members and that the cost of childcare services should be covered by family members. This supports earlier findings from the 2012 International Social Survey Programme (ISSP) survey discussed in chapter 3. Women appear to be the primary decision- makers on childcare arrangements; 76 percent of female respondents designate themselves as the primary decision makers, and 47 percent of male respondents designate their spouses (figure 6.7). About 44 percent of women and 40 percent of men are not willing to pay any money for high-quality childcare services. Reluctance to pay for childcare services is highest among women in low-income households, but a majority of women in high-income households also exhibit strong reluctance, with 62 percent indicating either an unwillingness to pay or a willingness to pay only up to PhP500 per week, compared to 50 percent of men in the same income group (figure 6.8). FIGURE 6.7 Decision-Maker on Childcare, FIGURE 6.8 Amount Willing to Pay per Week for High- Percent Quality Childcare, Percent Source: Women work and childcare survey, 2021. Note: HH = household; PhP = Philippine peso. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 121 CHAPTER 6 | Women, Care Work, Market Work, and the Pandemic: Some Preliminary Findings from a Rapid Survey Family remains the primary source of childcare. About 78 percent of respondents in households with children aged four years and below indicate that none of the children are enrolled in childcare facilities. The rates reach 85 percent in urban areas and 82 percent in high-income households. Both men and women cite the child’s being too young as the main reason for not being enrolled in a childcare facility, followed by closure due to the pandemic; women cite high cost as the third main reason, and men cite the lack of availability of childcare services (figure 6.9). The median age at which families enroll children in childcare facilities is four; enrolled children are slightly younger in rural areas than urban ones. About 20 percent of those who have children aged four or below enrolled in childcare facilities indicate receiving financial support for enrolling their children; the proportions are slightly higher among women in low-income households, at about 22 percent. This financial support is provided mainly by local governments. No matter the gender, location, or income group, family at home remains the primary resort for childcare, used by over 80 percent of respondents. After family at home, family nearby is the most prevalent form of childcare; less than 5 percent of households use public childcare services, and less than 1 percent use private ones (figure 6.10). FIGURE 6.9 Reason Why Children Ages 0-4 Years Are FIGURE 6.10 Primary Method of Childcare, Not Enrolled in Childcare, Percent Percent Source: Women work and childcare survey, 2021. Note: In figure 6.9, other reasons include operating hours, absence of birth certificate for the child, and inability to enroll. In figure 6.10, family & others include neighbors and helpers. Higher access to childcare services may not lead to a significant increase in women’s labor force participation. About 30 percent or less of nonworking women perceive their current childcare arrangements as a constraint to their working outside the home (figure 6.11, panel a). Among those feeling constrained, only about 31 percent think that increased access to childcare facilities will increase their participation in the labor market and 33 percent think it would help them engage in part-time work; 14 percent prefer not to work (figure 6.11, panel b). The proportion of women who prefer not to work is higher among richer households, which supports earlier findings that higher income security deters women’s labor supply. The prevalence of preferring not to work is also higher among women with better education, probably due to the fact that the proportion of working women is higher in this group in comparison to less educated women, and women who are not working are those who choose to do so. No matter the location, work status, education, or income group, the vast majority of women think the best work-childcare arrangements are for 122 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 6 | Women, Care Work, Market Work, and the Pandemic: Some Preliminary Findings from a Rapid Survey women to stay home and take care of children; the proportions vary from 74 percent among urban women to 82 percent among women with low education (figure 6.11, panel c). FIGURE 6.11 Women’s Opinions on Childcare Arrangements and Women’s Work, Percent a. Do current childcare arrangements act as a b. Would more access to childcare outside the home constraint to women’s work outside the home? help women join the labor market? c. What are the best work-childcare arrangements? Source: Women work and childcare survey, 2021. Note: Question in panel a is for nonworking women, question in panel b is for nonworking women constrained by childcare arrangements, and question in panel c is for all women. III. Attitudes toward Women’s Work Beliefs that children suffer when their mother works may dissuade women’s labor force participation. About 76 percent of women and 70 percent of men agree that the emotional and psychosocial development skills of children aged four years and below may be affected when their mother works outside the home. The rates reach 79 percent among rural women and those in higher-income households. These findings support—and proportions are even higher than—results from the 2019 World Values Survey discussed in chapter 3. Smaller proportions of men and women agree that the emotional and psychosocial development skills of a child aged five to nine years may be affected when his or her mother works outside the home, but proportions still exceed 70 percent among women. Similarly, more than 80 percent of men and women agree that a mother’s work outside the home affects the school performance of her child, and this belief is more prevalent among women in urban areas and in low education groups. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 123 CHAPTER 6 | Women, Care Work, Market Work, and the Pandemic: Some Preliminary Findings from a Rapid Survey Despite consensus that women should contribute to the income of the family, the contribution of their work to society is not well or widely understood yet. Over 75 percent of men and over 80 percent of women agree that a man’s job is to earn money and a woman’s job is to take care of the family and home, supporting the findings in chapter 3 about the persistent perception of men as breadwinners and of women as caregivers. However, 74 percent (71 percent of men and 76 percent of women) also agree that it is important and sometimes necessary for women to contribute to the income of the household. Over 50 percent of respondents agree that women contribute better to society and family by having a job rather than staying home. Agreement is highest among low-income households (figures 6.12 and 6.13). However, over 70 percent also agree that women contribute better to society by staying home to take care of family and children rather than working outside the home (figure 6.14). This suggests that the contribution of women’s work to society is not well understood or valued properly by a significant proportion of the population. FIGURE 6.12 Opinion on Whether Women FIGURE 6.13 Opinion on Whether Women Contribute Better to Society by Having a Job, Contribute Better to Family by Having a Job, Percent Percent Source: Women work and childcare survey, 2021. There are opportunities to encourage a more positive perception of women’s work outside the home among an important share of the population. Figure 6.15 shows that about 15 percent of respondents agree that women contribute better to society by staying home and disagree with the statement that women contribute better to society by having a job. This group is convinced that women should not work and would be difficult to convince otherwise. This group is composed mostly of rural, low-educated respondents. Another 20 percent agree that women contribute better to society by staying home and neither agree nor disagree with the idea that women contribute better to society by working. This group would tend to have a negative perception about women’s work and would also be difficult to convince to the contrary. Although less than 10 percent agree that women contribute better by working and disagree that they contribute better by staying home (that is, those fully convinced women should work instead of staying home), about 50 percent agree that women contribute better by working and also either agree that women contribute better by staying home or do not have an opinion on women staying home; the latter group could probably be convinced to be more supportive toward women working outside the home. This is further supported by the fact that over 75 percent of respondents agree that, if both parents are working, they should split domestic duties equally. 124 Overcoming Barriers To Women’s Economic Empowerment In The Philippines CHAPTER 6 | Women, Care Work, Market Work, and the Pandemic: Some Preliminary Findings from a Rapid Survey FIGURE 6.14 Opinion on Whether Women FIGURE 6.15 Hesitancy about the Contribution of Contribute Better to Society by Staying Home, Women’s Work to Society, Percent Percent Source: Women work and childcare survey, 2021. Perceptions about returning to work after a break to take care of children are mixed. Over 70 percent of respondents believe that part-time work for pay is available in the labor market. About 41 percent of women and 38 percent of men agree that it is easy for a mother to return to work after leaving the labor market to take care of children, and the proportion of those agreeing is slightly higher in urban areas and among groups with higher education and income. However, a large proportion of respondents see childcare, family obligations, and expectations of the child as major obstacles to a mother’s return to work (figure 6.16). No matter the location, gender, education, or income group of the household, over 35 percent of respondents believe family obligations are the main obstacle to a mother’s return to work, followed by a lack of childcare (over 20 percent) and expectations of the child (over 10 percent). About 8 percent of men and 14 percent of women see a gap in work experience as the main obstacle, with a slightly higher proportion among educated women than those with low education, and 18 percent of men and 13 percent of women cite low wages as the main obstacle. FIGURE 6.16 Main Obstacles to a Mother’s to Return to Work, Percent Source: Women work and childcare survey, 2021. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 125 CHAPTER 6 | Women, Care Work, Market Work, and the Pandemic: Some Preliminary Findings from a Rapid Survey Women tend to replicate their childhood parental model in organizing their adult work-family life when there is a child under school age. About 61 percent of women and 56 percent of men think that the mother should stay home and the father should work full-time (figure 6.17). This situation is not much different from the way respondents’ parents organized their life, though the proportion of men whose mother stayed home is slightly higher than that for women (Figure 6.18). Urban women and women with higher education are more likely to favor a work-life arrangement where the mother works part-time and the father works full-time. Less than 10 percent of respondents had both parents who worked full-time, and less than 9 percent of them think that both parents should work full-time. This indicates that Filipinos tend to replicate their parental model and that women tend to be even more conservative than their parents and than men. FIGURE 6.17 Opinion on Best Way to Organize FIGURE 6.18 Respondents’ Parents’ Family-Work Family-Work Life When There Is a Child Under Life When They Had a Child Under School Age, School Age, Percent Percent Source: Women work and childcare survey, 2021. 126 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Executive Summary Conclusion Overcoming Barriers To Women’s Economic Empowerment In The Philippines 127 Conclusion This report examines gender gaps in the Philippines in education, participation in the labor market, and employment and earnings, and additionally attempts to shed light on some of the factors that drive these gaps, with a focus on the role of childcare and social norms. Despite the Philippines’ progress in economic and human development, more than half of working-age women remain out of the labor force, and this situation has persisted over the past three decades. The findings of this report show that, although women outperform men in educational achievement and attainment, and despite a supportive legal framework for gender equality, the gender gap in labor force participation remains high, at over 27 percentage points. Lower women’s labor force participation represents a missed opportunity for economic growth and increased prosperity in the Philippines; a mere increase of women’s labor supply by 0.5 percentage points per year would increase GDP per capita by about 6 percent by 2040 and almost 10 percent by 2050, bringing the country closer to fulfilling its aspiration to achieve upper-middle-income status, according to our calculations. Fully understanding the barriers to women’s labor force participation is a daunting task, because several factors on the supply and demand sides are at play and often mutually reinforce each other. However, some important findings emerge from the analysis in this report. First, family circumstances (education of the household head, presence of a man with salaried employment, presence of an overseas working household member, and the number of young children in the household) strongly influence women’s labor supply. Although improving family circumstances through a reduced birth rate and increased ability to afford more help for domestic and childcare duties would raise women’s participation, many women seem to be working primarily out of economic necessity; increased income security of women’s households may actually depress their labor force participation. Education helps to advance women’s labor force participation, but social norms may partly offset its benefits. Advancing education would affect women’s participation positively, but evidence suggests that high returns to labor force participation occur only at the college level and above. Married women and those with young children have the lowest participation rates in the labor market, and they tend to invoke family duties as the main reason for staying out of the labor force. However, they seem to resist using childcare services outside the home and appear to strongly believe that their primary duty is to take care of children and family. Women in these groups also report fearing that the sociopsychological skills and school performance of their children will be affected if they work outside home, and they often perpetuate the parental model in which they were raised. Overall, gender norms and social stereotypes seem to inhibit women— particularly married ones—from taking a job outside the home. Although they tend to value the contribution of women’s work to society, both Filipino men and women continue to privilege the traditional family model where the father works full-time and the mother stays at home to take care of family and children. When they work, women appear to perform better than men on average, working at higher occupational and pay levels. However, this trend hides large discrepancies among women, who tend to cluster either in low-profile/low-pay occupations or in high-profile/high-pay occupations. Education remains the main factor advantaging them over men in terms of better occupations and higher returns at the upper level of the earnings distribution. At the lower level of the earnings distribution, however, they remain disadvantaged compared to men because of their inability to take up some jobs that offer more adequate returns to their skills. In part, women’s earning disadvantage at the lower end of the distribution may be due to their desire to work in jobs that offer more flexible hours and work arrangements, as well as to social stigmas that deem certain types of jobs to be men’s work. 128 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Conclusion COVID-19 presents some setbacks for women’s labor force participation, but it has also induced some promising changes that could benefit women’s work in the future. Although men were most affected by unemployment in the months immediately preceding the arrival of COVID-19 in the Philippines, rates of job loss were higher for women by the end of 2020. Despite some job recovery, women are regaining jobs lost during the pandemic at lower rates than men. In particular, the pandemic seems to have negatively affected the ability of women with young children in the home to return to work, which speaks to an increased domestic burden during the pandemic that includes the addition of overseeing children’s distance learning. Women are also much more likely than men to report joblessness due to caring for children and sick or elderly household members. However, the pandemic has also highlighted the potential of work-from-home opportunities and e-commerce, which both provide work flexibility that may be particularly helpful for women as they balance domestic and economic activities. Policy Considerations Although a more complete policy paper is forthcoming, the following are several suggested areas for shaping policies to support women’s labor force participation and economic activity. 1. Decrease time spent on domestic activities. Policies that aim to reduce the amount of time spent on domestic labor could free up time for women to increase participation in the labor market. Several examples of policies or trends related to time spent on domestic work that influenced rises in married women’s labor force participation are cited in the literature (Jayachandran 2021): (1) the expansion of electricity in South Africa after apartheid, which allowed for more households to engage in activities such as switching from wood-burning to electric stoves; (2) Indonesia’s Conversion to Liquified Petroleum Gas program, which resulted in households switching to labor-saving cooking fuel; and (3) an increase in household ownership of appliances in the United States. 2. Provide alternatives to childcare in the home. Much of the literature suggests that increased access to childcare allows mothers to participate in the labor market. Jayachandran (2021) points to increased availability of formal and informal childcare in Mexico City, a lottery for free childcare in Rio de Janeiro, increased access to subsidized day care in Nicaragua, vouchers for subsidized day care in Nairobi, and the expansion of public preschools in Indonesia as examples. However, it is important to caution that research cited in this report suggests that many households in the Philippines are reluctant to use childcare outside of the home, and willingness to pay appears to be low. As such, increasing childcare options alone is likely insufficient to raise female labor force participation in the Philippines. 3. Address gendered social norms that affect women’s participation in the labor market. Building on the previous recommendation, social norms that appear to contribute to depressed female labor force participation in the Philippines include a preference for mothers and family members to care for children (as opposed to formal childcare outside the home) and the belief in women’s role as the caregiver and men’s role as the breadwinner. Klasen et al. (2021) also support the sentiment that social norms play an important role; they note that, although factors such as women’s education and decreased fertility may positively influence women’s labor force participation, other barriers to women working outside the home exist in households and labor markets and are often tied to social norms. While addressing social norms themselves with policies is challenging, Jayachandran (2021) points to interventions Overcoming Barriers To Women’s Economic Empowerment In The Philippines 129 Conclusion such as media campaigns, behavior change communication, and attitude change interventions as possible solutions to changing opinions about masculinity, gender in the workplace, and gender roles. More specifically, the author cites (1) an intervention in India where families were shown informational videos aimed at improving familial support for women’s work outside of the home; (2) a study in Saudi Arabia that showed that men tended to overestimate other men’s opposition to women’s work and female students tended to underestimate their peers’ intentions to work but that both groups showed changes in opinion and behavior when presented with the correct statistics; and (3) a project in India where a nongovernmental organization held regular classroom discussions on gender roles and discrimination in schools, leading to increased support for gender equality and more gender-progressive behavior. In the Philippines, mixed results on the perceived contribution of women to society suggest that a significant portion of the population currently believes women play important roles in both the home and the workforce—this group could perhaps be convinced to support women’s labor force participation even when there are young children in the home. 4. Support more flexible work arrangements, particularly remote work models. The COVID-19 pandemic severely disrupted established work patterns, accelerating the adoption of remote work across various sectors of the economy. Findings from this report suggest that arrangements which allow women to work from home show promise in the Philippines, as it provides flexibility that is compatible with care work. In addition to the flexibility it provides, remote work—whether it be fully remote, partially remote or a hybrid model— also reduces the amount of time workers spend on the road commuting to and from work, freeing up more time for productive use. This is particularly useful for workers commuting within or into Metro Manila, one of the most congested cities in developing Asia (ADB 2019). The Philippine House of Representatives’ proposed amendment of the Telecommuting Act (Republic Act 11165) to expand WFH capabilities and support would be important step forward in this regard. To further support the ability of women to work remotely, internet connectivity would also need to be strengthened—these include expanding network coverage, improving the speed and quality of internet connectivity as well as decreasing the cost of internet access. 5. Strengthen support for women entrepreneurs, particularly in e-commerce. Along with the adoption of remote working arrangements, findings from this report also suggest that e-commerce could be a valuable tool in increasing labor force participation among women. Engaging in e-commerce allows women to set their work location and their work schedule, allowing them to grow their business while engaging in domestic activities and childcare. With Filipinos spending more time online than anywhere else in the world, averaging more than 9 hours per day among 16 to 64 year olds (We are Social 2020), and with the internet economy in the country still at a low level relative to peers, there is ample room for the industry to grow. This provides a window of opportunity for the emergence of more female entrepreneurs, with data from the Philippines showing that over two-thirds of sellers in e-commerce platform Lazada are women (IFC 2021). Support could take the form of basic skills training, which could include financial literacy, marketing and inventory management or expanding access to credit, particularly for women entrepreneurs engaged in e-commerce. Data from the sample of entrepreneurs on Lazada suggest that these could be highly impactful, given the strong reliance of women entrepreneurs on using personal savings as startup financing as well as the value that women place on the training and business support provided by the e-commerce platform. 130 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Conclusion 6. Support networking among working women and exposure to role models. Programs that enable women to network with other women, and potentially men, could help erode restrictive gendered norms and create a virtuous cycle to promote women’s economic activity. Networking among women can notably be valuable for female entrepreneurs. Similarly, exposure to counter- stereotypical role models can influence girls’ and women’s gender stereotypes and shift cultural norms; a study in Chile where business trainings for women added either mentorship by another female entrepreneur who became successful in her business or individually tailored consulting on how to improve business resulted in higher profitability for women entrepreneurs (Jayachandran 2021, citing La Fortune, Riutort, and Tessada 2018). Furthermore, e-commerce platforms could serve as important fora for such networking opportunities. 7. Scale efforts in reskilling and upskilling women, particularly in fields that have high growth potential. The disruptions caused by the pandemic has shifted how businesses operate, accelerating the adoption and raising the share of businesses which offer remote working arrangements. While this presents an important opportunity for women who would otherwise be unable to join the labor market due to domestic obligations, as well as women who were employed in low-skilled occupations and lost their jobs during the pandemic, these women may not be able to take advantage of this opportunity if they lack the skills or the knowledge required to benefit from this shift. Scaling efforts to provide women the opportunity to acquire new skills in fields such as ICT, and broadly, STEM-related fields, could help them find and secure more productive work in the new normal. This could also aid in a more inclusive recovery as the country begins to recover from the effects of the pandemic. It is worth noting that the Technical Education and Skills Development Authority (TESDA) already provides skills training programs in non-traditional trades as well as STEM-related fields, however, women comprised only 36 percent of graduates from these training courses in 2020.70 While these efforts are laudable, there is room to encourage more women to join these skills training programs. 8. Encourage firms to expand opportunities for women who wish to re-enter the labor force. Results from the Women work and childcare survey reveal that less than half of women agree that it is easy for a mother to return to work after leaving the labor market, with 14 percent of women viewing the gap in work experience as the main obstacle in returning to the labor force. This could be partly due to negative biases or perceptions that employers associate with extended gaps in work experience. Returnship programs, similar to the more common internship programs, typically provide women who wish to re-enter the labor force with skills training, mentorship as well as the potential to get hired after the training period (PwC 2016). While it is more common in other countries, companies in the Philippines could be encouraged to adopt this model given that it provides a channel for women to return to the workforce even after a period of absence. These efforts can also be complemented by services which help match women with returnship opportunities that are suited to their strengths, skills and previous work experience. Helping women re-enter the labor force could help break the persistence of gendered social norms, particularly as findings from this report reveal that Filipinos tend to replicate the parental model that they grew up with. As such, these efforts could act as a mechanism to improve women’s labor force participation for future generations by reinforcing the belief that taking a break from work should not become a major hindrance to women’s future career prospects. 70  Information based on PCW data. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 131 Conclusion 9. Address gender wage gaps at the bottom of the income distribution. As shown in this report, there are key differences in men’s and women’s wages at the bottom of the income distribution that largely result from the inability of women, notably those married and those with young children, to engage in jobs that offer higher returns. Policies aiming at supporting women in these groups to engage in higher-paying jobs (through encouraging hiring of women in certain industries, encouraging employers to offer flexible work arrangements, addressing social stigma associated with certain types of “male-dominated” jobs, and so forth) could help reduce gender inequalities in the labor market. 10. There is a need for greater data on childcare services. Data are limited on childcare service enrollment and quality—particularly for community-based and home-based facilities, nonprofit- run day care, and private nursery schools—making it difficult to connect childcare access and quality to labor market outcomes. 11. Although this report focuses on the interaction between women’s economic opportunities and specific areas such as education, employment, and childcare, there are additional factors pertinent to women’s work in the Philippines that merit research and policy focus. For example, the gender dynamics of migration and remittances from overseas foreign workers may affect the domestic burden and labor force participation of remaining household members. Additionally, the effects of recent legislation improving wages and rights of domestic helpers (kasambahay) need to be monitored for potential unintended consequences on female labor force participation, because demand for such help in lower-income households may decrease. Finally, analyses of the gender distribution across industries with the highest shares of employment—including high-skilled jobs in business process outsourcing and call centers— and of the gender distribution within manufacturing as compared with other countries in Southeast Asia merit further research. 132 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Appendices Overcoming Barriers To Women’s Economic Empowerment In The Philippines 133 Appendices Appendix A. The Effect of Increased Female Labor Force Participation on Long-Term Economic Growth in the Philippines This analysis was prepared by Steven Pennings (DECMG).71 It estimates the effect of higher female labor force participation (FLFP) on economic growth in the Philippines over the next 30 years, using the [standard] long-term growth model (LTGM).72 The LTGM is based on the celebrated Solow-Swan growth model, but adapted to developing countries, and solvable in a spreadsheet. Assumptions: Baseline FLFP is assumed to be constant at its 2020 value of 45 percent (based on Labor Force Survey data) over 2020–50, but in the scenario FLFP is assumed to increase by 0.5 percentage points (pp) per year to reach 60 percent by 2050 (figure A.1). This still leaves FLFP 12 pp below the LFP rate of males, which we assume is constant at its 2020 rate of 72 percent. The calibration of the rest of the LTGM follows that in World Bank (2018, chapter 5).73 Results: The increase in FLFP boosts gross domestic product (GDP) growth initially by 0.22 pp, rising to 0.35 pp by 2050 (figure A.2; the gain in GDP per capita [GDPPC] growth is the same). While these gains are small in annual percentage terms, they quickly accumulate over time. Figure A.3 shows that, in the scenario, GDPPC would be almost 6 percent higher than the baseline by 2040, and almost 10 percent higher by 2050. This translates into an increase in GDPPC by US$500 (real 2010 US dollars) by 2040 and US$1,325 by 2050. The percentage increases in total and per capita GDP levels are almost identical, such that total GDP is US$70 billion higher by 2040 and US$190 billion higher by 2050 (figure A.4, also real 2010 US dollars). The total cumulative dollar gains (adding up each year since 2020) are more than US$500 billion by 2040 and more than US$1.8 trillion by 2050 (figure A.5). To put those gains in context, that is around US$4,250 and US$12,500 (in 2040 and 2050, respectively) for each Filipino. FIGURE A.1 Female Labor Force Participation FIGURE A.2 Increment to GDP Growth from Higher FLFP, Percentage Points (Scenario less Baseline) Source: World Bank. Note: FLFP = female labor force participation. 71 Email: spennings@worldbank.org. The views expressed here are those of the author and are not necessarily those of the World Bank, its executive directors or the countries they represent.  72 The model is available for download from the www.worldbank.org/LTGM. 73 Growth and Productivity in the Philippines: Winning the Future. Most important, I assume that the labor share=0.5. The depreciation rate is 3.6 percent, capital-to-output ratio is 2.2, human capital growth rate is 0.5 percent, total factor productivity growth rate is 1.5 percent, and investment rate trends up from 26 percent of GDP initially to 30.1 percent by 2040. Population demographics are taken from the United Nations. Baseline GDP growth is initially about 5.5 percent, but slows to 4.5 percent by 2050. The male LFP rate is assumed constant at 72 percent. 134 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Appendices Mechanisms: The increase in growth is due to a direct effect through more labor and indirect effect through capital productivity, representing the two terms in equation (A.1) (where Δ_sb is the difference of scenario less baseline): First, the direct effect is that of higher FLFP on the number of workers producing GDP in equation (A.1)).74 Quantitatively, this term contributes about 0.2 pp to the boost to growth, with a slight downward trend over time. Analytically, its size depends on the importance of labor in producing output, which is governed by the size of the labor share =0.5 in the default calibration). Note that there is only one “type” of labor in the LTGM, so the productivity (and human capital) of new workers is the same as that of existing workers. The size of the effect also depends how big the shock to FLFP is—the size of the change in the growth rate of the participation rate ). With male LFP constant (and female participation constant in the baseline), the change in the growth rate of the participation rate depends on the increase in the female participation rate in pp (0.5 pp per year) relative to the total participation rate . Over time, the participation rate trends up slowly in the scenario (due to higher FLFP), from 0.586 initially to 0.66, which is why the contribution of this term falls slowly. Second is an indirect effect (second term in equation (A.1)) as higher labor supply makes capital more productive (more workers for each machine), which raises the effect of investment on growth. Quantitatively, the size of this effect is initially close to zero, as the marginal product of capital (MPK) is initially identical across baseline and scenario. But as the MPK increases in the scenario, the size of the indirect effect increases to 0.15 pp by 2050. As the investment ratio and capital share are the same in baseline and scenario, the dynamics are driven by the capital-to-output ratio (K/Y). Increasing FLFP in the scenario raises output and reduces the K/Y slowly over time, which increases the marginal product of capital in the scenario. FIGURE A.3 Increment to GDPPC Level from FIGURE A.4 Increment to Total GDP Level from Higher FLFP Higher FLFP Source: World Bank. Note: Figures show Scenario less Baseline in indicated year. FLFP = female labor force participation. 74 Higher in the production function , where each worker is assumed to have identical productivity. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 135 Appendices FIGURE A.5 Cumulative Gains to GDP from Higher FLFP Source: World Bank. Note: Figure shows Scenario less Baseline since 2020 (Real 2010 US dollars). FLFP = female labor force participation; GDPPC = GDP per capita. Robustness: In general, the size of the effect of higher FLFP on growth is quite robust to different model parameterizations, with the increment to growth generally lying between 0.2 pp and 0.4 pp. The increase in growth is most sensitive to how labor-intensive production is, which is governed by the labor share , and determines the strength of the direct effect above. In countries where GDP is more labor-intensive, the direct effect will be larger, shifting up the effect on growth for all horizons, but with a bigger effect in the short term (figure A.6). For example, a typical calibration for developed countries is =2/3, which would increase the size of the effect on growth by about 0.07 pp in the short term but by 0.02 pp in the long term. Likewise, reducing the labor share to =0.4, reduces the short-term effect by 0.04 pp, but the long-run effect by 0.02 pp. The reduced effect in the long term is because a lower labor share means a higher capital share, which increases the strength of the indirect effect through capital. The size of the effect of FLFP is fairly robust to other growth fundamentals like human capital growth, TFP growth or the investment rate (figure A.7). Despite these growth fundamentals having a large effect on the growth rate in the baseline and scenario, that is largely differenced out and leaving little effect on the size of the increment to growth from higher FLFP. Specifically, they have no impact in the short run (direct mechanism) and only a small effect through the indirect mechanism in the long run.75 75 However, if the investment rate is zero, the size of the indirect effect in equation (A.1) and figure A.2 also goes to zero. 136 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Appendices FIGURE A.6 Robustness 1: Extra GDP Growth FIGURE A.7 Robustness 2: Extra GDP Growth from Higher FLFP, Percentage Points from Higher FLFP, Percentage Points Source: World Bank. Note: Figures show Scenario less Baseline. FLFP = female labor force participation; TFP = total factor productivity. Comparison with 2018 McKinsey Global Institute (MGI) Report (Woetzel et al. 2018): The MGI report argues that, by 2025, GDP in the Philippines would be 7 percentage points (pp) higher than business as usual in an “improving gender equality to the best-in-region scenario” (Woetzel et al. 2018, exhibit 37). However, of this 7 pp, only 4 pp (58 percent × 7 pp in exhibit 37) is actually due to increasing FLFP.76 As such, the 4-pp increase in GDP in the MGI report could be compared with the increase in this note. But we need to calibrate the size of the increase in FLFP (that generates this 4-pp increase in GDP) to produce an apples-to-apples comparison. The MGI report assumes an 11-pp increase in FLFP participation from 50 percent to 61 percent over 2014–25 (Woetzel et al. 2018, exhibit 37, footnote 1), which is roughly double the pace of the FLFP scenario considered here (which is already very fast). As such, the size of the increase in FLFP in the MGI report is roughly equivalent to that considered here over 2020–40 (increasing FLFP 45 percent to 55 percent in figure A.1). By 2040, figure A.4 suggests GDP is about 6 percent above the baseline with higher FLFP. As such, the effect of FLFP on growth of 6 pp here is arguably stronger than the 4-pp increase in the MGI report. Why might this be case? The methodologies are quite different, and thus difficult to compare.77 On the one hand, the MGI report assumes that women enter the labor force with the average productivity of existing female workers based on the subsectors in which women work (which is presumably lower than for men). In contrast, the model in this note assumes that women and men are equally productive. On the other hand, the MGI report does not use a production function approach and so abstracts from “any drag on productivity due to changes in the supply of labor relative to capital,” which is equivalent to setting =1 in equation (A.1). These two factors can be expected to be partially offsetting.78 76 The remaining 3 pp are due to higher productivity from assuming a partial convergence of sectoral productivity by gender, which is a somewhat different exercise. 77 The MGI report multiplies LFP by a labor productivity term (and other adjustments) to get GDP (Woetzel et al. 2018, appendix). The labor productivity term is a weighted average of the labor productivity in different subsectors where men and women work. 78 In addition, the assumed rates of male LFP might be different (and follow a different path). The effect of β<1 only reduces the effect of higher FLFP on GDP in the medium term. In the very long term, it has no effect (assuming a constant long run K/Y). Overcoming Barriers To Women’s Economic Empowerment In The Philippines 137 Appendices Decomposing the Effect of Higher FLFP on Growth As the working-age population of males and females is usually similar, the participation rate is approximately the average of the participation rates of the two genders (note that we assume no unemployment in the LTGM): Assuming that the participation rate of males is constant over the time, ( is the change over time), the growth rate of the participation rate can be expressed as The log linear approximation of growth rate in the LTGM is Taking the difference of scenario less baseline ( ), all right-hand side terms except and cancel out. Substituting yields equation (A.1) in the main text: Analytically, the size of the direct effect pp initially, falling to pp.79 The second term is initially close to zero as the baseline and scenario have the same MPK. But by 2050, the MPK is 0.5 pp higher in the scenario (16.25 percent) than the baseline (15.75 percent). Multiplying by I/Y=30 percent, yields an increment to growth of 0.15 pp, similar to that in figure A.2. 79 The same increase in FLFP will generate a larger effect on growth in countries with a lower initial total participation rate. For example, if the initial participation rate was only 0.4, the direct effect would initially generate 0.31ppts of growth. 138 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Appendices Appendix B. Unconditional Quantile Decomposition of the Gender Pay Gap The recentered influence function (RIF) regression approach (Firpo, Fortin, and Lemieux 2009 2018) provides a simple regression-based procedure for performing a detailed decomposition of different distributional statistics such as quantiles, variance, and Gini coefficient. The RIF-regression model is called unconditional quantile regression when applied to the quantiles (percentiles, deciles, and so forth). The technique consists of decomposing the pay gap between men and women at various quantiles of the unconditional distribution into differences in endowment characteristics such as education, age, occupational level, and so forth (also called wage composition effects) and differences in the returns to these characteristics (called returns or wage structure effects). These components are then further decomposed to identify the specific attributes that contribute to the gender pay gap. The procedure is carried out in three steps. The first step consists in estimating the unconditional quantile regressions on log daily wages for men and women. The second step serves to estimate a counterfactual wage distribution for women—that is, the wage structure that would have been realized among women if they were paid the same returns for their labor market attributes and personal characteristics as men. The third step involves the comparison of the counterfactual and empirical distributions to estimate the part of the pay gap that is only attributable to differences in labor market endowments and characteristics between men and women and the part only explained by differences in returns to those characteristics. The endowment and return components can be further divided into the contribution of each specific characteristic variable. The method can be easily implemented as a standard linear regression, and an ordinary least squares (OLS) regression of the following form can be estimated: where is log daily wage, and is the RIF of the th quantile of estimated by computing the sample quantile and estimating the density of at that point by kernel methods: where is the marginal density function of and is an indicator function. is the regressors matrix including the intercept, is the regression coefficient vector, and ε is the error term. The regressors include seven groups of variables: (1) demographic characteristics, which include age, a categorical variable for marital status (single, married, widowed, and divorced/separated), and number of members aged 0–2 ¬years, 3–5, 6–8, 9–11, and 12–14; (2) education (six categories): no education, elementary, incomplete high school, high school graduate, postsecondary/some college, and college graduate and above; (3) employment sectors (six categories): agriculture, industry, construction, wholesale & retail trade, public administration and human development, and other services; (4) class of employment (four categories): private household, private establishment, Overcoming Barriers To Women’s Economic Empowerment In The Philippines 139 Appendices government employee, and family business with pay80; (5) employment nature (three categories): permanent job, seasonal job, and different employers on daily or weekly basis, a dummy for underemployed, and three categories for occupation level: high skilled, medium, and low skilled; (6) childcare assistance: dummy variable for whether children aged 0–4 are currently in childcare facility/kindergarten, whether the family benefits from domestic help, and number of extended members aged 10–60 years old; (7) geographic location fixed effects that capture external factors to the individual: urban, conflict zone, region exposed to climate hazards, Luzon region, Visayas, and Mindanao. Model (1) is estimated for the 10th to 90th quantiles and uses the below decomposition in which traditional Oaxaca-Blinder (OB) decompositions are applied to the wage distribution by percentile: where and are the th unconditional quantiles of log daily wage for men and women respectively, and the vectors of sample averages of characteristics, and the estimates of the unconditional quantile partial effect. is the counterfactual quantile representing the distribution of wage that would have prevailed for women if they received the same returns for their characteristics as men. The first term on the right-hand side of equation (B.2) represents the contribution of the differences in characteristics to the gender pay gap at the th unconditional quantile, or endowment effect. The second term of the right-hand side of the equation represents the gender pay gap due to differences (or discrimination) in returns to those characteristics at the th unconditional quantile. The endowment and return effects can be further decomposed into the contribution of individual specific characteristics (or group of some characteristics) as follows: where designates the individual specific household characteristics. Equation (B.1) is based on the standard linearity assumption between the dependent variable and the covariates used in the OB decomposition. When the linearity assumption does not hold, the model can lead to estimation errors (Firpo, Fortin, and Lemieux 2018; Fortin, Lemieux, and Firpo 2010). The problem can be addressed by using a reweighted regression approach and the reweighted-regression decomposition of the overall pay gap can be specified as follows: 80 Only classes with wage employees were considered; self-employed, employers, and unpaid family workers were excluded. Also, workers paid based on commissions or other nonregular/nonconventional ways were excluded.  140 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Appendices Where The composition effect in equation (B.4a) can be divided into a pure composition or endowment effect (first term in equation (B.4b)) and a component linking to the specification error in the linear model (second term in equation (B.4b)): Similarly, the structural effect in equation (B.4a) can be divided into a pure structural or returns effect (first term in equation (B.4c)) and a reweighting error component (second term in equation (B.4c)): If the model was truly linear, the specification error term would be equal to zero, as both the weighted and unweighted regressions would yield the same consistent estimates, where . Computing the specification error is thus important for checking whether the linear model is well specified, and for adjusting the composition effect in the case where the linear specification is inaccurate. When the reweighting factor is consistently estimated The reweighting factor is a simple function of that can be easily estimated using standard methods such as a logit or probit. Consider the dichotomous variable indicating the gender of wage workers: for men and for women. The reweighting factor can be expressed as Overcoming Barriers To Women’s Economic Empowerment In The Philippines 141 Appendices The conditional probabilities and can be estimated using a logit or probit specification and then used to estimate The decomposition of the gender pay gap in the reweighted model (as expressed in equation (B.4)) proceeds in the following steps: estimate the reweighted factor , compute the counterfactual quantiles using the reweighted wage distribution of men, and decompose gender pay gap into pure composition and structure effects as well as specification and reweighting errors at each selected quantile of the wage distribution. Another problem in the gender pay gap decomposition concerns the choice of the omitted group or base group for categorical covariates. For example, when considering the sectors of work, the underrepresentation of women in the industry sector would lead to significantly different returns effects when agriculture is considered as the omitted group compared to when services is the omitted one; in the first case, the underrepresentation of women in industry is priced at the relative returns in industry versus the agricultural sector, and in the other it is priced at the relative returns in industry versus the services sector. We address the problem by normalizing the coefficients of the categorical variables (sectors, employment class etc.) using the approach of (Firpo, Fortin, and Lemieux 2007, 2018; Rios-Avila 2019). Results for the reweighted model using a logit specification are presented in table 4.1 of chapter 4; results for the nonreweighted (linear) model and the reweighted model using a probit specification are in figures B.1 to B.4. Figures B.1 and B.2 compare the overall endowment effects obtained by a linear model and by reweighting. The pure endowment effect in the reweighted model shows a very similar pattern to the endowment effect displayed by the linear model. The specification error term is relatively small except for the 7th and 9th deciles. However, both models show different patterns for the returns effects. This means that the RIF-regression model allows a consistent estimation of the endowment effects, but the reweighting model allows a more accurate and robust estimation of the returns effects. FIGURE B.1 Aggregate Effects, Linear Model, FIGURE B.2 Aggregate Effects, Probit 2020 Reweighting Model, 2020 a. Aggregate Endowments Effect a. Pure Endowments and Specification Error Effects .4 .4 Difference in log daily wage Difference in log daily wage .2 .2 0 0 -.2 -.2 -.4 -.4 -.6 -.6 .1 .2 .3 .4 .5 .6 .7 .8 .9 .1 .2 .3 .4 .5 .6 .7 .8 .9 Quantiles Quantiles Total wage gap Pure Endowments Effect Total wage gap Total Endowments Effect Total Endowments Effect Specification Error 142 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Appendices b. Aggregate Returns Effect b. Pure Returns and Reweighting Error Effects .4 .4 Difference in log daily wage Difference in log daily wage .2 .2 0 0 -.2 -.2 -.4 -.4 .1 .2 .3 .4 .5 .6 .7 .8 .9 .1 .2 .3 .4 .5 .6 .7 .8 .9 Quantiles Quantiles Total wage gap Pure Returns Effect Total wage gap Total Returns Effect Total Returns Effect Reweighting error Source: Based on data from the 2020 Labor Force Survey. Source: Based on data from the 2020 Labor Force Survey. Figures B.3 and B.4 divide the endowment and returns effects into the contribution of five main groups of covariates.81 As for the aggregate effects, the reweighted model seems to better track the returns effects. FIGURE B.3 Detailed Effects, Linear Model, 2020 FIGURE B.4 Detailed Effects, Probit Reweighting Model, 2020 a. Detailed Endowments Effects a. Detailed Pure Endowments Effects .2 .2 Difference in log daily wage Difference in log daily wage .1 .1 0 0 -.1 -.1 -.2 -.2 -.3 -.3 .1 .2 .3 .4 .5 .6 .7 .8 .9 .1 .2 .3 .4 .5 .6 .7 .8 .9 Quantiles Quantiles Education Employment class & occupation Education Employment class & occupation Demographic characteristics Childcare support Demographic characteristics Childcare support Geographic location Geographic location 81 For simplicity the of the presentation the employment sectors, employment nature and occupational levels were combined in the same group. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 143 Appendices b. Detailed Returns Effects b. Detailed Pure Returns Effects .4 1 Difference in log daily wage Difference in log daily wage .2 .5 0 0 -.5 -.2 .1 .2 .3 .4 .5 .6 .7 .8 .9 .1 .2 .3 .4 .5 .6 .7 .8 .9 Quantiles Education Employment class & occupation Education Employment class & occupation Demographic characteristics Childcare support Demographic characteristics Childcare support Geographic location Geographic location Source: Based on data from the 2020 Labor Force Survey. Source: Based on data from the 2020 Labor Force Survey. At the top percentile, men earn about 5 percent more than women in daily wage. The pay gap is entirely driven by higher returns to men’s attributes than to women’s attributes. In this wage group women are overrepresented in government and public employment whereas men are overrepresented in the private sector. However, although in the private sector women and men (in the 99th wage percentile) receive equal returns to their skills and endowment, in the public sector women receive lower returns than men. FIGURE B.5 Endowment and Returns Gaps at Top Percentiles, 2020, Percent .4 Difference in log daily wage .2 0 -.2 -.4 -.6 -.9 -.91 -.92 -.93 -.94 -.95 -.96 -.97 -.98 -.99 Quantiles Total wage gap Pure Endowments Effect Pure Returns Effect Total Endowments Effect Total Returns Effect Source: Based on data from the 2020 Labor Force Survey. 144 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Appendices TABLE B.1 Unconditional Quantile Regression, by Gender, 2012 Men Women Q20 Q40 Q60 Q80 Q20 Q40 Q60 Q80 Geographic location fixed effects Region (Mindanao=omitted) Luzon -0.073** 0.056* 0.129*** 0.154*** 0.131** -0.109 0.191*** 0.124* Visayas -0.274*** -0.123*** -0.074*** -0.056** -0.069 -0.255*** 0.050 -0.040 Conflict-affected zone -0.193*** -0.095*** -0.100*** -0.068** -0.125** -0.296*** -0.082* -0.067 Natural disaster–affected zone 0.008 -0.081*** -0.103*** -0.065*** -0.065** -0.159*** -0.187*** -0.063* Urban 0.076*** 0.181*** 0.197*** 0.170*** 0.165*** 0.325*** 0.285*** 0.096*** Demographic characteristics Age 0.003*** 0.003*** 0.004*** 0.003*** 0.003*** 0.005*** 0.002** 0.009*** Marital status (separated = omit- ted) Single -0.118*** -0.107** -0.091** -0.043 -0.117*** -0.149** -0.013 -0.069 Married 0.040 0.088** 0.045 0.063 -0.034 -0.049 0.047 0.077 Widowed -0.034 -0.053 -0.077 -0.034 -0.067 -0.100 -0.009 -0.006 Number members aged 0–2 0.008 -0.035** -0.028** -0.022 0.000 -0.001 0.013 -0.028 Number members aged 3–5 0.016 -0.005 0.001 0.004 0.014 0.019 -0.015 -0.012 Number members aged 6–8 -0.001 -0.010 0.008 0.010 -0.012 -0.013 -0.005 -0.002 Number members aged 9–11 -0.004 0.001 -0.003 0.002 -0.014 0.004 -0.014 0.036* Number members aged 12–14 -0.020** -0.012 -0.001 0.014 -0.010 -0.015 -0.012 0.019 Number of extended members aged -0.014*** -0.011** -0.007* -0.014*** -0.002 0.006 0.004 -0.015 10–60 Young child in school 0.011 0.022 0.004 -0.009 -0.023 0.024 -0.002 -0.000 Domestic help -0.164*** -0.083*** 0.030 0.293*** -0.062* -0.241*** -0.067*** 0.335*** Education level (none = omitted) Elementary 0.114*** 0.062*** 0.014 -0.045*** 0.114** 0.071 -0.057* -0.012 Incomplete high school 0.131*** 0.090*** 0.022 -0.021 0.115** 0.144** 0.015 0.091*** High school graduate 0.235*** 0.220*** 0.136*** 0.065*** 0.220*** 0.373*** 0.139*** 0.073*** Postsecondary 0.349*** 0.404*** 0.292*** 0.274*** 0.333*** 0.827*** 0.491*** 0.276*** College and above 0.450*** 0.608*** 0.553*** 0.833*** 0.475*** 1.312*** 1.052*** 1.368*** Sector of employment (agriculture = omitted) Industry 0.227*** 0.537*** 0.360*** 0.102*** -0.111*** 1.002*** 0.698*** 0.016 Construction 0.446*** 0.702*** 0.383*** -0.022* -0.009 1.218*** 0.660*** -0.021 Wholesale & Retail Trade 0.133*** 0.351*** 0.181*** 0.025 -0.165*** 0.583*** 0.270*** -0.059* Public Admin & HD 0.247*** 0.552*** 0.321*** 0.196*** -0.150*** 0.817*** 0.553*** 0.461*** Other Services 0.176*** 0.423*** 0.254*** 0.150*** -0.167*** 0.618*** 0.363*** 0.102** Total hrs worked all jobs 0.001** 0.001** -0.001* -0.001*** -0.003*** -0.006*** -0.004*** -0.002*** Underemployed -0.015 -0.040*** -0.040*** -0.035*** 0.034 0.029 -0.044* -0.180*** Employment class (private HH = omitted) Private Establishment 0.488*** 0.324*** 0.214*** 0.244*** 0.506*** 0.805*** 0.391*** 0.273*** Government/Gov. corporation 0.397*** 0.231*** 0.197*** 0.436*** 0.491*** 0.697*** 0.381*** 0.756*** Family-owned business with pay 0.368*** 0.062 -0.015 0.075 0.295* 0.076 -0.019 0.311* Overcoming Barriers To Women’s Economic Empowerment In The Philippines 145 Appendices Men Women Q20 Q40 Q60 Q80 Q20 Q40 Q60 Q80 Employment nature (different employers = omitted) Permanent job 0.015 0.208*** 0.129*** 0.123*** -0.218*** -0.033 0.075* 0.141*** Seasonal/short-term -0.001 0.078*** 0.010 -0.019 -0.261*** -0.143** -0.119*** -0.143*** Constant 4.174*** 4.050*** 4.872*** 5.288*** 4.377*** 3.519*** 4.463*** 4.889*** Observations 23723 23723 23723 23723 15221 15221 15221 15221 Adjusted R2 0.190 0.377 0.372 0.348 0.227 0.436 0.466 0.443 Source: Based on data from the 2012 Labor Force Survey. Note: HD = human development ; HH = household; Q = quantile. Significance level: * = 10 percent, ** = 5 percent, *** = 1 percent, based on bootstrap standard deviations using 100 replications. TABLE B.2 Unconditional Quantile Regression, by Gender, 2020 Men Women Q20 Q40 Q60 Q80 Q20 Q40 Q60 Q80 Geographic location fixed effects Region (Mindanao=omitted) Luzon -0.002 0.194*** 0.274*** 0.090*** 0.080 0.196*** 0.184*** 0.046 Visayas -0.118*** 0.049** -0.087*** -0.037** 0.024 0.103*** -0.073* -0.112* Conflict-affected zone -0.146*** -0.011 -0.112*** -0.032* -0.143** -0.017 -0.142*** -0.148** Natural disaster–affected zone 0.012 -0.017 -0.020 -0.008 -0.084** -0.052** -0.055** -0.006 Urban 0.071*** 0.167*** 0.271*** 0.086*** 0.271*** 0.234*** 0.247*** 0.114*** Demographic characteristics Age 0.001 0.001* 0.004*** 0.001** -0.002 -0.003*** -0.000 0.003** Marital status (separated = omitted) Single -0.046* -0.039 -0.120** -0.018 -0.028 -0.077* -0.049 -0.101 Married 0.019 0.034 -0.004 0.028 0.018 -0.039 0.032 0.068 Widowed 0.025 -0.036 -0.045 -0.011 -0.018 -0.025 -0.052 -0.097 Number members aged 0–2 0.001 -0.001 -0.021 0.009 0.018 -0.027 -0.028 0.013 Number members aged 3–5 -0.016* -0.005 -0.000 -0.003 -0.028 -0.028 -0.016 -0.004 Number members aged 6–8 -0.002 -0.007 -0.010 0.000 0.009 -0.009 0.002 -0.004 Number members aged 9–11 0.003 0.002 -0.003 0.002 -0.026 -0.033* -0.041** -0.023 Number members aged 12–14 -0.001 -0.005 -0.003 -0.004 -0.070*** -0.049*** -0.032* 0.001 Number of extended members aged 0.001 0.004 0.002 -0.007* 0.019* 0.022*** 0.005 -0.010 10–60 Young child in school 0.009 -0.015 -0.028 -0.018 0.052 0.062* -0.026 -0.012 Domestic help -0.074** -0.022 0.024 0.126*** -0.210** -0.186*** -0.045 0.194** Education level (none = omitted) Elementary 0.062*** 0.041** -0.008 -0.014 0.164* 0.043 -0.019 -0.034 Incomplete high school 0.090*** 0.056*** 0.045* 0.015 0.127 0.021 -0.024 0.014 High school graduate 0.176*** 0.145*** 0.095*** 0.014 0.398*** 0.172*** 0.032 -0.022 Postsecondary 0.213*** 0.190*** 0.223*** 0.111*** 0.685*** 0.448*** 0.373*** 0.030 146 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Appendices Men Women Q20 Q40 Q60 Q80 Q20 Q40 Q60 Q80 College & above 0.283*** 0.293*** 0.479*** 0.396*** 0.836*** 0.615*** 0.696*** 0.719*** Sector of employment (agriculture = omitted) Industry 0.233*** 0.196*** -0.057* -0.081*** 0.408*** 0.367*** -0.131*** -0.271*** Construction 0.401*** 0.275*** 0.220*** -0.011 0.428*** 0.260*** 0.204* -0.094 Wholesale & Retail Trade 0.185*** 0.070*** -0.116*** -0.162*** 0.194* 0.120** -0.087* -0.353*** Public Admin & HD 0.238*** 0.185*** 0.132*** 0.085*** 0.266*** 0.282*** 0.228*** -0.016 Other Services 0.204*** 0.141*** 0.021 -0.059*** 0.261** 0.148*** 0.042 -0.155*** Total hrs worked all jobs 0.001*** 0.001*** 0.001 0.001** -0.005*** -0.002*** -0.002*** -0.000 Underemployed -0.006 -0.032** -0.065*** -0.012 0.028 -0.022 -0.037 -0.068 Employment class (private HH = omitted) Private Establishment 0.231*** 0.161*** 0.090 0.031 0.652*** 0.178*** 0.119*** 0.008 Government/Gov. corporation 0.173*** 0.150*** 0.108 0.073* 0.693*** 0.204*** 0.227*** 0.741*** Family owned business with pay 0.053 0.102 0.029 0.108 -0.028 -0.144 0.008 0.037 Employment nature (different employers =nomitted) Permanent job 0.062*** 0.026 0.033 0.023* -0.280*** -0.019 0.056 0.092*** Seasonal/short-term 0.024 -0.051*** -0.103*** -0.042*** -0.453*** -0.163*** -0.173*** -0.175*** Constant 5.055*** 5.262*** 5.773*** 6.336*** 4.507*** 5.338*** 5.862*** 6.853*** Observations 22917 22917 22917 22917 14293 14293 14293 14293 Adjusted R2 0.268 0.364 0.374 0.391 0.309 0.447 0.483 0.457 Source: Based on data from the 2020 Labor Force Survey. Note: HD = human development; HH = household; Q = quantile. Significance level: * = 10 percent, ** = 5 percent, *** = 1 percent, based on bootstrap standard deviations using 100 replications. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 147 Appendices Appendix C. Estimates from High Frequency Survey FIGURE C.1 Share of Men and Women, by Sector, FIGURE C.2 Proportion of Jobs Lost by Sector 2020, Percent and Gender, 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey. FIGURE C.3 Employment, by Sector, Gender, and FIGURE C.4 Employment, by Sector, Gender, and Region, 2020, Percent Quintiles, 2020, Percent Source: Based on data from the 2020 Household High-Frequency Survey. 148 Overcoming Barriers To Women’s Economic Empowerment In The Philippines References Overcoming Barriers To Women’s Economic Empowerment In The Philippines 149 References Abrigo, M. R. M., and K. Francisco-Abrigo. 2019. Pacific Gender Innovation Lab Policy Brief, “Counting Women’s Work in the Philippines.” Issue 2, World Bank, Washington, DC. Discussion Paper Series, No. 2019-02, Philippine Institute for Development Studies. Cho, Y., J. Avalos, Y. Kawasoe, D. Johnson, and R. Rodriguez. 2021. “Mitigating the Asian Development Bank. 2015. Key Indicators Impact of COVID-19 on the Welfare of Low for Asia and the Pacific 2015 (Issue 2015). Income Households in the Philippines: The Manila: Asian Development Bank. https:// Role of Social Protection.” COVID-19 Low www.adb.org/publications/key-indicators-asia- Income HOPE Survey Note No. 1, World Bank, and-pacific-2015 Washington, DC. Asian Development Bank. 2019. Asian Cho, Y., D. Doan, A. Kuddo, and R. Rodriguez. Development Outlook 2019 Update: Fostering 2021. “Revisiting Philippine Labor Market Growth and Inclusion in Asia’s Cities. Manila: Regulations in the Time of COVID-19.” World Asian Development Bank. https://www.adb. Bank, Washington, DC. org/sites/default/files/publication/524596/ ado2019-update.pdf Cho, Y., D. Johnson, Y. Kawasoe, J. Avalos, and R. Rodriguez. 2021. “The Impact of the Bayudan-Dacuycuy, C. 2019. “Examining the COVID-19 Crisis on Low Income Households Women’s Low Labor Market Participation Rate in the Philippines: Deepening Distress despite in the Philippines: Is Housework the Missing Rebounding Economy.” COVID-19 Low Income Link?” Discussion Paper Series, No. 2019-05, HOPE Survey Note No. 2. World Bank, Philippine Institute for Development Studies. Washington, DC. Beylis, F., R. F. Jaef, M. Morris, A. R. Sebastian, Coen-Pirani, D., A. Leon, and S. Lugauer. 2010. and R. Sinha. 2020. “Going Viral: COVID-19 “The Effect of Household Appliances on and the Accelerated Transformation of Jobs Female Labor Force Participation: Evidence in Latin America and the Caribbean.” Latin from Microdata.” Labour Economics 17: 503– American and Caribbean Studies, World Bank, 13. Washington, DC. Dang, H. H., M. Hiraga, and C. V. Nguyen. Brinkman, S. A., A. Hasan, H. Jung, A. Kinnell, 2019. “Childcare and Maternal Employment: N. Nakajim, and M. Pradhan. 2016. “The Evidence from Vietnam.” Policy Research Role of Preschool Quality in Promoting Child Working Paper, No. 8856, World Bank, Development: Evidence from Rural Indonesia.” Washington, DC. Policy Research Working Paper 7529, World Bank Group, Washington, DC. Du, F., and X. Dong. 2013. “Women’s Employment and Child Care Choices in Clark, S., C. W. Kabiru, S. Laszlo, and S. Urban China during the Economic Transition.” Muthuri. 2019. “The Impact of Childcare on Economic Development and Cultural Change Poor Urban Women’s Economic Empowerment 62 (1): 131–55. https://www.jstor.org/ in Africa.” Demography 56: 1247–72. https:// stable/10.1086/671714. doi.org/10.1007/s13524-019-00793-3. Epetia, C. F. 2019. “Explaining the Gender Gap Chowdhury, I., H. Johnson, A. Mannava, and in Labor Force Participation in the Philippines.” E. Perova. 2018. “Gender Gap in Earnings in Japan Labor Issues 3 (17): 30–38. Vietnam: Why Do Vietnamese Women Work in Lower Paid Occupations?” East Asia and 150 Overcoming Barriers To Women’s Economic Empowerment In The Philippines References Eviota, E. U. 1992. The Political Economy of Grant Thornton International. 2020. Women in Gender: Women and the Sexual Division of Business 2020. Grant Thornton International. Labor in the Philippines. London: Zed Books. Grantham, G. 2012. “Occupational, Marital, Firpo, Sergio, Nicole M. Fortin, and Thomas and Life-Cycle Determinants of Women’s Lemieux. 2007. Decomposing Wage Labor Force Participation in Mid Nineteenth- Distributions using Recentered Influence Century Rural France.” Feminist Economics 18 Functions Regressions. In Mimeo. Vancouver: (4): 97–119. https://doi.org/10.1080/13545701 University of British Columbia. .2012.737007. Firpo, S., N. Fortin, and T. Lemieux. 2009. Halim, D., H. Johnson, and E. Perova. 2017. “Unconditional Quantile Regressions.” “Could Childcare Services Improve Labor Econometrica 77 (3): 953–73. Market Outcomes in Indonesia?” East Asia and Pacific Gender Innovation Lab Policy Brief, Firpo, S., N. Fortin, and T. Lemieux. 2018. Issue 1, World Bank, Washington, DC. “Decomposing Wage Distributions Using Recentered Influence Function Regressions.” Halim, D., H. Johnson, and E. Perova. 2019. Econometrics 6 (2): 28. https://doi. “Preschool Availability and Female Labor org/10.3390/econometrics6020028. Participation: Evidence from Indonesia.” Policy Research Working Paper 8915, World Bank, Fitzpatrick, M. D. 2008. “Preschoolers Enrolled Washington, DC. and Mothers at Work? The Effects of Universal Pre-Kindergarten.” SIEPR Discussion Hanushek, E. A., and L. Woessmann. 2015. Paper, No. 08-01, Stanford Institute for Universal Basic Skills: What Countries Stand Economic Policy Research. to Gain. Paris; OECD Publishing. https://doi. org/10.1787/9789264234833-en. Fortin, N. 2015. “Gender Role Attitudes and Women’s Labor Market Participation: IFC (ILO (International Labor Organization). Opting-Out, AIDS, and the Persistent Appeal 2018. Global Wage Report 2018/19: What Lies of Housewifery.” Annals of Economics and behind Gender Pay Gaps. Geneva: ILO. Statistics 117/118 (Special Issue on the Economics of Gender): 379–401. Institute of Labor Studies (2020). Assessing regulatory measures toward realizing decent Fortin, N., T. Lemieux, and S. Firpo. 2010. work for domestic workers. Issue paper 2020- “Decomposition Methods in Economics.” NBER 01. https://ils.dole.gov.ph/issue-papers/. Working Paper 16045. National Bureau of Economic Research, Cambridge, MA. https:// International Finance Corporation. 2021. doi.org/10.3386/w16045 Women and E-commerce in Southeast Asia. Washington, DC: IFC. Goldin C. 1995. “The U-Shaped Female Labor Force Function in Economic Development and Jayachandran, S. 2021. “Social Norms as a Economic History.” In Schultz TP Investment Barrier to Women’s Employment in Developing in Women’s Human Capital and Economic Countries.” Department of Economics, Development, edited by T. P. Schultz, 61–90. Northwestern University. University of Chicago Press. https://scholar. harvard.edu/goldin/publications/u-shaped- Karamollaoglu, N., and B. Soybilgen. 2020. female-labor-force-function-economic- “Determinants of Turkish Female Labour development-and-economic-history. Force Participation: An Analysis with Overcoming Barriers To Women’s Economic Empowerment In The Philippines 151 References Manufacturing Firm-Level Data.” Applied Moser, C. O. N. 1993. Gender Planning and Economics Letters 27 (19): 1607–10. DOI: Development: Theory, Practice, and Training. 10.1080/13504851.2019.1707757. London: Routledge. Klasen, S. 2018. “What Explains Uneven Mukhopadhyay, T., C. Rivera, and H. Tapia. Female Labor Force Participation Levels and 2019. “Gender Inequality and Multidimensional Trends in Developing Countries?” Discussion Social Norms.” Working Paper, United Paper No. 246, Georg-August-Universität Nations Development Programme, Human Göttingen, Courant Research Centre – Poverty, Development Report Office, New York. Equity and Growth (CRC- PEG), Göttingen. http://hdl.handle.net/10419/175179. Mulugeta, G. 2021. The Role and Determinants of Women Labor Force Participation for Klasen, S., T. T. N. Le, J. Pieters, and M. Santo Household Poverty Reduction in Debre Birhan Silva. 2021. “What Drives Female Labour Town, North Shewa Zone, Ethiopia.” Cogent Force Participation? Comparable Micro-level Economics & Finance 9 (1). https://doi.org/10.1 Evidence from Eight Developing and Emerging 080/23322039.2021.1892927. Economies.” Journal of Development Studies 57 (3): 417–42. https://doi.org/10.1080/0022 Navarro, J. 2020. “Can BPOs Sustain a 0388.2020.1790533. Work from Home Model?” PwC Philippines. https://www.pwc.com/ph/en/taxwise-or- Klasen, S., and J. Pieters. 2015. “What otherwise/2020/can-bpos-sustain-a-work- Explains the Stagnation of Female Labor Force from-home-model.html. Participation in Urban India?” World Bank Economic Review 29 (3): 449–78. NEDA (Republic of the Philippines National Economic and Development Authority). Lafortune, J., J. Riutort, and J. Tessada. 2019. “Determinants of Female Labor Force 2018. “Role Models or Individual Consulting: Participation in the Philippines.” The Impact of Personalizing Micro- Entrepreneurship Training.” American Ortiz Rodríguez, J., and V. Kumara Pillai. Economic Journal: Applied Economics 10: 2019. “Advancing Support for Gender 222–45. Equality among Women in Mexico: Significance of Labor Force Participation.’ Lavy, V., G. Lotti, and Z. Yan. 2016. International Social Work 62 (1): 172–84. DOI: “Empowering Mothers and Enhancing Early 10.1177/0020872817717323 Childhood Investment: Effect on Adults Outcomes and Children Cognitive and Non- Oxfam International. 2019. Understanding cognitive Skills.” NBER Working Paper 22963, Norms around the Gendered Division of National Bureau of Economic Research, Labour: Results from Focus Group Discussion Cambridge, MA. http://www.nber.org/papers/ in the Philippines. Women’s Economic w22963. Empowerment and Care (WE-Care) Report. Oxford: Oxfam International. Lundin, D., E. Mörk, and B. Öckert. 2008. “How Far Can Reduced Childcare Prices Push Female Pennings, S. 2021. “The Effect of Increased Labor Supply?” Labour Economics 15: 647–59. Female Labor Force Participation on Long doi:10.1016/j.labeco.2008.04.005. Term Economic Growth in the Philippines.” World Bank, Washington, DC. 152 Overcoming Barriers To Women’s Economic Empowerment In The Philippines References PwC. 2016. Women Returners: The £1 Billion UNDP (United Nations Development Career Break Penalty for Professional Women. Programme). 2020b. Human Development https://www.pwc.co.uk/economic-services/ Perspectives: Tackling Social Norms: A Game women-returners/pwc-research-women- Changer for Gender Inequalities. New York: returners-nov-2016.pdf. United Nations. Rios-Avila, F. 2019. “A Semi-Parametric UN Women. 2020. From Insight to Action: Approach to the Oaxaca-Blinder Gender Equality in the Wake of COVID-19. Decomposition with Continuous Group New York: United Nations. Variable and Self-Selection.” Economics Working Paper Archive wp_930, Levy We Are Social. 2020. “Digital 2020: Global Economics Institute. https://ideas.repec.org/p/ Digital Overview” https://wearesocial.com/ lev/wrkpap/wp_930.html. digital-2020. Rothe, C. 2015. “Decomposing the Composition WEF (World Economic Forum). 2020. The Effect: The Role of Covariates in Determining Global Gender Gap Report 2020. Geneva: WEF. Between-Group Differences in Economic Outcomes.” Journal of Business & Economic Widarti, D. 1998. “Determinants of Labour Statistics 33. https://doi.org/10.1080/07350 Force Participation by Married Women: The 015.2014.948959 Case of Jakarta.” Bulletin of Indonesian Economic Studies 34 (2): 93–120. https://doi. Taşseven, O., D. Altaş, and T. Ün. 2016. org/10.1080/00074919812331337350. “The Determinants of Female Labor Force Participation for OECD Countries.” Uluslararası Woetzel, J., A. Madgavkar, K. Sneader, O. Ekonomik Araştırmalar Dergisi 2 (2). Tonby, D. Lin, J. Lydon, S. Sha, M. Krishnan, K. Ellingrud, and M. Gubieski. 2018. “The Power Thompson, M. 2020. “COVID-19 and of Parity: Advancing Women’s Equality in Asia the Philippines’ Outsourcing Industry.” Pacific.” McKinsey Global Institute. McKinsey London School of Economics (LSE) Blog, & Company. https://www.mckinsey.com/ September 22, 2020. https://blogs.lse. featured-insights/gender-equality/the-power- ac.uk/seac/2020/09/22/covid-19-and-the- of-parity-advancing-womens-equality-in-asia- philippines-outsourcing-industry/. pacific#. UNDP (United Nations Development World Bank. 2012. World Development Report Programme). 2019. Human Development 2012: Gender Equality and Development. World Report 2019: Beyond Income, Beyond Bank. Averages, Beyond Today—Inequalities in Human Development in the 21st Century. New World Bank. 2018. “Growth and Productivity York: United Nations. http://hdr.undp.org/ in the Philippines: Winning the Future.” World en/2019-report. Bank, Washington, DC. UNDP (United Nations Development World Bank. 2020a. Philippines Country Programme). 2020a. Human Development Gender Action Plan. Washington, DC: World Report 2020: The Next Frontier—Human Bank. Development and the Anthropocene. New York: United Nations. Overcoming Barriers To Women’s Economic Empowerment In The Philippines 153 World Bank. 2020b. Philippines Digital Economy Report 2020: A Better Normal Under COVID-19—Digitalizing the Philippine Economy Now. World Bank, Washington, DC. World Bank. 2020c. The Human Capital Index 2020 Update: Human Capital in the Time of COVID-19. Washington, DC: World Bank. World Bank. 2021. Women, Business and the Law 2021. Washington, DC: World Bank. doi:10.1596/978-1-4648-1652-9. World Bank Group. 2019a. Country Partnership Framework for Republic of the Philippines. Washington, DC: World Bank. World Bank Group. 2019b. Systematic Country Diagnostic of the Philippines: Realizing the Filipino Dream for 2040. Washington, DC: World Bank. World Bank Group. 2021. “Philippine Basic Education System: Strengthening Effective Learning During the COVID-19 Pandemic and Beyond.” Philippines COVID-19 Monitoring Survey Policy Notes, May 2021, World Bank, Washington, DC. 154 Overcoming Barriers To Women’s Economic Empowerment In The Philippines Produced by the Philippine Women’s Economic Empowerment Program of the World Bank with support from the Australian Government