Women’s Economic Empowerment in the Democratic Republic of the Congo: Obstacles and Opportunities Source: https://plantwithpurpose.org/anastazie/ 1 Contents Acknowledgements......................................................................................................................... 4 Executive Summary ........................................................................................................................ 6 1. Introduction.............................................................................................................................. 9 2. Gender Disparities in Economic Opportunities and Outcomes in the DRC.......................... 13 Gender gaps in labor force participation ................................................................................... 13 Gender gaps in agricultural outcomes ....................................................................................... 15 Gender gaps in entrepreneurship profits ................................................................................... 18 Gender gaps in wage earnings................................................................................................... 19 3. What are the Drivers of the Gender Disparities in the DRC?................................................ 23 Drivers of gender gaps in labor force participation ................................................................... 24 Drivers of gender gaps in agricultural outcomes ...................................................................... 25 Drivers of gender gaps in entrepreneurship profits ................................................................... 28 Drivers of gender gaps in wage earnings .................................................................................. 29 4. Diving Deep on Three Constraints ........................................................................................ 32 Deep Dive 1: Control over Land ............................................................................................... 32 Deep Dive 2: Agency ............................................................................................................... 37 Deep Dive 3: Risk and uncertainty .......................................................................................... 44 5. Evidence Review and Policy Recommendations .................................................................. 49 5.1 To Increase Women’s Agricultural Productivity ................................................................ 57 Increasing the Adoption of Cash Crops Among Female Farmers ......................................... 57 Increasing Land Tenure Security ........................................................................................... 57 Increasing Adoption and Use of Agricultural Inputs............................................................. 58 Increasing Female Farmer’s Use of More Productive Labor ................................................ 59 5.2 To Increase Women’s Educational Attainment and Skills.................................................. 59 Increasing Women and Girl’s Educational Attainment ......................................................... 59 Increasing Women’s Vocational, Entrepreneurial, and Socioemotional Skills ..................... 59 5.3 To Increase Women’s Access to Capital............................................................................. 60 Strengthening Financial Inclusion ......................................................................................... 60 Supporting Productive Inclusion to Increase Income ............................................................ 61 5.4 To Increase Women’s Physical Security and Agency ........................................................ 62 Increasing Knowledge of Rights and Laws to Prevent Gender-Based Violence .................. 62 Decrease Women’s Risk of Intimate Partner Violence ......................................................... 63 Increase Women’s Reproductive Agency ............................................................................. 63 2 Decreasing Women’s Burden of Care ................................................................................... 64 6. Appendix................................................................................................................................ 65 Appendix A: Data Sources ........................................................................................................ 65 Nationally Representative Datasets ....................................................................................... 65 Impact Evaluation Datasets ................................................................................................... 65 Qualitative Case Study in Goma............................................................................................ 66 Appendix B: Descriptive Statistics ........................................................................................... 67 Appendix C: Decomposition Methods...................................................................................... 80 Appendix D: Decompositions ................................................................................................... 84 7. Endnotes .............................................................................................................................. 106 3 Acknowledgements This report was prepared by a team of authors from the World Bank’s Africa Gender Innovation Lab (GIL) and the Social Sustainability & Inclusion (SSI) Global Practice, comprised of Nono Ayivi-Guedehoussou, Ervin Dervisevic, Aletheia Donald, Helena Hwang, Lisette Meno Khonde, Chloé Lewis, Laurel Morrison, Verena Phipps, Rachael Pierotti, and Julia Vaillant. The authors would like to gratefully acknowledge Natalia Agapitova, Patricia Fernandes and Lucia Hanmer, who peer reviewed the report. The authors would like to thank team members from the SME Growth and Development project, the DRC Western Growth Poles project and the Engaging Men through Accountable Practice project. We would also like to express our gratitude to the skilled interviewers who collected the qualitative data in Goma, Alfred Banga and Rébecca Bora, without whom the research would not have been possible. 4 List of Figures Figure 1: Gender Gaps in the Labor Force ................................................................................... 13 Figure 2: Gender Gaps in the Labor Force by Region .................................................................. 14 Figure 3: Gender Gaps in Agricultural Yields .............................................................................. 15 Figure 4: Gender Gaps in Agricultural Yield by Region .............................................................. 16 Figure 5: Gender Gaps in Agricultural Production ....................................................................... 17 Figure 6: Difference in Agricultural Production by Region ......................................................... 17 Figure 7: Gender Gaps in Business Profits ................................................................................... 18 Figure 8: Gender Gaps in Business Profits by Region ................................................................. 19 Figure 9: Gender Gaps in Wage Earnings .................................................................................... 20 Figure 10: Gender Gaps in Wage Earnings by Region ................................................................. 20 Figure 11: Wage Earning Gaps by Sector..................................................................................... 21 Figure 12: Gender Gaps in Plot Ownership .................................................................................. 33 Figure 13: Women & Men’s Pathways to Land Tenure ............................................................... 35 Figure 14: Differential Educational Aspirations ........................................................................... 38 Figure 15: Household Decision Making by Domain .................................................................... 41 Figure 16: Gendered Division of Domestic Responsibilities ....................................................... 43 Figure 17: Attitudes Towards Violence ........................................................................................ 45 Figure 18: Women aged 15+ Experience of Violence in Lifetime ............................................... 46 Figure 19: Evidence Map .............................................................................................................. 51 List of Tables Table 1: Significant drivers of the labor force participation gender gap ..................................... 24 Table 2: Significant drivers of the agricultural productivity gender gap ...................................... 26 Table 3: Significant drivers of the agricultural production gender gap ........................................ 27 Table 4: Significant drivers of business profitability gender gap ................................................. 29 Table 5: Significant drivers of gender gap in wage earnings........................................................ 30 List of Boxes Box 1: Measurement of Gender Gaps........................................................................................... 14 Box 2: Decomposition Methodology ........................................................................................... 23 5 Executive Summary The aim of this report is to identify binding constraints to women’s economic empowerment in the DRC and identify promising entry points to unbind these constraints. This report makes three core contributions. It provides: i) a comprehensive picture of gender gaps across the country, ii) an in- depth analysis of underlying drivers of the observed gender gaps, and iii) concrete policy and programmatic guidance on how to close the gender gaps. A comprehensive picture of gender gaps across the country is drawn from six quantitative datasets, including the nationally representative 1-2-3 Survey and Demographic and Health Survey (DHS) as well as four household datasets collected within impact evaluations conducted by the World Bank’s Africa Gender Innovation Lab (GIL). What these four surveys lack in representativeness, they gain in explanatory power, allowing the report to examine gender gaps not only across households, but also within households. ➢ Gender Gap in Labor Force Participation: Nationally, women are 6.2% to 8.2% less likely to work than their male counterparts. ➢ Gender Gaps in Agricultural Outcomes: Within agriculture, which employs over two-thirds of women in the DRC, the production of women farmers is 18% lower than that of men and their productivity is 11% lower. Gender gaps are even larger when comparing men and women in the same households. ➢ Gender Gaps in Entrepreneurship Profits: Women’s business profits are nearly 67% lower than those of men, and women’s businesses have 70% fewer paid workers. ➢ Gender Gaps in Wage Earnings: Less than 10% of women are wage or salaried workers compared to nearly 25% of men. Nationally, there is a 77% gap in wage earnings between men and women and gender disparities are particularly pronounced in the private sector with nearly an 80% wage gap. The report also finds important regional variation in these gender gaps, with the largest disparities generally observed in the northeastern, northwestern and southern regions of the DRC. An in-depth investigation of the observed gender gaps utilizes the Kitagawa-Oaxaca-Blinder decomposition methodology to isolate factors contributing to gender gaps. The methodology decomposes the gender gap into two main components: an endowment effect and a structural effect. The endowment effect captures the difference in the levels of resources that women have relative to men, while the structural effect refers to the portion of the gender gap that exists because of differences in the returns to these resources. While the results cannot be interpreted as causal, we denote variables that are both theoretically and empirically linked to greater gender disparities as ‘drivers’. ➢ Drivers of Gender Gap in Labor Force Participation: A higher number of dependents, capturing women’s higher burden of care work, is identified as the primary driver of women’s lower labor force participation. ➢ Drivers of Gender Gaps in Agricultural Outcomes: Women’s lower agricultural yields are driven are primarily driven by women’s lower cultivation of cash crops. Women’s lower agricultural production is driven by a variety of interlocking constraints, including a higher 6 dependency ratio, lower cell phone ownership, less land, lower number of workers and a lower likelihood of rice cultivation. ➢ Drivers of Gender Gaps in Entrepreneurship Profits: The lower number of workers emerges as the single most consistent driver of the entrepreneurship gender gap, though businesswomen’s lower levels of capital compared to their male counterparts also impede their profits. ➢ Drivers of Gender Gaps in Wage Earnings: Lower educational attainment is the primary driver of women’s lower wage earnings. In addition to these sector-specific drivers, women’s lower access to certain key assets and lower returns to marriage are consistent drivers across sectors. Deep dive explorations of three important underlying constraints are included: lack of control over land, agency, and risk and uncertainty. The report includes findings from both quantitative analysis and original qualitative research to provide a deeper a understanding of underlying constraints. Control over Land: Within the DRC, women make up more than half of the agricultural workforce, yet men own three quarters of agricultural plots. Women’s cultivation of a lower number of plots and smaller plot sizes are significant drivers of the gender gap in agricultural production. The deep dive shows that women’s weak control over land influences their agricultural outcomes by disincentivizing female farmers’ investment in agricultural production. Lack of land tenure could act as a barrier to women accessing credit, further dampening their agricultural production as well as other entrepreneurial endeavors. Agency: Women are consistently less empowered than men to make decisions affecting their lives. Men are the primary household decisionmakers, while women are responsible for 83% of household chores and 64% of all childcare. Research summarized in this report shows that women’s lack of agency impacts their use of time, their fertility and their participation in income generating activities. Risk and Uncertainty: Women face increased vulnerability to shocks, which range from physical, to financial, to institutional insecurity. Gender based violence is prevalent and domestic abuse is a widely accepted practice. Female entrepreneurs mistrust financial institutions yet are unable to keep profits in the household for fear their husband will take them. Moreover, 52% of women report experiencing physical violence, damaging their economic participation and human flourishing. Concrete policy and programmatic guidance conclude the report. The recommendations are carefully tailored both to the specific constraints identified in the quantitative and qualitative data, and to implementation feasibility in the DRC context. There is a lack of evidence from the DRC on effective strategies to close gender gaps. Additionally, there is a disparity between evidence available for the conflict-affected, eastern region of the DRC and the rest of the country. While studies conducted in the DRC are prioritized, evidence from other countries in Sub-Saharan Africa are included, highlighting those which are from fragile or conflict-affected contexts. More high- quality data collection and research will be needed to move the needle for women’s econom ic empowerment in the DRC. 7 Policy Priority: Increase Women's Agricultural Productivity Driver Addressed Policy Options • Conditionally subsidized land registration • Increase women's inheritance rights through attainment of official birth and marriage Insecure Land Tenure certificates • Government driven national land regularization program • Female extension service agents Low Adoption and Returns • Extension services targeting couples to Agricultural Inputs • Utilization of digital technology for agricultural extension information Low Adoption of Cash • Engaging men to change norms around gendered crops Crops • Women Farmer Groups Low use and Productivity • Mechanization to reduce farm labor of Farm Labor • Cash Transfer Policy Priority: Increase Women’s Educational Attainment and Skills Driver Addressed Policy Options • Adolescent girls empowerment programming Low Educational • Providing scholarships to female students Attainment • Ensuring availability of affordable and quality preschools • Utilizing mobile technology to improve adult education • Adolescent girls empowerment programming Low Levels of Skills • Livelihood skills-based training Training • Socioemotional skills training Policy Priority: Increase Women’s Access to Capital Driver Addressed Policy Options • Same-gender microfinance agent banking Low Financial Inclusion • Village Savings and Loans • Secure savings mechanisms Low Income Generation • Cash & Productive Asset Transfers Policy Priority: Increase Women's Physical Security and Household Agency Driver Addressed Policy Options Low Knowledge of Rights • Decrease bribery and GBV by increasing understanding of legal rights and procedures and Laws between women and police High Levels of IPV • Shift social norms and behaviors through couples-based discussion groups Low Levels of • Increase agency regarding fertility decision-making through life skills training Reproductive Agency • Encourage school enrollment, delayed marriage and delayed birth through UCTs & CCTs • Provision of Childcare Services High Burden of Care • Decrease women's time constraints by increasing men's participation in household & childcare responsibilities 8 1. Introduction The Democratic Republic of Congo (DRC) is a post-conflict and fragile country with a rapidly growing population. Its turbulent history, along with weak governance, has limited successive governments’ ability to establish stable institutions and improve the population’s living standards. Macroeconomic performance improved until mid-2015 and was marked by strong economic growth.1 The ongoing COVID-19 pandemic has slowed economic growth from a pre-pandemic level of 4.4% in 2019 to about 0.8% in 2020. Non-mining sectors have contracted as a result of mobility restrictions, diminished trading, and constrained government spending. i However, pre- pandemic growth has failed to translate into a significant reduction in poverty and inequality. Inequalities have fallen slightly—the Gini coefficient improved from 38 in 2005 to 35 in 2012— but large portions of the population remain trapped in extreme poverty, being made worse by the COVID-19 pandemic. While the proportion of people living below the poverty line declined from 69.3% to 64% between 2005 and 2012, the absolute number of poor increased by 7 million during the same period.ii The DRC ranked 175th out of 188 countries in the 2019 Gender Inequality Index , which benchmarks national gender gaps using economic, political, education, and health criteria. iii Although important gains have been made in the health and education domains over the past two decades, as well as in terms of legislation that addresses gender inequalities, persistent socio- cultural disparities restrict women’s engagement in social and economic life and public decision - making. Women’s representation in politics is limited, currently occupying about 4.6% of seats in the Senate and 10.3% in the National Assembly (compared with an average of 20.6% among low- income countries).iv Within the DRC, women face significant barriers to economic opportunities and empowerment. The DRC has severe gender disparities in education, especially as students progress to the secondary level. 28% fewer women have completed primary school than men, and only 16.8% of women have completed secondary school overall, which is less than half the rate of completion for men. Correspondingly, the literacy rate of women trails that of men by more than 22%.v Women face significant inequality regarding decision-making ability and income generation. Only 60% of women report participating in decisions about major household purchases, and less than half of women report participating in decisions about their own healthcare. Early marriage and high fertility rates are significant challenges for women in the DRC. Women without any education have a fertility rate double that of women who completed secondary school. Throughout the country, the average age at first birth is 19.9 years and 37% of women aged 20-24 were married before their eighteenth birthday, as compared to 6% of men in the same age group. Gender-based violence is a pervasive problem throughout the country, in both urban and rural areas. Of women in the DRC, 52% have experienced physical violence, 27% of all women report experiencing sexual violence in their lifetimes, and 57% of women report experiencing intimate partner violence.vi Additionally, the COVID-19 pandemic, and efforts to contain it, can increase 1 Ranging from 5.6 to 6.2 percent between 2002 and 2008. 9 the risk of gender-based violence.vii Physical security is further jeopardized in areas of the country affected by conflict. Legal reforms and amendments over the last five years have removed several discriminatory provisions restricting women’s economic empowerment. Notably, these reforms include prohibiting gender discrimination in employment and guaranteeing equal remuneration (2017); prohibiting gender discrimination in financial services (2017), amending the law to allow women to make legal commitments, open a bank account, or register a business without their husband’s consent (2018), allowing women to get a job without their husband’s consent (2018) , and no longer requiring women to obey their husbands (2018). However, much remains to be done to ensure that such legislation is enforced, and that women and men are equal in both agency and opportunity. Women’s labor force participation rate in the DRC is estimated at almost 62%. Most women (69.7%) work in agricultural production, with the next-highest category of employment being entrepreneurship (20.5%). This share is low compared to other countries: the 2013 World Bank Enterprise Survey found an average of 34% of women working in entrepreneurship in surveyed countries.viii An even smaller share of women are wage or salaried workers, in stark contrast with male employment: 23.9% of working men are in wage and salaried employment, against only 6.4% of women.ix Moreover, while women’s labor force participation is relatively high, their earnings are much lower than that of men’s.x Women own fewer assets, in particular land: only 7.6% of women own land alone compared to 21.8% of men.xi While women make up most of the workers in the agricultural sector (53%), their access to land and credit remains constrained, limiting productivity.xii Moreover, according to the ILO, there is a 20 percentage point gender gap in the share of employment classified as ‘vulnerable’: 90.1% of women’s employment versus 69.6% of men. To make progress on reducing gender disparities in economic opportunities for the women of the DRC, it is crucial to better understand the magnitude and drivers of gender gaps in key economic sectors. This is particularly the case in the face of the COVID-19 pandemic, which has disproportionately impacted women. During the pandemic, women-run businesses reduced the rates of their employee wages more than male-run enterprises by almost 10%. In October 2020, 3.5% of households reported a household member having lost their job. In Kinshasa, of those households which reported a job loss, female household members were 38% less likely to have gotten a new job than male members of the household. xiii Additionally, lockdown measures, loss of income, and increased stress can contribute to an increased risk of gender-based violence and intimate partner violence.xiv The pandemic also increases many barriers faced by adolescent girls in the DRC, such as increased risks of early marriage, poor education outcomes, and GBV, which all contribute to increased risk of poverty and insecurity. xv This report presents data from multiple quantitative and qualitative sources—compiled together for the first time, and in the case of the qualitative data, collected specifically for this report —to provide a comprehensive overview of existing gender gaps in the DRC across labor force participation, agricultural production, enterprise profits and wage earnings. This report makes three core contributions. First, it draws on six quantitative datasets to provide a comprehensive picture of gender gaps across the country. These are the nationally 10 representative 1-2-3 Survey and Demographic and Health Survey, as well as four household surveys conducted as part of impact evaluations conducted by the World Bank’s Africa Gender Innovation Lab (GIL)2 , described under Data Sources in Appendix A. What these four surveys lack in representativeness, they gain in explanatory power, allowing the report to examine gender gaps not only across households, but also within households. The data shows that in the DRC, women are 6 to 8% less likely to work than similar men. Moreover, there are striking differences in earnings outcomes. The largest gaps are found within wage earnings, with women earning 77% less than men on average, followed by business profits: female-run businesses have 67% lower profits on average. Agriculture, which occupies over two-thirds of women in the DRC, also has significant disparities, with women farmers producing 18% less revenue and having 11% lower productivity. Second, this report provides an in-depth investigation of the drivers of the observed gender gaps. Using quantitative decomposition analysis, the report identifies women’s higher care burdens as the primary driver of women’s lower labor force participation. Women’s lower agricultural production is driven by a variety of interlocking constraints, including a higher dependency ratio, lower cell phone ownership, less land, lower number of workers and a lower likelihood of rice cultivation. Women also face lower returns to both labor and fertilizer. Attaining gender parity in Congolese agriculture will thus not only require that women have the same access to productive inputs as men, but also that their returns to these inputs are equalized. When looking at agricultural yields instead of production, women’s lower cultivation of cash crops (such as palm oil, coffee, cocoa, sugar cane, and rubber) emerges as the main driver of women’s lower productivity. In entrepreneurship, a lower number of workers emerges as the single most consistent driver of the gender gap, though businesswomen’s lower levels of capital compared to their male counterparts also impede their profits. Meanwhile, lower educational attainment is the primary driver of women’s lower wage earnings. In addition to the dependency ratio, women’s lower access to certain key assets as well as lower returns to marriage are consistent drivers across sectors. To investigate these constraints further, this report presents deep dive explorations of three important underlying constraints to women’s economic empowerment in the DRC: lack of control over land, decision-making power, and risk and uncertainty. Finally, using analysis of original qualitative data from Goma in North Kivu, the report closely examines the constraints and opportunities for urban women’s economic advancement. This case study offers detailed illustrations of the contextual factors and social processes that underlie the gender gaps documented in the quantitative analysis. Third, this report proposes concrete policy and programmatic guidance to advance women’s economic empowerment in the DRC. Using the entry points identified in the decomposition results and the deep dives, the evidence base from both the DRC and other comparator countries on what works to close gender gaps is summarized in Section 5. Lastly, this report provides an 2 The World Bank’s Africa Gender Innovation Lab (GIL) conducts impact evaluations of development interventions and leads policy research to generate evidence on how to close gender gaps in earnings, productivity, assets, and agency. With these findings, GIL equips project teams and policy makers to design innovative and scalable interventions to address gender inequality. The GIL team is currently working on over 70 impact evaluations in nearly 30 countries with the aim of building an evidence base with lessons for the region. 11 evidence map of tested, emerging, and frontier interventions to strengthen gender equality in the country. 12 2. Gender Disparities in Economic Opportunities and Outcomes in the DRC As noted in the introduction, women have relatively high rates of labor force participation in the DRC. Women in the qualitative case study in Goma indicated that they are motivated to work by a desire to support their families and earn respect from their husbands, the precarity of their husbands’ economic contributions to the household, a desire to avoid dependence on others, and an affinity for their work. Nearly all women rejected the idea of leaving the labor force if their husband earned enough to support the household. Their responses are reflected in the sentiment shared by one interview participant who runs a hair salon and a moto taxi business, who said: “There is no wealth of a man that would prevent a woman from working to earn money.� Nonetheless, gender disparities do exist. To investigate potentially important regional variation in the estimated gender gaps, the report additionally shows each gender gap separately for each of the 25 provinces of the DRC, plus Kinshasa, in Appendix Table B1, computed using 1-2-3 Survey data. These provinces are mapped to larger geographic groupings (shown in the first column of Table B1) and included for each of the outcomes below. Gender gaps in labor force participation Using the quantitative data, this section first examines gender patterns in labor force participation (defined in Box 1 along with all other outcome variables presented in this report). Throughout this report, a positive gender gap indicates that men produce or earn more than women. Figure 1: Gender Gaps in the Labor Force 13 Data from the nationally representative 1-2-3 Survey indicate that women are 8.2% less likely to work than men. This figure is 6.4% in the nationally representative DHS survey. Figure 2 shows that these gender gaps in labor force participation are highest in the south, northeast and west of the country, while in the central, northwest and southwest parts of the country men are not significantly more likely to work than women. Figure 2: Gender Gaps in the Labor Force by Region 14 Box 1: Measurement of Gender Gaps How are gender gap measured? This report uses a variety of data sources to provide a rich picture of gender inequality in the DRC. � Labor Force Participation: This is defined as being engaged in the production of goods and services, and includes those who seek work in the 1-2-3 Survey. The analysis uses the gender of all household members aged 15 or older, and compares men and women of the same age and region of residence. � Agricultural earnings: Production is measured as the total value of the agricultural harvest, while yields result from dividing this value of production by the area of land used (owned or rented). The 1- 2-3 Survey analysis uses the gender of the household head to compute gaps, while the Childcare and Growth Pole surveys use the gender of plot managers, defined as the individual who makes most of the decisions on a plot. The analysis compares men and women with the same land area and region of residence. � Entrepreneurship profits: Business profits are defined as the difference between total revenues from the sale of goods and services minus total costs (such as utilities, rent, and labor). The analysis uses the gender of business managers (the individual primarily responsible for running the business) and compares men and women with the same age, household size and region of residence. � Wage earnings: This is defined as earnings from paid employment for someone outside the household. Similarly to labor force participation, the analysis uses the gender of all household members aged 15 or older, and compares men and women of the same age and region of residence. Gender gaps in agricultural outcomes Within the DRC, women make up the majority (57%) of the agricultural workforce. Moreover, agriculture makes up over two-thirds of women’s employment , with 70% of working women employed in agriculture (1-2-3 Survey). Yet women farmers produce less agricultural output per hectare compared to male farmers (Figure 3). In addition to the nationally representative estimate from the 1-2-3 Survey, which only allows for gender disaggregation by household head, two additional survey estimates are included. These allow us to look more precisely at within- household gender differences (see Appendix A for a description of each survey and eligibility criteria). Though non-significant in the 1-2-3 Survey, the gender gap in agricultural productivity amounts to 28.2% in the Childcare survey and 47.6% in the Growth Poles survey. This is a stark finding given the importance of agriculture for the Congolese economy and for women’s employment in particular. 15 Figure 3: Gender Gaps in Agricultural Yields Figure 4: Gender Gaps in Agricultural Yield by Region Similarly to labor force participation, there is substantial regional variation in the observed gender differences in agricultural productivity. Male farmers tend to have substantially higher agricultural yields than female farmers in the northeast, northwest and south of the country, while in the southwest and west women are as or more productive than men (though not significantly so). 16 Moreover, gender gaps exist not only in agricultural yields, but also in agricultural production. Yields, namely the value of output per hectare of land, allow us to examine the efficiency of production, while the total value of production is an indication of farm earnings. In contexts where women farm smaller amounts of land, yield gaps can underestimate the magnitude of gender differences. Moreover—beyond productivity—agricultural earnings are an important development and policy-relevant outcome, and women’s production may be subject to different constraints compared to their productivity. The gender gaps in the value of agricultural farm output produced are large and positive (to the disadvantage of women) in every survey, whether looking at differences between households or within households. They range from 11.1% to 21.6%. Figure 5: Gender Gaps in Agricultural Production The regional patterns observed are similar to those for agricultural yields (with the largest gender gaps in the northeast, northwest and south of the country), though Figure 6 shows substantial gender disparities in the value of agricultural production are also present in the eastern part of the DRC. 17 Figure 6: Difference in Agricultural Production by Region Gender gaps in entrepreneurship profits Non-agricultural entrepreneurship is the second-largest sector of employment in the DRC, and thus an important area in which to examine gender disparities. Gender gaps in business profits are large and significant across all data sources (Figure 7). They range from 45.9% lower profits for women in the SME survey to 66.5% lower profits in 1-2-3 Survey. 18 Figure 7: Gender Gaps in Business Profits Appendix Figure 2 and Appendix Figure 3 depict gender gaps in business revenues and costs respectively. They show that while women’s business costs are lower than men’s, their cost - savings is outweighed by their much lower revenue. This leads to a large and significant profit gap for women. Lower costs for female-run businesses are due both to their smaller size and to differences in the type of operations. In the 1-2-3 Survey, women businesses have 70% fewer paid workers. Women are also more likely to operate home-based businesses and not pay rent and utilities for operating spaces. Turning to regional differences in business profit gender gaps, Figure 8 shows that they are geographically concentrated in somewhat different areas of the country compared to agricultural gender gaps. While similarly to the agriculture sector, gender disparities in business profits are markedly high in the southern and eastern regions, substantial gender gaps in entrepreneurship are also observed in the western and central regions of the country where agricultural gender gaps are low. 19 Figure 8: Gender Gaps in Business Profits by Region Gender gaps in wage earnings As mentioned in the introduction, less than 10% of women in the DRC are wage or salaried workers, compared to nearly 25% of men. Wage employment is thus an important sector in which to investigate gender differences. The gender gap in wage earnings is significant in the nationally representative 1-2-3 Survey, where it stands at 77%. The estimate is similar and only somewhat lower in the Growth Poles survey, where the gap is approximately 68%. The gap is lower in the Childcare survey, which was conducted in rural villages in which only 4% of either men or women had any wage earnings (Table B8). 20 Figure 9: Gender Gaps in Wage Earnings Unlike for the other outcomes considered in this report, Figure 10 shows that there is no meaningful variation in wage earnings gender gaps, with men earning substantially more than women across the entirety of the DRC. Figure 10: Gender Gaps in Wage Earnings by Region 21 How do these gender gaps in wage earning vary by sector? Among those employed and working for non-household firms and organizations in the 1-2-3 Survey sample, the largest number of women are employed in private enterprises or in public administration (39.7% and 38.3% of the total, respectively). A further 17.6% of non-household employed women are employed in public enterprises, while 2.9% are employed in different associations. Among men working for non- household firms and organizations, the largest number are also in private enterprises or in public administration (39.6% and 36.7%, respectively). A further 16.9% of men are employed in public enterprises, while 5.2% are employed in different associations. Figure 11 shows that the gender gaps in wage earnings are largest in private enterprises, followed by associations (which include NGOs, cooperatives and religious organizations). Public enterprises have smaller gender gaps of 10% or under. This may be because there is better enforcement of pay regulation in the public sector, a pattern we observe internationally. Indeed, de jure protections in the Constitution as well as the labor code prohibit discrimination in the right to work and renumeration on the basis of gender. The largest gender gaps in wage earnings— averaging nearly 80%—are observed in the private sector. Programming aimed at shrinking gender inequality in the DRC would thus do well to integrate a strong focus on the private sector. Figure 11: Wage Earning Gaps by Sector Summary 22 • Significant gender disparities exist in the DRC in labor force participation, as well as in outcomes across the three main economic sectors: agriculture, entrepreneurship and wage employment. • Considering only nationally representative data, women are 6 to 8% less likely to work compared to similar men. • Within agriculture, which employs over two-thirds of women in the DRC, the production of women farmers is 18% lower than that of men and their productivity is 11% lower. • The most striking disparities are in off-farm earnings outcomes. The largest gaps are found within wage earnings (77%) followed by business profits (67%). Gender disparities are particularly pronounced in the private sector. Next, we turn to what shapes these differences in outcomes for women vs. men in the DRC. 23 3. What are the Drivers of the Gender Disparities in the DRC? To understand the factors driving the gender gaps in economic outcomes described in Section 2, this section uses Kitagawa-Oaxaca-Blinder decompositions. This analytical approach isolates factors contributing to gender gaps and thus determines whether different access to and quantities used of resources like credit, assets, inputs, and education —or differences in the returns to these resources—drive the gaps in women’s economic outcomes. Identifying the main factors driving disparities between men and women is key to determining which constraints should be prioritized by policymakers. While the results cannot be interpreted as causal—since they are not experimental in nature—we denote variables that are both theoretically and empirically linked to greater gender disparities as ‘drivers’ of the gaps. This is in line with conventionxvi. We discuss our choice of variables and the implications of this choice further in Appendix C. Box 2: Decomposition Methodology The Kitagawa-Oaxaca-Blinder decomposition methodology is widely used in economic analysis to isolate the factors contributing to gender gaps in agricultural productivity and wages, among other outcomes. The methodology decomposes the gender gap into two ma in components: an endowment effect and a structural effect. The endowment effect captures the difference in the levels of resources that women have relative to men, such as education, fertilizer, or amount of credit. Policies and programs may diminish the endowment effect by ensuring equal access to and use of the resources across genders. However, even when men and women have access to the same quantity and quality of resources, they may not achieve the same results: the structural effect refers to the portion of the gender gap that exists because of differences in the returns on resources. For example, the structural effect captures the difference in agricultural output per hectare that women and men obtain for every additional unit of inputs used on the land, given the same levels of education, equivalent use of fertilizer, or equal amounts of credit. Discrimination, social norms, and institutional constraints all perpetuate the structural effect. In this report, the endowment effect is referred to as “levels,� while the structural effect is referred to as “returns.� Appendix C provides additional technical details on the Kitagawa-Oaxaca-Blinder decomposition. This decomposition method is used to identify the key variables associated with the economic gaps described in Section 2: labor force participation, agricultural production and productivity, profitability of businesses, and wage earnings. The tables below show the results of the decomposition analysis, differentiating between variables that matter because women have less of a productive resource (‘levels’) and variables that matter because women fac e a lower economic benefit compared to men even when they have the same level of that resource (‘returns’). The distinction between these two types of drivers is explained in more detail in Box 2 and Appendix C. Only variables that widen the gender gap in at least one survey—to the disadvantage of women—are included. 24 In each table, teal is used to highlight variables that widen the gender gap in a given decomposition analysis. Summary statistics for each variable are provided in Appendix B. Grey is used to highlight any variables that were not included in a survey-specific analysis. Drivers of gender gaps in labor force participation Table 1: Significant drivers of the labor force participation gender gap 1-2-3 Survey DHS Levels Returns Levels Returns Individual married/cohabiting Individual attended or completed high school Female HH head HH size HH dependency ratio HH owns radio HH owns television HH owns cellular phone HH owns bicycle/motorcycle Table 1 highlights the drivers of gender gaps in labor force participation. As summarized in Section 2, there is a 6-8% gender gap between men and women in labor force participation. Being married, having completed high school, a higher dependency ratio, and owning a radio, television and cell phone emerge as significantly associated with larger gender gaps in labor force participation. Across the two nationally representative surveys that exhibit a gender gap in labor force participation, the dependency ratio—that is, the ratio of children and the elderly relative to working-age adults in the household —is the primary driver of women’s lower engagement in the labor market. Indeed, women are more likely to face a higher dependency ratio: Table B6 shows that women in the DRC live in households with a dependency ratio that is 17% higher compared to men. This higher dependency ratio can translate into increased care burdens and time constraints for women, reducing their labor force participation. Women also face lower returns to marriage compared to men. This means that being married increases men’s labor force participation more than it increases women’s. This emerges as an important driver in both surveys. Indeed, in the DHS survey, women’s labor force participation can almost entirely be explained by women’s lower returns to marriage, and this same factor is the 25 largest contributor to the gender gap in the 1-2-3 Survey. Women may face lower economic returns to marriage because they are expected to adopt or intensify their role as caregivers upon marriage—even given the same number of children —or generally have decreased economic control. On the other hand, men may obtain a higher economic gain from marriage as they are more able to work when there is an additional person in their household taking on care work and chores. This important result for women’s economic empowerment in the DRC is explored further in the deep dive on women’s agency. Finally, women also face lower returns to their education in terms of their labor force participation, as well as to the ownership of assets such as cell phones in the household. That is, men participate more in the labor market as a result of schooling and household wealth compared to women, even when they have the same level of education. These associations may be undergirded by gendered social norms. For example, even when women are as educated as men or live in wealthier households, the men are still expected to be the primary breadwinner. It could also suggest that women are less able to reap returns from phone ownership because they have smaller networks they can leverage or because they have lower digital literacy. Drivers of gender gaps in agricultural outcomes As highlighted in Section 2, women’s agricultural productivity is 11 to 48% lower than men’s in the DRC (depending on the data source). Table 2 presents the drivers of this gap, which include marriage status and education levels. Being married and having attended or completed high school are positively related to agricultural productivity in the DRC, but female farmers are significantly less likely to be married or have attended high school. For example, in our Growth Poles sample (Table B2) female farmers are half as likely to have attended high school compared to men: 32% versus 64%. Having a female household head also emerges as a major driver of lower agricultural productivity in the Childcare Survey. Female farmers also lower returns to the number of paid workers per plot. Women may face these lower returns to agricultural labor because workers work less hard for them, because they cannot afford to pay as much as men for effective farm workers, or because they are less productive supervisors (perhaps because they often multitask farm supervision with taking care of their children).xvii Overall, the largest contributor to the gender gap in agricultural productivity is w omen’s concentration in food crops versus higher-value cash crops (defined as palm oil, coffee, cocoa, sugar cane, rubber, butter fruit, acacia, and mango, orange, and avocado fruit trees). Indeed, women’s rates of cash crop cultivation vary from half that of men in the 1-2-3- Survey and Childcare survey to three-quarters that of men in the Growth Poles survey (Tables B1, B4 and B7). These lower rates of cash crop cultivation are linked to lower agricultural productivity for women in every single data source. 26 Table 2: Significant drivers of the agricultural productivity gender gap 1-2-3 Survey Growth Poles Childcare Levels Returns Levels Returns Levels Returns Manager married/cohabiting Female HH head HH head attended or completed high school HH owns bicycle/motorcycle HH number of plots Plot surface (ha) Number of paid workers per plot Use of pesticides Cash crops plot Table 3 presents the drivers of the gender gap in agricultural production. While women’s efficiency relative to men is important, so are their incomes, which are better captured through agricultural production. In terms of differential levels, the higher dependency ratio, lower cell phone ownership, lower number of plots, lower number of unpaid or household workers and lower likelihood of rice cultivation emerge most frequently as associated with a larger gender gap. Unlike for agricultural productivity, the dependency ratio and the number of unpaid laborers systematically emerge as drivers, while women’s lower returns to hired labor remains a constraint. While women faced lower returns to cell phone ownership in terms of labor force participation, within agricultural production lower levels matter too, perhaps because cell phones are an important source of agricultural price and extension information. Moreover, the number of plots a household has is naturally positively linked to agricultural production, and women report significantly fewer plots compared to (approximately 0.06 fewer in the 1-2-3 Survey, Table B5). The gender gap in agricultural production is also exacerbated by a lower number of unpaid workers or household members working on plots. Naturally, having a higher number of workers results in higher agricultural production; when a man is the head of a household, plots have an average of 3 unpaid workers working on them, versus 2.4 unpaid workers for women. Lastly, the proportion of female-headed households farming rice, an increasingly profitable crop in the DRC xviii, is lower: 15.3% versus 11.4%. 27 Lower ownership of bicycles and motorcycles emerges as related to the gender gap in agriculture. 27% of male household heads own a bicycle or motorcycle, compared to 8% of female household heads (Table B5) which can result in transportation constraints and exacerbate existing time burdens, as plots are on average over 40 minutes away.3 Table 3: Significant drivers of the agricultural production gender gap 1-2-3 Survey Growth Poles Childcare Levels Returns Levels Returns Levels Returns Manager attended or completed high school HH size HH Dependency ratio HH owns cellular phone HH owns bicycle/motorcycle HH number of plots Plot surface (ha) Number of unpaid//HH workers on plots Number of paid workers on plots Use of pesticides Use of fertilizers Cash crops plots Peanut plots Rice plots The size of farmed plots is another significant driver of women’s lower agricultural production in the Growth Poles survey. Women’s farmed plot surfaces are substantially smaller on average: 0.27 ha versus 0.5 ha (Table B2). This issue of women’s lower control over land is explored more in - depth in Section 4. Women are also less likely to use non-labor inputs such as pesticide. The use of pesticide is strongly linked to the value of agricultural production in the DRC. While pesticide use is generally very low, being used on only 2.8% of plots managed by men in the Growth Poles survey (Table B2), this proportion is still four times higher than for plots managed by women. Lastly, women are less likely to farm cash crops. Though crop choice is not as consistent a driver 3 This information comes from the Growth Poles survey, since it is the only data source with precise distance information. 28 in agricultural production compared to agricultural productivity, it does widen the gender gap in the Childcare survey. Male-managed plots are twice as likely as female-managed plots to contain a cash crop (Table B8). Instead, women are more likely to farm food crops like cassava. Similar to labor force participation, female farmers witness less economic benefit from having the same level of productive resources compared to men (that is, they face lower returns to productive inputs). While the patterns in lower returns are less systematic than the patterns of lower access and use reviewed above, this result is observed for three types of variables. First, women face lower returns to farming common crops (like peanuts and rice). Second, and relatedly, men obtain more agricultural production from the same rate of fertilizer use. This could be because women apply incorrect quantities of fertilizer (or in an incorrect way) due to lower agricultural skills, or because they are sold lower quality fertilizer. Third, women achieve lower agricultural production from the same number of paid agricultural workers compared to their male counterparts. Drivers of gender gaps in entrepreneurship profits Table 4 shows the drivers of the gender gap for entrepreneurship profits. As shown in Section 2, gender gaps in this sector are very large, ranging from 46% to 67% depending on the data source. Women employing fewer workers emerges as the single most consistent driver of the gender gap. As in agriculture, a higher number of workers is associated with higher profits. Women are systematically disadvantaged on this metric across all surveys (the sample inclusion criteria are described in Appendix A). In the SME survey, women employ 2.8 workers compared to men’s four workers. When looking at unpaid workers in the 1-2-3 Survey, women also employ significantly fewer (Table B6), though this difference is smaller. Lower values of physical capital, utilities and intermediate inputs also contribute to widening the gender gap in business profits. For example, the value of physical capital (land, buildings, machinery) in female-run businesses is CDF 51,000, compared to CDF 168,000 in male-run businesses (Table B6). Likewise, the value of intermediate inputs, which are goods bought by businesses and processed for sale to the public, is CDF 26,638 in women’s businesses compared to men’s CDF 41,152. Similar to the agriculture sector, women face lower returns to the number of workers as well as to cell phone ownership in the household. That is, they not only have fewer employees working for them, but they also get less profit from the same number of employees. Lastly, they also face lower returns to marriage in terms of their business profits. A large body of work shows how women's business decisions are influenced by members of their household, especially their spouse, and that these intrahousehold dynamics contribute to gender gaps in entrepreneurship outcomes. xix Qualitative findings from Goma underlined that while there are some reports of men’s support and positive impact on their w ife’s business, more often they are associated with undermining her success through ill temper towards her or coopting her business profits. Responses highlight a common belief that this behavior stems from husbands’ insecurity about his wife’s income as a threat to his role as dictated by gender norms. 29 Table 4: Significant drivers of business profitability gender gap4 Growth Poles SME 1-2-3 Survey Levels Returns Levels Returns Levels Returns Manager married/cohabiting HH owns a cellphone Business location in a building/rental Number of unpaid workers (all workers in SME) Value of paid labor (million CDF) Value of intermediate inputs (million CDF) Value of paid utilities (million CDF) Value of physical capital Drivers of gender gaps in wage earnings Table 5 shows the drivers of the gender gaps in the third-largest sector for women’s employment in the DRC, after agriculture and entrepreneurship: wage earnings. The variables most consistently related to the gender gap in wage earnings levels are women's lower attendance or completion of high school and lower radio, cell phone and bicycle/motorcycle ownership. Having attended high school is strongly linked to higher wage earnings, but across all data sources, women are around 24 percentage points less likely to have attended or completed high school (Table B4, B7 and B9). In addition to differences in educational attainment across men and women, in the 1-2-3 Survey women also face lower returns to education. This means that the same level of education translates less into increased wages for women. 4 Detailed information on costs and revenues of househo ld businesses is not available in Childcare and DHS survey, so this part of the analysis uses 1-2-3 Survey, Growth Poles, and SME survey. Most of the variables are the same across datasets, with some differences. For example, breakdown of business costs is not available in SME survey. Also, since SME survey was focused on households with female businesses, this analysis uses only households where there are at one female and one male business. 30 Table 5: Significant drivers of gender gap in wage earnings 5 1-2-3 Survey Growth Poles Childcare Levels Returns Levels Returns Levels Returns Individual married/cohabiting Individual attended or completed high school HH dependency ratio HH owns radio HH owns television HH owns cellular phone HH owns bicycle/motorcycle Moreover, w omen’s lower earnings are associated with lower asset ownership (radio, cell phone and bicycle/motorcycle). This may indicate that women’s wage-earning opportunities are dampened by lower access to information (radio and cell phones) or transportation (bicycle/motorcycle). Decomposition analysis is however descriptive in nature, and therefore lower asset ownership may be a consequence of women’s lower earnings rather than a cause. We also observe that women face lower returns from these assets, suggesting that lower agency or lower complementary skills may limit how much women’s employment benefits from househ old assets and wealth. Though only present in one data source, another important driver that emerges is the household dependency ratio. A higher dependency ratio is strongly linked to lower wage earnings in the 1-2- 3 Survey, and women face a 17% higher dependency ratio (Table B7). Lastly, and as observed for other sectors, women witness systematically lower returns to marriage in terms of their wage earnings. As discussed above, this may be due to norms around women and men’s roles in marriage and women’s relatively lower agency in their households. These underlying constraints to women’s economic participation are explored in the next section of this report. The main drivers of the gender wage gap are robust to the inclusion or exclusion of household assets, occupational sector, belonging to a union, and religion. Unsurprisingly, the association between the gender wage gap and education disappears when the level of qualification on the job 5Wage employment is defined as working in a paid employment for someone outside the household. In 1-2-3 Survey, wage income is the total monthly wage in current employment. In Childcare survey, wage income is the monthly wage over the 12 months preceding the survey, while wage income in Growth Poles is the total wage o ver the preceding 12 months. The DHS and SME survey do not provide information on wages. 31 (manager, qualified, semi-qualified, etc.) is included, indicating that this relationship is mediated by the level of qualification.6 Summary • Women in the DRC face sizable constraints limiting their labor force participation, agricultural production and productivity, business profits and wage earnings. • A higher dependency ratio—the number of children and the elderly relative to working- age adults— is the primary driver of women’s lower labor force participation, highlighting t he importance of tackling women’s higher care burdens and time constraints. • The main driver of women’s lower agricultural productivity is their lower cultivation of cash crops (such as palm oil, coffee, cocoa, sugar cane, and rubber). Turning to earnings instead of efficiency, w omen’s lower value of agricultural production is driven by a wider range of factors. These drivers include women’s higher dependency ratios, lower cell phone ownership, less land, lower number of workers and a lower likelihood of rice cultivation. Women also face lower returns to both labor and fertilizer. Attaining gender parity in Congolese agriculture will thus not only require that women have the same access to productive inputs as men, but also that their returns to these inputs are equalized. • In entrepreneurship, a lower number of workers emerges as the single most consistent driver of the gender gap. Businesswomen’s lower levels of capital compared to their male counterparts also matter greatly. • Lower educational attainment is the primary driver of women’s lower wage earnings. • In addition to the dependency ratio, women’s lower access to certain key assets as well as lower returns to marriage are consistent drivers across sectors. 6 Given the potential sensitivity of wage-earning decompositions to the inclusion of particular covariates, Table D11 presents variations on the analysis adding and removing certain variables additional that may be highly correlated with each other. The results show that the association between gender gaps in wage earnings on the one hand, and education and the dependency ratio on the other, does not depend on the inclusion (or removal) of household assets in the analysis. In addition, women’s education levels, as well as the household dependency ratio and cell phone ownership, remain as drivers when including the worker’s occupational sector and other characteristics ( such as religion and union membership). Lastly, the relationship between educational attainment and the earnings gender gap appears to be mediated by the worker’s level of seniority (whether the worker is a manager or supervisor and their degree of qualification, shown in the last column of Table D11). As might be expected, the worker’s qualification in the labor market is strongly linked to whether they attended or completed high school: 95% of managers and supervisors have completed high school, while only 88% of qualified workers, 77% of semi-qualified workers and only 65% of non- qualified workers have. 32 4. Diving Deep on Three Constraints Underlying many of the gender gaps examined in the previous two sections are fundamental constraints related to women’s low level of control over land, limited agency, and exposure to many sources of risk and uncertainty. Women manage and own less land than men, which, as seen above, has important implications for agricultural productivity. Restrictions on women’s agency hamper their ability to make important decisions in their own lives and can limit their opportunities for economic advancement. Finally, women face physical, financial, and institutional risks that lower the returns they receive from their investments and can prevent them from engaging in economic activities in the first place. These three drivers are explored in more detail in what follows. It is important to note that the deep dives do not necessarily represent the full extent of priorities above and beyond the drivers identified in the decomposition analysis. Rather, they appear as key constraints underlying some of the gaps identified in the econometric analysis, which the quantitative analysis signaled as important but was unable to explore further given a lack of data. Deep Dive 1: Control over Land Women have less control over land compared to men in the DRC In the DRC, as throughout the developing world, women own less land and are less able than men to assert equal property rights, constituting a significant barrier to economic empowerment, gender equality, financial inclusion, and food security. Within the DRC, women make up more than half of the agricultural workforce, yet men represent 91% of agricultural landholders in the country. As shown in Section 3, women’s cultivation of a lower number of plots and smaller plot s izes are significant drivers of the gender gap in agricultural production. These differences are explored further in the data using the nationally representative 1-2-3 Survey. While there is a significant but small difference in the number of plots cultivated by female- versus male-headed households (a 0.06 difference, as mention in Section 3), differences in plot-level ownership are striking. Indeed, women only own about a quarter of plots (28%) in the nationally representative 1-2-3 Survey, while men own 72% of plots (Figure 7). While the value of agricultural production is only available at the household-level in the 1-2-3 Survey, this stark gender disparity in plot ownership may indicate that national agricultural gender gaps are likely larger than those shown in Figure 3 and Figure 5. Further investigation demonstrates that women might not be able to invest fully in their land. For instance, the same survey shows that women and their spouses jointly manage 80% of all plots. By contrast, only 18% of plots are solely managed (either by the woman or her spouse alone). This indicates that women are doing a significant portion of work on plots, despite only owning a small subset of these plots. Moreover, as previous research has indicated, women who jointly manage land with their partner might not necessarily have an equal voice or decision-making power regarding the day-to-day management of plots, which further limits their agency.xx 33 Figure 12: Gender Gaps in Plot Ownership 80% 72% 70% 60% 50% 40% 30% 28% 20% 10% 0% Female Male Female Male Source: 1-2-3 Nationally representative survey conducted in 2012. Why does women’s land tenure matter? Land is a valuable productive asset that serves as a primary source of food and income for many rural farmers. In addition to the immediate value of land for economic and food security, secure land rights are a vital element of women’s economic empowerment and security. Research has demonstrated the positive impact that secure land tenure has on increasing investments into land through agricultural inputs and improvements by both male and female farmers. xxi Possession of valuable and immovable assets, such as land, are frequently preferred by banks for use as security interests, xxii and required for individuals to access credit which can be used to re-invest in agriculture and grow entrepreneurial businesses. xxiii Without an asset such as land, many women are ineligible for credit and are unable to productively invest in agricultural production or other businesses.xxiv In addition to financial gains, women’s land rights can increase bargaining power within the household and decrease women’s dependency on their partner. xxv Evidence shows that land insecurity discourages farmers from investing in agricultural production and discourages off -farm activity for fear of losing control of their land. xxvi For example, married couples in Zambian villages where a widow does not inherit the land upon her husband's death are likely to invest less in their land while the husband is alive, suggesting a preference for potentially more "secure" investments.xxvii 34 Our analysis highlights this tenure insecurity, showing that close to 80% of plots lack any written documentation regarding property ownership. Plots owned or managed by women are more likely to not have any form of documentation (85% for women versus 81% for men). Interestingly, both male and female plot owners feel secure that no one can contest their land ownership or use, despite the lack of documentation on land rights. However, women are slightly less confident than men in the possibility of winning a dispute over their land rights or use. In the 1-2-3 Survey, 53% of women responded that it was implausible that someone would challenge the use of ownership of their land compared to 60% of men. The lack of land tenure security is especially problematic in conflict-affected areas. Secure land rights both ensure one’s ability to return to their homes and property after fleeing violence, and reduce the likelihood of further conflict over disputed property upon their return. xxviii Most smallholder farmers in the DRC have no legal documentation of their land tenure. Residents who flee violence and return often find others occupying their land. The lack of standard, enforceable governing land principles, lack of land records, and the widespread nature of displacement can instigate conflict over land tenure.xxix Women who fled violence to return later often face the increased challenge of male relatives as well as other community members claiming their land.xxx How do women own and secure land in the DRC? Purchase The constitution of the DRC grants equal rights to property for both women and men. Additionally, The Family Code dictates that daughters and sons have equal rights to inherit land and that both male and female spouses have an equal right to inheritance. xxxi However, the Family Code further states that “management� or control of marital property is entrusted to the husband, unless spouses specifically agree to manage their property separately at the time of legal marriage registration.xxxii As a result of such contradictory and discriminatory legal protections, gender disparities in land tenure are prevalent under both statutory and customary law. After DRC gained its independence, all land in the country was nationalized as government property, of which the state can make land concessions on a case-by-case basis. The Land Act of 1973 permitted customary law, enacted through traditional authorities such as chiefs, who retain power through patrilineal succession, to govern unallocated rural land's use rights and grant land concessions to individuals..xxxiii Throughout the country, up to 97% of the land is distributed at the customary/community level.xxxiv The 1-2-3 Survey confirms that customary land ownership is the most prevalent type of ownership. Only 8% of plots have concessionary ownership granted by the government. The majority of plot owners’ land rights are granted by local chiefs (60%), although women are at a slight disadvantage compared to men; 57% of female-owned plots are recognized by local chiefs, versus 62% of male owned-plots. Of those with land tenure, the ways in which both men and women acquired their land rights is depicted below in Figure 13. 35 Figure 13: Women & Men’s Pathways to Land Tenure Customary concessions require petitioning the customary chief, although women's rights are protected under the constitution and other laws, the traditional chiefs regularly enforce customary rulings which undermine women’s constitutional rights. Through this system, women are most often able to gain access to land through their male relatives or their husband, yet are denied ownership or control.xxxv Women are marginalized in seeking land tenure at the customary level by customary practices, social norms, lack of economic resources, and limited access to seek justice.xxxvi An additional factor contributing to women's lack of tenure is a dearth of awareness by women and communities about women's rights, undermining their ability to advocate. xxxvii As opposed to customary land rights obtained through a chief, formal land concessions granted by the government are better documented and offer increased land tenure security if ownership is ever contested. To apply for a land concession directly from the national government, the individual must bear a significant transaction cost. Distance is a significant factor as the applicant must travel great distances to apply at the official land administration offices. Applications are highly discouraging for illiterate individuals, placing a higher burden on women as the literacy rate of women is 22% lower than that of men. xxxviii Once the application and documents have been submitted the land administration requires an in-person investigation followed by a survey of the land. The survey is cost prohibitive for most with set prices of $50 for residential plots and $100 per day for agricultural land.xxxix Finally, throughout the application process, fees are required for services and approvals, making the process very expensive. xl In particular, women are less likely than men to have the financial capital, political influence, and literacy to register land through the state-regulated system. xli Due to the significant expense of obtaining a land concession, landowners may try to capitalize on their investment through sharecropping and employing predatory contractual conditions. However, purchasing access rights to agricultural and residential land by renting is generally seen as a culturally acceptable option for women who have had to leave their husbands. For internally displaced (IDPs) women, renting is one of the only ways to access land to cultivate while they are displaced. xlii 36 Family Allocation The 1-2-3 Survey shows that across men and women, only 2% of plots were acquired through marriage. This low acquisition rate is due in part to the legal framework surrounding marriage and property ownership. Upon marriage, couples must choose one of three formal marital practices to ensure equal property ownership between spouses. The practices, or ‘regimes’ include: Separation of Property, in which each maintains their assets; Communal Acquisition, in which each maintains ownership of property acquired before marriage and shares assets acquired in marriage; and Universal Community Property, in which both share all assets acquired before and during the marriage. However, regardless of which regime is chosen, it is presumed that assets obtained in marriage, such as land, are registered in the husband’s name and are under his control, except for items intended for the wife’s personal use, such as clothes, jewelry, and work tools of little value.xliii In eastern DRC and other rural areas, women are usually granted access to their husband's land once married, yet often they cannot make agricultural decisions. In some rural areas, women are only allowed to purchase land in their name if they are unmarried. xliv If married, custom dictates that she registers the land in her husband's name regardless of who pays for it. In some areas of South Kivu, this practice extends until after a woman is widowed. She is expected to put the land in either a son's name or that of her deceased husband .xlv 7 The 1-2-3 Survey further shows that 30% of plots were acquired through inheritance. The Family Code affords female children and female widows equal rights as their male counterparts to inheritance. However, customary practice consistently discriminates against both. Daughters, often viewed as temporary family members that will be married into, and receive land from, her husband’s family, are left out of family inheritance in favor of the sons of the household. In situations where a daughter is allowed to inherit, her rights are less than those of her male siblings, often restricted to use rights rather than tenure. xlvi Additionally, parents in the DRC often do not obtain a birth certificate when their children are born. Without a registered birth certificate, women and girls are unable to advocate for their right of inheritance. xlvii Once married, women continue to be seen as temporary family members by their husband’s family. If their husband dies, the widow is often left with nothing while her in-laws claim all property and assets. xlviii Often marriages, especially in eastern DRC, follow traditional customs and do not receive marriage certificates, undermining widows’ ability to plead legal rights of inheritance. xlix As legal protections are in place, promising equal rights to inheritance and land ownership, it is imperative that women are able to advocate for and realize their legal and constitutional rights. A significant barrier to women’s ability to inherit land rests on a lack of birth certificates and marriage certificates, used to verify their rights. There are currently initiatives in the DRC encouraging parents to register their children l and spouses to register their marriagesli in order to obtain birth and marriage certificates. Since many in the DRC are unaware that customary land rights are at odds with national law, the previously mentioned initiatives are reinforced by campaigns raising awareness regarding land rights and how to assert those rights. lii These grassroots campaigns are backed by evidence from an impact evaluation pilot in Côte d’Ivoire.liii Other evidence from Uganda suggests that using incentives such as waiving land registration fees for co-tilting between husbands and wives can be a promising policy option for increasing 7In some areas of the country, widowed women cannot inherit property because the property must go to the deceased husband's children, even if born out of wedlock (US State Department, 2019). 37 women’s tenure security and agricultural investment.liv Section 5 provides further details on these interventions. Deep Dive 2: Agency Women’s have lower agency than men in the DRC Women in the DRC are consistently less empowered than men to make decisions affecting their lives. Agency is defined as “the ability to define goals and act on them, free of violence, retribution or fear�.lv As detailed below, documented constraints to agency include, among others, gender- based violence, lack of control over reproductive and sexual rights, lack of property ownership, lack of political representation, and lower educational attainment. lvi Conflict compounds issues of agency in the DRC, further endangering women by increasing their risk of land and food insecurity, sexual violence, rape, and exploitation.lvii Inequality Under the Law Women’s agency is vital to development outcomes as a critical driver of gender equality, economic development, and health outcomes. In the DRC, many legal protections exist to protect women's agency, yet due to customary and traditional practices, those legal protections are insufficient. Article 14 of the constitution certifies gender equality. The article bans discrimination against women, pledges to fight violence against women, and guarantees gender parity in its national, provincial and local institutions. However, alongside constitutional protections of gender equality, exist contradictory and discriminatory laws. For example, Article 444 of the Family Code, which has since been amended and is discussed below, delegated husbands as household heads and stipulated that wives were legally obligated to obey them. lviii In addition to contradictory statutory laws, customary and traditional practices often supersede statutory law in practice, especially in rural areas where over half of the population lives. lix In 2010, the national gender policy action plan was validated, which included comprehensive measures promoting gender equality within the family, the community and the economy, and in decision making. The policy specifically highlighted that economic development policies must be designed and implemented with gender parity and equal access to beneficiaries in mind. Yet, there is a severe disparity in political representation between men and women. lx As mentioned in Section 2, laws under the constitution and the Labor Code protect workers from discrimination and guarantee equal remuneration for equal work. However, gender-based discrimination in employment persists along with a gender wage gap and a disproportionate number of women in the informal sector.lxi Unfortunately, amendments are not always sufficient and are often underenforced. For example, in the family code, Article 444 delegating men as heads of households and requiring women to obey their husbands, was amended so that women are no longer required to obey their husbands and could enter legal contracts without their consent. lxii Yet, as men are still recognized as the heads of household, discriminatory practices remain. For example, married women in public service careers cannot receive equal social benefits for themselves or their family because it is the head of household who is entitled to extend social benefits from his work to his family. lxiii 38 Inequality in access to education A significant impediment to realizing gender equality is the severe gender gap in education which produces fewer qualified women than men for employment outside of the agricultural sector and negatively impacts advocacy outcomes as fewer women are informed or are able to access information regarding their legal rights. The DRC has one of the worst gender disparities in education in the world at the secondary level, impacting women’s wage earnings and other important economic outcomes as shown in Section 3. Girls’ education in the DRC has long been compromised by conflict, poverty, and gender discrimination.lxiv Specific barriers to girls include gendered social norms, risk of violence, and both financial and opportunity costs of education. Traditional gender roles for women do not necessarily require formal education, and consequently education is less valued for daughters than it is for sons. Some of these gender roles and expectations might be propagated by mothers. Indeed, our analysis of EMAP survey data in Figure 14 shows that women are more likely to want a higher level of education for their sons compared to their daughters. For instance, 62% of women desire a college education for their daughters versus 72% for their sons. Figure 14: Differential Educational Aspirations 39 In 2015 the government passed Act No. 15/013, prescribing equal access to education and vocational services.lxv Despite the government’s commitment to fund education for all children, parents remain responsible for purchasing materials, uniforms and subsidizing teachers’ pay. One possibility is that poor households often prioritize a son’s education because it is p erceived to be a better investment, and girls are, therefore, frequently withdrawn to support household chores or income generation. Finally, parents may also choose early or forced marriage for their daughter in exchange for a dowry.lxvi Social norms that constrain women’s agency & economic advancement One of the fundamental constraints to women’s agency are social norms that define restrictive roles and responsibilities for men and women. The qualitative case study of women in the economy in Goma identified three interrelated social norms that constrain women’s economic opportunities and shed light on why women consistently face lower economic returns to being married compared to men. First, there is a common and core expectation that men ought to be the primary household breadwinner. This means that their earnings are expected to pay for major household expenses, but it does not mean that men are expected to the be sole provider. In fact, women in Goma reported that they earned respect by contributing income to the household. For example, a wholesaler of plastic shoes explained, “a woman is respected by her husband if she also brings money to meet household needs.� This social norm, therefore, does not completely discourage women’s work, but it can have important implications for intrahousehold dynamics. Given the current economic context, many men fail to achieve the breadwinner ideal. It is common for women to contribute substantially to meeting their household needs. There is, therefore, a mismatch between ideas about what ought to be, and the realities of how things really are. This mismatch can create tension within households and families, as well as hinder women’s opportunity for higher productivity work. Businesswomen talked about some husbands feeling threatened by their wife’s relative economic success. A woman who owns a tailoring business and is the president of several women’s business associations described her own husban d as a good model, but noted that, “many men have complexes in front of their wives. They think that if his wife has more money than him, she will dominate him. So men prefer to hinder their wives because of that.� The data contain many instances where men used their wives’ profits without their consent, preventing women from reinvesting in their businesses or increasing their savings. Women in a focus group discussion of cross-border traders complained, “There are some women who are helped by their husbands, even if this category represents only 20 percent. The other 80 percent [of husbands] are rather there to cause their wife’s business to fail. These are the husbands who steal their wives’ money, whose only job is to thunder around the house to demonst rate their power as household head.� At the same time, while they may not be perceived as the majority, women also told stories of men helping by providing capital infusions and/or encouragement. Overall, it is clear that the behavior of husbands facilitat es or constrains their wives’ businesses. A second consequence of the norm that associates men with the breadwinner role is skepticism regarding women’s financial management capabilities. Doubts about women’s ability to manage large budgets, combined with their unequal access to collateral, restrict their access to capital for 40 their businesses. Even though women are legally allowed to engage in financial transactions without the consent of their spouse, women reported facing doubts and requests for thei r husband’s guarantee to access loans. Relatedly, men are accorded the social status of household head , meaning that they are viewed as entitled to be the final decisionmaker. As a young entrepreneur stated: “no one here denies that it’s men who are the head of the family.� This is reinforced by the practice of paying bride price to the wife’s family, and the notion that the woman moves to her husband’s home, and not the other way around. The constitution affords the right to marry the person of their choosing to both men and women, and the Family Code defines marriage as a consensual act between a man and woman of legal age, recently amended to be 18 years old. Legally, men are required to pay a dowry to the bride and her family. lxvii Dowry payments are believed to be a significant contributor to the continued prevalence of child marriage,lxviii an extremely harmful practice which causes health risks and higher fertility, and in the DRC is correlated with lower educational attainment, lower wealth, and higher labor force participation. lxix If the couple divorces, the woman is legally required to repay the dowry to her husband, creating a potential economic barrier to divorce on the woman's part.lxx Once married, women often have limited control over household decisions. Collaborative decision-making is seen by some to undermine men’s social status. A newly married young woman pharmacist whose husband is a teacher reported that she and her husband plan together, as instructed by their church. She thinks this is uncommon among households in Goma. When asked what people think of couples who collaborate, she said, “that depends. There are people who will congratulate those couples for living in harmony and others who will oppose them by saying that the wife has used sorcery to dominate her husband because they do not understand how a husband could sit with his wife to plan together....� Some people believe that intrahousehold collaboration undermines the man’s status as household head. Others believe that collaborati on can improve household relationships and welfare and does not necessarily undermine men’s status as household head. Thus, there is disagreement over whether collaborative planning undermines the man’s social status. At the same time, there is little or no rejection of the male household head norm. This is consistent with what has been documented in other studies, for example the EMAP impact evaluation.lxxi Indeed, data from our EMAP survey indicates that women's autonomy in household decision- making is relatively low. Figure 15 shows that husbands are most likely to make final decisions in all domains of decision-making. Specifically, 70% of women report that their partner is the main decisionmaker for their own healthcare needs. This is followed by her choice of making visits to a family member where a little over 50% of women report that their partner decides when she should visit her parents or family members. Finally, 45% of women reported that their partner made decisions regarding money that she made and major household investments. 41 Figure 15: Household Decision Making by Domain Source: EMAP survey. The category “other or another household member makes decisions� correspond to less than 3% of all responses at all levels. Furthermore, intrahousehold collaboration is discouraged by differences in men and women’s spending priorities, which reflect gender-specific social pressures. Research participants report that most women control their own earnings and that both husbands and wives hide income from each other so that they can allocate their income according to their own priorities. Women offered several different reasons to explain this phenomenon. Women claimed to hide income so that they could support their natal family without having to ask their husband for permission or assistance. Others indicated that some men withdraw support for the household if they know that their wife has income, so women have an incentive to hide earnings. Women complained that men hide money so that they can spend it on mistresses and beer. Moreover, a husband’s relatives may discourage him from collaboratively managing resources with his wife out of fear that she will advocate for a reduction in the level of support he offers. This type of social pressure was explained by a young unmarried woman with an agro-processing business. She reported that her sister is the main breadwinner for her household and while her brother-in-law accepts the arrangement, his family creates problems for them. Men also face social pressure from their peers to socialize outside of work hours and they are hassled when they are perceived to be spending too much time at home or collaboratively managing resources with their wives. These disincentives to collaboration and lack of transparency can have important implications for women’s businesses. Women often prioritize secrecy in their financial management strategies. Some women noted that they prefer not to store money at home for fear that their husband will steal from them or interfere in their business. Given the mistrust of formal financial institutions and the fees associated with mobile money, this means that many women resort to carrying their earnings and operating capital on their body, which poses a security risk. Other women enlist the help of friends or family for secretly saving or investing. In a focus group discussion with women entrepreneurs who participate in the same Village Savings and Loan Association (VSLA), the 42 entrepreneurs reported that many members do not tell their husbands when they have received cash from the group. To conceal the money, they will entrust it to another person to save or to invest in a business on their behalf. They report that this is a risky strategy, however, because by its nature the agreement is secret and that leaves women with little recourse when their savings or investments are lost or stolen by their designee. A similar strategy, which presents a similar set of risks, is for women to purchase land or housing in someone else’s name, often a brother. In all of these descriptions of the ways that women hide income, it is evident that women are often willing to incur costs or tolerate certain risks if it means maintaining greater control over their earnings. The norm of male headship, however, does not always translate into a lack of transparency or collaboration. When asked to estimate the proportion of households where the husband and wife sit together to plan, women consistently guessed around 20 percent of households. They noted that it was uncommon, but not unheard of for a husband and wife to have a mutually supportive relationship. Women attributed collaboration to higher levels of education, to positive role models in the family, to teachings on marital relationships from churches or NGO programs, or to a good quality relationship based on love. The interviews with entrepreneurs further suggest that cooperative relationships may be easier to establish and maintain when men have a stead y source of income and, therefore, the breadwinner ideal is not threatened by collaborative decision-making. Overall, the findings of the qualitative research suggest that, at least in the short term, it may be more feasible to decouple the male breadwinner norm from specific patterns of uncollaborative decision-making than to undermine or change the norm. The third social norm with important implications for women’s businesses is the idea that housework and care work are a woman’s responsibility. These norms can dictate women's responsibilities in the household, such as childcare, domestic responsibilities, and growing food for household consumption. lxxii As observed in Section 3, women’s higher care burdens are a consistent driver of economic gender gaps. Unlike the norm about men’s role as household breadwinners, the norm of women’s caregiving is reflected both in actual behavior and in behavioral ideals. Women in the DRC spend more time than men on domestic chores. This limits the time women can invest in their own productive work. Figure 16 shows how couples in our EMAP data share household chores and child-rearing. In Figure 16, the outer circle depicts the how couples share household chore responsibilities while the inner circle demonstrates how childcare is shared between couples. Among couples, 83% of women responded that they were the primary person to perform all household chores, and 64% reported being solely responsible for childcare. Only 13% of men were reported as primarily responsible for household chores, with only 9% for childcare responsibilities. Childcare is shared among 27% of couples, and household chores are shared by 4%. 43 Figure 16: Gendered Division of Domestic Responsibilities 4% 27% Household Chores Childcare Responsibilities 13% 9% 64% 83% Woman Both Partner Several of the female entrepreneurs who participated in the qualitative study reported that domestic work constrains women’s economic activities because of competition for their time. Others took for granted their responsibility for domestic work and therefore any tradeoffs went essentially unnoticed. Those women who did comment on the challenges of combining income generation and domestic work spoke about difficulties at specific times in their lives. In particular, women mentioned that they are pulled away from work most often when pregnant or caring for young children. This challenge is especially acute for women who cannot afford to hire domestic help and do not have family nearby. A tailor noted that, “domestic responsibilities can affect women’s ability to earn income because those tasks require a lot of women’s time. Especially those who have small children and no one to care for them.� Several women also mentioned that they had to take time out of work or make adjustments to their work during and/or immediately after their pregnancies. For example, a new mother with a salon and moto taxi service noted that she felt the moto driver and hairdressers were cheating her because she wasn’t supervising them closely after she gave birth. She decided to change the nature of their contracts so that they would pay her a fixed amount, which required less supervision. Women also reported challenges with arriving at work early enough or remaining open late enough, given their responsibilities at home. Some women noted that domestic responsibilities were one reason to operate a business from home, although more reported the exact opposite—that operating a business at home was a bad idea because of all of the distractions. Another aspect of the attribution of responsibility for caregiving to women is the expectation that women will use the money that they control to meet daily need s. To feed their families, women need regular access to cash. This can influence their business and financial management strategies. Women report, for example, that while men can invest and wait to receive a good price for their products or services, sometimes women feel compelled to lower their prices so they have at least some daily income to buy what their family needs. Participants in the key informant interviews 44 cited the need to meet the daily needs of the family as a key challenge that women face in their income generating activities. Indeed, several respondents noted that women will sell goods at a loss if they do not have enough to feed their family that day, and also observed that men do not seem to have the same pressure to sell goods quickly. The norm that housework and care work are women’s responsibility is widely enforced and strongly influences behavior. In a focus group discussion, the female participants estimated that a maximum of 1 out of 10 men help their wives with domestic work. When asked how people respond when men do help, they said, “everyone will say that the wife dominates her husband.� In another focus group, women explained, “Even if it is the husband himself who wanted it [to help], people will not accept it.� Women mostly said that they would welcome help, but all expected that people would respond very negatively to men’s participation in domestic work. Deep Dive 3: Risk and uncertainty Over the last 20 years, the government of the DRC has passed significant amendments to previously discriminatory laws. Many of these amendments were designed to decrease the risk and insecurity faced by women, and to increase their capacity for economic advancement. These new laws and amendments aimed to protect women’s physica l and economic security by outlawing sexual harassment at the workplace, prohibiting gender discrimination in seeking employment and accessing savings and credit, while others have provided maternity leave, equalized the retirement age, and eliminated restrictions to work at night. Legally, Congolese women are now equal to men in their ability to register businesses, sign contracts, access credit, open bank accounts, and get a job.lxxiii However, while legal protections for women have grown, women still face significant barriers to economic success. For example, while in 2017 the government passed legislation to increase credit access for women and prohibit gender discrimination in financial services, lxxiv resistance to women’s economic empowerment, especially in rural areas governed by customary practice, results in women often experiencing economic discrimination, especially in accessing credit. lxxv Stories from the qualitative case study in Goma illustrate many of the different types of risks that affect women’s economic activities. These risks exist inside and outside of women’s households, and range from physical, to financial, to institutional. Drawing attention to these myriad insecurities highlights the precariousness of the landscape women entrepreneurs in Goma navigate and illustrates their short- and longer-term impacts on women’s economic pursuits. As women entrepreneurs in Goma often operate informally, they have few avenues for recourse in the event of a setback. Women entrepreneurs in Goma have long operated in a climate of uncertainty, absorbing and adapting to shocks of various kinds. Such shocks, be they related to conflict, natural disasters, or individual losses, make it hard to plan and grow income generating activities. As one entrepreneur noted : “there will always be unforeseen events, you just never know.� The current global pandemic represents just one, albeit largescale, example of such a shock. The border closures resulting from the pandemic led to a dramatic reduction in cross-border trade and movement, which had direct impacts on the businesses of women in commerce. It also led to food shortages and price increases 45 that indirectly affected much of the population in Goma. Women dominate small scale cross- border trade and are mostly responsible for feeding their families, so they have been hard hit. The manager of a microfinance institution in Goma reported that, “the [COVID -19] lockdown hurt women the most because they most often cross the border to obtain their merchandise…. Men have more capital and are still able to obtain their merchandise through agents who were authorized to transport only goods across the border.� Physical insecurity is a pervasive feature of women’s day -to-day lives, which affects their income generating activities. Despite constitutional protections and national action plans to eliminate violence against women, there is no law prohibiting domestic or intimate partner violence (IPV). Nationally, almost 70% of Congolese women report having suffered at least one form of physical, sexual, or emotional intimate partner violence in their lives. lxxvi IPV is generally accepted socially, and law enforcement rarely intervenes, seeing it as a private family matter. lxxvii Data from the EMAP survey (Figure 17) shows the prevalence of beliefs that GBV is justified, even among women themselves. Almost 70% of women reported that a partner was justified in using violence on a woman if she is unfaithful. A little less than 50% of women also said that violence was warranted if she had disobeyed her partner, and slightly fewer women reported that violence was justified if she refused to engage in sexual intercourse with her partner (42%). The only category where less than 30% agreed that violence was justified is when the woman burns the food (17%). Figure 17: Attitudes Towards Violence As is the case with much of the legal framework designed to protect women, the laws have a limited impact on the realities within which women operate. The constitution protects against sexual violence, and it is considered a criminal offense. However, marital rape is not a prosecutable offense in the DRC law.lxxviii Both sexual violence and rape are pervasive throughout the country and increase drastically in conflict-affected areas where GBV and rape are used as weapons of war.lxxix In 2019, documented cases of conflict-related sexual violence increased by 34% from the previous year, with the vast majority of perpetrators belonging to both non-state and state armed forces.lxxx Figure 18 depicts the experiences of physical and sexual violence of women age 15 and above in their lifetime to date, comparing urban and rural areas. 46 Figure 18: Women aged 15+ Experience of Violence in Lifetime 60% 52% 52% 50% 40% 30% 30% 22% 20% 10% 0% Urban Rural Experienced phy sical violence Experienced sexual violence Conflict-related insecurity in the territories surrounding Goma is a main constraint particularly to the success of businesses in the agriculture value chain. Women with agro-processing businesses face consistent problems with access to inputs and risk the loss of capital when cash or products are stolen in transport. One of the interview participants who has a groundnut and sesame processing business explained, “the biggest challenge that we have is related to transport because our inputs come from Rutshuru. Because of the insecurity that reigns, we send money via mobile money and suppliers send products. En route, the products are stolen and our businesses face losses.� Similarly, a fruit juice producer who sources her fruit from Beni, explained th at the insecurity and poor state of the roads had multiple impacts on her business, including ruptures in the supply chain. In addition, the drivers, who face enormous risks to transport the produce, sometimes increase their prices, which, in turn, affects business profitability. In large part because of security challenges, women in the markets in Goma report that most of the produce they sell is imported from Rwanda, where, pre-pandemic, they could access consistent and organized supply. During a key informant interview, the president of an association of agricultural product vendors was asked how she thinks money should be invested to improve women’s empowerment. She responded, “First of all, security and the construction of roads. There are agricultural deserts because these are absent. We even have difficulty finding food. For the moment, we live on agricultural products that come from other countries.� Insecurity within Goma encourages women to minimize travel outside of their homes when it is dark. This is one factor discouraging work in the hospitality sector, which often requires evening 47 work hours. Risks associated with transit are further compounded by risks of harassment at work. In a focus group of women who own small restaurants in Goma, women described how “some men think we’re prostitutes because we sell alcoholic drinks, even if you’re married they solicit a debt from you and when you ask them to pay, it’s a problem, either they give you a place to meet and when you get there they start to proposition you in ways that dishonor you and if you refuse, they refuse to pay you and this negatively affects the business.� In this vein, the manager of a renowned restaurant in Goma noted that it is difficult to hire women service staff. As they have to stay overnight for their shifts, husbands will generally not permit their wives to work in this restaurant. Nearly all women who participated in the qualitative case study indicated that it is better to be self-employed than to work for someone else. This preference is linked to grave risks of harassment, as well as uncertainty regarding regularity of pay, job tenure security, and a desire to control one’s own working hours and conditions. In addition to physical insecurity, women entrepreneurs discussed risks in the institutional environment, specifically related to financial institutions and government structures. Many women reported that they do not trust most financial institutions because they have witnessed or experienced loss of savings when MFIs and cooperatives have failed and not paid out to account holders. Women also talked about the problem of theft of money from mobile money accounts through stolen SIM cards or stolen numbers. The owner of a clothing shop reported having lost her savings that way. While Congolese can replace SIM cards using the de facto identification card (carte d’électeur) linked to their account, they will not be able recover money that was withdrawn. Finally, entrepreneurs report facing unpredictable and extensive demands for the payment of a wide variety of taxes. The owner of a tailoring business who also has a leadership role in a number of associations complained, “Here, when you open a business, the next day the parade of taxation services begins.� She explained that entrepreneurs do not have information on official tax policies, which means that demands for payment of taxes can come as a negative shock to the business and can also be subject to corruption. Finally, there is insecurity linked to women’s disadva ntaged social position. As described in the section on social norms, women can feel compelled to hide money from their husband out of fear that he will try to steal or control the income, or fear that he will reduce his contributions to the household when her income grows. A cross-border trader reported that husbands can undermine their wives’ business: “The reason that women do not earn much is that they have bad husbands who steal their money.� This can lead to risky savings and investment strategies that are more easily concealed, such as leaving cash with extended family members. While only 38% of women in the DRC report being able to come up with emergency funds if needed (versus 44% of men), they are likelier than men to draw on family or friends as a source of these funds: 31% of women versus 21% of men. Men are more likely to report sourcing emergency funds from their earnings or formal loans.lxxxi In addition, women’s businesses are often more at risk of temporary closure due to a household illness because women are the primary caretakers. At the extreme, when abandoned by their husbands with responsibility for their dependents, women face challenges protecting their business capital while managing to meet daily needs at home. Many of the women who participated in the qualitative interviews reported feeling abandoned because their husband no longer provides, or of actually being abandoned by a husband who left. 48 Participation in associations and savings groups is one of the primary ways that women manage these many risks, though still relatively rare. Only 9% of women in the DRC report saving using a savings club or a person outside the family, versus 12% for men.lxxxii Associations can provide members with benefits including market linkages, access to information, advocacy, and access to VSLAs. VLSAs are highly valued as a relatively secure way to save, as well as a source of loans, emergency support for business or household needs, information relevant to business operations, and social solidarity. A woman who runs a clothing shop explained, “VSLAs have many advantages. First, they enable a weekly meeting among friends. They help us have access to credit and to keep our savings. We also have a petty cash account so we can help each other. As for inconveniences, I don’t see any.� The structure of VSLAs and other types of associations means that, unlike formal financial services, they can flexibly respond to help members facing a variety of difficulties. For example, a wholesaler of plastic shoes explained that she received assistance from a VSLA when she was sick. Similarly, a tailor who is struggling partly because she has recently separated from her husband noted that she borrowed from a VSLA when her child needed medical care. She noted, “VSLAs are advantageous because they permit rapid access to credit and facilitate mutual aid.� As noted, one of the main benefits of VLSAs is that they provide a relatively secure method of saving outside of the household, and often outside of the view of other household members. Interviews with entrepreneurs and with key informants from business associations highlighted the importance of good savings practices for women’s business success. In a focus group discussion with women entrepreneurs, the women agreed: “the first thing that facilitates women’s economic success is her ability to know how to save.� Given the mistrust of many formal financial institutions and mobile money, VSLAs provide an attractive method of saving. The presidents of business associations who participated in the Goma case study as key informants described some of the other benefits that they offer their members. In addition to savings and loans, associations can provide services and access to information. The president of an association of vendors of agricultural products described mobilizing members to advocate for a reduction in taxes at the border. The president of a solidarity group (Ligue pour la Solidarité Congolaise, LSC), described how the group translates laws on women’s rights and provides information and training to members on budgeting and taxation. The association of hardware vendors (L’association de quincaillers, ASQUIGO) helps its members advocate for tax reform, as their members currently have to pay more than 15 different taxes. On the Ground is a cooperative that links smallholder coffee grower cooperatives to the export market and pays its members a higher wage than Fair Trade rates. Finally, in addition to these organized services, many women entrepreneurs cited informal exchange within associations as important sources of information for their businesses. Of course, associations themselves are not removed from the risky environment, and some women commented on the possibility of losing savings when associations fail. For some, it is safer to rely only on themselves and their family members. Most of the case study participants, however, expressed appreciation for the many ways that association membership can be helpful. 49 5. Evidence Review and Policy Recommendations This report identifies key drivers of economic gaps in labor force participation, in agricultural production and productivity, in entrepreneurship, and in wage employment. Drivers of the gender gap that emerged across sectors include: a higher dependency ratio in the household and lower returns to marriage, women’s lower ownership and use of assets, having lower levels of paid or unpaid labor, and having less capital. The gender gap in wage earnings is particularly driven by lower levels of education, while women’s lower rates of cash crop cultivation are the primary driver of their lower agricultural productivity. In addition, we find that women have lower returns to certain productive inputs, both in farming and in entrepreneurship, indicating that even with equal access to inputs, additional interventions, such as education and skills development, may be necessary to increase their earnings and productivity. Based on these findings, four policy priority areas emerge: 1) increasing women’s agricultural productivity through increased land tenure security and improved use and returns to productive inputs, 2) investing in women’s formal education and skills development , 3) facilitating and improving women’s access to capital, and 4) increase women’s physical security and agency . It is important to note that the decomposition analysis used to identify these drivers cannot determine causality. Nevertheless, decomposition methods do document the relative quantitative importance of factors in explaining an observed gap, thus suggesting priorities for further analysis and policy interventions. In this spirit, the evidence review below draws on rigorous evidence provided only by studies which used reliable methods of measuring the effect of specific programs or policies. Such studies are also known as experimental and quasi-experimental. This means that in all the studies cited, the measured effect of a specific program or policy can be precisely attributed to the program. While studies conducted in the DRC are prioritized, this section also includes research from other countries in Sub-Saharan Africa, highlighting those which are from fragile or conflict-affected contexts. These were included because there remains a lack of evidence in the DRC on effective strategies to close gender gaps. Additionally, there is a disparity between evidence available for the conflict-affected, eastern region of the DRC and the rest of the country. Given the diverse socio-political environments of the DRC, going forward, it will be important to prioritize women’s economic empowerment research in both conflict-affected and non-conflict-affected regions of the country. Finally, further research on the economic opportunities and barriers women face throughout all life stages would be valuable in order to better target policies specific to women of all ages. This section provides rigorous evidence to guide policies and programs to help close economic gender gaps in the DRC. The evidence map below (Figure 19Error! Reference source not found.) summarizes the policies and interventions described in this section. This map is a visual guide to the current state of regional evidence addressing the identified policy priorities. Evidence is qualified as credible (blue) if more than one rigorous impact evaluation has produced promising results within Sub-Saharan Africa. Evidence is qualified as emerging (orange) if at least one 50 rigorous impact evaluation has produced promising results. Finally, evidence is qualified as frontier (grey) if found to be a promising idea that remains untested. Furthermore, this map identifies which policy evidence has been tested within the DRC, underlining the lack of rigorous evidence currently available from within the country. Going forward , it will be vital to increase research efforts within the DRC addressing the identified policy priorities. 51 Figure 19: Evidence Map Policy Driver DRC Sources Referenced for Policy Option Key Findings for Policy Priority Addressed Evidence Police Recommendations • (Cherchi et al., 2019) Conditionally Demand for co-tilting of land increased by 50% Insecure Land subsidized land when offered fully subsidized land titles on the Tenure registration condition that the wife’s name is included. • (Agyei-Holmes et al., 2020) • (Ali et al., 2011) • (Goldstein et al., 2015) Government Government formalization and registration of Increase Driven National Insecure Land land rights with corresponding policies to Women's Land Tenure equalize right access to land access between Agricultural Regularization men and women. Productivity Program Eliminate Removing the transaction costs of a civil • (Donald et al, 2020) transaction cost of marriage could encourage spouses married formalizing Insecure Land under customary practices to formalize their marriage and birth Tenure marriage through civil registration, thereby registration to enabling the wife to seek inheritance if she enable becomes widowed. inheritance. 52 • (Kondylis et al., 2016) • (BenYisha et al., 2016) Female farmers were more likely to adopt a technology from female extension agents for Low Adoption and improved productivity. Further evidence Returns to Female extension demonstrates that gender bias prevented male Agricultural service agents farmers from readily accepting information Inputs about new technologies from female extension agents without assurance of their credibility. • (Lambrecht et al., 2016) Joint participation of male and female Low Adoption and Extension services household members in extension programming Returns to targeted to Yes led to the highest technology adoption rates Agricultural couples from the compared to women and men attending Inputs same household. extension training separately. Local language voice and video agricultural • (Aker & Ksoll, 2016) extension messaging have been shown to • (Lecoutere et al., 2019) Utilization of increase women farmer's participation in • (World Bank, 2017) Low Adoption and digital technology agricultural decision making and improve Returns to for agricultural production outcomes. Additionally, farmers’ Agricultural extension increased capacity to use mobile phones resulted Inputs information. in the increased diversity of crops grown and an increase in marginal cash crops grown by women. Couple creating • (World Bank, 2020) joint action plans Couples who created the joint action plans made to encourage Low Adoption of more joint agricultural decisions and women women's Cash Crops managed more cash-crop tasks than non- engagement in the participating couples. cash crop value chain 53 Engaging men to Engaging men and women through couple's • (Ambler et al., 2018) training encouraged female adoption and Low Adoption of change norms participation of more valuable "male crops". Cash Crops around gendered The adoption of "male crops" is vital to crops women's agricultural productivity. Female farming groups were able to collectively • (Mumbeya et al, 2020) address gender specific challenges in cash crop Low Adoption of Women Farmer value chains and engagement in collective Yes Cash Crops Groups bargaining enabled women to improve their value chain participation and increased their yields. • (Daum et al., 2019) Female farmers benefit from mechanization, but Low use and • (Fischer et al., 2018) Mechanization to often they do not have the capital for purchase, Productivity of reduce farm labor or social norms lead men to adopt mechanized Farm Labor responsibilities. • (Daidone et al., 2015) • (Beaman et al., 2010) Provision of cash or asset transfers increased Low use and Cash/Asset spending on hired labor and resulted in Productivity of Transfer increased agricultural output while decreasing Farm Labor time women spent on farm activities. • (Bandiera et al., 2019) • (Buehren et al., 2017) Increase • (Adoho et al., 2014) Adolescent Girls Providing adolescent girls with life skills, Women’s Low Educational Empowerment & livelihood training and credit for income Educational Attainment Safe Space generating activities. Had the effect of Attainment Programming mitigating negative impacts from shocks. and Skills 54 • (Randall & Garcia, Providing financial assistance for education 2020) Providing Low Educational through mechanisms such as scholarships scholarships to Yes Attainment successfully improved female students' female students outcomes. Utilizing mobile Including a mobile phone component as an • (Aker et al., 2012) Low Educational technology to educational tool improved test results for Attainment improve adult women and men. education • (Bandiera et al., 2019) Low Levels of • (Buehren et al., 2017) Education, Consisting of life skills and livelihood trainings, • (Adoho et al., 2014) Vocational, Adolescent girl’s participants were more likely to be involved in Entrepreneurial, empowerment income generating activities, stay in school and programming longer, have increased savings and had Socioemotional increased control of their money. Skills • (World Bank, 2013) Low Levels of Vocational, Technical and life skills training increased Entrepreneurial, Livelihood skills- Yes employment, earnings and investment into and based training businesses. Socioemotional Skills Low Levels of • (World Bank, 2021) Vocational, Psychology-based entrepreneurial training, such • (Salman et al., 2019) Entrepreneurial, Socioemotional as personal initiative training, increased firm and skills training profits more than traditional training. Socioemotional Skills 55 Observational data showed preference among • (Chamboko et al., 2020) Same-gender women to bank with an agent of the same Low Financial microfinance gender. However, additional evidence also Inclusion agent banking showed that female loan officers are at least as gender biased as men. Low Financial Village Savings Village savings and loans groups have been • (Bass et al., 2014) Yes shown to increase savings and productive Inclusion and Loans investments. • (Dupas & Robinson, 2013) • (Bastian et al., 2018) Secure savings mechanisms enabled female Low Financial Secure savings microentrepreneurs to set aside earnings and • (Carranza et al., 2018) Inclusion mechanisms Increase increase investment into their business. Access to Capital • (FAO, 2016) • (Glass et al., 2017) • (FAO, 2017) • (Gastian, 2017) Cash and productive asset transfers, on their Cash & own or as part of as part of larger programs, • (Noble et al., 2020) Low Income Productive Asset Yes such as the graduation program, have • (Banerjee et al., 2015) Generation Transfers demonstrated significant impact on the ultra- poor. Decrease bribery • (Croke et al., 2020) and GBV by Increase increasing Women's Low Knowledge Clear understanding of individual rights and understanding of Physical of Rights and Yes legal responsibilities has been shown to legal rights and Security Laws decrease bribe payments and instances of GBV. procedures and Agency between women and police 56 Interventions targeting either male or female • (Doyle et al., 2018) Couples-based program participants and holding group based • (Gupta et al., 2013) High Levels of discussion groups discussions which include the participants IPV targeting couples partner have produced results showing shifting social norms and reduced IPV. Participants, on average, improved decision- • (Bandiera et al., 2020) Low Levels of Increase agency making ability and control of their bodies, • (Bandiera et al., 2019) Reproductive through life skills reduces rates of childbirth, cohabitation, and Agency training having nonconsensual sex, and were more likely to engage in income-generating activities. Encourage school Both conditional cash transfers (CCTs) and • (Baird et al., 2011) enrollment, unconditional cash transfers (UCTs) resulted in Low Levels of delayed marriage fewer girls dropping out of school. UCTs Reproductive and delayed birth resulted in significantly fewer early marriages Agency through UCTs & and early births than either the CCT or the CCTs control group. Enrolling young children in preschool and • (Martinez et al., 2013) childcare significantly decreased the number of • (Clark et al., 2019) High Burden of Provision of hours caregivers, mostly women, spent on Care Childcare Services childcare and increased the likelihood of women working outside of the home. Increasing men's • (Vaillant et al., 2020) High Burden of participation in Gender transformative interventions engaging • (Doyle et al., 2018) household & Yes men have increased men's contribution to Care childcare household responsibilities. responsibilities 57 5.1 To Increase Women’s Agricultural Productivity Increasing the Adoption of Cash Crops Among Female Farmers As shown in Section 3, women’s low adoption of cash crops is the primary driver of their lower agricultural productivity in the DRC. Female farmers face significant barriers to the adoption cash crops, often seen as “male crops�, and to participating in larger value chains. Increasing women’s participation in cash crop value chains could positively influence the economic recovery following the pandemic. In Côte d’Ivoire, couples were encouraged to create joint action plans regarding their cash crop production, participating couples made more joint agricultural decisions and women managed more cash-crop tasks than non-participating couples.lxxxiii Emerging evidence from an ongoing study in Uganda shows that engaging men through cooperation-based couples’ training and encouraging them to register sugarcane block contracts to their wife increased women’s participation in the sugarcane value chain. However, more educated and w ealthier men or households in which the wife was already highly involved in sugarcane production were the most likely to agree to register their wives. lxxxiv An intervention in South-Kivu, DRC, studied the economic impact of innovation platforms (IPs), within the maize value chain, on female farmers. Specifically, the study analyzed the economic performance of female farmers that were part of all- female farmer IPs against the economic performance of female farmers that made up about 20% of mixed-gender farmer IPs. Evidence showed that participants in the all-female farmers IP were better able to address challenges, improve their value chain participation, and increase their yields compared to women from the mixed-gender farmers IP.lxxxv More research is needed to know which interventions are effective at promoting women’s inclusion in cash crop value chains. Increasing Land Tenure Security Land insecurity is pervasive in the DRC, especially among women, most of whom work on plots owned by a husband or male family member. Land insecurity constrains female producers from accessing financial resources such as credit, limiting their ability to invest in new technologies and productive assets. Currently, there is no rigorous evidence regarding the impact of formalizing land rights for women in the DRC. The evidence described below shows that formalizing land rights can increase women’s land tenure security, agricultural investment, and economic empowerment. A land registration program in Ghana led to increased land tenure security and a corresponding reallocation of women’s labor investments away from agricultural to off -farm economic activities, resulting in increased business profits.lxxxviA nation-wide land tenure regularization program in Rwanda, which formalized land rights and supplied titles to land holders, provided legally married women, and later unmarried women, with equal rights to land access and property ownership as men, and strengthened inheritance rights without gender bias. The program resulted in significantly higher investment into soil conservation by female land holders. lxxxvii In Benin, a program that formalized land ownership and provided land certificates to owners, resulted in increased investments in cash crops and fertilizer use by female-headed households. Widowed beneficiaries were more likely to be able to remain in their dwelling after their husbands' death and the program led to more gender-inclusive inheritance patterns.lxxxviii 58 Marriage and birth certificates are often required in order to exercise one’s right to inherit land from a deceased spouse or parent. Rigorous evidence of an innovative pilot project from Côte d’Ivoire suggests that removing the transaction costs of a civil marriage could encourage spou ses married under customary practices to formalize their marriage through civil registration, thereby enabling the wife to seek inheritance if she becomes widowed. Further, qualitative results show that concerns about identification requirements were one of the main concerns preventing couples from entering into civil marriages. lxxxix One potential solution that has proved effective in increasing women’s property ownership in Uganda has been increasing the demand for co-titling property between husbands and wives. Demand for co-titling of land increased by 50% when households were offered fully-subsidized land titles on the condition that the wife’s name is included, and demand increased by 25% when participants were exposed to an educational video about the benefits of joint titling.xc Increasing Adoption and Use of Agricultural Inputs Increasing the adoption of improved agricultural inputs and technology is essential to increasing women's agricultural production in the DRC. Globally, agricultural extension services are often of limited use to women because they do not target, or have not been adapted to address, the specific needs and barriers of female farmers. xci There is evidence that the gender of extension workers can significantly impact outcomes for female farmers. Programming from Mozambique shows increased adoption of and demand for agricultural technologies among female farmers whose village has a female extension worker.xcii A study in Malawi found that female extension workers could transmit information more successfully and that their students experienced greater yields than their male counterparts. Yet, due to gender perceptions, both male and female farmers believed female trainers to be less capable and were less receptive to their training. xciii Minimal evidence exists from the DRC regarding agricultural extension services targeting women. Some evidence suggests that targeted participation in agricultural extension programming may significantly impact new technology uptake. Experimental evidence from South-Kivu, DRC demonstrates that joint participation of male and female household members in extension programming led to the highest technology adoption rates compared to women and men attending extension training separately. Additionally, female farmers' participation from female-headed households produced significantly higher adoption rates than those of female farmers from male- headed households. The results suggest that due to a lack of decision-making power, women in male-headed households are less likely to adopt capital-intensive technologies without the participation of a male member of the household.xciv Digital technology can be a promising option to enhance women farmer’s use of extension services. Critically, digital extension services offer an invaluable opportunity for farmers to engage with, and seek assistance from, agricultural extension services while person-to-person contact is limited. Program data from an adult education intervention in Niger suggests that access to mobile phones and an increased capacity to use them resulted in an increased diversity of crops grown and in marginal cash crops grown by women. xcv A randomized control trial in Uganda, providing video extension messaging services, found that women who watched the videos had more knowledge about cultivation practices and adoption of inputs, played a larger role in agricultural decision-making, and had higher production and sales.xcvi 59 Increasing Female Farmer’s Use of More Productive Labor Women’s agricultural production is further constrained by time and labor available to invest in economic activities. Financial limitations or restrictive social norms often prevent female farmers from hiring more productive labor, leaving them dependent on unpaid/household labor which may be less productive. When women do hire labor it is less productive than labor hired by male farmers, possibly due to financial constraints which result in hiring cheaper labor. xcvii An evaluation of a program in Zambia showed that spending on hired labor increased by four times when households with children under 5 received cash transfers. xcviii Access to machinery that would reduce required labor for farm activities could help women overcome time constraints. Yet, women often lack sufficient financial resources making agricultural mechanization less accessible to women than men and exacerbating gender inequalities. xcix However, improving mechanization does not guarantee that women will be the beneficiaries, the availability of the machinery can itself impact gender roles in households and communities. One study in Tanzania found that as a traditional female agricultural role was mechanized, it was adopted by men, leaving women dependent on men’s willingness to perform the task. c 5.2 To Increase Women’s Educational Attainment and Skills Increasing Women and Girl’s Educational Attainment Gender disparities in education have long-term implications for both economic and wellbeing outcomes. Ensuring female students continue through upper primary school and perform well enough to continue their education is essential to improving the gender ratio in secondary schools. In particular, efforts to mitigate the negative impacts of shocks and crises are vital to keeping girls in school. In Sierra Leone, an impact evaluation of an adolescent girl empowerment program was ongoing at the outset of the Ebola epidemic. While the program was designed to provide life skills, livelihood training and credit for income generating activities, the program had a strong positive impact on slowing the school-to-work transition by keeping young girls in school and mitigating long-term impacts of the crisis. In villages highly impacted by Ebola, girls that had been exposed to the program improved their numeracy and literacy levels and were twice as likely to maintain school enrolment compared to nonparticipants.ci In addition to making sure girls are physically able to attend school, strengthening the academic performance of female students as they progress through primary school is critical to maintaining enrollment rates of female students in secondary school. Experimental evidence from the DRC suggests that financial assistance such as scholarships successfully improves female students' outcomes, while interventions supporting their studies, such as providing tutoring services, are less impactful.cii Adult populations are also able to benefit from education interventions which can have significant impacts on both educational as well as economic outcomes for participants. For example, evidence from Niger demonstrates improved test results for women and men by including a mobile phone component, as a motivational and educational tool, to an adult education intervention.ciii Increasing Women’s Vocational, Entrepreneurial, and Socioemotional Skills Livelihood skills and vocational training can help women and girls start successful businesses and take advantage of economic opportunities to transition into the workforce. In the face of the 60 disproportionate negative impacts of the pandemic on women’s employment , it is vital that skills trainings and livelihood programs exist to help women gain, or regain, employment. Evidence from the DRC regarding vocational and entrepreneurial skills training is thin and more research is needed. Adolescent empowerment programs for girls, generally consisting of a combination of safe spaces, life skills training, and vocational skills training, are effective interventions to increase young women’s labor force participation and economic outcomes. A study in South Sudan found that adolescent female participants in an intervention consisting of life skills and livelihood trainings were more likely to be engaged in income generating activities and were more likely to have savings and control over their money.civ An impact evaluation of an intervention in Liberia, which provided classroom-based technical and life skills training coupled with support to enter the workforce to adolescent girls, produced strong evidence of increased employment by 47%, increased earnings by 80%, increased access to and control of financial resources, and improved self-confidence.cv Experimental evidence from a program in Uganda, which provided vocational and life-skills training to adolescent girls, found that participants were more likely to engage in income-generating activities and, on average, increased their consumption expenditures.cvi Livelihood skills training can also be effective for women following instability, such as in post- conflict situations. Income generation and livelihood opportunities are often considered essential in discouraging the remilitarization during the reintegration process of ex-combatants. Experimental evidence from Maniema, DRC shows that an income-generating and budget management activity, as part of gender-targeted programming to support female combatants' socio-economic reintegration, produced positive results. Two years after the program, 85% of participants were still engaged in the income generating activities initiated by the project , participants were more likely to own their homes and more likely to own a piece of land than non- participants within their community. Additionally, participants were more likely to have purchase assets valued over $20 in the previous year and owned significantly more assets than non- participants.cvii Evidence shows that outside of formal education and more traditional technical and business skills training, psychology-based, mindset-oriented training can improve entrepreneurs' business outcomes. An impact evaluation in Togo compared the impacts of a business training program and a personal initiative training on microentrepreneurs. Women who received the personal initiative training increased profits by 40%, while those who received the business training increased earnings by only 5%.cviii A study from Ethiopia confirmed that psychological skills are important for female entrepreneurs' success and that such skills can be transferred through training, and evidence further suggests that training is more impactful when delivered by a trainer who previously owned a busines and can relate to the challenges faced by entrepreneurs.cix 5.3 To Increase Women’s Access to Capital Strengthening Financial Inclusion Access to credit can enable female entrepreneurs to start or grow their businesses, but there is a shortage of evidence from the DRC regarding formal financial services or microfinance. In the face of the pandemic, access to credit and financing can provide needed liquidity to women-owned 61 businesses. cx Agent banking is employed to increase the geographic presence of banks and microfinance institutions, especially in more rural areas. Observational findings, using FINCA DRC data, show that clients are more likely to choose to interact with an agent of their gender and that women are significantly more likely to make transactions with female agents. This suggests that increasing the number of women working in financial institutions may be a promising option for enhancing women’s financial inclusion.cxi Access to formal and informal savings mechanisms can also increase women’s access to capital, but to be effective they should be designed with the customer’s needs and constraints in mind . Since the 1990s, Village Savings and Loans Associations (VSLAs) have been frequently included in development interventions, both on their own and in combination with other mechanisms, such as technical skills training or as a part of graduation programs. cxii A VSLA program in South Kivu, DRC, targeted women who had experienced or witnessed sexual violence and had significant mental health problems. The VSLA program provided a savings vehicle, increased financial inclusion, and provided a support network for participants. Participants significantly increased per capita food consumption and reported having more animals for breeding than non-participants.cxiii As connectivity and cellphone use grow, mobile platforms are becoming increasingly promising alternatives to traditional banking. These platforms can be particularly appealing to women entrepreneurs in the DRC that have concerns regarding the security of their money both at home and in MFIs. A study from Kenya among women market vendors and men bicycle taxi drivers showed that the women entrepreneurs were more likely to sign up for a formal savings account than the male entrepreneurs, possibly because men were better able to save money safely at home. Women who opened an account increased their savings and made more productive investments in their business.cxiv Experimental evidence from Tanzania shows that women microentrepreneurs save significantly more by using the mobile savings accounts and obtain more microloans from the mobile financial service provider.cxv Like entrepreneurs, female wage workers also have specific concerns that must be addressed by policies and interventions in order to increase uptake. An intervention in Côte d’Ivoire introduced direct-deposit commitment savings accounts in a cashew factory. The experimental study produced evidence of a 10% increase in productivity and earnings for participants. The study found that the uptake of accounts was concentrated among women who faced higher levels of redistributive pressure and that demand for the savings accounts was dependent on its existence remaining private to the individual. These findings identify that the expropriation of women’s earnings by family members and acquaintances is a significant barrier to women generating savings or benefitting from their productive efforts. This research underscores the value of privacy in account ownership which enabled beneficiaries to save more money. cxvi Supporting Productive Inclusion to Increase Income Cash and productive asset transfers, along with productive inclusion interventions, provide both a vehicle for income generation, as well as training for how to best capitalize on the asset or cash transfer, in an effort to ensure long-term earnings. Providing transfers directly to women can be used in emergency response to increase the use of the provided input, increase economic activities, encourage dietary diversity and shift time-use patterns. Direct cash and asset transfers to women is an effective emergency response and stimulant of economic recovery. 62 Regular cash and/or one-time productive asset transfers of income generating assets such as of livestock, farm equipment, trade supplies or land, can be effective strategies within fragile and conflict-affected contexts as mechanisms to aid in recovery and stability. Productive asset transfers specifically offer an opportunity to diversify income sources and start a virtuous cycle of income generation, economic empowerment, asset ownership, food security, and improved nutrition.cxvii A hybrid microcredit and livestock asset transfer program in Eastern DRC aimed to improve economic and health outcomes and reduce intimate partner violence. The program provided a pig to primarily female participants. A year and a half after receiving the productive asset, 24.7% fewer recipients than non-recipients had outstanding loans, and recipients experienced subjective improvements in health and mental health.cxviii A program in Burkina Faso combined unconditional cash transfers with the provision of in-kind livestock inputs to women. The program produced experimental evidence of increased household revenue by 27% and increased savings, asset ownership, and improved food security.cxix A randomized control trial, implemented in Northern Nigeria, provided unconditional cash transfers to the primary female decisionmakers of households. The study found that recipients were 14% more likely to be in the labor force and were 11% more likely to work in a nonfarm business. The participants working in a nonf arm business invested more into that business, and profits were 80% higher than those of nonrecipients.cxx Productive inclusion programs, also known as graduation or multi-faceted programs, can have a significant impact on increasing the income generating capacity of participants. This model is a more expansive and multifaceted intervention, including a combination of productive asset and cash transfers, technical and life skills training, regular in-person support, and a savings vehicle. An economic and social empowerment program in the DRC, which included training, a monthly stipend, a savings vehicle (VSLA), and social networking through community-groups, positively impacted women. Participants were twice as likely to have savings, which were likely to be significantly higher than those of non-participants. Additionally, earnings of participants were 1.6 times higher than those of non-participants and they were 40% more likely to have become self- employed. cxxi A large scale randomized control trial of a BRAC graduation program was conducted in six countries, including two in Sub-Saharan Africa, Ethiopia and Ghana. The approach included a productive asset transfer of livestock or supplies for petty trade, skills training, food or cash support, savings account access, home visits, and health education. Across all six countries the program resulted in increased consumption spending, 16.4% in Ethiopia and 6.9% in Ghana, as well as increased productivity through self-employment. cxxii 5.4 To Increase Women’s Physical Security and Agency Increasing Knowledge of Rights and Laws to Prevent Gender-Based Violence Women’s ability to engage in economic activities can be compromised by harassment, corruption, and the threat of gender-based violence (GBV). The risk of GBV, especially in conflict-affected areas, can make engaging in, and traveling to and from, work dangerous for women. A traders’ empowerment training on corruption and GBV trained small-scale cross-border traders, many being women, on procedures, tariffs, and their legal rights in order to facilitate safe passage in cross-border trading on the Rwandan border. The evaluation of the program found that participating in the training reduced bribe payments by 27.5% and reduced the incidence of GBV 63 by 30.5% among the participants. However, these results were driven by a change in behavior among traders to avoid situations associated with bribe requests, not a reduction in bribes being requested. These findings signal a need for clear regulations, improved service delivery at border- crossings, and improved awareness of individual rights and legal responsibilities. cxxiii Decrease Women’s Risk of Intimate Partner Violence Impacts of the COVID-19 pandemic, including increased poverty, insecurity and restricted movement have helped elevate the risk of gender-based violence for women, especially within the home. Intimate partner violence (IPV) can have a significant physical, emotional, and economic impact on women. Economic shocks from IPV can occur due to injury, medical bills, and reduced productivity due to distraction and absenteeism. cxxiv Changing social norms regarding gender roles and acceptability of IPV by engaging men is a promising policy option. A gender-transformative couples' intervention in Rwanda engaged expectant/current fathers and their partners in participatory small group discussions with the aim of shifting perceived gender norms and lowering IPV. Women reported experiencing less physical and sexual IPV over the last year than their counterparts in the control group.cxxv Another group-based intervention targeted conflict- affected female VSLA participants and their male partners in Côte d’Ivoire by adding Gender Dialogue Groups to VSLA programs. Participants of the program experienced red uced physical, emotional, and sexual IPV as compared with the control group, as well as statistically significant increased control of household economic resources.cxxvi Increase Women’s Reproductive Agency During crises, when schools close and economic activities slow, adolescent girls are often increasingly vulnerable. Downstream effects of the pandemic and of the economic hardships caused by the pandemic increase the risk faced by adolescent girls of early marriage, early birth, and higher fertility as well as increased risk of sexual exploitation and abuse. Women who marry early or have children at a young age are likely to attain lower education and lower wealth while facing more significant health risks and higher fertility rates than women who do not.cxxvii Evidence shows a close relationship between childbearing and the labor force participation of women, with data indicating a two-year reduction of women’s labor supply per birth. This indicates that lower fertility rates may increase women’s ability to be economically productive.cxxviii Finally, in addition to the negative economic impacts, girls married before age 18 are 22% more likely to experience IPV.cxxix Poverty and education are closely correlated with early marriage and efforts to delay both must address these underlying constraints. A study in Malawi compared the impact of conditional cash transfers (CCTs) and unconditional cash transfers (UCTs) provided to households with teenage girls, on teenage pregnancy rates, early marriage rates, and school dropout rates. Provision of the CCTs was contingent on the school attendance of the girl. The study found that recipients of the UCTs substantially and statistically significantly delayed marriage and childbearing as compared to both the control group and recipients of the CCTs. cxxx In addition to findings of improved income generation, a program in Uganda providing vocational and life-skills training to adolescent girls improved their decision-making ability and control of their bodies. The rate of early childbearing was reduced by 26%, the rate of marriage/cohabitation was decreased by 58%, and the rate of adolescent girls reporting having had sex unwillingly reduced by 6%. cxxxi 64 In light of the ongoing COVID-19 pandemic it is imperative to take into account the influence public health crises such as COVID-19 and Ebola can have on IPV and implement programing and policies which avoid negative outcomes. In Sierra Leone during the Ebola epidemic, a confluence of factors, such as economic hardship, movement restrictions, and limited services, contributed to a drastic increase in teenage pregnancy. cxxxii The adolescent girls empowerment program, described in section 5.2, proved to be an effective buffer for participants against the large-scale shock, mitigating long-term negative impacts and safeguarding participants’ decision - making power. In the face of schools closed due to the crisis, program participants spent less time with men and were less likely to become pregnant than nonparticipants. cxxxiii Decr easing Women’s Burden of Care Women disproportionately lack agency in allocating their time due to social norms within which they are responsible for most unpaid domestic and childcare obligations. As a result, women have less time available to spend on their agricultural plots or working at their businesses, instead spending significantly more time caring for children while farming or going to the market than men.cxxxiv One possible solution is to make childcare services accessible to women. The previously mentioned randomized control trial of a preschool intervention in Mozambique demonstrated that, in addition to educational outcomes, caregivers were spared over 15 hours of childcare duties per week and were 26% more likely than the control group to have worked in the previous 30 days.cxxxv Experimental evidence, produced by a randomized control trial in Kenya, shows that women who received vouchers for subsidized childcare were 8.5% more likely to be employed. Single mothers specifically benefited by shifting to jobs with more regular hours and less loss to their income.cxxxvi Another option to increase time available for women to pursue economic activities is for the husband to contribute more to household and childcare responsibilities. In Eastern DRC, a men’s discussion group series to change attitudes and behaviors about gender and reduce intimate partner violence (IPV) increased men’s participation in housework (but did not change women’s experience of IPV).cxxxvii Both men and women in the gender-transformative couples' intervention in Rwanda, mentioned above, reported higher levels of men's participation in childcare and household tasks in addition to reductions in IPV.cxxxviii Whether they are farmers, entrepreneurs, or wage employees, Congolese women earn much less than Congolese men for the same amount of work. Not all barriers and constraints to women’s economic empowerment can be addressed at once, and this report has identified priority constraints that should be addressed to increase women’s earnings. This stock -taking exercise has identified several effective interventions that contribute to enhancing women’s earnings and agency, but more work is needed in the DRC to generate new evidence of effective interventions that can work at scale. 65 6. Appendix Appendix A: Data Sources Nationally Representative Datasets The 1-2-3 Survey is the first national household survey in the DRC focusing on Employment, the Informal Sector and Household Expenditures. In our analysis, we use the second wave of the survey from 2012, which was implemented by the National Institute of Statistics in the DRC and covers 21,454 households in all 26 provinces of the DRC. Sampling in this survey was conducted to select approximately equal number of households in all provinces, with the exception of the capital Kinshasa being oversampled. Datasets can be accessed at: https://www.uantwerpen.be/en/projects/great-lakes-africa-centre/national-datasets-livelihoods- drc/national-household-s/. The Demographic and Health Survey (DHS) is designed to provide data for monitoring the population and health situation in the DRC. In our analysis, we use the second wave of the survey which provides nationally representative data on labor outcomes with a sample of 27,483 adult men and women who were interviewed from November 2013 till February 2014. The sampling frame for the DHS was the same as used in the 1-2-3 survey. Datasets can accessed at: https://dhsprogram.com/methodology/survey/survey-display-421.cfm Impact Evaluation Datasets The Growth Poles survey uses the baseline survey of the World Bank Western DRC Growth Poles project impact evaluation, conducted in the province of Kongo Central in December 2015. Surveys were conducted with 2,931 households within the project intervention zone, defined as a 40 km radius from six “growth poles� within Kongo Central as well as control regions within and outside of Kongo Central. The sampling frame was established through a two-step selection: randomly selecting villages among a list of villages with a farmer association (stratified by distance to road and the presence of cassava, rice or palm oil cultivation) and then randomly selecting 20 households in each village (10 with a female plot manager and 10 with a male plot manager). The Growth Poles Project was implemented by the government of the DRC with funding from the World Bank. The impact evaluation was led by the World Bank’s Africa Gender Innovation Lab. The Childcare Pilot Survey was conducted as the baseline for the impact evaluation of community-based childcare centers within the Kongo Central province of the DRC. 138 villages were identified by the local NGO REPAFE as fulfilling two main eligibility criteria: having interest is hosting a childcare center and having the adequate infrastructure to do so. 2,450 households having a child between the ages of 1 and 5 and interesting in sending their child to a daycare center were surveyed across these villages. Surveys were administered to household heads, with particular sections reserved for mothers and children. Both the impact evaluation and intervention are led by the World Bank’s Africa Gender Innovation Lab, in partnership with the 66 Congolese NGO REPAFE, Save the Children US, and the DRC's Ministry of Primary, Secondary and Technical Education (EPST). The EMAP survey collected baseline data for the impact evaluation of the Engaging Men through Accountable Practice (EMAP) project in 2016. 1387 men who were enrolled in EMAP were interviewed as well as their partners: 1220 women. Eligibility criteria for men’s participation in the study included: being aged 18 years and older, having lived in the community for at least 6 months with plans to continue living there for at least an additional 6 months, ability to actively participate in the group, non-involvement with an ongoing evaluation of adolescent girl programming that was operational in some sites and committing to not perpetrate violence for the duration of the intervention. Female partners of men were interviewed if they were above 15 years of age. For polygamous households, the first wife was interviewed . Data were collected in 28 communities of North and South Kivu provinces, in Eastern DRC. EMAP was designed and implemented by the International Rescue Committee (IRC), with funding from the World Bank. The impact evaluation was a collaboration between the IRC and the World Bank’s Africa Gender Innovation Lab. The SME Growth survey was collected to provide a baseline for the impact evaluation of the SME Growth and Development project in the DRC. 3604 women entrepreneurs who registered in the project were interviewed in Goma and Lubumbashi (1804 and 1800 respectively) in November and December 2020. Eligibility criteria included: 1.) completed registration, 2.) provided at least 1 valid contact, 3.) no more than 60 years of age, 4.) for either the primary or secondary business activity reported during registration, the women A.) is either an employer or self -employed, B.) has no more than 10 employees, C.) has been running the business for at least 1 year, D.) works in an eligible sector (value added, see full list attached) . We also surveyed the husbands of partnered women if the husband could be successfully tracked and agreed to participate in the interview. Of the 2414 partnered women in the sample, 1799 husbands were interviewed. The SME Growth and Development project is implemented by the Government of the DRC with funding from the World Bank. The impact evaluation is a collaboration between the World Bank’s Africa Gender Innovation and Leuphana University. Qualitative Case Study in Goma In addition to the survey data analysis, this report includes findings from a qualitative case study of women’s economic activities in Goma. The qualitative data collection began with seven focus group discussions with women in different economic sectors, including cross-border traders, restaurant/food stall owners, vendors of second-hand clothing, and retailers of a variety of agricultural and non-agricultural products. Focus group participants were recruited through snowball sampling either recruiting in the marketplace or through business association leaders. The focus groups were used to examine social norms regarding women’s work and to refine the data collection tools for the in-depth interviews. In-depth interviews were conducted with approximately 25 entrepreneurs who had been randomly selected from the database of entrepreneurs who had expressed interest in participating in an ongoing project offering support to small businesses. The entrepreneurs were asked to describe their businesses and the challenges that they face. Finally, the research team conducted approximately 20 key informant interviews with business association leaders, microfinance institution representatives, and private sector 67 employers such as hotel and restaurant owners. The key informants were asked about hiring women as employees, sourcing products from women entrepreneurs, and the challenges faced by women in the economy. Appendix B: Descriptive Statistics Appendix Figure 1 68 Appendix Figure 2 Appendix Figure 3 69 Table B1: Regional differences in gender gaps (%) Regional Income in Grouping Agricultural Agricultural Labor Force Business paid output productivity Participation profits employment DRC 0.178*** 0.109 0.082*** 0.773*** 0.665*** Provinces: Bas-Uele Northeast 0.189 0.184 0.084* 0.780*** 0.923 Equateur West 0.129 0.070 0.052 0.847*** 0.616 Haut-Katanga South 0.441*** 0.716** 0.188*** 0.794*** 0.850*** Haut-Lomami Southwest -0.006 0.493 0.066 0.704*** 1.422 Haut-Uele Northeast 0.407*** 0.636** -0.023 0.711*** 0.648 Ituri Northeast 0.442 -0.187 0.244*** 0.558** 0.489* Kasai Southwest 0.134 -0.035 -0.019 0.879*** 0.121 Kasai-Central Central 0.157* 0.158 -0.006 0.712*** 0.282 Kasai-Oriental Central 0.262 0.365 -0.005 0.893*** 0.552* Kinshasa West 0.504 1.557* 0.293*** 0.716*** 0.745** Kongo Central West 0.058 -0.242 0.026 0.779*** 0.718* Kwango Southwest -0.205 -0.589 -0.097*** 0.690*** 0.379 Kwilu West -0.043 -1.206 0.019 0.786*** 1.311** Lomami Central 0.100 -0.393* -0.042** 0.857*** 0.925* Lualaba South 0.443** 0.518 0.169*** 0.868*** 0.926** Mai-Ndombe West -0.003 0.718 -0.027 0.861*** 1.225** Maniema East 0.305 -1.189 0.169** 0.807*** 0.591 Mongala Northwest 0.517** 0.885*** 0.114** 0.787*** 1.571 Nord-Kivu East 0.437*** 0.344* 0.032 0.775*** 0.683** Nord-Ubangi Northwest 0.036 0.466 -0.027 0.843*** -0.888 Sankuru Central 0.023 0.512** -0.046 0.769*** 1.017 Sud-Kivu East 0.324* -0.036 0.059 0.853*** 0.890 Sud-Ubangi Northwest 0.275 0.571** 0.001 0.906*** 0.660* Tanganyika Southwest 0.112 0.348 0.025 0.873*** -0.168 Tshopo Northeast 0.555*** 0.721 0.174*** 0.763*** -0.151 Tshuapa Northwest 0.454** 0.560 0.012 0.856*** 0.592 *p<0.1 **p<0.05 ***p<0.01 Table B2: Summary statistics for agricultural households in Growth Poles survey Male Female Male Female Difference managers managers managers managers (Male- obs. obs. mean mean Female) Production value (million 3533 1640 0.286 0.205 0.080*** CDF) Plot yields (million CDF/ha) 3450 1594 4.454 3.593 0.861 Manager age 3533 1638 46.138 48.426 -2.287*** Manager married/cohabiting 3524 1640 0.918 0.387 0.532*** 70 Manager attended or 3533 1640 0.639 0.324 0.315*** completed high school Female HH head 3533 1640 0.015 0.670 -0.655*** HH head age 3533 1640 46.595 51.633 -5.038*** HH head married/cohabiting 3530 1640 0.918 0.401 0.517*** HH head attended or 3533 1640 0.630 0.387 0.243*** completed high school Household size 3533 1640 6.044 5.124 0.921*** Dependency ratio 3533 1640 1.254 1.563 -0.309*** HH owns radio 3533 1640 0.523 0.278 0.244*** HH owns television 3533 1640 0.075 0.036 0.039*** HH owns cellular phone 3533 1640 0.389 0.248 0.141*** HH owns bicycle/motorcycle 3533 1640 0.260 0.113 0.146*** HH number of plots 3533 1640 4.721 4.633 0.088 Plot surface (ha) 3533 1640 0.500 0.286 0.214*** Number of unpaid workers 3533 1640 15.691 14.171 1.520*** on the plot Number of paid workers on 3533 1640 2.863 2.323 0.540*** the plot Use of pesticides 3533 1640 0.008 0.004 0.004* Use of fertilizers 3533 1640 0.019 0.013 0.006 Distance to plot from the dwelling (walking minutes) 3512 1636 36.634 42.330 -5.697*** Cash crops plot 3533 1640 0.475 0.359 0.117*** Cassava plot 3533 1640 0.794 0.838 -0.044*** Corn plot 3533 1640 0.584 0.592 -0.008 Peanut plot 3533 1640 0.463 0.549 -0.086*** Rice plot 3533 1640 0.095 0.040 0.055*** * p<0.1 ** p<0.05 *** p<0.01 71 Table B3: Summary statistics for business households in Growth Poles survey Male Female Male Female Difference managers managers managers managers (Male- obs. obs. mean mean Female) Profits in last month (million 310 123 0.192 0.054 0.137*** of CDF) Manager age 306 121 42.944 43.240 -0.295 Manager married/cohabiting 309 122 0.887 0.574 0.313*** Manager attended or completed high school 310 123 0.713 0.496 0.217*** Female HH head 310 123 0.039 0.455 -0.417*** Household size 310 123 6.261 5.439 0.822*** Dependency ratio 310 123 1.236 1.324 -0.088 HH owns radio 310 123 0.648 0.455 0.193*** HH owns television 310 123 0.174 0.114 0.060* HH owns cellular phone 310 123 0.535 0.463 0.072 HH owns bicycle 310 123 0.300 0.163 0.137*** Business location in 310 123 0.068 0.049 0.019 building/rental Number of unpaid workers 310 123 1.719 1.431 0.288** Value of paid labor 310 123 7.877 1.924 5.953*** (thousand CDF) Value of intermediate inputs 310 123 45.152 26.638 18.514** (thousand CDF) Value of paid utilities (thousand CDF) 310 123 0.009 0.006 0.003 * p<0.1 ** p<0.05 *** p<0.01 72 Table B4: Summary statistics for individual labor sample in Growth Poles survey Male HH Female HH Male HH Female HH Difference members members members members (Male- obs. obs. mean mean Female) Labor force participation 4085 4502 0.674 0.771 -0.097*** Currently in wage employment 3988 4473 0.056 0.032 0.024*** Wage employment in past 4085 4502 0.154 0.078 0.077*** 12m Wage income in past 12 4085 4502 73.968 23.278 50.690*** months (thousand CDF) Individual age 3999 4358 35.124 36.760 -1.635*** Individual 4133 4547 0.523 0.515 0.008 married/cohabiting Individual attended or 4147 4560 0.685 0.445 0.241*** completed high school Female HH head 4147 4560 0.125 0.283 -0.158*** Household size 4147 4560 6.477 6.313 0.164*** Dependency ratio 4147 4560 1.041 1.271 -0.230*** HH owns radio 4147 4560 0.502 0.427 0.075*** HH owns television 4147 4560 0.080 0.068 0.012** HH owns cellular phone 4147 4560 0.386 0.340 0.046*** HH owns 4147 4560 0.233 0.200 0.033*** bicycle/motorcycle * p<0.1 ** p<0.05 *** p<0.01 73 Table B5: Summary statistics for agricultural households in 1-2-3 Survey Difference Male HH Female HH Male HH Female HH (Male- heads obs. heads obs. heads mean heads mean Female) Production value (million 11067 2207 0.344 0.301 0.043** CDF) Plot yields (million CDF/ha) 10248 2003 1.705 1.592 0.113 HH head age 11067 2207 42.769 48.366 -5.597*** HH head married/cohabiting 11067 2207 0.934 0.244 0.690*** HH head attended or 11067 2207 0.540 0.176 0.364*** completed high school HH size 11067 2207 5.268 3.701 1.567*** HH dependency ratio 11067 2207 1.197 1.928 -0.731*** HH owns a radio 11067 2207 0.380 0.096 0.284*** HH owns a TV 11067 2207 0.028 0.009 0.019*** HH owns a cellular phone 11067 2206 0.188 0.085 0.104*** HH owns a bicycle/motorcycle 11067 2207 0.266 0.080 0.186*** Average distance to HH plots 11013 2192 0.877 0.823 0.054* Number of HH plots 11067 2207 1.688 1.627 0.061* Total plots area (ha) 11067 2207 1.752 1.236 0.517*** Number of paid workers per 11067 2207 0.531 0.342 0.189*** plot Number of unpaid workers per plot 11067 2207 3.014 2.379 0.634*** HH uses fertilizers 11067 2207 0.025 0.020 0.005 HH uses pesticides 11067 2207 0.009 0.001 0.008* Proportion of HH plots with cash crops 11050 2204 0.044 0.019 0.025*** Proportion of HH plots with 11050 2204 0.782 0.806 -0.024 cassava Proportion of HH plots with corn 11050 2204 0.465 0.463 0.002 Proportion of HH plots with peanut 11050 2204 0.213 0.221 -0.008 Proportion of HH plots with 11050 2204 0.153 0.114 0.039*** rice * p<0.1 ** p<0.05 *** p<0.01 74 Table B6: Summary statistics for business households in 1-2-3 Survey Male Female Male Female Difference managers managers managers managers (Male- obs. obs. mean mean Female) Total profits (million CDF) 1522 2557 0.249 0.080 0.169*** Manager age 1522 2557 37.957 38.073 -0.116 Manager married/cohabiting 1522 2557 0.732 0.680 0.052* Manager attended or completed high school 1522 2557 0.816 0.665 0.151*** Female individual 1522 2557 0.061 0.292 -0.231*** HH size 1522 2557 6.197 6.409 -0.212 HH dependency ratio 1522 2557 0.968 1.126 -0.158*** HH owns a radio 1522 2557 0.554 0.497 0.057** HH owns a TV 1522 2557 0.412 0.402 0.010 HH owns a cellular phone 1522 2557 0.707 0.677 0.030 HH owns a bicycle/motorcycle 1522 2557 0.243 0.165 0.078*** Business location in a 1522 2557 0.097 0.035 0.062*** building/rental Business location at home 1522 2557 0.446 0.630 -0.183*** Number of unpaid workers 1522 2557 1.288 1.152 0.136*** Number of paid workers 1522 2557 0.268 0.082 0.187*** Total value of physical 1522 2557 168.012 50.680 117.333*** capital (thousand CDF) Value of intermediate inputs 1522 2557 31.781 59.120 -27.339*** (million CDF) Value of paid fines (million 1522 2557 0.210 0.085 0.125*** CDF) Value of paid utilities (million CDF) 1522 2557 13.073 4.283 8.789*** * p<0.1 ** p<0.05 *** p<0.01 75 Table B7: Summary statistics for individual labor sample in 1-2-3 Survey Female HH Male HH Female HH Difference Male HH members members members (Male- members obs. obs. mean mean Female) Labor force participation 28087 30149 0.747 0.677 0.070*** Currently in wage 28087 30148 0.169 0.042 0.127*** employment Monthly income from 28087 30148 11.984 2.647 9.337*** wages (thousand CDF) Individual age 28623 30520 35.185 34.330 0.855*** Individual 28554 30465 0.592 0.589 0.003 married/cohabiting Individual attended or 28623 30520 0.667 0.435 0.232*** completed high school Female HH head 28623 30520 0.092 0.231 -0.139*** HH size 28623 30520 6.039 6.028 0.011 HH dependency ratio 28623 30520 0.955 1.112 -0.158*** HH owns a radio 28621 30518 0.438 0.400 0.038*** HH owns a TV 28620 30516 0.199 0.201 -0.002 HH owns a cellular phone 28618 30518 0.375 0.362 0.013*** HH owns a 28623 30520 0.224 0.200 0.024*** bicycle/motorcycle * p<0.1 ** p<0.05 *** p<0.01 76 Table B8: Summary statistics for agricultural households in Childcare survey Male Female Male Female Difference managers managers managers managers (Male- obs. obs. mean mean Female) Plot production (million 2661 995 0.346 0.298 0.048*** CDF) Plot yields (million CDF/ha) 2661 995 55.801 38.403 17.398*** Manager age 2661 995 41.124 39.233 1.890*** Manager married/cohabiting 2661 995 0.960 0.545 0.415*** Manager attended or 2660 994 0.721 0.385 0.335*** completed high school Female HH head 2661 995 0.011 0.446 -0.435*** HH head age 2661 993 41.324 47.753 -6.429*** HH head married/cohabiting 2661 995 0.962 0.563 0.399*** HH head attended or 2660 988 0.715 0.548 0.167*** completed high school HH size 2661 995 5.839 5.953 -0.114 HH dependency ratio 2661 995 1.290 1.467 -0.176*** HH owns a radio 2661 995 0.488 0.332 0.157*** HH owns a TV 2661 995 0.056 0.040 0.015** HH owns a cell phone 2661 995 0.521 0.437 0.084*** HH owns a bicycle/motorcycle 2661 995 0.142 0.082 0.060*** HH number of plots 2661 995 2.021 2.257 -0.236*** Plot surface (ha) 2661 995 0.029 0.040 -0.012** Number of HH workers on 2661 995 2.318 2.149 0.170*** plot Number of non-HH workers 2661 995 0.381 0.392 -0.011 on plot Fertilizer use on plot 2661 995 0.032 0.007 0.025*** Pesticides use on plot 2661 995 0.028 0.007 0.021*** Cash crops plot 2661 995 0.042 0.021 0.021*** Cassava plot 2661 995 0.810 0.842 -0.032** Corn plot 2661 995 0.070 0.079 -0.010 Peanut plot 2661 995 0.167 0.184 -0.017 Rice plot 2661 995 0.012 0.009 0.003 * p<0.1 ** p<0.05 *** p<0.01 77 Table B9: Summary statistics for individual labor sample in Childcare survey Male HH Female HH Male HH Female HH Difference members members members members (Male- obs. obs. mean mean Female) Labor Force Participation 2931 3658 0.918 0.944 -0.026*** Currently in wage employment 2929 3653 0.036 0.038 -0.003 Wage employment in past 2931 3659 0.040 0.043 -0.003 12 months Monthly wage income 2928 3658 0.123 0.116 0.007 (thousand CDF) Individual age 3044 3735 35.842 32.186 3.656*** Individual 3044 3735 0.710 0.612 0.098*** married/cohabiting Individual attended or 3036 3724 0.708 0.455 0.254*** completed high school Female HH head 3044 3735 0.059 0.180 -0.121*** HH size 3044 3735 6.274 6.207 0.066 HH dependency ratio 3044 3735 1.170 1.231 -0.061*** HH owns a radio 3044 3735 0.463 0.428 0.034*** HH owns a TV 3044 3735 0.049 0.048 0.001 HH owns a cell phone 3044 3735 0.510 0.485 0.024** HH owns a 3044 3735 0.145 0.130 0.014* bicycle/motorcycle * p<0.1 ** p<0.05 *** p<0.01 78 Table B10: Summary statistics for individual labor sample in DHS survey Male HH Female HH Male HH Female HH Difference members members members members (Male- obs. obs. mean mean Female) Labor force participation 8656 18827 0.822 0.723 0.099*** Currently in wage employment 8656 18827 0.242 0.055 0.186*** Wage employment in past 8656 18827 0.261 0.060 0.201*** 12 months All year work in past 12m 8655 18827 0.170 0.041 0.129*** Seasonal work in past 12m 8655 18827 0.040 0.007 0.033*** Occasional work in past 12m 8655 18827 0.050 0.012 0.038*** Individual age 8656 18827 31.413 28.083 3.329*** Individual 8655 18826 0.579 0.633 -0.053*** married/cohabiting Individual attended or completed high school 8656 18827 0.728 0.476 0.252*** Female HH head 8656 18827 1.102 1.238 -0.136*** HH size 8656 18827 6.673 6.780 -0.107 HH dependency ratio 8656 18827 1.048 1.288 -0.240*** HH owns a cell phone 8648 18813 0.482 0.464 0.018** HH owns a 8651 18813 0.069 0.061 0.008** bicycle/motorcycle HH owns a radio 8651 18816 0.514 0.474 0.040*** HH owns a TV 8653 18822 0.195 0.209 -0.014* * p<0.1 ** p<0.05 *** p<0.01 79 Table B11: Summary statistics for business households in SME survey Male Female Male Female Difference managers managers managers managers (Male- obs. obs. mean mean Female) Profits in last month (million 136 136 1.565 0.851 0.715** of CDF) Manager age 136 136 43.529 38.699 4.831*** Manager attended or 136 136 0.926 0.860 0.066* completed high school HH head female 136 136 0.154 0.154 0.000 Household size 136 136 7.368 7.368 0.000 HH dependency ratio 136 136 1.201 1.201 0.000 HH owns a 136 136 0.294 0.294 0.000 bicycle/motorcycle HH owns a cellular phone 136 136 1.000 1.000 0.000 HH owns a radio 136 136 0.846 0.846 0.000 HH owns a TV 136 136 0.860 0.860 0.000 Number of workers 136 136 4.000 2.824 1.176** Value of physical capital (million of CDF) 136 136 117.858 34.315 83.544 * p<0.1 ** p<0.05 *** p<0.01 80 Appendix C: Decomposition Methods The main purpose of decomposition methods is to partition the overall difference of a given distribution between two groups, group A and group B 8 . In this diagnosis, Group A will be women (women plot managers, enterprise managers or employees) and Group B will be men (men plot managers, enterprise managers and employees). The standard assumption used in these decompositions is that the outcome variable Y is linearly related to the covariates, X, and that the error term υ is conditionally independent of X. 𝐾 𝑌𝑔𝑖 = 𝛽𝑔0 + ∑ 𝑋𝑖𝑘 𝛽𝑔𝑘 + 𝑣𝑔𝑖 , 𝑔 = 𝐴, 𝐵. 𝑘=1 Where E(𝑣𝑔𝑖 |𝑋𝑖 ) = 0 and X is the vector of covariates. The overall difference in average outcomes between group B and A is ̂ 𝜇 = ̅̅̅ 𝛥 𝑦𝐴 , 𝑦𝐵 − ̅̅̅ 𝑂 which means 𝐾 𝐾 ̂ 𝜇 = (𝛽 𝛥 ̂𝐵𝑂 − 𝛽 ̅𝐵𝑘 (𝛽 ̂ 𝐴𝑂) + ∑ 𝑋 ̂𝐴𝑘 ) + ∑(𝑋 ̂𝐵𝑘 − 𝛽 ̅𝐴𝑘 ) 𝛽 ̅𝐵𝑘 − 𝑋 ̂𝐴𝑘 𝑂 𝑘=1 𝑘=1 where 𝐾 𝜇 ̂ 𝑠 ̂𝐴𝑂) + ∑ 𝑋 ̂𝐵𝑂 − 𝛽 ̂𝐵𝑘 − 𝛽 ̅𝐵𝑘 (𝛽 ̂ 𝐴𝑘) 𝛥 (𝑈𝑛𝑒𝑥�𝑙𝑎𝑖𝑛𝑒𝑑) = (𝛽 𝑘=1 𝐾 ̂ 𝜇 (𝐸𝑥�𝑙𝑎𝑖𝑛𝑒𝑑) = ∑(𝑋 𝛥 ̅𝐴𝑘 ) 𝛽 ̅𝐵𝑘 − 𝑋 ̂𝐴𝑘 𝑋 𝑘=1 And 𝛽̂𝑔𝑂and 𝛽 ̂ 𝑔𝑘 (k = 1, . . . K) are the estimated intercept and slope coefficients, respectively, of the regression models for groups A and B. The overall decomposition 𝛥 ̂ 𝜇 and its two components 𝛥 ̂ 𝜇 s structural effect (unexplained) and 𝛥X ̂ 𝜇 𝑂 composition effect (explained by differences in covariates) are identified with the aggregate decomposition. The detailed decomposition involves subdividing both components into the respective contributions of each ̂ 𝜇 and 𝛥 covariate, 𝛥 ̂ 𝜇 for k=1….K. 𝑆,𝑘 𝑋,𝑘 Kitagawa-Oaxaca-Blinder -decomposition of mean productivity differentials 𝑌𝑔 = 𝑋𝛽𝑔 + 𝑣𝑔 where 𝐸[𝑣𝑔 |𝑋] = 0 . Letting 𝐷𝐵 = 1 be an indicator of group B (Men plot managers) and taking the expectations over X, the overall mean productivity gap 𝛥 ̂ 𝜇 can be written as 𝑂 ̂ 𝜇 = 𝐸[𝑌𝐵 |𝐷𝐵 = 1] − 𝐸[𝑌𝐴|𝐷𝐵 = 0] 𝛥𝑂 ̂ 𝜇 = 𝐸[𝐸(𝑌𝐵 |𝑋. 𝐷𝐵 = 1)|𝐷𝐵 = 1] − 𝐸[ 𝐸(𝑌 𝛥 𝐴 |𝑋, 𝐷𝐵 = 0)|𝐷𝐵 = 0] 𝑂 ̂ 𝜇 = 𝐸[𝑋|𝐷𝐵 = 1]𝛽𝐵 + 𝐸[𝑣𝐵 |𝐷𝐵 = 1] − ( 𝐸[𝑋|𝐷𝐵 = 0]𝛽𝐴 + 𝐸 [𝑣𝐴 |𝐷𝐵 = 0]) 𝛥𝑂 8 This appendix is based, mainly, on the analysis presented in the Handbook of Labor Economics. Volume 4, Part A, 2011, Pages 1-102. Chapter 1 - Decomposition Methods in Economics Authors. Nicole Fortin, Thomas Lemieux, Sergio Firp,. and the report Levelling the field. (World Bank, 2014). 81 Where 𝐸 [𝑣𝐴 |𝐷𝐵 = 0] = 𝐸[𝑣𝐵 |𝐷𝐵 = 1] = 0 . Adding and subtracting the average counterfactual productivity that group B (women) would have obtained under the productivity structure of group A (men), 𝐸[𝑋|𝐷𝐵 = 1]𝛽𝐴, the expression becomes: 𝜇 ∆𝑂 = 𝐸[𝑋|𝐷𝐵 = 1]𝛽𝐵 − 𝐸[𝑋|𝐷𝐵 = 1]𝛽𝐴 + 𝐸[𝑋|𝐷𝐵 = 1]𝛽𝐴 − 𝐸[𝑋|𝐷𝐵 = 0]𝛽𝐴 𝜇 ∆𝑂 = 𝐸[𝑋|𝐷𝐵 = 1](𝛽𝐵 − 𝛽𝐴 ) + (𝐸[𝑋|𝐷𝐵 = 1] − 𝐸[𝑋|𝐷𝐵 = 0]𝛽𝐴 𝜇 𝜇 𝜇 ∆𝑂 = ∆𝑆 + ∆𝑋 Replacing the expected value of the covariates E[X|𝐷𝐵 = 𝑑] , for d=0,1, by the sample averages 𝑋 ̅𝑔 , the decomposition is estimated as ̂ 𝜇 = 𝑋 ∆ ̅𝐵 𝛽 ̂𝐵 − 𝑋 ̅𝐵 𝛽 ̅𝐵 𝛽 ̂ 𝐴 + 𝑋 ̅𝐴 𝛽 ̂𝐴 − 𝑋 ̂𝐴 𝑂 𝜇 ̂ = 𝑋 ̅𝐵 (𝛽̂𝐵 − 𝛽 ̅̅̅𝐵 − 𝑋 ̂ 𝐴 ) + (𝑋 ̂𝐴 ̅𝐴 )𝛽 ∆ 𝑂 ̂ 𝜇 ̂ 𝜇 ̂ 𝜇 ∆𝑂 = ∆𝑆 + ∆𝑋 ̂ 𝜇 The first term in the equation ∆𝑆 is the structure effect (unexplained, or the part due to discrimination) while the second term is the composition effect ∆ ̂ 𝜇 . 𝑋 In practice, it is computed by plugging in the sample means and the OLS estimates 𝛽 ̂𝑔 in the above formula. 82 The Kitagawa-Oaxaca-Blinder decomposition is a statistical method that explains the difference in the means of a dependent variable between two groups by decomposing this gap into two parts: The Endowment Effect (Explained) The Structural Effect (Unexplained) Captures the return to resources. The differences in Exp lains the differences between men and what is obtained from a given amount of a factor of wo men in terms of factors of production production, i.e. the difference in productivity that the such as years of experience, total inputs or men obtain compared with wo men who have exactly access to credit. It refers to the differences the same years of experience or who use the same total in the quantities or levels of resources used amount of inputs. in plots by male managers compared to women managers In other words, this is the portion of the Even when men and wo men plot managers have gender gap attributable to the quantity or access to the same quantities of resources, they do level of resources that can be reduced by not achieve the same results. Provid ing wo men ensuring that women receive the resources farmers with the same resources will not they lack, relative to men. necessarily reduce the structural portion of gender gap. Policies need to address broader issues of constraints faced by women managers. In other words, this part captures a discrimination component and the unobservable variables. 83 Choice of covariates A limitation of our analysis is that while decompositions are useful for quantifying, purely in an accounting sense, the contribution of various factors to a difference in an outcome across groups (or a change in an outcome for a particular group over time), they are based on correlations, and hence cannot be interpreted as estimates of underlying causal parameters, as noted by Fortin et al. (2011). The use of phrases in this report such as “drivers of the gender gap� should therefore be viewed in this light. Whether variables included in Oaxaca-Blinder-Kitagawa decompositions can be labelled as ‘drivers’ thus depends on theory and previous experimental evidence linking the covariate to the outcome. The analysis presented in this report uses a common set of explanatory variables across datasets and sectors to maximize comparability, which were selected on the basis of existing theory on drivers of the gender gaps, and similar analysis conducted previously (World Bank and ONE Campaign 2014, World Bank 2019). • Marriage status is conceptualized as a predictor of earnings—and in its interplay with gender norms, as a specific predictor of women’s economic outcomes—in most economic models going back to Becker (1973) and Lundberg and Pollack (1993). While causal evidence linking marriage to women’s economic outcomes is hard to come by, existing quasi-experimental work finds a negative impact on labor force participation (e.g., Assaad 2017). • Educational levels are likewise included as predictors of earnings in all standard economic models. Recent empirical work validates this theoretical inclusion for developing economies, finding that a marginal year of schooling in developing economies raises enterprise income by an average of 5.5 percent, which is close to the average return in industrial countries (van der Sluis 2005). The same study finds higher returns for women by about 4 percentage points. Moreover, a broad range of other micro data studies showing that girls have a higher marginal return to education, which is even higher if the impact of female education on fertility and education of the next generation is included (Hill and King 1995; World Bank 2001; Elizabeth M. King, Stephan Klasen, and Maria Porter 2008). • Assets are likewise included in economic models as a theoretical predictor of an individual’s earnings. This reflects that individuals in households with more accumulated wealth may have more resources to draw on or engage in different behaviors from poorer households (Cesarini et al. 2015). Most empirical work on the topic finds that greater household wealth as a driver of individual earnings, though the relationship can clearly be bi-directional (with increased household wealth contributing to increased individual earnings and vice-versa). • Dependency ratio: dependency ratio is conceptualized as an important driver of women’s earnings (and as a result, a country’s income and growth rates) in both micro- and macroeconomic models. Moreover, a range of new causal empirical work documents the impact that care burdens—reflected in greater dependency ratios—have as a driver of earnings (e.g., Clark et al. 2017). 84 Appendix D: Decompositions Table D1: Decomposition of agricultural outcomes in Growth Poles survey (1) (2) (3) (4) (5) (6) Production Production Production Plot yields Plot yields Plot yields value value value (million (million (million (million (million (million CDF/ha) CDF/ha) CDF/ha) CDF) CDF) CDF) Decomposition Prediction for male plot 0.2860*** 0.2860*** 0.2860*** 4.4646*** 4.4646*** 4.4646*** managers (0.010) (0.010) (0.010) (0.779) (0.780) (0.780) Prediction for female plot 0.2054*** 0.2054*** 0.2054*** 3.5959*** 3.5959*** 3.5959*** managers (0.012) (0.012) (0.012) (0.593) (0.594) (0.595) Difference 0.0806*** 0.0806*** 0.0806*** 0.8687 0.8687 0.8687 (0.016) (0.016) (0.016) (0.979) (0.980) (0.981) Explained -0.0068 0.0164 0.0158 -1.3129 -2.1096 -1.7522 (0.026) (0.026) (0.025) (1.478) (1.518) (1.606) Unexplained 0.0874*** 0.0642** 0.0648** 2.1816 2.9783** 2.6209* (0.029) (0.028) (0.028) (1.464) (1.458) (1.509) Explained gap Manager age -0.0021 -0.0007 -0.0010 -0.0009 -0.0606 0.0033 (0.004) (0.004) (0.004) (0.103) (0.107) (0.120) Manager married/cohabiting -0.0192 -0.0187 -0.0199 1.6052* 1.6910* 1.7807* (0.031) (0.031) (0.031) (0.943) (0.974) (1.020) Manager attended or completed -0.0030 0.0004 0.0008 -1.8401** -1.6677** -1.5155** high school (0.016) (0.015) (0.015) (0.727) (0.703) (0.686) Female HH head -0.0166 -0.0014 -0.0012 -2.1232 -2.0437 -1.4216 (0.031) (0.030) (0.030) (1.595) (1.622) (1.789) HH head age -0.0025 -0.0042 -0.0029 -0.3485 -0.2546 -0.3528 (0.008) (0.008) (0.008) (0.258) (0.264) (0.303) HH head married/cohabiting 0.0209 0.0193 0.0194 0.2408 0.1397 -0.0587 (0.031) (0.032) (0.032) (0.993) (1.049) (1.098) HH head attended or completed 0.0021 -0.0013 -0.0006 1.3627*** 1.2502** 1.1543** high school (0.012) (0.012) (0.012) (0.521) (0.491) (0.491) Household size 0.0009 -0.0003 -0.0000 -0.3567* -0.3364 -0.3481* (0.003) (0.003) (0.003) (0.211) (0.210) (0.209) Dependency ratio -0.0005 -0.0017 -0.0016 -0.2432 -0.2489 -0.2456 (0.002) (0.002) (0.002) (0.225) (0.224) (0.224) HH owns radio 0.0080* 0.0069 0.0070 0.4637 0.4147 0.3943 (0.005) (0.005) (0.005) (0.290) (0.274) (0.270) HH owns television 0.0034 0.0031 0.0030 0.4075 0.4209 0.4352 85 (0.002) (0.002) (0.002) (0.278) (0.273) (0.275) HH owns cellular phone 0.0070*** 0.0065*** 0.0060** 0.0346 0.0393 0.0148 (0.003) (0.003) (0.003) (0.132) (0.130) (0.131) HH owns bicycle/motorcycle -0.0049 -0.0065* -0.0055 -0.0005 0.0287 0.0372 (0.003) (0.003) (0.003) (0.319) (0.312) (0.307) HH number of plots -0.0002 -0.0002 0.0205 0.0221 (0.000) (0.000) (0.026) (0.028) - - Plot surface (ha) 0.0090** 0.0103*** 1.3184*** 1.3223*** (0.004) (0.004) (0.232) (0.239) Number of unpaid workers on 0.0050*** 0.0043*** 0.1087* 0.0850 the plot (0.002) (0.002) (0.057) (0.054) Number of paid workers on the 0.0019* 0.0017 0.1313 0.1255 plot (0.001) (0.001) (0.083) (0.082) Use of pesticides 0.0003 0.0003 0.0127 0.0135 (0.001) (0.001) (0.017) (0.017) Use of fertilizers 0.0008 0.0008 -0.0046 -0.0034 (0.001) (0.001) (0.018) (0.018) Cash crops plot -0.0021 0.2780* (0.002) (0.143) Cassava plot -0.0000 -0.0651 (0.001) (0.058) Corn plot -0.0003 0.0089 (0.001) (0.036) Peanut plot -0.0042** -0.1922** (0.002) (0.092) Rice plot 0.0035** -0.0797 (0.002) (0.061) Unexplained gap Manager age 0.4784*** 0.4150*** 0.4103*** -0.7421 0.7850 -0.7073 (0.152) (0.147) (0.145) (3.956) (4.079) (4.621) Manager married/cohabiting 0.0346 0.0222 0.0336 -0.4375 -1.2069 -1.4097 (0.075) (0.076) (0.076) (2.308) (2.385) (2.499) Manager attended or completed 0.0938* 0.1006** 0.0937** 1.9840 2.4279 2.6539 high school (0.049) (0.046) (0.046) (2.238) (2.165) (2.114) Female HH head 0.0308 0.0185 0.0191 -1.0598 -0.8795 -1.2287 (0.033) (0.032) (0.031) (1.669) (1.697) (1.871) - HH head age -0.3759** -0.3566** 2.6125 1.7503 3.9500 0.4260*** (0.157) (0.153) (0.151) (4.767) (4.879) (5.597) HH head married/cohabiting 0.0033 0.0168 0.0031 -0.2410 0.6692 0.6359 (0.080) (0.081) (0.081) (2.523) (2.664) (2.789) HH head attended or completed -0.0549 -0.0638 -0.0571 -0.1158 -0.3580 -0.4674 high school (0.052) (0.049) (0.049) (2.192) (2.063) (2.071) 86 Household size 0.0209 0.0024 -0.0001 -4.3518* -4.2483* -4.1008 (0.037) (0.037) (0.038) (2.559) (2.544) (2.533) Dependency ratio 0.0010 0.0078 0.0083 2.3828 2.2832 2.3601 (0.019) (0.019) (0.019) (1.996) (1.987) (1.988) HH owns radio 0.0107 0.0110 0.0115 -1.4379 -1.2108 -1.1291 (0.015) (0.015) (0.015) (0.939) (0.888) (0.875) HH owns television -0.0106 -0.0108 -0.0106 -1.5397** -1.5221** -1.5295** (0.007) (0.007) (0.007) (0.741) (0.724) (0.729) HH owns cellular phone 0.0004 -0.0020 0.0006 0.0055 0.0782 0.1493 (0.011) (0.011) (0.011) (0.587) (0.579) (0.584) HH owns bicycle/motorcycle 0.0093 0.0094 0.0072 1.9598** 1.8586** 1.8130** (0.009) (0.008) (0.008) (0.830) (0.810) (0.799) HH number of plots -0.0091 -0.0082 -1.1251 -1.1835 (0.030) (0.030) (1.943) (2.020) Plot surface (ha) 0.0295** 0.0232* 1.4003* 1.5035* (0.014) (0.014) (0.796) (0.822) Number of unpaid workers on 0.0190 0.0257 -0.2209 -0.0789 the plot (0.019) (0.019) (0.893) (0.896) Number of paid workers on the -0.0100 -0.0078 -0.6183 -0.6331 plot (0.009) (0.009) (0.674) (0.675) Use of pesticides -0.0003 -0.0003 -0.0230 -0.0094 (0.002) (0.002) (0.050) (0.049) Use of fertilizers 0.0038 0.0039 0.0465 0.0482 (0.003) (0.003) (0.107) (0.105) Cash crops plot 0.0125 -0.5332 (0.013) (1.005) Cassava plot -0.0312 0.8342 (0.032) (2.136) Corn plot -0.0077 -1.8032 (0.020) (1.629) Peanut plot -0.0309* 0.7370 (0.018) (1.043) Rice plot 0.0061 0.0007 (0.004) (0.145) Observations 5162 5162 5162 5033 5033 5033 Robust standard errors in parentheses p<0.1 ** p<0.05 *** p<0.01 Table D2: Decomposition of agricultural outcomes in 1-2-3 Survey (1) (2) (3) (4) (5) (6) Production Production Production Plot yields Plot yields Plot yields value value value (million (million (million (million (million (million CDF/ha) CDF/ha) CDF/ha) CDF) CDF) CDF) Decomposition 87 Prediction for male HH heads 0.3438*** 0.3438*** 0.3442*** 1.7051*** 1.7051*** 1.7064*** (0.018) (0.018) (0.017) (0.338) (0.336) (0.328) Prediction for female HH heads 0.3014*** 0.3014*** 0.3018*** 1.5926*** 1.5926*** 1.5932*** (0.022) (0.022) (0.022) (0.373) (0.373) (0.369) Difference 0.0424** 0.0424** 0.0424** 0.1125 0.1125 0.1133 (0.019) (0.019) (0.018) (0.223) (0.223) (0.223) Explained 0.0324 0.0259 0.0358* -0.3231 -0.3926 -0.0732 (0.020) (0.020) (0.020) (0.239) (0.253) (0.281) Unexplained 0.0100 0.0165 0.0067 0.4356 0.5051 0.1864 (0.023) (0.023) (0.022) (0.317) (0.323) (0.317) Explained gap HH head age -0.0097** -0.0084** -0.0085** -0.0545 -0.0669 -0.0642 (0.004) (0.004) (0.004) (0.060) (0.062) (0.060) HH head married/cohabiting 0.0113 0.0039 0.0116 0.0575 0.0456 0.2204 (0.016) (0.015) (0.014) (0.276) (0.276) (0.293) HH head attended or completed 0.0038 0.0039 0.0026 -0.1960** -0.1837** -0.1810** high school (0.009) (0.009) (0.009) (0.093) (0.091) (0.088) HH size 0.0184*** 0.0093* 0.0084 -0.0788 -0.0221 -0.0308 (0.006) (0.005) (0.005) (0.069) (0.064) (0.063) HH dependency ratio 0.0067** 0.0047* 0.0053* 0.0168 0.0267 0.0300 (0.003) (0.003) (0.003) (0.037) (0.035) (0.039) HH owns a radio 0.0177** 0.0114 0.0095 -0.0932 -0.0415 -0.0922 (0.009) (0.008) (0.008) (0.062) (0.055) (0.068) HH owns a TV -0.0005 -0.0001 0.0001 0.0326 0.0335 0.0334 (0.002) (0.002) (0.002) (0.021) (0.022) (0.023) HH owns a cellular phone -0.0026 -0.0021 -0.0015 -0.0330 -0.0320 -0.0245 (0.004) (0.004) (0.004) (0.034) (0.033) (0.030) HH owns a bicycle/motorcycle 0.0120** 0.0115** 0.0105** 0.0340 0.0463 0.0450 (0.005) (0.005) (0.005) (0.032) (0.033) (0.034) Number of HH plots 0.0050* 0.0054* -0.0466** -0.0454** (0.003) (0.003) (0.022) (0.020) Total plots area (ha) 0.0016 0.0008 -0.0689** -0.0754** (0.002) (0.002) (0.030) (0.033) Number of paid workers per 0.0011 0.0008 -0.0241 -0.0257 plot (0.001) (0.001) (0.019) (0.018) Number of unpaid workers per 0.0052** 0.0039** -0.0131 -0.0258 plot (0.002) (0.002) (0.021) (0.024) HH uses fertilizers -0.0000 0.0001 -0.0021 -0.0022 (0.000) (0.000) (0.004) (0.004) HH uses pesticides 0.0004 0.0004 -0.0174 -0.0123 (0.001) (0.001) (0.016) (0.013) Proportion of HH plots with 0.0045 0.1981* cash crops (0.003) (0.114) 88 Proportion of HH plots with -0.0006 0.0113 cassava (0.001) (0.026) Proportion of HH plots with 0.0003 -0.0028 corn (0.003) (0.009) Proportion of HH plots with -0.0009 -0.0035 peanut (0.002) (0.007) Proportion of HH plots with 0.0041** 0.0181 rice (0.002) (0.047) Unexplained gap HH head age 0.0201 0.0229 0.0196 -1.3311 -1.3476 -1.4638 (0.053) (0.055) (0.056) (1.047) (1.067) (1.074) HH head married/cohabiting 0.0380 0.0340 0.0279 0.3595 0.3214 0.4658 (0.028) (0.028) (0.027) (0.558) (0.544) (0.576) HH head attended or completed 0.0029 -0.0002 -0.0010 0.1347 0.1513 0.1978 high school (0.014) (0.014) (0.014) (0.151) (0.154) (0.155) HH size 0.0017 0.0033 0.0097 0.2769 0.2094 0.2501 (0.031) (0.030) (0.029) (0.348) (0.328) (0.323) HH dependency ratio -0.0061 -0.0044 -0.0037 0.0822 0.0694 0.0959 (0.011) (0.010) (0.011) (0.189) (0.185) (0.193) HH owns a radio -0.0056 -0.0045 -0.0041 0.1916 0.1404 0.1579 (0.016) (0.015) (0.015) (0.120) (0.111) (0.121) HH owns a TV -0.0004 -0.0009 -0.0012 0.0304 0.0306 0.0294 (0.004) (0.004) (0.004) (0.046) (0.047) (0.044) HH owns a cellular phone -0.0050 -0.0066 -0.0054 0.0426 0.0588 0.0372 (0.009) (0.009) (0.009) (0.085) (0.089) (0.088) HH owns a bicycle/motorcycle -0.0109 -0.0118 -0.0116 -0.1789* -0.1716* -0.1594* (0.008) (0.008) (0.008) (0.098) (0.094) (0.096) Number of HH plots 0.0315 0.0170 0.2998 0.3303 (0.036) (0.035) (0.340) (0.332) Total plots area (ha) -0.0207** -0.0173** 0.0000 0.0435 (0.009) (0.008) (0.052) (0.062) Number of paid workers per 0.0077* 0.0068 0.1021** 0.1071** plot (0.004) (0.004) (0.047) (0.049) Number of unpaid workers per 0.0019 0.0000 -0.0624 0.0481 plot (0.010) (0.010) (0.120) (0.129) HH uses fertilizers 0.0060** 0.0055** 0.0069 0.0046 (0.003) (0.003) (0.010) (0.014) HH uses pesticides -0.0006 -0.0005 -0.0093 -0.0062 (0.001) (0.001) (0.014) (0.012) Proportion of HH plots with -0.0038 0.0329 cash crops (0.004) (0.120) 89 Proportion of HH plots with -0.0174 0.1337 cassava (0.028) (0.487) Proportion of HH plots with -0.0209 0.1141 corn (0.016) (0.199) Proportion of HH plots with 0.0055 0.0208 peanut (0.008) (0.106) Proportion of HH plots with 0.0168* -0.3185 rice (0.010) (0.240) Observations 13273 13273 13253 12250 12250 12238 Robust standard errors in parentheses p<0.1 ** p<0.05 *** p<0.01 Table D3: Decomposition of agricultural outcomes in Childcare survey (1) (2) (3) (4) (5) (6) Plot Plot Plot Plot yields Plot yields Plot yields production production production (million (million (million (million (million (million CDF/ha) CDF/ha) CDF/ha) CDF) CDF) CDF) Decomposition Prediction for male plot 55.7656** 55.7656** 55.7656** 0.3456*** 0.3456*** 0.3456*** managers * * * (0.006) (0.006) (0.006) (1.181) (1.182) (1.182) Prediction for female plot 38.1140** 38.1140** 38.1140** 0.2971*** 0.2971*** 0.2971*** managers * * * (0.008) (0.008) (0.008) (1.331) (1.333) (1.335) 17.6517** 17.6517** 17.6517** Difference 0.0485*** 0.0485*** 0.0485*** * * * (0.010) (0.010) (0.010) (1.779) (1.781) (1.784) 13.6114** 13.4196** Explained 0.0484*** 0.0663*** 0.0439*** 6.9330*** * * (0.012) (0.013) (0.013) (2.059) (3.327) (3.289) 10.7187** Unexplained 0.0001 -0.0178 0.0047 4.0403 4.2320 * (0.015) (0.015) (0.014) (2.728) (3.694) (3.622) Explained gap Manager age 0.0021 0.0042** 0.0019 -0.0167 0.7655** 0.6481* (0.002) (0.002) (0.001) (0.309) (0.379) (0.364) Manager married/cohabiting 0.0500*** 0.0186 0.0373** 2.7199 -2.8045 -2.8536 (0.014) (0.018) (0.018) (2.616) (5.678) (5.297) Manager attended or completed 0.0005 0.0127 0.0059 2.0228 4.8781** 4.7525** high school (0.008) (0.009) (0.008) (1.640) (2.160) (2.120) Female HH head 0.0193 0.0239* 0.0171 6.9451*** 9.4140*** 9.3564*** (0.012) (0.013) (0.012) (2.650) (2.987) (2.983) HH head age -0.0020 0.0053 -0.0019 -0.7507 1.5232 1.1086 (0.005) (0.005) (0.005) (1.127) (1.199) (1.196) 90 - HH head married/cohabiting -0.0207 -0.0344* -4.0880 -0.8687 -0.7959 0.0465*** (0.016) (0.019) (0.019) (3.184) (5.583) (5.235) HH head attended or completed 0.0017 -0.0056 -0.0022 -0.8020 -2.3497** -2.1750** high school (0.004) (0.005) (0.004) (0.818) (1.096) (1.072) HH size 0.0000 -0.0003 -0.0002 0.0208 -0.1019 -0.1042 (0.000) (0.000) (0.000) (0.059) (0.092) (0.093) HH dependency ratio 0.0025** 0.0027** 0.0023** -0.0099 0.2161 0.2093 (0.001) (0.001) (0.001) (0.184) (0.172) (0.168) HH owns a radio 0.0017 0.0027* 0.0029* -0.1740 0.0208 0.0836 (0.002) (0.002) (0.002) (0.349) (0.333) (0.324) HH owns a TV 0.0001 0.0001 -0.0001 0.0458 0.0456 0.0220 (0.000) (0.000) (0.000) (0.087) (0.084) (0.078) HH owns a cell phone 0.0038*** 0.0037*** 0.0039*** -0.2961* -0.2942* -0.2825* (0.001) (0.001) (0.001) (0.176) (0.168) (0.165) HH owns a bicycle/motorcycle 0.0006 -0.0000 -0.0002 0.1397 0.1552 0.1415 (0.001) (0.001) (0.001) (0.187) (0.179) (0.171) HH number of plots 0.0056*** 0.0040*** 1.0587*** 0.9763*** (0.002) (0.001) (0.299) (0.279) Plot surface (ha) -0.0034** -0.0031** 0.9409** 0.9716** (0.002) (0.001) (0.387) (0.400) Number of HH workers on plot 0.0009 0.0003 -0.6188** -0.6007** (0.001) (0.001) (0.270) (0.262) Number of non-HH workers on -0.0002 -0.0001 -0.0780 -0.0743 plot (0.001) (0.001) (0.446) (0.424) Fertilizer use on plot -0.0026 -0.0017 0.1429 0.1473 (0.002) (0.002) (0.209) (0.210) Pesticides use on plot 0.0063** 0.0067*** 0.7212*** 0.7070*** (0.003) (0.003) (0.273) (0.257) Cash crops plot 0.0059*** 0.9473*** (0.002) (0.295) Cassava plot -0.0045** -0.3406** (0.002) (0.167) Corn plot -0.0021 -0.0591 (0.002) (0.066) Peanut plot -0.0034 -0.2174 (0.003) (0.186) Rice plot 0.0002 0.1106 (0.000) (0.161) Unexplained gap - Manager age -0.0361 0.0757 0.0156 34.6352** -1.5286 -3.7739 * (0.069) (0.063) (0.060) (13.436) (14.212) (14.145) 26.4624** Manager married/cohabiting 0.1201** 0.0242 0.1038 1.7292 1.3506 * 91 (0.050) (0.063) (0.065) (9.443) (20.501) (19.125) Manager attended or completed 0.0131 0.0485 0.0312 1.4111 12.3701* 11.3783 high school (0.026) (0.030) (0.026) (5.383) (7.055) (6.927) Female HH head -0.0116 -0.0146 -0.0200 -3.7204 -3.5183 -4.0265 (0.013) (0.014) (0.012) (2.773) (3.118) (3.114) HH head age -0.0464 -0.1697** -0.0864 26.3495* -13.0327 -9.0471 (0.075) (0.070) (0.066) (15.600) (16.540) (16.532) HH head married/cohabiting -0.1236** -0.0265 -0.1299* -12.8417 8.6917 6.1722 (0.060) (0.071) (0.071) (12.104) (21.252) (19.927) HH head attended or completed - - -0.0084 -0.0531 -0.0345 -3.6939 high school 17.6449** 16.3006** (0.030) (0.035) (0.030) (6.142) (8.057) (7.901) HH size 0.0575** 0.0778*** 0.0807*** 0.0369 5.7619 5.3244 (0.029) (0.030) (0.028) (5.721) (5.781) (5.773) HH dependency ratio -0.0029 -0.0082 -0.0034 3.9723 2.3524 2.7703 (0.015) (0.014) (0.013) (3.027) (2.718) (2.656) HH owns a radio 0.0127 0.0086 0.0066 -0.3841 0.1646 1.1071 (0.008) (0.008) (0.008) (1.786) (1.708) (1.663) HH owns a TV -0.0027 -0.0022 -0.0005 0.0577 0.0768 0.0888 (0.003) (0.003) (0.002) (0.532) (0.514) (0.487) - - HH owns a cell phone -0.0219** -1.2210 -0.8016 -1.0557 0.0253*** 0.0280*** (0.010) (0.009) (0.009) (1.869) (1.766) (1.741) HH owns a bicycle/motorcycle 0.0045 0.0061 0.0054 1.2011* 1.2546* 1.2501* (0.004) (0.004) (0.004) (0.708) (0.674) (0.647) - - HH number of plots -0.0082 0.0077 8.7314*** 8.2820*** (0.013) (0.013) (2.730) (2.643) Plot surface (ha) 0.0058 0.0048 -0.4450 -0.6477* (0.004) (0.004) (0.362) (0.379) - - Number of HH workers on plot -1.8106 -1.4504 0.0604*** 0.0587*** (0.020) (0.019) (4.040) (3.939) Number of non-HH workers on 0.0111** 0.0116*** 0.2764 0.3339 plot (0.004) (0.004) (0.801) (0.781) Fertilizer use on plot 0.0044 0.0044 0.5242 0.2387 (0.004) (0.003) (0.334) (0.331) Pesticides use on plot -0.0027 -0.0022 0.9227** 0.8984** (0.004) (0.004) (0.404) (0.377) Cash crops plot -0.0009 1.0809** (0.002) (0.455) Cassava plot -0.0035 -9.1314** (0.024) (4.209) Corn plot 0.0054 -0.5886 (0.004) (0.500) Peanut plot 0.0096* 1.5504 92 (0.005) (0.971) Rice plot 0.0008 -0.1066 (0.001) (0.152) Observations 3648 3648 3648 3648 3648 3648 Robust standard errors in parentheses p<0.1 ** p<0.05 *** p<0.01 Table D4: Decomposition of business profits in Growth Poles survey (1) (2) Profits in last month Profits in last month (million of CDF) (million of CDF) Decomposition Prediction for male managers 0.1916*** 0.1916*** (0.026) (0.026) Prediction for female managers 0.0551*** 0.0551*** (0.015) (0.015) Difference 0.1365*** 0.1365*** (0.030) (0.030) Explained 0.0314 0.0288 (0.021) (0.020) Unexplained 0.1051*** 0.1077*** (0.030) (0.029) Explained gap Manager age 0.0001 0.0001 (0.001) (0.001) Manager married/cohabiting -0.0264 -0.0224 (0.017) (0.016) Manager attended or completed high school -0.0172* -0.0168* (0.009) (0.009) Female HH head 0.0508** 0.0368 (0.025) (0.023) Household size 0.0076 0.0073 (0.008) (0.007) Dependency ratio 0.0005 0.0002 (0.001) (0.001) HH owns radio -0.0018 -0.0045 (0.006) (0.005) HH owns television -0.0005 -0.0042 (0.004) (0.005) HH owns cellular phone 0.0093 0.0073 (0.006) (0.005) HH owns bicycle 0.0034 0.0029 (0.005) (0.005) Business location in building/rental -0.0003 (0.001) Number of unpaid workers 0.0001 (0.004) 93 Value of paid labor (million CDF) -0.0003 (0.003) Value of intermediate inputs (million CDF) 0.0155* (0.009) Value of paid utilities (million CDF) 0.0014 (0.003) Unexplained gap Manager age 0.1030 0.1069 (0.119) (0.115) Manager married/cohabiting -0.0270 -0.0574 (0.077) (0.075) Manager attended or completed high school -0.0768 -0.0655 (0.047) (0.044) Female HH head -0.0237 -0.0232 (0.029) (0.027) Household size 0.1258 0.0579 (0.094) (0.091) Dependency ratio -0.0103 -0.0110 (0.024) (0.022) HH owns radio -0.0107 -0.0269 (0.032) (0.031) HH owns television -0.0007 0.0032 (0.014) (0.015) HH owns cellular phone 0.0802*** 0.0518* (0.031) (0.028) HH owns bicycle 0.0129 0.0220 (0.016) (0.016) Business location in building/rental 0.0114 (0.008) Number of unpaid workers 0.0395 (0.033) Value of paid labor (million CDF) 0.0141** (0.006) Value of intermediate inputs (million CDF) 0.0212 (0.027) Value of paid utilities (million CDF) -0.0025 (0.002) Observations 425 425 Robust standard errors in parentheses p<0.1 ** p<0.05 *** p<0.01 Table D5: Decomposition of business profits in 1-2-3 Survey (1) (2) Total profits (million Total profits (million CDF) CDF) Decomposition 94 Prediction for male managers 0.2491*** 0.2491*** (0.031) (0.031) Prediction for female managers 0.0798*** 0.0798*** (0.024) (0.024) Difference 0.1693*** 0.1693*** (0.030) (0.030) Explained 0.0042 0.0824*** (0.019) (0.025) Unexplained 0.1651*** 0.0869** (0.035) (0.036) Explained gap Manager age 0.0000 0.0000 (0.000) (0.000) Manager married/cohabiting 0.0030 0.0018 (0.002) (0.002) Manager attended or completed high school -0.0020 -0.0030 (0.006) (0.006) Female HH head -0.0087 -0.0063 (0.014) (0.014) HH size -0.0009 -0.0006 (0.001) (0.001) HH dependency ratio 0.0019 0.0015 (0.003) (0.003) HH owns a radio 0.0018 0.0019 (0.002) (0.002) HH owns a TV 0.0002 -0.0001 (0.001) (0.000) HH owns a cellular phone 0.0015 0.0009 (0.001) (0.001) HH owns a bicycle/motorcycle 0.0004 0.0004 (0.002) (0.002) Business location in a building/rental 0.0033 (0.004) Business location at home 0.0087 (0.006) Business location in the street -0.0007 (0.004) Number of unpaid workers 0.0078* (0.004) Number of paid workers 0.0083 (0.005) Value of physical capital (million CDF) 0.0266** (0.011) Value of intermediate inputs (million CDF) 0.0010 (0.004) Value of paid fines (million CDF) 0.0016 95 (0.005) Value of paid utilities (million CDF) 0.0244** (0.012) Unexplained gap Manager age -0.0682 -0.0656 (0.080) (0.081) Manager married/cohabiting 0.0943** 0.0774* (0.043) (0.044) Manager attended or completed high school -0.0724 -0.0622 (0.059) (0.059) Female individual 0.0173 0.0147 (0.024) (0.024) HH size 0.0330 0.0380 (0.048) (0.048) HH dependency ratio -0.0106 -0.0152 (0.031) (0.031) HH owns a radio 0.0120 0.0130 (0.040) (0.039) HH owns a TV 0.0169 0.0059 (0.028) (0.033) HH owns a cellular phone 0.0208 0.0153 (0.039) (0.039) HH owns a bicycle/motorcycle 0.0191* 0.0148 (0.011) (0.011) Business location in a building/rental 0.0127* (0.007) Business location at home -0.0745** (0.030) Business location in the street -0.0212 (0.023) Number of unpaid workers 0.1254** (0.062) Number of paid workers 0.0096 (0.010) Value of physical capital (million CDF) -0.0199 (0.021) Value of intermediate inputs (million CDF) -0.0163 (0.012) Value of paid fines (million CDF) -0.0143 (0.012) Value of paid utilities (million CDF) -0.0581*** (0.021) Observations 4079 4079 Robust standard errors in parentheses p<0.1 ** p<0.05 *** p<0.01 Table D6: Decomposition of business profits in SME survey 96 (1) (2) Profits in last month Profits in last month (million of CDF) (million of CDF) Decomposition Prediction for male managers 1.5654*** 1.5654*** (0.296) (0.295) Prediction for female managers 0.8507*** 0.8507*** (0.138) (0.136) Difference 0.7147** 0.7147** (0.326) (0.325) Explained 0.0123 0.2734 (0.133) (0.188) Unexplained 0.7025** 0.4413* (0.303) (0.266) Explained gap Manager age -0.0124 -0.0178 (0.069) (0.054) Manager attended or completed high school 0.0247 0.0141 (0.024) (0.020) HH head female 0.0000 0.0000 (0.000) (0.004) Household size 0.0000 0.0000 (0.060) (0.040) HH dependency ratio 0.0000 0.0000 (0.029) (0.022) HH owns a bicycle/motorcycle 0.0000 0.0000 (0.021) (0.021) HH owns a radio 0.0000 0.0000 (0.000) (0.005) HH owns a TV 0.0000 0.0000 (0.010) (0.016) Number of workers 0.1537* (0.090) Value of physical capital (million of CDF) 0.1235 (0.093) Unexplained gap Manager age -0.5523 -0.5700 (1.172) (0.919) Manager attended or completed high school 0.5482 0.3265 (0.541) (0.492) HH head female 0.0493 0.0128 (0.154) (0.137) Household size 1.3244 0.6474 (1.089) (0.864) 97 HH dependency ratio -0.1599 -0.0183 (0.374) (0.344) HH owns a bicycle/motorcycle 0.0596 0.0273 (0.228) (0.201) HH owns a radio 0.4228 0.4103 (0.534) (0.446) HH owns a TV 0.0498 0.2813 (0.484) (0.496) Number of workers 0.3671 (0.291) Value of physical capital (million of CDF) -0.1185* (0.070) Observations 272 272 Robust standard errors in parentheses p<0.1 ** p<0.05 *** p<0.01 Table D7: Decomposition of labor outcomes in Growth Poles survey (1) (2) (3) (4) Currently in Wage Wage income in Labor force wage employment in past 12 months participation employment past 12m (thousand CDF) Decomposition Prediction for male HH members 0.6793*** 0.0579*** 0.1571*** 75.3768*** (0.007) (0.004) (0.006) (5.023) Prediction for female HH members 0.7773*** 0.0318*** 0.0779*** 23.7892*** (0.006) (0.003) (0.004) (2.450) Difference -0.0980*** 0.0261*** 0.0791*** 51.5876*** (0.010) (0.005) (0.007) (5.588) Explained -0.0384*** 0.0083*** 0.0086*** 4.7094** (0.007) (0.002) (0.003) (2.027) Unexplained -0.0597*** 0.0178*** 0.0705*** 46.8782*** (0.008) (0.005) (0.007) (5.784) Explained gap Individual age -0.0094*** -0.0014*** -0.0000 0.2839 (0.002) (0.000) (0.000) (0.363) Individual married/cohabiting 0.0038 0.0000 0.0008 0.3562 (0.006) (0.001) (0.001) (0.522) Individual attended or completed -0.0089*** 0.0098*** 0.0077*** 2.3923 high school (0.002) (0.001) (0.002) (1.477) Female HH head -0.0222*** -0.0043*** -0.0044*** 0.0659 (0.003) (0.001) (0.002) (1.135) Household size -0.0022*** 0.0001 -0.0006* -0.5926** (0.001) (0.000) (0.000) (0.268) Dependency ratio -0.0011* 0.0004 -0.0001 0.0355 (0.001) (0.000) (0.001) (0.518) HH owns radio -0.0005 0.0010*** 0.0005 0.5292 (0.001) (0.000) (0.001) (0.472) 98 HH owns television 0.0004 0.0006** 0.0002 -0.2079 (0.000) (0.000) (0.000) (0.165) HH owns cellular phone 0.0002 0.0013*** 0.0017*** 0.7726** (0.000) (0.000) (0.001) (0.359) HH owns bicycle/motorcycle 0.0003 0.0006** 0.0007* 0.2565 (0.000) (0.000) (0.000) (0.280) Unexplained gap Individual age 0.0382* 0.0268** -0.0126 -13.1212 (0.021) (0.013) (0.020) (16.481) Individual married/cohabiting -0.0015 0.0155** 0.0585*** 33.1100*** (0.011) (0.006) (0.010) (7.920) Individual attended or completed 0.0247*** -0.0008 -0.0045 -1.8797 high school (0.009) (0.006) (0.009) (7.079) Female HH head -0.0360*** -0.0044* -0.0165*** -6.7403** (0.006) (0.002) (0.004) (2.896) Household size -0.0854*** -0.0007 -0.0497*** -38.6068*** (0.017) (0.012) (0.016) (11.926) Dependency ratio -0.0126* -0.0057 -0.0065 1.6604 (0.007) (0.005) (0.007) (5.458) HH owns radio -0.0004 0.0039 0.0126* 5.7904 (0.007) (0.005) (0.007) (5.895) HH owns television -0.0034* -0.0002 -0.0013 -1.5184 (0.002) (0.002) (0.002) (1.430) HH owns cellular phone 0.0073 0.0119*** 0.0077 6.7024 (0.006) (0.004) (0.006) (5.040) HH owns bicycle/motorcycle 0.0023 0.0054* 0.0075* 3.5957 (0.004) (0.003) (0.004) (3.572) Observations 8233 8123 8233 8233 Robust standard errors in parentheses p<0.1 ** p<0.05 *** p<0.01 Table D8: Decomposition of labor outcomes in 1-2-3 Survey (1) (2) (3) Monthly income Labor force Currently in wage from wages participation employment (thousand CDF) Decomposition Prediction for male HH members 0.7473*** 0.1691*** 11.9999*** (0.007) (0.005) (0.491) Prediction for female HH members 0.6769*** 0.0417*** 2.6487*** (0.008) (0.002) (0.185) Difference 0.0704*** 0.1274*** 9.3512*** (0.007) (0.004) (0.404) Explained -0.0188*** 0.0203*** 1.5841*** (0.003) (0.002) (0.158) 99 Unexplained 0.0893*** 0.1071*** 7.7671*** (0.007) (0.005) (0.396) Explained gap Individual age 0.0025*** 0.0014*** 0.1227*** (0.000) (0.000) (0.024) Individual married/cohabiting 0.0012 0.0003 0.0200 (0.002) (0.000) (0.031) Individual attended or completed high -0.0096*** 0.0187*** 1.2652*** school (0.002) (0.001) (0.086) Female HH head -0.0136*** -0.0019** 0.0358 (0.001) (0.001) (0.086) HH size -0.0001 -0.0000 -0.0024 (0.000) (0.000) (0.008) HH dependency ratio 0.0009** 0.0007*** 0.0817*** (0.000) (0.000) (0.018) HH owns a radio -0.0002 0.0010*** 0.0558*** (0.000) (0.000) (0.016) HH owns a TV 0.0001 -0.0001 -0.0137 (0.000) (0.000) (0.038) HH owns a cellular phone -0.0004** 0.0009*** 0.0745*** (0.000) (0.000) (0.020) HH owns a bicycle/motorcycle 0.0000 -0.0001 -0.0264*** (0.000) (0.000) (0.009) Unexplained gap Individual age 0.0080 0.0422*** 4.4215*** (0.013) (0.009) (0.695) Individual married/cohabiting 0.0571*** 0.0750*** 5.6758*** (0.009) (0.006) (0.478) Individual attended or completed high 0.0167*** 0.0269*** 2.1176*** school (0.006) (0.004) (0.272) Female HH head -0.0317*** -0.0092*** -0.8602*** (0.003) (0.002) (0.188) HH size -0.0539*** -0.0180*** -1.7084*** (0.013) (0.007) (0.618) HH dependency ratio -0.0125*** -0.0014 -0.5335*** (0.004) (0.003) (0.204) HH owns a radio 0.0009 0.0123*** 0.7732*** (0.004) (0.003) (0.208) HH owns a TV 0.0121*** 0.0148*** 2.5789*** (0.004) (0.003) (0.314) HH owns a cellular phone 0.0143*** 0.0255*** 2.3739*** (0.005) (0.004) (0.342) HH owns a bicycle/motorcycle -0.0033 -0.0023 -0.3328*** (0.003) (0.002) (0.127) Observations 58107 58106 58106 Robust standard errors in parentheses 100 p<0.1 ** p<0.05 *** p<0.01 Table D9: Decomposition of labor outcomes in Childcare survey (1) (2) (3) (4) Wage Monthly wage Labor Force Currently in wage employment in income (thousand Participation employment past 12 months CDF) Decomposition Prediction for male HH 0.9176*** 0.0356*** 0.0400*** 0.1235*** members (0.005) (0.003) (0.004) (0.021) Prediction for female HH 0.9435*** 0.0382*** 0.0428*** 0.1168*** members (0.004) (0.003) (0.003) (0.019) Difference -0.0259*** -0.0026 -0.0028 0.0067 (0.006) (0.005) (0.005) (0.028) Explained 0.0118*** 0.0066*** 0.0063*** 0.0143 (0.004) (0.002) (0.002) (0.010) Unexplained -0.0377*** -0.0092* -0.0091* -0.0076 (0.007) (0.005) (0.005) (0.029) Explained gap Individual age 0.0052*** 0.0021** 0.0016* -0.0028 (0.001) (0.001) (0.001) (0.005) Individual married/cohabiting 0.0121*** -0.0010 -0.0010 -0.0043 (0.002) (0.001) (0.001) (0.004) Individual attended or -0.0006 0.0050*** 0.0046*** 0.0123* completed high school (0.002) (0.001) (0.001) (0.007) Female HH head -0.0024 -0.0003 0.0001 0.0019 (0.002) (0.001) (0.001) (0.006) HH size -0.0011 -0.0002 -0.0002 -0.0009 (0.001) (0.000) (0.000) (0.001) HH dependency ratio -0.0016*** -0.0005 -0.0004 0.0013 (0.001) (0.000) (0.000) (0.001) HH owns a radio 0.0005* 0.0006** 0.0008** 0.0031** (0.000) (0.000) (0.000) (0.002) HH owns a TV -0.0000 0.0000 0.0000 0.0001 (0.000) (0.000) (0.000) (0.001) HH owns a cell phone 0.0000 0.0005* 0.0005* 0.0010 (0.000) (0.000) (0.000) (0.001) HH owns a 0.0003 0.0000 0.0000 -0.0001 bicycle/motorcycle (0.000) (0.000) (0.000) (0.001) Unexplained gap Individual age 0.0275 -0.0066 -0.0105 -0.0129 101 (0.022) (0.014) (0.014) (0.085) Individual married/cohabiting 0.0205 -0.0146 -0.0098 -0.0108 (0.015) (0.009) (0.009) (0.057) Individual attended or 0.0162** -0.0010 -0.0011 0.0204 completed high school (0.008) (0.005) (0.006) (0.033) Female HH head -0.0085* -0.0040** -0.0040** -0.0123 (0.005) (0.002) (0.002) (0.012) HH size -0.0435* -0.0114 -0.0058 0.0156 (0.024) (0.015) (0.016) (0.082) HH dependency ratio 0.0308** 0.0029 0.0015 0.0015 (0.012) (0.010) (0.011) (0.033) HH owns a radio 0.0083 0.0036 0.0032 -0.0032 (0.005) (0.004) (0.005) (0.028) HH owns a TV -0.0004 -0.0010 -0.0010 -0.0132 (0.001) (0.002) (0.002) (0.011) HH owns a cell phone -0.0092 -0.0063 -0.0057 -0.0123 (0.006) (0.005) (0.005) (0.026) HH owns a 0.0013 -0.0008 -0.0006 0.0021 bicycle/motorcycle (0.002) (0.002) (0.002) (0.013) Observations 6572 6565 6573 6569 Robust standard errors in parentheses p<0.1 ** p<0.05 *** p<0.01 Table D10: Decomposition of labor outcomes in DHS survey (1) (2) (3) Labor force Currently in wage Wage employment in participation employment past 12 months Decomposition Prediction for male HH members 0.8221*** 0.2416*** 0.2609*** (0.008) (0.009) (0.010) Prediction for female HH members 0.7228*** 0.0555*** 0.0599*** (0.010) (0.003) (0.003) Difference 0.0993*** 0.1861*** 0.2010*** (0.011) (0.009) (0.009) Explained 0.0064 0.0223*** 0.0244*** (0.005) (0.003) (0.004) Unexplained 0.0929*** 0.1638*** 0.1766*** (0.011) (0.009) (0.010) Explained gap Individual age 0.0298*** 0.0098*** 0.0096*** (0.002) (0.001) (0.001) Individual married/cohabiting -0.0093*** -0.0019*** -0.0017*** (0.002) (0.001) (0.001) Individual attended or completed -0.0100*** 0.0163*** 0.0174*** high school (0.002) (0.002) (0.002) 102 Female HH head -0.0032** -0.0025* -0.0021 (0.002) (0.001) (0.001) HH size 0.0008 0.0003 0.0002 (0.001) (0.000) (0.000) HH dependency ratio -0.0020* -0.0001 0.0004 (0.001) (0.001) (0.001) HH owns a cell phone -0.0002 0.0006* 0.0007* (0.000) (0.000) (0.000) HH owns a bicycle/motorcycle -0.0001 0.0003 0.0002 (0.000) (0.000) (0.000) HH owns a radio 0.0003 0.0010*** 0.0010** (0.000) (0.000) (0.000) HH owns a TV 0.0002 -0.0004 -0.0005 (0.000) (0.000) (0.000) Unexplained gap Individual age -0.1273*** 0.0430** 0.0425** (0.021) (0.020) (0.020) Individual married/cohabiting 0.0693*** 0.0502*** 0.0487*** (0.012) (0.011) (0.010) Individual attended or completed 0.0076 0.0142 0.0165* high school (0.009) (0.009) (0.009) Female HH head -0.0431 -0.0198 -0.0233 (0.028) (0.025) (0.027) HH size -0.0265 -0.0265 -0.0221 (0.018) (0.018) (0.019) HH dependency ratio -0.0192* 0.0145 0.0091 (0.010) (0.011) (0.010) HH owns a cell phone -0.0051 0.0046 0.0050 (0.009) (0.007) (0.008) HH owns a bicycle/motorcycle 0.0007 0.0032** 0.0025 (0.002) (0.002) (0.002) HH owns a radio -0.0001 0.0149** 0.0129* (0.008) (0.007) (0.007) HH owns a TV 0.0049 0.0081 0.0089 (0.006) (0.006) (0.006) Observations 27434 27434 27434 Robust standard errors in parentheses p<0.1 ** p<0.05 *** p<0.01 103 Table D11: Decomposition of wage outcomes in 1-2-3 Survey (1) (2) (3) Monthly income Monthly income Monthly income from wages from wages from wages (thousand FC) (thousand FC) (thousand FC) Decomposition Prediction for male HH members 11.9999*** 11.9999*** 11.9999*** (0.498) (0.489) (0.497) Prediction for male HH members 2.6487*** 2.6487*** 2.6487*** (0.195) (0.194) (0.198) Difference 9.3512*** 9.3512*** 9.3512*** (0.399) (0.391) (0.405) Explained 5.5867*** 5.2508*** 7.0857*** (0.267) (0.260) (0.313) Unexplained 3.7644*** 4.1004*** 2.2655*** (0.327) (0.316) (0.272) Explained gap Individual age 0.0214** 0.0177** 0.0166** (0.009) (0.009) (0.008) Individual married/cohabiting 0.0008 0.0010 -0.0005 (0.002) (0.002) (0.001) Individual attended or completed high school 0.7877*** 0.5213*** 0.0539 (0.074) (0.060) (0.038) Female HH head 0.1059 0.0919 0.0600 (0.065) (0.077) (0.057) HH size -0.0004 -0.0015 -0.0004 (0.002) (0.005) (0.001) HH dependency ratio 0.0651*** 0.0372** 0.0224* (0.015) (0.015) (0.013) Catholic -0.0006 -0.0006 -0.0004 (0.007) (0.007) (0.005) Protestant -0.0205** -0.0218*** -0.0189*** (0.008) (0.008) (0.007) Kimbanquist 0.0002 0.0003 0.0002 (0.001) (0.001) (0.001) Muslim -0.0008 -0.0013 0.0026 (0.003) (0.003) (0.002) Other Christian -0.0117 -0.0100 -0.0083 (0.007) (0.006) (0.005) Public administration 1.9674*** 1.9642*** -0.5027*** (0.144) (0.143) (0.158) Public and para -public enterprises 1.0987*** 1.0835*** 0.0293 104 (0.112) (0.111) (0.076) Private entreprises 0.0753 0.0814 0.0386 (0.048) (0.051) (0.024) Associations (co-ops/NGOs/church) 0.2207*** 0.2143*** -0.0092 (0.036) (0.035) (0.019) Labor union exists 1.6071*** 1.5354*** 0.7691*** (0.195) (0.193) (0.109) Member of labor union -0.2484*** -0.2402*** -0.1450** (0.086) (0.085) (0.063) HH owns a radio 0.0066 -0.0042 (0.012) (0.008) HH owns a TV -0.0108 -0.0070 (0.030) (0.020) HH owns a cellular phone 0.0457*** 0.0213*** (0.014) (0.007) HH owns a bicycle/motorcycle -0.0305*** -0.0211*** (0.008) (0.006) Manager or supervisor 3.2567*** (0.240) Qualified worker 2.4394*** (0.217) Semi-qualified worker 1.0890*** (0.110) Unexplained gap Individual age 0.9140 0.8889 1.0984** (0.558) (0.564) (0.502) Individual married/cohabiting 0.7181* 0.8538** 0.0377 (0.397) (0.415) (0.315) Individual attended or completed high school 1.7065*** 1.0670*** 0.2652 (0.229) (0.205) (0.179) Female HH head -0.3367*** -0.3058** -0.1929 (0.122) (0.137) (0.117) HH size 0.0707 -0.9665* -0.0331 (0.557) (0.496) (0.420) HH dependency ratio -0.5232*** -0.2583 -0.2216 (0.174) (0.176) (0.159) Catholic 1.1950*** 1.2061*** 0.8349*** (0.240) (0.239) (0.219) Protestant 0.7088*** 0.7520*** 0.5822*** (0.183) (0.180) (0.173) Kimbanquist 0.0624 0.0778** 0.0479 (0.042) (0.040) (0.031) 105 Muslim 0.0211 0.0160 0.0290 (0.021) (0.022) (0.019) Other Christian 0.8146*** 0.7369*** 0.5115*** (0.205) (0.196) (0.156) Public administration 0.0773 0.0760 -0.0566 (0.135) (0.134) (0.202) Public and para -public enterprises 0.1400* 0.1233 0.1456 (0.082) (0.082) (0.103) Private entreprises 4.1617*** 4.4417*** 2.3195*** (0.479) (0.476) (0.258) Associations (co-ops/NGOs/church) 0.0695 0.0569 -0.0656* (0.044) (0.044) (0.034) Labor union exists 0.1813 0.0878 0.3994** (0.204) (0.200) (0.162) Member of labor union 0.0077 0.0080 -0.1180 (0.103) (0.101) (0.084) HH owns a radio 0.1569 -0.1633 (0.200) (0.177) HH owns a TV 2.0889*** 1.3928*** (0.266) (0.203) HH owns a cellular phone 1.5030*** 0.8416*** (0.281) (0.232) HH owns a bicycle/motorcycle -0.3082*** -0.1965* (0.102) (0.107) Manager or supervisor -0.0228 (0.203) Qualified worker -0.0113 (0.197) Semi-qualified worker 0.3501*** (0.078) Observations 58106 58106 58106 Robust standard errors in parentheses p<0.1 ** p<0.05 *** p<0.01 106 7. Endnotes i World Bank, “The World Bank in the DRC, Overview,� Text/HTML, World Bank, accessed June 24, 2021, https://www.worldbank.org/en/country/drc/overview. ii World Bank Group. 2018. Democratic Republic of Congo Systematic Country Diagnostic : Policy Priorities for Poverty Reduction and Shared Prosperity in a Post-Conflict Country and Fragile State. World Bank, Washington, DC. World Bank. https://openknowledge.worldbank.org/handle/10986/30057 License: CC BY 3.0 IGO. iii UNDP Human Development Reports, http://hdr.undp.org/en/content/gender-inequality-index. iv http://archive.ipu.org/wmn-e/classif.htm v “World Development Indicators | DataBank,� accessed April 13, 2021, https://databank.worldbank.org/reports.aspx?source=world-development-indicators#. vi “Congo Democratic Republic DHS, 2013 -14 - Key Findings (English, French),� ac cessed April 13, 2021, https://dhsprogram.com/publications/publication-sr218-summary-reports-key-findings.cfm. vii Lindsay Stark et al., “The Syndemic of COVID -19 and Gender-Based Violence in Humanitarian Settings: Leveraging Lessons from Ebola in the Dem ocratic Republic of Congo,� BMJ Global Health 5, no. 11 (November 1, 2020): e004194, https://doi.org/10.1136/bmjgh-2020-004194. viii World Bank Enterprise Survey Results : https://data.worldbank.org/data-catalog/enterprise-surveys. ix World Development Indicators, accessed at https://databank.worldbank.org/source/world -development-indicators. x World Development Indicators, accessed at https://databank.worldbank.org/source/world -development-indicators. xi Congo Democratic Republic, 2013-2014 Demographic and Health Survey, https://dhsprogram.com/methodology/survey/survey-display-421.cfm. xii 1-2-3 Survey on Employment, the Informal Sector, and Household Living Conditions (2013 -2014). xiii “Publications | Institut National de la Statistique,� accessed June 25, 2021, http://www.ins.cd/publication/. xiv John D. Ditekemena et al., “Intimate Partners Violence against Women during a COVID -19 Lockdown Period: Results of an Online Survey in 7 Provinces of the Democratic Republic of Congo,� International Journal of Environmental Research and Public Health 18, no. 10 (May 12, 2021): 5108, https://doi.org/10.3390/ijerph18105108. xv International Rescue Committee UNICEF, “COVID -19 - GBV Risks to Adolescent Girls and Interventions to Protect and Empower Them,� n.d. xvi Reports using similar terminology include the World Bank’s 2019 Ethiopia Gender Diagnostic Report and Profiting from Parity Report, and the World Bank and ONE Camp aign’s 2014 Levelling the Field Report, among many examples. xvii Donald, Aletheia; Vaillant, Julia; Campos, Francisco; Cucagna, Maria Emilia. 2018. Caring about Carework : Lifting Constraints to the Productivity of Women Farmers in the Democratic Republic of the Congo. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/1 0986/30118. xviii Cornell University, SRI international Network and Resources Center, http://sri.ciifad.cornell.edu/countries/drcongo/index.html. xix Friedson-Ridenour, Sophia; Pierotti, Rachael S.. 2018. Competing Priorities : Women's Microenterprises and Household Relationships. Policy Research Working Paper;No. 8550. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/30240. xx Mariola Acosta et al (2020) What does it Mean to Make a ‘Joint’ Decision? Unpacking Intra -household Decision Making in Agriculture: Implications for Policy and Practice, The Journal of Development Studies, 56:6, 1210- 1229, DOI: 10.1080/00220388.2019.1650169. xxi “7 Reasons for Land and Property Rights to Be at the Top of the Global Agenda,� accessed April 12, 2021, https://blogs.worldbank.org/voices/7-reasons-land-and-property-rights-be-top-global-agenda. xxii Almodovar-Reteguis, Nayda; Kushnir, Khrystyna; Meilland, Thibault.2013.“Mapping the Legal Gender Gap in Using Property and Building Credit,� Text/HTML, World Bank, accessed March 23, 2021, https://documents.worldbank.org/en/publication/documents- reports/documentdetail/403341468149370071/Mapping-the-legal-gender-gap-in-using-property-and-building-credit. xxiii Almodovar-Reteguis, Nayda; Kushnir, Khrystyna; Meilland, Thibault.2013.“Mapping the Legal Gender Gap in Using Property and Building Credit,� Text/HTML, World Bank, accessed March 23, 2021, https://documents.worldbank.org/en/publication/documents- reports/documentdetail/403341468149370071/Mapping-the-legal-gender-gap-in-using-property-and-building-credit. 107 xxiv MADRE, “Gender-Based Violence and Discrimination against Women and Girls In the Democratic Republic of the Congo,� MADRE, July 19, 2018, https://globalpressjournal.com/africa/democratic -republic-of-congo/awareness- campaigns-encourage-drc-citizens-register-birth-certificates/. xxv Richardson, Amanda and Hughes, Ailey, “The Link Between Land and Gender -Based Violence,� LandLinks (blog), March 10, 2015, https://land-links.org/2015/03/the-link-between-land-and-gender-based-violence/. xxvi Siba, Eyerusalem, “Considering Land Tenure Security for Structural Transformation of African Agriculture,� Brookings (blog), November 30, 1AD, https://www.brookings.edu/blog/africa -in-focus/2015/12/30/considering- land-tenure-security-for-structural-transformation-of-african-agriculture/. xxvii Dillon, Brian and Voena , Alessandra , “Widows’ Land Rights and Agricultural Investment,� Journal of Development Economics 135 (November 1, 2018): 449 – 60, https://doi.org/10.1016/j.jdeveco.2018.08.006. xxviii “7 Reasons for Land and Property Rights to Be at the Top of the Global Agenda,� accessed April 12, 2021, https://blogs.worldbank.org/voices/7-reasons-land-and-property-rights-be-top-global-agenda. xxix Landlinks, “Democratic Republic Of Congo,� Landlinks (Blog), 2010, Https://Land-Links.Org/Country- Profile/Democratic-Republic-Congo/. xxx Un-Habitat, “Improving Women’s Access To Land In Eastern Drc: Challenges And Emerging Opportunities,� 2015, Https://Unhabitat.Org/Sites/Default/Files/Download -Manager- Files/Improving%20womens%20access%20drc.Pdf. xxxi OECD, “DRC Social Institutions And Gender Index,� 2019, Https://Www.Genderindex.Org/Country -Profiles/. xxxii “Code De La Famille,� Accessed March 18, 2021, Http://Www.Leganet.Cd/Legislation/Code%20de%20la%20famille/Table.Htm. xxxiii Huggins, Chris And Mastaki, Christol Paluku “The Political Economy Of Land Law And Policy Reform In The Democratic Republic Of Congo: An Institutional Bricolage Approach,� Canadian Journal Of Development Studies / Revue Canadienne D’études Du Développement 41, No. 2 (April 2, 2020): 260 – 78, Https://Doi.Org/10.1080/02255189.2019.1683519; Landlinks, “Democratic Republic Of Congo�; Dunia P. Zongwe, François Butedi, And Phebe Mavungu Clément, “UPDATE: The Legal System Of The Democratic Republic Of The Congo (DRC): Overview And Research - Globalex,� 2020, Https://Www.Nyulawglobal.Org/Globalex/Democratic_Republic_Congo1.Html. xxxiv MADRE, “Gender-Based Violence and Discrimination against Women and Girls In the Democratic Republic of the Congo,� M ADRE, July 19, 2018, https://globalpressjournal.com/africa/democratic-republic-of-congo/awareness- campaigns-encourage-drc-citizens-register-birth-certificates/. xxxv Huggins, Chris And Mastaki, Christol Paluku “The Political Economy Of Land Law And Policy Reform In The Democratic Republic Of Congo: An Institutional Bricolage Approach,� Canadian Journal Of Development Studies / Revue Canadienne D’études Du Développement 41, No. 2 (April 2, 2020): 260 – 78, Https://Doi.Org/10.1080/02255189.2019.1683519. xxxvi OECD, “DRC Social Institutions And Gender Index,� 2019, Https://Www.Genderindex.Org/Country -Profiles/. xxxvii Un-Habitat, “Analyse Du Régime Foncier Et Social Dans Les Provinces Du Nord Kivu, Du Sud Kivu Et De L’ituri En Répub lique Démocratique Du Congo | Un-Habitat,� 2019, Https://Unhabitat.Org/Analyse -Du- R%C3%A9gime-Foncier-Et-Social-Dans-Les-Provinces-Du-Nord -Kivu-Du-Sud-Kivu-Et-De-L%E2%80%99ituri- En. xxxviii World Bank | Databank, “World Development Indicators | Databank,� 2016, Https://Databank.Worldbank.Org/Reports.Aspx?Source=World -Development-Indicators. xxxix Un-Habitat, “Improving Women’s Access To Land In Eastern Drc: Challenges And Emergin g Opportunities,� 2015, Https://Unhabitat.Org/Sites/Default/Files/Download -Manager- Files/Improving%20womens%20access%20drc.Pdf. xl Un-Habitat, “Improving Women’s Access To Land In Eastern Drc: Challenges And Emerging Opportunities,� 2015, Https://Unhabitat.Org/Sites/Default/Files/Download -Manager- Files/Improving%20womens%20access%20drc.Pdf. xli Huggins, Chris And Mastaki, Christol Paluku “The Political Economy Of Land Law And Policy Reform In The Democratic Republic Of Congo: An Institutional Bricolage Approach,� Canadian Journal Of Development Studies / Revue Canadienne D’études Du Développement 41, No. 2 (April 2, 2020): 260 – 78, Https://Doi.Org/10.1080/02255189.2019.1683519. 108 xlii Un-Habitat, “Improving Women’s Access To Land In Eastern Drc: Challenges And Emerging Opportunities,� 2015, Https://Unhabitat.Org/Sites/Default/Files/Download -Manager- Files/Improving%20womens%20access%20drc.Pdf. xliii “Code De La Famille De La Republique Democratique Du Congo.Pdf,� Accessed March 18, 2021, Https://Www.Leganet.Cd/Legislation/Code%20de%20la%20famille/Cdf.2017.Pdf. xliv Un-Habitat, “Improving Women’s Access To Land In Eastern Drc: Challen ges And Emerging Opportunities,� 2015, Https://Unhabitat.Org/Sites/Default/Files/Download -Manager- Files/Improving%20womens%20access%20drc.Pdf. xlv Un-Habitat, “Improving Women’s Access To Land In Eastern Drc: Challenges And Emerging Opportunities,� 2015, Https://Unhabitat.Org/Sites/Default/Files/Download -Manager- Files/Improving%20womens%20access%20drc.Pdf. xlvi Un-Habitat, “Improving Women’s Access To Land In Eastern Drc: Challenges And Emerging Opportunities�; Esther Nsapu, “Challenging Family And Custom, Women Assert Legal Right To Inherit Land,� Glob al Press Journal, July 2, 2017, Https://Globalpressjournal.Com/Africa/Democratic-Republic-Of-Congo/Challenging-Family- Custom-Women-Assert-Legal-Right-Inherit-Land/; Oecd, “Drc Social Institutions And Gender Index,� 2019, Https://Www.Genderindex.Org/Country-Profiles/. xlvii Merveille Luneghe, “Awareness Campaigns Encourage Drc Citizens To Register For Birth Certificates,� Global Press Journal, July 19, 2018, Https://Globalpressjournal.Com/Africa/Democratic-Republic-Of-Congo/Awareness- Campaigns-Encourage-Drc-Citizens-Register-Birth-Certificates/. xlviii OOECD, “DRC Social Institutions And Gender Index,� 2019, Https://Www.Genderindex.Org/Country - Profiles/.� xlix Women For Women International, “‘Men Can Take Many Wives; We Can’t Register Our Children.’ | Women For Women International,� 2019, Https://Www.Womenforwomen.Org/Blogs/Men -Can-Take-Many-Wives-We- Cant-Register-Our-Children. l Merveille Luneghe, “Awareness Campaigns Encourage Drc Citizens To Register For Birth Certificates,� Global Press Journal, July 19, 2018, Https://Globalpressjournal.Com/Africa/Democratic-Republic-Of-Congo/Awareness- Campaigns-Encourage-Drc-Citizens-Register-Birth-Certificates/. li Women For Women International, “‘Men Can Take Many Wives; We Can’t Register Our Children.’ | Women For Women International,� 2 019, Https://Www.Womenforwomen.Org/Blogs/Men-Can-Take-Many-Wives-We-Cant- Register-Our-Children. lii Esther Nsapu, “Challenging Family And Custom, Women Assert Legal Right To Inherit Land,� Global Press Journal, July 2, 2017, Https://Globalpressjournal.Com/Africa/Democratic-Republic-Of-Congo/Challenging-Family- Custom-Women-Assert-Legal-Right-Inherit-Land/. liii Donald, Aletheia; Goldstein, Markus; Hartman, Alexandra; La Ferrara, Eliana; O'Sullivan, Michael; Stickler, Mercedes. 2020. What’s Mine is Yours : Pilot Evidence from a Randomized Impact Evaluatio n on Property Rights and Women’s Empowerment in Côte d’Ivoire . World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/34081 . liv Cherchi, Ludovica; Goldstein, Markus; Habyarimana, James; Montalvao, Joao; O'Sullivan, Michae l; Udry, Chris; Gruver, Ariel. 2019. Empowering Women Through Equal Land Rights : Experimental Evidence From Rural Uganda. Gender Innovation Lab Policy Brief;No. 33. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/31513. lv Donald, Aletheia; Koolwal, Gayatri; Annan, Jeannie; Falb, Kathryn; Goldstein, Markus. 2017. Measuring Women's Agency. Policy Research Working Paper;No. 8148. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/27955 License: CC BY 3.0 IGO. lvi Klugman, Jeni; Hanmer, Lucia; Twigg, Sarah; Hasan, Tazeen; McCleary -Sills, Jennifer; Santamaria, Julieth. 2014. Voice and Agency : Empowering Women and Girls for Shared Prosperity. Washington, DC: World Bank Group. © World Bank. https://openknowledge.worldbank.org/handle/10986/19036. lvii Conciliation Resources, “Unable to Stay in Place,� 2017, https://www.c -r.org/resource/unable-stay-place. lviii OECD, “DRC Social Institutions and Gender Index,� 2019, https://www.genderindex.org/country -profiles/. lix World Bank | DataBank, “World Development Indicators | DataBank,� 2016, https://databank.worldbank.org/reports.aspx?source=world-development-indicators. lx ILO, “Comments - Direct Request (CEACR) - Adopted 2018, Published 108th ILC Session (2019),� 2019, https://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:13100:0::NO::P13100_COMMENT_ID:3934978. 109 lxi OECD, “DRC Social Institutions and Gender Index,� 2019, https://www.genderindex.org/country -profiles/. lxii ILO, “Democratic Republic of the Congo - Loi N° 16/008 Du 15 Juillet 2016 Modifiant et Complétant Le Loi N°87-010 Du 1er Août 1987 Portant Code de La Famille.,� 2016, https://www.ilo.org/dyn/natlex/natlex4.detail?p_lang=&p_isn=102954; World Bank, “Women Business and the Law - Reforms.� lxiii WILPF, “Submission to the Committee on the Elimination of Discrimin ation Against Women - Democratic Republic of the Congo.� 2019, https://www.wilpf.org/portfolio-items/submission-to-the-committee-on-the- elimination-of-discrimination-against-women-democratic-republic-of-the-congo/. lxiv Winthrop, Rebecca ; Greubel, Lauren; Ackerman, Xanthe, “The New Push for Education R eform in the Democratic Republic of the Congo,� Brookings (blog), 2013, https://www.brookings.edu/blog/up- front/2013/03/01/the-new-push-for-education-reform-in-the-democratic-republic-of-the-congo/. lxv ILO, “Comments - Direct Request (CEACR) - Adopted 2018, Published 108th ILC Session (2019),� 2019, https://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:13100:0::NO::P13100_COMMENT_ID:3934978. lxvi “Barriers to Education for Girls in the Democratic Republic of Congo,� Resource Centre, accessed March 22, 2021, https://resourcecentre.sa vethechildren.net/node/17718/pdf/750_barriers_to_education_for_girls_in_drc.pdf. lxvii OECD, “DRC Social Institutions and Gender Index,� 2019, https://www.genderindex.org/country -profiles/. lxviii OECD, “DRC Social Institutions and Gender Index,� 2019, https://www.genderindex.org/country -profiles/.; US State Department, “Democratic Republic of the Congo.� lxix Male, Chata; Wodon, Quentin. 2016. Basic Profile of Child Marriage in the Republic of Congo. Health, Nutrition and Population Knowledge Brief;. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/25471. lxx OECD, “DRC Social Institutions and Gender Index,� 2019, https://www.genderindex.org/country -profiles/. lxxi Rachael Pierotti, Milli Lake, and Chloé Lewis. “Equality on His Terms: Doing and Undoing Gender through Men’s Discussion Groups,� Gender & Society 32, no 4 (August 2018): 540 -562. https://doi.org/10.1177%2F0891243218779779. lxxii OECD, “DRC Social Institutions and Gender Index�; Raymond Achu Samndong and Darley Kjosavik, “Gendered Forests: Exploring Gender Dimensions in Forest Governance and REDD+ in Équateur Province, Democratic Republic of Congo (DRC),� Ecology and Society 22, no. 4 (December 5, 2017), https://doi.org/10.5751/ES-09753-220434. lxxiii World Bank, “Women Business a nd the Law - Reforms,� Text/HTML, World Bank, 2020, https://wbl.worldbank.org/en/reforms. lxxiv World Bank, “Women Business and the Law - Reforms,� Text/HTML, World Bank, 2020, https://wbl.worldbank.org/en/reforms. lxxv OECD, “DRC Social Institutions and Gender Index,� 2019, https://www.genderindex.org/country -profiles/. lxxvi Julia Vaillant et al., “Engaging Men to Transform Inequitable Gender Attitudes and Prevent Intimate Partner Violence: A Cluster Randomised Controlled Trial in North and South Kivu, Democratic Republic of Congo,� BMJ Global Health 5, no. 5 (May 27, 2020), https://doi.org/10.1136/bmjgh -2019-002223. lxxvii OECD, “DRC Social Institutions and Gender Index,� 2019, https://www.genderindex.org/country -profiles/. lxxviii Peterman, A., Palermo, T., & Bredenkamp, C. (2011). Estimates and determinants of sexual violence against women in the Democratic Republic of Congo. American journal of public health , 101 (6), 1060 – 1067. https://doi.org/10.2105/AJPH.2010.300070 lxxix OECD, “DRC Social Institutions and Gender Index,� 2019, https://ww w.genderindex.org/country-profiles/. lxxx United Nations Office of the Special Representative of the Secretary -General on Sexual Violence in Conflict, “Democratic Republic of the Congo,� United Nations Office of the Special Representative of the Secretary-General on Sexual Violence in Conflict (blog), 2020, https://www.un.org/sexualviolenceinconflict/countries/democratic - republic-of-the-congo/. lxxxi World Bank Group, “Global Findex Database�, 2017, https://globalfindex.worldbank.org/ . lxxxii World Bank Group, “Global Findex Database�, 2017, https://globalfindex.worldbank.org/ . lxxxiii “Top Policy Lessons in Agriculture,� accessed June 25, 2021, https://openknowledge.worldbank.org/handle/10986/33493. lxxxiv Ambler, Kate; Jones, Kelly M.; and O'Sullivan, Michael. 2018. What is the role of men in connecting women to cash crop markets? Evidence from Uganda. IFPRI Discussion Paper 1762. Washington, DC: International Food Policy 110 Research Institute (IFPRI), https://www.ifpri.org/publication/what-role-men-connecting-women-cash-crop-markets- evidence-uganda. lxxxv Paulin Njingulula Mumbeya et al., “Can Innovation Platforms (IPs) Improve Rural Women Participation in Maize Value Chain? Evidence from the Eastern DR Congo,� European Journal of Agriculture and Food Sciences 2, no. 3 (Ma y 21, 2020), https://doi.org/10.24018/ejfood.2020.2.3.16. lxxxvi Andrew Agyei-Holmes et al., “The Effects of Land Title Registration on Tenure Security, Investment and the Allocation of Productive Resources : Evidence from Ghana,� n.d., https://openknowledge.worldbank.org/handle/10986/34423. lxxxvii Ali, Daniel Ayalew; Deininger, Klaus; Goldstein, Markus. 2011. Environmental And Gender Impacts Of Land Tenure Regularization In Africa : Pilot Evidence From Rwanda. Africa Region Gender Practice Policy Brief;No. 2. World Bank, Washington, Dc. © World Bank. https://openknowledge.worldbank.org/handle/10986/25527 . lxxxviii Markus Goldstein Et Al., Formalizing Rural Land Rights In West Africa: Early Evidence From A Randomized Impact Evaluation In Benin , Policy Research Working Papers (The World Bank, 2015), Https://Doi.Org/10.1596/1813-9450-7435. lxxxix Donald, Aletheia et al.,. 2020. What’s Mine is Yours : Pilot Evidence from a Randomized Impact Evaluation on Property Rights and Women’s Empowerment in Côte d’Ivoire . World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/34081. xc Cherchi, Ludovica et al.,. 2019. Empowering Women Through Equal Land Rights : Experimental Evidence From Rural Uganda. Gender Innovation Lab Policy Brief;No. 33. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/31513. xci Ileana I. Diaz And Dina Najjar, “Gender And Agricultural Extension: Why A Gender Focus Matters,� Journal Of Gender, Agriculture And Food Security (Agri-Gender), 2019, Https://Doi.Org/10.22004/Ag.Econ.301204. xcii Florence Kondylis Et Al., “Do Female Instructors Reduce Gender Bias In Diffusion Of Sustainable Land Managemen t Techniques? Experimental Evidence From Mozambique,� World Development 78 (February 2016): 436 – 49, Https://Doi.Org/10.1016/J.Worlddev.2015.10.036. xciii Ariel Benyishay Et Al., Are Gender Differences In Performance Innate Or Socially Mediated? (World Bank, Washington, Dc, 2016), Https://Doi.Org/10.1596/1813-9450-7689. xciv Isabel Lambrecht, Bernard Vanlauwe, And Miet Maertens, “Agricultural Extension In Eastern Democratic Republic Of Congo: Does Gender Matter?,� European Review Of Agricultural Economics 43, No. 5 (December 2016): 841 – 74, Https://Doi.Org/10.1093/Erae/Jbv039. xcv Jenny C. Aker And Christopher Ksoll, “Can Mobile Pho nes Improve Agricultural Outcomes? Evidence From A Randomized Experiment In Niger,� Food Policy 60 (April 2016): 44 – 51, Https://Doi.Org/10.1016/J.Foodpol.2015.03.006. xcvi Els Lecoutere, David J. Spielman, and Bjorn Van Campenhout, “Empowering Women with Digital Extension in Uganda: Effects of Information and Role Models,� AgEcon Search, 2019, h ttps://doi.org/10.22004/ag.econ.295694. xcvii Michael O’Sullivan et al., “Levelling the Field : Improving Opportunities for Women Farmers in Africa,� Text/HTML, World Bank, 2014, https://documents.worldbank.org/en/publication/documents - reports/documentdetail/579161468007198488/Levelling-the-field-improving-opportunities-for-women-farmers-in- Africa. xcviii Daidone, Silvio & Davis, Benjamin & Dewbre, Joshua & Gonzalez-Flores, Mario & Handa, Sudhanshu & Seidenfeld, David & Tembo, Gelson. (2015). Zambia’s Child Grant Programme: 24 -month impact report on productive activities and labour allocation. http://www.fao.org/reduce-rural-poverty/resources/resources- detail/en/c/468332/. 10.13140/RG.2.1.5007.8566. xcix “Impacts o f Agricultural Mechanization: Evidence From Four African Countries by Thomas Daum, Ygué P. Adegbola, Geoffrey Kamau, Alpha O. Kergna, Christogonus Daudu, Roch C. Zossou, Géraud F. Crinot, Paul Houssou, Lawrence Moses, Yarama Ndirpaya, A Wahab, Oliver Kirui, Fatunbi A. Oluwole :: SSRN,� accessed March 10, 2021, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3672085. c Gundula Fischer et al., “Gender and Mechanization: Exploring the Sustainability of Mechanized Forage Chopping in Tanzania,� Journal of Rural Studies 64 (November 1, 2018): 112 – 22, https://doi.org/10.1016/j.jrurstud.2018.09.012. ci Oriana Bandiera et al., “Empowering Adolescent Girls in a Crisis Context : Lessons from Sierra Leone in the Time o f Ebola,� n.d., https://openknowledge.worldbank.org/handle/10986/32115. 111 cii Jennifer Randall and Alejandra Garcia, “Let’s Go Girls!: Evaluating the Effectiveness of Tutoring and Scholarships on Primary School Girls’ Attendance and Academic Perform ance in the Democratic Republic of the Congo (DRC),� FIRE: Forum for International Research in Education 6, no. 3 (October 19, 2020): 19 – 35, https://doi.org/10.32865/fire202063222. ciii Aker, Jenny C., Christopher Ksoll, and Travis J. Lybbert. 2012. "Can Mobile Phones Improve Learning? Evidence from a Field Experiment in Niger." American Economic Journal: Applied Economics , 4 (4): 94-120. https://www.aeaweb.org/articles?id=10.1257/app.4.4.94 civ “Adolescent Girls’ Empowerment in Conflict -Affected Settings: Experimental Evidence from South Sudan,� BRAC Institute of Governance and Development (blog), accessed March 22, 2021, https://bigd.bracu.ac.bd/publications/adolescent-girls-empowerment-in-conflict-affected-settings-experimental- evidence-from-south-sudan/. cv Franck Adoho et al., The Impact of an Adolescent Girls Employment Program: The EPAG Project in Liberia , Policy Research Working Papers (The World Bank, 2014), https://doi.org/10.1596/1813 -9450-6832. cvi Oriana Bandiera et al., “Women’s Empowerment in Action: Evidence from a Randomized Control Trial in Africa,� American Economic Journal: Applied Economics 12, no. 1 (January 1, 2020): 210 – 59, https://doi.org/10.1257/app.20170416. cvii World Bank, “Reintegration Opportunities for Vulnerable Households of Ex -Combatants Demobilized through PNDDR : An Overview of the Socioeconomic Situation of Pilot Project Beneficiaries Two Years Following Project Closure - Maniema, Democratic Republic of Congo,� 2013, https://openknowledge.worldbank.org/handle/10986/17859?locale-attribute=en. cviii “New Mindset, Increased Profits: Lessons from a n Innovative Entrepreneurial Training in Togo,� World Bank, accessed March 18, 2021, https://www.worldbank.org/en/news/feature/2018/01/18/new-mindset-increased-profits- lessons-from-an-innovative-entrepreneurial-training-in-togo. cix Alibhai, Salman; Buehren, Niklas; Frese, Michael; Goldstein, Markus; Papineni, Sreelakshmi; Wolf, Kathrin. 2019. Full Esteem Ahead? Mindset-Oriented Business Training in Ethiopia. Policy Research Working Paper;No. 8892. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/31905 License: CC BY 3.0 IGO. cx World Bank. 2020. Supporting Women Throughout the Coronavirus (COVID-19) Emergency Response and Economic Recovery. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/33612 License: CC BY 3.0 IGO.� cxi Richard Chamboko et al., The Role of Gender in Agent Banking: Evidence from the Democratic Republic of Congo , Policy Research Working Papers (The World Bank, 2020), https://doi.org/10.1596/1813 -9450-9449. cxii Victoria Reichel, “Financial Inclusion for Women and Men in Artisanal Gold Mining Communities: A Case Study from the Democratic Republic of the Congo,� The Extractive Industries and Society 7, no. 2 (April 1, 2020): 412 – 19, https://doi.org/10.1016/j.exis.2019.05.003. cxiii Judith Bass et al., Study of Effectiveness of a Social-Economic Intervention for Sexual Violence Survivors in Eastern DRC (World Bank, 2014), https://doi.org/10.1596/26551. cxiv Pascaline Dupas and Jonathan Robinson, “Savings Constraints and Microenterprise Development: Evidence from a Field Experiment in Kenya,� American Economic Journal: Applied Economics 5, no. 1 (January 1, 2013): 163 – 92, https://doi.org/10.1257/app.5.1.163. cxv Gautam Bastian et al., “Short -Term Impacts of Improved Access to Mobile Savings, with and without Business Training: Experimental Evidence from Tanzania,� 2018, Center for Global Development, https://www.cgdev.org/publication/short-term-impacts-improved-access-mobile-savings-business-training. cxvi Carranza, Eliana; Donald, Aletheia; Grosset, Florian; Kaur, Supreet. 2018. Working Under Pressure : Improving Labor Productivity through Financial Innovation. Gender Innovation Lab Policy Brief;No. 31. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/31029 License: CC BY 3.0 IGO. cxvii FAO, “Innovative Productive Transfers (CASH+) in Mali,� 2016, http://www.fao.org/emergencies/resources/documents/resources- detail/en/c/463032/#:~:text=The%2018%2Dmonth%20programme%20combines,villages%20of%20the%20Ka yes% 20region. 112 cxviii Nancy Glass et al., “Randomised Controlled Trial of a Livestock Productiv e Asset Transfer Programme to Improve Economic and Health Outcomes and Reduce Intimate Partner Violence in a Postconflict Setting,� BMJ Global Health 2, no. 1 (February 2017): e000165, https://doi.org/10.1136/bmjgh -2016-000165. cxix FAO, “Cash+, FAO’s Approach,� 2 017, http://www.fao.org/3/i7864e/i7864e.pdf. cxx Gastian, Gautam; Goldstein, Markus; Papineni, Sreelakshmi. 2017. Are Cash Transfers Better Chunky or Smooth? : Evidence from an Impact Evaluation of a Cash Transfer Program in Northern Nigeria. Gender Innovation Lab Policy Brief;No. 21. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/28434. cxxi “Impact Evaluation in the Democratic Republic of Congo | Women for Women International,� accessed April 13, 2021, https://www.womenforwomen.org/impact-evaluation-democratic-republic-congo. cxxii “Building Stable Livelihoods for the Ultra Poor,� Innovations for Poverty Action, September 8, 2015, https://www.poverty-action.org/publication/building-stable-livelihoods-ultra-poor. cxxiii Kevin Croke et al., Up before Dawn: Experimental Evidence from a Cross-Border Trader Training at the Democratic Republic of Congo – Rwanda Border (World Bank, Washington, DC, 2020), https://doi.org/10.1596/1813- 9450-9123. cxxiv “Intimate Partner Violence : Economic Costs and Implications for Growth and Development,� Text/HTML, World Bank, accessed June 24, 2021, https://documents.worldbank.org/en/publication/documents- reports/documentdetail/412091468337843649/Intimate-partner-violence-economic-costs-and-implications-for- growth-and-development. cxxv Kate Doyle et al., “Gender-Transformative Bandebereho Couples’ Intervention to Promote Male Engagement in Reproductive and Maternal Health and Violence Prevention in Rwanda: Findings from a Randomized Controlled Trial,� PloS One 13 (April 4, 2018): e0192756, https://doi.org/10.1371/journal.pone.0192756. cxxvi Gupta, Jhumka; Falb, Kathryn; Kpebo, Denise; Seban, Juliette; Annan, Jeannie.2013. Village savings and loans associations, gender dialogue groups, and gender-based violence against women in Côte d’Ivoire : findings from a randomized community trial (English). enGender Impact : the World Bank's Gender Impact Evaluation Database Washington, D.C. : World Bank Group. http://documents.worldbank.org/curated/en/494681468235490374/Village-savings-and-loans-associations-gender- dialogue-groups-and-gender-based-violence-against-women-in-Cote-dIvoire-findings-from-a-randomized- community-trial cxxvii Malé and Wodon, “Basic Profile of Child Marriage in the Democratic Republic of the Congo.� cxxviii Shubha Chakravarty, Smita Das, and Julia Vaillant, Gender and Youth Employment in Sub-Saharan Africa: A Review of Constraints and Effective Interventions (World Bank, Washington, DC, 2017), https://doi.org/10.1596/1813-9450-8245. cxxix “Intimate Partner Violence : Economic Costs and Implications for Growth and Development,� Text/HTML, World Bank, accessed June 24, 2021, https://documents.worldbank.org/en/publication/documents- reports/documentdetail/412091468337843649/Intimate-partner-violence-economic-costs-and-implications-for- growth-and-development. cxxx “Cash or Condition? Evidence from a Cash Transfer Experiment * | The Quarterly Journal of Economics | Oxford Academic,� accessed June 24, 2021, https://academic.oup.com/qje/article- abstract/126/4/1709/1922509?redirectedFrom=fulltext. cxxxi Oriana Bandiera et al., “Women’s Empowerment in Action: Evidence from a Randomized Control Trial in Africa,� American Economic Journal: Applied Economics 12, no. 1 (Januar y 1, 2020): 210 – 59, https://doi.org/10.1257/app.20170416. cxxxii “Assessing Sexual and Gender Based Violence during the Ebola Crisis in Sierra Leone | UNDP in Sierra Leone,� accessed February 29, 2020, https://www.sl.undp.org/content/sierraleone/en/home/library/crisis_prevention_and_rec overy/assessing-sexual-and- gender-based-violence-during-the-ebola -cris.html. cxxxiii Bandiera, Oriana; Buehren, Niklas; Goldstein, Markus; Rasul, Imran; Smurra, Andrea. 2019. Empowering Adolescent Girls in a Crisis Context : Lessons from Sierra Leone in the Time of Ebola. Gender Innovation Lab Policy Brief;No. 34. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/32115 License: CC BY 3.0 IGO. 113 cxxxiv Aletheia Donald et al., “Caring about Carework : Lifting Constraints to the Productivity of Women Farmers in the Democratic Republic of the Congo,� World Bank, 2018, https://openknowledge.worldbank.org/handle/10986/30118. cxxxv Martinez, Sebastian; Naudeu, Sophie; Pereira, Vitor.2013. “The promise of preschool in Africa : a randomized impact evaluation of early childhood development in rural Mozambique � (English). enGender Impact : the World Bank's Gender Impact Evaluation Database Washington, D.C. : World Bank Group. http://documents.worldbank.org/curated/en/799391468059930056/The-promise-of-preschool-in-Africa -a- randomized-impact-evaluation-of-early-childhood-development-in-rural-Mozambique. cxxxvi Clark, S., Kabiru, C.W., Laszlo, S. et al. The Impact of Childcare on Poor Urban Women’s Economic Empowerment in Africa. Demography 56, 1247 – 1272 (2019). https://doi.org/10.1007/s13524-019-00793-3. cxxxvii Vaillant J, Koussoubé E, Roth D, et al ., “Engaging men to transform inequitable gender attitudes and prevent intimate partner violence: a cluster randomised controlled trial in North and South Kivu, Democratic Republic of Congo,� BMJ Global Health 2020; 5:e002223. https://gh.bmj.com/content/5/5/e002223. cxxxviii Kate Doyle et al., “Gender-Transf ormative Bandebereho Couples’ Intervention to Promote Male Engagement in Reproductive and Maternal Health and Violence Prevention in Rwanda: Findings from a Randomized Controlled Trial,� PloS One 13 (April 4, 2018): e0192756, https://doi.org/10.1371/journal.pone.0192756. 114