Closing Gaps, Increasing Opportunities: A Diagnostic on Women’s Economic Empowerment in Nigeria 1 © 2022 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; email: pubrights@worldbank.org. Cover design: Blossom The Nigeria Gender Innovation Lab is a country-level initiative of the regional Africa Gender Innovation Lab, which conducts rigorous impact evaluations across Sub-Saharan Africa to build the evidence base on innovative interventions that promote women’s economic empowerment. The goal of the Nigeria GIL is to equip policy makers and project teams in Nigeria with new evidence on what does and does not work in addressing the underlying causes of gender inequality, and how best to close gender gaps in earnings, productivity, and assets. 2 Closing Gaps, Increasing Opportunities: A Diagnostic on Women’s Economic Empowerment in Nigeria 3 Table of Contents Acknowledgments 6 Preface 8 Executive Summary 10 Introduction 18 Part I: Women in the Workforce 20 Issues around Women’s Labor Force Participation in Nigeria 25 Part II: Measuring the Gender Gaps 28 Agriculture 30 Self-Employment 36 Wages 40 Part III: Deep Dives 44 Deep Dive I: Land Tenure and Titling 46 Deep Dive II: Livestock 51 Deep Dive III: Occupational Segregation 55 Part IV: What Are the Costs of Economic Gender Gaps in Nigeria? 60 Part V: Policy Priorities for Closing Gender Gaps in Economic Outcomes in Nigeria 64 Appendices 76 Appendix 1: Correlates of Labor Supply 77 Appendix 2A: Kitagawa-Oaxaca-Blinder Decomposition Methods 79 Appendix 2B: Agriculture 81 Appendix 2C: Self-Employment 94 Appendix 2D: Wages 107 Appendix 3: Technical Appendix 118 Endnotes 121 5 Acknowledgments This report was prepared by a team of authors from (FMWA); the Federal Ministry of Industry, Trade, and the Africa Gender Innovation Lab (GIL), led by Julia Investment (FMITI); the Federal Ministry of Finance, Vaillant, and included Andrew Brudevold-Newman, Budget, and National Planning (FMFBNP); the Central Amy Copley, Ayodele Fashogbon, Paula Gonzalez, Bank of Nigeria (CBN); Nigeria Incentive-Based Risk Laurel Morrison, Patricia Paskov, and Abhilasha Sahay. Sharing System for Agricultural Lending (NIRSAL), the Zuzana Johansen provided editorial assistance, and National Bureau of Statistics, the Small and Medium Blossom (Blossoming.it) led the design and layout of Enterprises Development Agency of Nigeria (SMED- the report. AN), the state governments of Ekiti, Edo, Kaduna, and Kebbi, state-level nongovernmental organizations, The team gratefully acknowledges the guidance and and civil society organizations. The team is grateful leadership of Markus Goldstein, Head of the Africa for the guidance received from Dr. Adeyemi Dipeolu, Gender Innovation Lab. The team would like to thank Special Adviser to the President for Economic Matters Tara Vishwanath and Olumide Okunola, who peer-re- in the Office of the Vice-President; Karima Babangida, viewed this report. The team thanks Kehinde Ajayi, then Director of Extension Services at FMARD; Aisha Sybil Chidiac, Audu Grema, Rachael Pierotti, Alan Ren- Ahmad, Deputy Governor, Financial Systems Stabili- nison, and Vicki Wilde for their valuable comments ty Directorate at CBN; and Bisi Adeleye-Fayemi, First and guidance. Support and guidance from Shubham Lady of Ekiti State. The authors also thank the follow- Chaudhuri, World Bank Country Director for Nigeria, ing development partners, nongovernmental organ- Rachid Benmessaoud, his predecessor as World Bank izations, and civil society organizations met during Country Director for Nigeria, and the World Bank Ni- consultation meetings in Abuja: United Nations–Ni- geria Country Management Unit are gratefully ac- geria, European Union (EU) delegations to Nigeria and knowledged. Economic Community of West African States (ECOW- AS), International Fund for Agricultural Development This report has benefited from the feedback of over (IFAD), Agence Française de Développement (AFD), two hundred experts and policy makers during both The Deutsche Gesellschaft für Internationale Zusam- individual and group consultations. The team is grate- menarbeit (GIZ), Mercy Corps, Catholic Relief Services ful to the participants of the consultation meetings (CRS), The Foreign, Commonwealth and Development that were held in April 2019, July 2019, and September Office (FCDO), Global Affairs Canada, ActionAid Nige- 2019 with the office of the Vice President of the Fed- ria, Oxfam Nigeria, and Nigeria Governors’ Forum. eral Ministry of Agriculture and Rural Development (FMARD); the Federal Ministry of Labor and Employ- ment (FMLE); the Federal Ministry of Women Affairs 6 The team would also like to acknowledge the gener- ous support of the Umbrella Facility for Gender Equal- ity (UFGE). The UFGE is a multidonor trust fund admin- istered by the World Bank to advance gender equality and women’s empowerment through experimenta- tion and knowledge creation to help governments and the private sector focus policy and programs on scalable solutions with sustainable outcomes. The UFGE is supported with generous contributions from Australia, Canada, Denmark, Finland, Germany, Ice- land, Latvia, the Netherlands, Norway, Spain, Sweden, Switzerland, the United Kingdom, the United States, and the Bill and Melinda Gates Foundation. 7 Preface Nigeria is Africa’s most populous country and largest resources, which prevent women from achieving equal economy. Its abundance of natural resources, large outcomes to those of men. youth population, and vibrant private sector suggest that it has the potential to be a global economic pow- The Government of Nigeria is working strategically to erhouse. Yet, economic growth has lagged population remove the obstacles that prevent women from equal- growth in recent years, hampering efforts to reduce ly participating in and benefiting from the economy. We poverty and boost shared prosperity. have enacted policies to support women's economic participation, facilitate access to education for women To realize the country’s vast potential, and to reach the and girls, especially in states affected by insurgency, ambitious target of lifting 100 million Nigerians out of and refocus sector budgets to address gender gaps in poverty by 2030, we must make significant progress the provision of social safety nets, among others. Spe- against an array of development challenges. Advancing cifically, through the National Gender Policy, we work gender equality and promoting women and girls’ em- across agencies and sectors, analyzing the impacts of powerment are among the foremost issues to address gender inequality in Nigeria to develop policy reforms, since they have broad implications for accelerating in- promote gender education and capacity building, and clusive economic growth. Women’s economic empow- enact economic reforms that secure equal rights and erment cannot only support inclusive development opportunities for women and men. Through the Eco- through women’s employment, but also by decreasing nomic Recovery and Growth Plan (ERGP), we are also fertility rates, thereby supporting the critical demo- striving to implement gender-responsive programming. graphic dividend agenda. For example, to accelerate inclusive growth in non-oil sectors, the ERGP is investing in increasing women’s Notably, more than half of Nigerian women work in economic participation through education, job crea- agriculture, entrepreneurship, or wage employment; tion, social inclusion, and improving the ease of doing however, their work is more likely to be less secure and business. less profitable than that of men. Women's lower earn- ings are not because women are less capable farm- Like the government, the World Bank is committed to ers, entrepreneurs, or employees. Rather, it is due to enhancing women’s economic empowerment in Nige- numerous gender-specific barriers to their economic ria with the conviction that human capital development participation and productivity, such as lower access to and the economic empowerment of adolescent girls 8 are central to Nigeria harnessing its demographic divi- closing gender gaps is critical to fueling growth, pover- dend. The World Bank is supporting the government of ty reduction, and structural transformation in Nigeria. Nigeria in its efforts to achieve gender equality through Current gender gaps in key economic sectors prevent a coordinated, multisectoral, multistakeholder, and Nigeria from realizing gains up to an additional 5.8 per- evidence-based approach to improve livelihoods and cent of GDP. employment opportunities for women. This report supports the drive for action by highlighting To help achieve this aim, the Bank is supporting a wide effective and evidence-based solutions to target the range of operations across sectors that target the spe- most pressing challenges facing women. With the right cific obstacles faced by women farmers, entrepreneurs, policies, partners, and investments, we can help unlock and employees. the potential of Nigerian women and girls and trans- form our economy and society. To better identify those constraints, the World Bank’s latest report, "Closing Gaps, Increasing Opportunities: A Diagnostic on Women's Economic Empowerment in Ni- geria," identifies and measures gender gaps in econom- ic sectors, analyzes the drivers of the gaps, estimates the cost of gender gaps to the gross domestic product (GDP) and offers policy guidance to target the drivers of inequality effectively. As highlighted in this report, Dame Pauline K. Tallen, OFR, KSG Shubham Chaudhuri Prince Clem Agba Honourable Minister, Country Director, Nigeria Honourable Minister of State, Federal Ministry of Women Affairs Western and Central Africa Region Federal Ministry of Finance, The World Bank Budget, and National Planning 9 Executive Summary As Nigeria faces the immediate challenge of stimulat- Gender Gaps in Labor ing economic recovery in the wake of the COVID-19 (coronavirus) pandemic and corresponding economic Participation and Labor Supply shocks, it also has the opportunity to address the siz- Significant gender gaps exist in labor force participa- able gender gaps that undermine women’s economic tion and labor supply in Nigeria, varying significantly empowerment and hinder inclusive economic growth. between the northern and southern regions of the Gender disparities in earnings not only hold back the country. At the national level, fewer women work than Nigerian economy, they also represent an opportuni- men: 55 percent of working-age women work, which ty: closing the gender gaps in key economic sectors is 14 percentage points fewer than the 69 percent of could yield additional gains of US$9.3 billion or up to men. Women also work roughly 5.5 fewer hours per US$22.9 billion. Women’s economic empowerment week than men. Moreover, women face occupation- will also be key to accelerating a demographic transi- al segregation, with women and men being equally tion and reaping the gains of a demographic dividend. likely to be self-employed, yet women are 12 percent- age points less likely to participate in agriculture and Drawing on data from the most recent Nigeria General 7 percentage points less likely to have a wage-paying Household Survey (2018–2019), this report makes five job. critical contributions: (1) highlighting the gender gaps in labor force participation; (2) documenting the mag- nitude and drivers of the gender gaps in key econom- ic sectors; (3) diving deep into three contextual con- Gender Gaps across Key Economic straints: land, livestock, and occupational segregation; (4) measuring the costs of the gender gaps; and (5) of- Sectors and Their Drivers fering policy and programming recommendations of In agriculture, the value of the output per hectare on innovative options to close the gender gaps. female-managed plots is 30 percent lower than com- parably sized male-managed plots (figure 1). Mean- while, almost half of income-earning women spend most of their time in entrepreneurship-related activi- ties, yet their profits are 66 percent less than those of men. 10 Gender disparities in earnings not only hold back the Nigerian economy, they also represent an opportunity: closing the gender gaps in key economic sectors could yield additional gains of US$9.3 billion or up to US$22.9 billion. Figure 1. Only 12 percent of Nigerian women primarily earn wages, and those who do earn 22 percent less than Gender gaps across key sectors their male counterparts. 80 Further analysis using the Kitagawa-Oaxaca-Blinder 71% decomposition approach identifies significant driv- 70 ers of the gender earnings gaps in these economic 66% sectors, highlighting priority areas where policy and 60 program innovation may be particularly effective at closing these gaps. 50 Percent 40 30% 30 28% 22% 20 16% 10 0 Agriculture Self-employment Wage work Unconditional Conditional Source: Nigeria General Household Survey data. 11 Agriculture Women use fewer inputs. Women use less productive labor. Male farmers use over 8 times more fertilizer and 50 Female plot managers work more than their male percent more herbicide per hectare than their female peers and use more household and hired labor. How- counterparts. Lower use of these inputs constrains the ever, male labor used by female plot managers is sig- production of female plot managers because dou- nificantly less productive than male labor used by bling the quantities of fertilizer and herbicide used on male farmers. a plot in Nigeria increases agricultural productivity by 6 percent and 18 percent, respectively, on average. Self-Employment Women farm less valuable crops. Women operate less-capitalized firms. There are consistent value differences in both the Self-employed women operate firms with significantly North and the South across common crop types. On less capital than firms operated by men. More capi- average, households farming yams attain significantly talized firms have higher profits, yet, on average, the higher values per hectare while those farming other value of the equipment owned by women-operated roots and tuber crops attain significantly lower values firms is only 16 percent of the total value for firms op- per hectare. In both the North and the South, wom- erated by men. This lower capitalization may be relat- en consistently farm the less valuable roots and tuber ed to women’s lower access to credit. crops, restricting their agricultural productivity rela- tive to men farmers. 12 Deep Dives on Three More Constraints to Women’s Economic Outcomes Differential access to and use of land hold back on-farm and off-farm Women almost exclusively sell to final consumers. opportunities for women. Selling to final consumers generates 46 percent lower Men are five times more likely than women to own profits than selling to traders or small businesses. De- land. Some 70 percent of plots are owned by individ- spite its lower profitability, nearly all female entrepre- ual men, while individual women own only 8 percent neurs (95 percent) operate in this lower profit space of plots. Land assets are critical in generating income and are half as likely (5 percent) to sell to traders or via agricultural and livestock activities as well as land small businesses as men entrepreneurs (10 percent). rental or sale. They also help store and accumulate wealth, and provide collateral to access credit and fa- Wage Work cilitate future investments. Gender differences in livestock holdings constrain women’s wealth accumulation. Animal ownership is higher in Nigerian male-headed households than in female-headed households. Wom- Women consistently work in lower-paying sectors. en livestock owners are more likely to own lower-val- This is seen most clearly in the education sector, ue animals than men livestock owners. The quantity where women are more than twice as likely to work of livestock owned by men also tends to be greater. relative to men. This stark sectoral segregation holds Livestock is a particularly useful, versatile, and profita- back women’s earnings because education is the low- ble asset, especially for poor households: it can store est-paying sector in Nigeria, with earnings averaging 28 wealth, serve as collateral, buffer against shocks, and percent lower than in other sectors, after controlling provide complementary inputs for crop production. for other individual and occupational characteristics. 13 Occupational segregation results in concen- Figure 2. Cost of the gender gaps trations of women in lower-profit sectors and lower-earning positions within sectors. $22.9 billion (5.8% of GDP) Occupational segregation impacts women working in agriculture through their choice of crops, self-em- ployed women through their position in the value chain, and women wage workers through sectoral 8.1 choice. Such segregation has far-reaching conse- quences in terms of overall productivity and earnings in the country. The misallocation of high-ability wom- en into low-return occupations may reduce econom- ic growth through the suboptimal allocation of labor across the sectors of the economy and segments of the value chain. $9.3 billion Billion USD (2.3% of GDP) The Cost of Gender Gaps to the 2.3 13.2 Nigerian Economy The total estimated forgone earnings stemming from gender gaps in agricultural productivity, firm profits, 6.2 and wage earnings amount to US$9.3 billion, equiva- lent to 2.3 percent of overall gross domestic product (GDP). This is a lower bound estimate, which corre- 0.8 1.6 sponds to the amount of additional earnings generat- At least With multipliers ed if every woman’s productivity, profits, and wage level were equalized with those of men. Using GDP multipli- Agricultural productivity gap Self-employement profits gap ers, which take into account potential spillover effects Wage earnings gap across sectors of the economy, the costs could be as Source: Nigeria General Household Survey data. high as US$22.9 billion, or 5.8 percent of overall GDP. 14 Policy Recommendations for The evidence map reveals that more work needs to be done to generate new evidence about effective Closing Gender Gaps in Economic interventions across contexts and at scale in Nigeria. Outcomes Moreover, translating this knowledge into significant progress toward closing the gender gaps will require a Despite the vibrant role of Nigerian women in agricul- profound change in the approach to programming for ture, entrepreneurship, and the labor market, many gender equality. constraints still hold back women, and in turn, the Nigerian economy. Recognizing that not all issues can Innovation in policy and programming will be a central be addressed with limited resources, this report helps facet of this approach. Generating innovative gender identify priority constraints that play an important role programming ideas will entail exploring and designing in equalizing earnings between women and men in Ni- interventions that are not “business as usual” and can geria. Each of the drivers of the gaps identified in the re- effectively address the factors that are keeping wom- port inform a policy priority that is backed by research en from performing as well as men in key economic and could help policy makers narrow gender gaps in sectors. Conducting rigorous impact evaluations to earnings. Adopting an evidence-driven approach to understand the effects of these interventions will be policy and program design for these priority areas can needed to further build the evidence base to inform help ease constraints, close gender gaps, and acceler- scale-up efforts of the government and its develop- ate economic gains for Nigerian women. The evidence ment partners. Notably, new knowledge produced map that follows (Table 1) summarizes what we know by Nigeria has the potential to move the emerging about what works to address those key constraints. regional and global evidence base forward on what It is important to note that as women face multiple re- works to address the constraints to women’s econom- inforcing constraints, considering complementarities ic empowerment. between interventions will be key to unlocking their potential. In addition, the policy options laid out in Nigerian policy makers are already making demonstra- this section must be complementary to a comprehen- ble strides toward narrowing gender gaps. Equipped sive approach to closing gaps in human capital. With- with additional evidence on the most effective pro- out equal access to health, education, and reduced gramming, they can accelerate progress toward reach- fertility, gender equality will likely be out of reach. ing inclusive development and economic recovery targets while advancing women’s empowerment and providing valuable lessons for the rest of the world. 15 TABLE 1 Note: The categories referenced in the “State of the evidence” column in table 1 are defined as follows: credible indicates that more than one impact evaluation from Sub-Saharan Africa demonstrates consistent, positive impacts of an intervention; emerging indicates that just one impact evaluation (from Sub-Saharan Africa or another developing context) shows positive impacts or multiple impact evaluations show mixed or not exclusively positive results; frontier indicates that there are no impact evaluations showing strong positive impacts, but other non experimental What works to close gender gaps in economic outcomes in Nigeria? evidence suggests that the intervention could address the given constraint; and not promising indicates that at least one impact evaluation shows no or negative impacts of an intervention. A review of the evidence Policy priority/ Policy Main Evidence State of the Policy priority/ Policy Main Evidence State of the constraint option conclusions from Nigeria evidence constraint option conclusions from Nigeria evidence Providing farmers with subsidized, improved seed varieties Subsidies for In-kind grants to female entrepreneurs can increase can increase crop yields and incomes for both women and improved seed Yes Emerging business profits, in some cases more effectively than cash men farmers, but may not necessarily result in gains for In-kind grants No Frontier varieties grants, given the pressure women face in redistributing cash women without specific efforts to target them. to household expenses. Engaging men to Engaging men and women through couples’ trainings can change norms encourage women's adoption and production of more No Frontier Large cash grants—provided to growth-oriented female around gendered valuable crops, typically farmed by men. entrepreneurs in the context of business plan competitions, crops Large grants Yes Credible for example—increase the likelihood of firm survival and Unlocking firm boost sales and profits of women-owned firms. Availability of Health insurance combined with weather index insurance owners’ access to Promoting women insurance products has the potential to increase women’s willingness to switch growth capital farmers’ choice of No Frontier higher-value crops suited to women to more valuable crops by decreasing the risk farmers they face. Secure savings mechanisms enable female Secure savings microentrepreneurs to set aside earnings and increase No Credible mechanisms investment in their businesses. Trainings to boost women’s socioemotional skills could Socioemotional skills facilitate their adoption of cash crops, given the strong No Frontier training correlation between noncognitive skills and take-up of more valuable crops, particularly in patriarchal societies. Programming that merges skills training and social Social network network building can increase earnings among female No Emerging building entrepreneurs, although the impacts of these programs tend to dissipate after they end. Employing more female extension service agents can increase women farmers’ use of productivity-enhancing Female extension technologies, since women in certain settings are more No Emerging service agents likely to adopt new technologies from female extension agents. Promoting women’s Psychology-based entrepreneurial training increases engagement in greater Psychology-based firm profits for both male and female entrepreneurs at Promoting women No Credible value addition business training significantly higher rates than traditional training, with farmers’ choice of women registering even greater gains than men. higher-value crops Local language voice and video agricultural extension and enhancing women Digital technology for messaging can increase women farmer’s participation No Emerging farmers’ use of farm agricultural extension in agricultural decision-making, boost their cultivation inputs of cash crops, and improve production outcomes. Job training in Job training in professional, male-dominated sectors can higher-return male- increase the likelihood of women entering the sector while Yes Emerging dominated sectors also changing gender-biased perceptions of the sector. Provision of subsidized fertilizer improves women Subsidies for inputs/ farmers’ fertilizer use and increases agricultural output, Yes Credible fertilizer especially when women farmers are specifically targeted. Enhancing women farmers’ use Access to accurate information regarding earnings in Information about male-dominated sectors can increase women’s interest in of farm inputs No Emerging earnings entering the sector, but may not sustain their participation in the sector, potentially due to challenging gender norms. Subsidies to While mechanization of agriculture has been shown to Decreasing mechanize farm benefit women relatively more than men, gender roles No Emerging labor can constrain women’s access to and use of technology. occupational segregation Access to subsidized daycare can help mothers shift from Childcare services lower-wage jobs with more flexible hours to higher-paying No Emerging jobs with fixed hours. Facilitating access Provision of cash transfers to households with children to farm labor and under five can increase spending on hired labor, increase mechanization Cash transfers No Emerging agricultural output, and decrease the time women spend on farm activities. Programs that Interventions aiming to expand women’s social networks build networks, or introduce them to role models and mentors could have No Frontier role modelling, and positive effects on shifting gender norms around women’s mentorship sectoral choice. Microcredit does not serve as sufficient capital to significantly impact women’s business outcomes; however, Microcredit Yes Not promising it has demonstrated other important impacts on women’s vulnerability, empowerment, and labor force participation. Enrolling young children in preschool can significantly decrease the number of hours caregivers, mostly women, Childcare services spend on childcare and increase the likelihood of women No Credible working outside of the home as well as increasing the Mesocredit/mid- Mid-sized loans for growth-oriented women-owned firms likelihood of women engaging in higher-paid work. No Emerging Unlocking firm sized loans can accelerate business growth. owners’ access to growth capital Easing women’s Gender norms and behavior training for men and young time constraints Engaging men Cash transfers and grants delivered within holistic couples can increase men’s participation in housework, to participate in No Emerging Cash transfers productive packages boost ultra-poor women’s business No Credible although women’s time spent doing housework does not housework start-ups and revenue. necessarily decrease as a result. 16 17 Introduction In recent years, Nigeria’s economic growth story has sizable gains, thus accelerating progress toward recov- been one of adaptation, resilience, and recovery in ery and inclusive development. the face of shocks. Following the 2014–2016 global oil price slump and the subsequent recession in Nigeria, Yet, several binding constraints underpin the gender the Nigerian government has been working to accel- gaps in economic outcomes in Nigeria. Gaps in human erate recovery through structural reforms aiming to capital, voice and agency, and the enabling environ- boost productivity and growth. These efforts, howev- ment limit a woman’s quality of life and her ability to er, have not yet translated into broad-based economic fully participate in the economy. While a number of gains for Nigerians. Nigeria ranked 152nd of 157 coun- policy and programming options are already under- tries in the World Bank’s 2018 Human Capital Index, way working to lift the barriers to women’s economic highlighting underinvestment in public infrastructure empowerment, to fully unlock women’s potential to and services as a persistent challenge.1 Nigerian wom- advance Nigeria’s development agenda, Nigerian pol- en face particularly acute challenges including some icy makers and their development partners must take of the world’s highest fertility and maternal mortality a strategic approach to addressing these gaps, identi- rates, and socioeconomic outcomes that consistently fying the evidence-based programs and policies with lag those of Nigerian men. the greatest promise for impact and the key partners poised to deliver them. Women in Nigeria are nearly half as likely as men to be salaried employees, twice as likely as men to be One of Nigeria’s greatest current challenges is achiev- contributing family workers, and 12 percentage points ing a demographic transition and creating the condi- more likely to work in vulnerable employment than tions for a demographic dividend, which occurs when men.2 Women-owned farms and firms are also typi- fertility and mortality drop and change the age struc- cally less productive and lucrative than those of their ture of the population. Fertility remains very high in male counterparts. These productivity and earnings Nigeria and is yet to reach a point which would ena- gaps constitute critical challenges for the Nigerian ble the economy to benefit from a large active, young economy, given women’s sizable role in these sectors. population, relative to its inactive age groups. Labor For instance, while over half of the firms are female market opportunities for women have the potential to owned, monthly profits of women-owned businesses contribute significantly to the demographic dividend are 66 percent lower than those of their male coun- agenda since they can both decrease fertility and con- terparts. If Nigeria were able to realize the full poten- tribute to economic growth through the employment tial of women—and equalize profits between working of women. On the other hand, high fertility and early men and women—it could translate these gaps into marriage and childbearing are holding women’s eco- 18 If Nigeria were able to realize the full potential of women—and equalize profits between working men and women—it could translate these gaps into sizable gains, thus accelerating progress toward recovery and inclusive development. nomic opportunities back. There is thus a virtuous cy- South regions of Nigeria, the report presents disaggre- cle by which women’s economic empowerment and gated results by region whenever possible. marriage and fertility decisions can positively impact each other. While this report recognizes the urgency Third, this report performs in-depth reviews of issues of the demographic transition agenda, it focuses on critical to enriching the understanding of gender gaps women’s economic empowerment, which will be key in key economic sectors. The deep dives in part III to achieving a demographic dividend, simultaneously focus on the evidence around three key issues that with a drop in fertility. are not captured comprehensively in the analysis in part II—land tenure security, participation in livestock The Nigeria Gender Diagnostic uses the latest 2018/19 value chains, and occupational segregation—yet hold Nigeria General Household Survey data to quantify promise for shaping women’s economic outcomes gender gaps in key economic sectors, their underly- and overall growth in Nigeria. ing drivers, and the costs of these gaps to the Nigerian economy. In doing so, this report provides the follow- Fourth, this report measures the costs of the gender ing contributions: gap in terms of forgone gross domestic product (GDP) in agriculture, entrepreneurship, and wage employ- First, this report examines the Nigerian labor force, the ment. Estimates in part IV suggest that closing gen- extent to which individuals work, and in which sector der gaps in key sectors of the Nigerian economy could they work. Part I highlights the gender gaps in labor yield additional gains of US$9.3 billion (2.3 percent of force participation, how women are less likely than overall GDP) or up to US$22.9 billion (5.8 percent of men to work on average, and how when they do work, overall GDP) accounting for spillovers to other sectors. they work fewer hours in less remunerative sectors. Fifth, this report reviews the evidence base from Nige- Second, this report documents the magnitude and ria and elsewhere on what works and does not work to drivers of the gender gaps in key economic sectors— support women farmers, business owners, and wage agriculture, entrepreneurship, and wage employ- workers in closing the gender gaps in their respective ment—using Kitagawa-Oaxaca-Blinder decomposition sectors. Part V provides targeted recommendations analysis. Part II highlights the immense gender gaps in for policy makers and their development partners, productivity and earnings in these sectors, amounting suggesting that policy action, innovative gender pro- to 30 percent in agriculture, 66 percent in self-em- gramming, and further research are needed to realize ployment, and 22 percent in wage employment. Given a more inclusive, gender-equal society in Nigeria. the well-known differences between the North and 19 Part I Women in the Workforce There are significant gender gaps in labor force par- FIGURE 3 ticipation and labor supply in Nigeria. Working-age women in Nigeria are 14 percentage points less likely Gender gaps in labor force to work than men, with 55 percent of women working compared to 69 percent of men, a gender gap larger participation and labor supply than the 6 percentage point gap estimated for Sub-Sa- haran Africa (box 1).I,3 The gender gap extends beyond 100 40 participation in the labor market to intensity in the la- 38 36 bor supply as well: among working individuals, women 35 34 work over 5.5 hours less per week than men.II 80 31 30 There are notable differences in women’s labor sup- 69% 27 68% 69% 71% ply between Nigeria’s North and South. In the South, Mean hours worked women are no less likely to work than men, although 60 Percent working 55% working women spend 4 fewer hours working per 20 46% week. In contrast, women in the North are 22 percent- age points less likely to work and those who do work 40 spend an average of 8 fewer hours working than men (figure 3). 10 20 0 0 National North South Women working Men working I The gender gap in labor force participation decreases slightly after Hours worked (women) Hours worked (men) controlling for individual and household characteristics, although women are still 9 percentage points less likely to work than men. Source: Nigeria General Household Survey data. II The gender gap in labor supply decreases to around 3 hours less per week than men after controlling for individual and household characteristics. 21 BOX 1 Data and definitions This report follows the National Bureau of Statistics (NBS) and defines working-age individuals as individuals aged between 15 and 65, and working individuals as any working-age individual who spent one hour or more in the past week on agricul- tural activities; nonagricultural business activities; casual, part-time, or temporary labor; or any other work for which they received a wage or salary. This section of the report uses data from the 2018/19 Nigeria General Household Survey (GHS) conducted by NBS in collab- oration with the World Bank Living Standards Measurement Study team. The survey data include data on 10,967 working-age individuals, which are weighted to form a nationally representative sample. Working status is calculated using the posthar- vest survey module, which was conducted in January–February 2019. This report presents subnational estimates for the North and the South. While subnational estimates may obscure important within-region or state-level differences, it is not possible to disaggregate them further due to sample size constraints and lack of representativeness at the state level of the GHS. The North encompasses the geopolitical zones of North East, North Central, and North West, and the South includes the zones of South East, South South, and South West. There are also significant differences in the sectors Table 2 displays the individual and household char- in which women participate and their labor supply acteristics that—holding all else constant—predict within each sector. Nationally, women and men are whether, and the extent to which, men and women equally likely to participate in an entrepreneurial ac- work. Factors highlighted in green are positively as- tivity, but women are 12 percentage points less likely sociated with the probability of working while factors to participate in agriculture and 7 percentage points highlighted in orange are negatively associated with less likely to have a wage-paying job.III There are im- the outcomes. For example, columns 1 and 3 of the portant regional differences in occupational choice: first row show that the likelihood of working at least the national gender gap in agricultural participation is one hour increases with age for both men and wom- driven entirely by the gap in the North where women en, while columns 2 and 4 show that the total number are 21 percentage points less likely than men to work of hours worked increases as age increases for both in agriculture. In contrast, the wage gap is equivalent working men and working women. The analysis iden- across the country as women are 8 and 7 percentage tifies three main characteristics correlated with adults points less likely than men to hold a wage job in the working in Nigeria: age, education, and marital status. South and in the North, respectively. Notably, while there is no gender gap in entrepreneurship participa- tion at the national level, this obscures regional varia- tion: relative to men in their respective regions, wom- en in the South are 4 percentage points more likely to have a self-employed income-generating activity, while women in the North are 3 percentage points less likely to do so. III Businesses include both formal and informal enterprises operated on either an own-account basis or together with hired employees or family labor. 22 TABLE 2 Correlates of working status WOMEN MEN Worked Total Worked Total Factors at least number at least number one hour of hours one hour of hours Age + + + + Age squared - - - - Household head + + + Married + + Polygamous household + Divorced + Primary education + + Junior secondary education - Senior secondary education + + + Postsecondary education + + - + Wealth index + - + Dependency ratio + Note: Geopolitical zone controls included in all specifications. The table presents only the statistically significant variables. The full list of variables included in the analysis is presented in appendix 1. Dependency ratio defined as number of dependents (children and adults over age 65) divided by the number of working-age adults in a household. Factors highlighted in green are positively associated with the probability of working while factors highlighted in orange are negatively associated with the outcomes. 23 BOX 2 Characteristics of working women are different in the North and in the South When comparing working women in the North and in the South, it is important to recognize that the character- istics of working women in each of these regions are different. On average, working women in the North are 4 years younger, 17 percentage points more likely to be married, and 29 percentage points more likely to have never attended school than working women in the South. Working women in the North also live in less-wealthy house- holds that have an average of 3 more people than working women in the South. Similarly, it is worth noting that the working women in each of these regions are also different than their nonwork- ing peers. Relative to their nonworking counterparts, working women in the North are, on average, 4 years older, 20 percent more likely to be married, but have equivalent education and wealth. On average, working women in the South are 10 years older, almost twice as likely to be married, as well as slightly more educated and less wealthy than their nonworking peers. FIGURE 4 Probability of attending school or working by age 1 0.87 0.86 0.87 0.87 0.84 0.80 0.80 0.8 0.78 0.79 0.78 0.77 0.77 Proportion working or attending school 0.69 0.69 0.68 0.68 0.65 0.65 0.64 0.60 0.61 0.59 0.6 0.53 0.50 0.4 0.2 0.0 5 10 15 20 25 30 35 40 45 50 55 60 Age Female Male Source: Nigeria General Household Survey data. 24 Issues around Women’s school enrollment rates.4 These gaps are consequen- tial to women’s future job prospects, since an addi- Labor Force Participation tional year of postsecondary education is estimated to increase the probability of a woman’s participation in Nigeria in the wage market by 15 percent in Nigeria.5 Throughout Nigeria, households prioritize educational investments in boys over girls, with a particularly large Age and Labor Market Entry disparity among low-income families.6 This disparity may be driven by gender norms that assume men as Age is an important correlate of working status: fig- breadwinners and women as the homemakers, or by ure 4 looks at the relationship between age and the differential returns to education for women and men, probability of either attending school or working, for of which gender norms are also an underlying factor. men and women. We combine the fraction attending An International Labour Organization survey on youth school with the fraction working because younger, aged 15–29 across eight Sub-Saharan African countries working-age individuals may still be in school: focus- indicates that 28.5 percent of young women with no ing only on those workers would incorrectly imply a schooling attribute their lack of education to paren- far higher rate of inactivity among the younger co- tal decisions, relative to 17.2 percent of young men.7 horts. The figure shows that boys and girls are equally Gender preferences are stronger among certain reli- likely to either attend school or work until the age of gions and ethnicities: some interpretations of the Is- 14, after which women’s participation drops, precip- lamic religion in the North-East of the country may itating the emergence of a persistent gap between discourage girls’ education, while ethnic groups like men and women. The timing of the emergence of this the Hausa-Fulani and the Kanuri restrict girls’ access to gap at the end of basic schooling highlights the fact formal education.8 Box 3 details the acute education that a significant portion of Nigerian women are una- challenges posed by the regional Boko Haram insur- ble to make the school-to-work transition. Early mar- gency. riage and childbearing are among key factors explain- ing this phenomenon: the adolescent fertility rate is very high at 106 per 1,000 women ages 15 to 19 per annum. The magnitude of the gap is relatively stable, at around 15 percentage points, across the remainder of the age distribution. Education and Parental Investments Education is a key early determinant of long-term Education is a key early economic opportunity with gender gaps in schooling determinant of long-term precipitating later gender gaps in economic outcomes. economic opportunity with Although Nigeria’s Universal Basic Education, initiated by the 1999 Constitution, establishes six years of free gender gaps in schooling primary schooling and three years of free secondary precipitating later gender gaps in schooling for children in Nigeria, a gender gap per- economic outcomes. sists in educational enrollment. In 2016, for example, there was a four-percentage-point gap between the male (44 percent) and female (40 percent) secondary 25 BOX 3 The threat and impact of terrorism deters girls’ pursuit of education Boko Haram’s ongoing insurgency has displaced over 2.7 million people in the Lake Chad region, including 1.9 million internally displaced persons in Northeast Nigeria.9 Between 2014 and 2017, Boko Haram destroyed over 1,500 schools; killed at least 1,280 students and teachers; and prevented thousands of children from continuing their education.10 The Global Coalition to Protect Education from Attack (GCPEA) estimates that the group has also abducted 600 women and girls from schools, often subjecting them to forced conversion from other faiths to Islam,11 forced marriage, rape, subsequent childbirth, or a fate as a suicide bomber.12 In interviews by GCPEA, female students from attacked schools reported suspending or permanently abandoning their education after school attacks due to destroyed or damaged schools, their parents’ inability to pay for school expenses, and be- cause of their or their parents’ fear. The negative impacts of conflict on education extend beyond those directly impacted: quantitative analysis of households in Northeastern Nigeria suggests that households within 5 kilome- ters of conflict experience significant decreases in years of education and probability of school enrollment, and increases in the probability of dropout.13 Marriage, Childbearing, and Domestic Responsibilities ic characteristics.17 Childbirth has both supply- and Nigeria has some of the world’s highest rates of ear- demand-side effects on women’s occupational out- ly marriage for women and evidence from the region comes. On the supply side, it introduces heightened suggests that marriage is linked to leaving the labor responsibilities and time constraints for women, as force. Nationwide, 18 percent and 43 percent of girls social norms designate the responsibility of domestic marry by age 15 and 18, respectively.14 Rates are par- work and childcare to women. On the demand side, ticularly high in the Northwest, where the median age employers may view female employees as generally of marriage for women is 15.8 years old.15 Early mar- more expensive and riskier. For these reasons, among riage negatively affects labor market outcomes: a others, global estimates on 97 countries indicate that study in four Sub-Saharan countries found secondary over the course of a woman’s reproductive life, each education to be the main determinant of wage sec- birth reduces her total labor supply by about two tor employment for women before marriage but that years.18 Both the number of recent births and short marriage completely negates this effect.16 Initiatives birth spacing have substantial negative effects on aiming to use education to increase women’s occu- women’s employment.19 Education can play a pivotal pational outcomes must consider and address gender role in decreasing fertility: a study has estimated that norms related to marriage and domestic responsibil- an additional year of schooling reduces fertility by 11 ities (box 4). to 19 percent in Nigeria and, in particular, that Nigeria’s Universal Primary Education (UPE) program was asso- One of the key ways marriage impacts labor market ciated with a 16 percent reduction of early births.20 outcomes is through childbirth: women who marry before age 18 have approximately 20 percent more births over their lifetime relative to women who mar- ry after age 18, after controlling for socioeconom- 26 BOX 4 Labor market opportunities for women can contribute to lowering desired fertility and delaying marriage One of Nigeria’s biggest current challenges is achieving a demographic transition, that is, a drop in mortality rates followed by a drop in fertility. The demographic transition changes the population age structure and creates a youth bulge. With the right conditions in place, this youth bulge can lead to large economic gains, called the demographic dividend. This occurs when the dependency ratio plummets and the large young population is gainfully employed. At 5.3 children per woman, fertility is very high in Nigeria, especially in the northern states, where it can reach 7 or more children per woman.21 High fertility is driven by an array of factors, from a high desired number of chil- dren, a young age at marriage, to unmet contraceptive needs and low levels of women’s empowerment. Fertility reduction requires a multipronged approach, combining the supply of contraceptives and quality family plan- ning services with women’s increased empowerment and earnings.22 Currently, the level of modern contraceptive methods use is 12%, but as low as 6% and 8% in North West and North East Nigeria respectively. The desired number of children is a function of many factors, including labor market opportunities. Improved income and employment prospects increase the opportunity cost of childbearing for women, who may choose to delay childbearing to avoid earning losses. In addition, increased economic empowerment can improve women’s bargaining power in the household and enhance their ability to negotiate the timing of childbearing and use of contraceptives. Finally, being gainfully employed can expand a woman’s ability to contribute economically to the household and may alleviate pressures to marry.23 Although the negative relationship between women’s econom- ic empowerment and fertility has been established, there is a dearth of empirical evidence on the causal impact of women’s increased labor force participation on their desired and realized fertility. Several studies, however, including in the Nigerian setting, have demonstrated that increased educational outcomes reduce fertility.24 Later marriage and fewer children also contribute to promoting women’s participation in the labor market, thus creating a virtuous cycle through which women’s economic empowerment and lower fertility positively impact each other, and setting the conditions for Nigeria to reap a demographic dividend. 27 Part II Measuring the Gender Gaps In addition to women working less than men, as de- FIGURE 5 tailed in part I, those active in the workforce also con- sistently earn less than men, even after holding labor Gender gaps across key sectors supply constant (figure 5). This section of the report documents the magnitude and drivers of the gender 80 gaps in agricultural productivity, business profits, and hourly wages. The analysis uses the Kitagawa-Oaxa- 71% 70 ca-Blinder decomposition methodology (see box 5) 66% to determine whether differential access to resourc- es—such as credit, assets, and education—or differ- 60 ential returns to these resources drive the observed gender gaps. 50 Percent 40 30% 30 28% 22% 20 16% 10 0 Agriculture Self-employment Wage work Unconditional Conditional Source: Nigeria General Household Survey data. 29   BOX 5  Examining differences in levels of and returns to resources 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 method- ology 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, 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. The structural effect also captures potential differential selection into the various sectors, which may be due to differential comparative advantage, discrimination, or social norms. For example, if women face additional barri- ers to enter a given sector, the analysis compares the set of women who were able to overcome those barriers to a set of men that did face those barriers in the first place. The characteristics that helped women overcome the barriers may be observed in the data (such as education) or unobserved (such as perseverance). The estimated difference in returns to these characteristics for men and women combines the effect due to gender and the effect due to the differential composition of the samples. In this report, the endowment effect may be referred to as “levels,” while the structural effect may be referred to as “returns.” Annex 2A provides additional technical details on the Kitagawa-Oaxaca-Blinder decomposition. Agriculture Who is a Plot Manager? Over 20 percent of working-age individuals in Nigeria This report defines plot managers as individuals listed participate in the agricultural sector that produces 21 by the household as the main decision-makers for a percent of the country’s GDP.25 The sector is character- given plot. ized by relatively low female participation, women be- ing less likely to manage their own plot, and a produc- This section uses data from the 2018/19 Nigeria Gen- tivity gap between male and female plot managers. eral Household Survey. The survey data include 5,922 Women are almost 10 percentage points less likely to plots managed by 2,852 farm managers, of whom 21 work in agriculture—either on their own or someone percent are women. Female plot managers have very else’s plot—than men. Women are also 25 percentage different household characteristics than their male points less likely to manage an agricultural plot than counterparts: a majority of female plot managers are men, and those who manage their own plot have sig- widowed, separated, or divorced (63 percent), con- nificantly lower productivity than their male counter- trasting with male plot managers who are almost ex- parts. This section characterizes the productivity gap clusively married (94 percent). Relative to male plot between male and female plot managers and exam- managers, female plot managers are also four years ines the differences in resources and returns to re- older, 13 percentage points more likely to have never sources that drive the gap. attended school, and reside in households with an av- erage of 0.64 fewer adults. 30 BOX 6 How do we measure agricultural productivity? In this report, agricultural productivity is defined as the average value of agricultural output produced per unit of land managed (in hectares). The value of the output is based on farmer-reported estimated values for each crop on each plot harvested during the 2018/2019 agricultural season. Land area is measured by GPS devices or, in a small number of cases, estimated by the farm manager. FIGURE 6 Gender gap in agricultural productivity 30% National 16% 35% North 23% 25% South 22% 0 20 40 Gender gap (percent) Unconditional Conditional Source: Nigeria General Household Survey data. Note: The unconditional gender gap controls for plot size. Decomposition of the istics of the farmers, their households, and their crops, Agricultural Productivity Gap and the other half is attributable to structural charac- teristics, such as differential returns to inputs. Notably, Female plot managers in Nigeria produce significant- there are not drastic differences between the North ly less per hectare than their male counterparts (fig- and the South, with almost equivalent gender gaps of ure 6). Nationally, the value of the output per hectare 23 percent and 22 percent, respectively although the on female-managed plots is 30 percent lower than factors driving these gaps differ between the North comparably sized male-managed plots (box 6).IV This and the South (see box 8).V agricultural productivity gap between male- and fe- male-managed plots shrinks to 16 percent after con- V The national conditional average is smaller than the gap in either trolling for individual-, household-, and plot-level the North or the South, due to the composition of the sample. characteristics. This suggests that almost half of the Productivity is higher in the South for both men and women, but the ratio of men to women farmers is much higher in the North gender gap is attributable to the differing character- than in the South. When producing a national gender gap, the analysis compares a sample of men, the majority of whom are from the North (with slightly lower productivity), to a sample of IV In this report, we use the terms female plot managers and women women equally represented in the North and in the South (with farmers interchangeably. relatively higher productivity). 31 BOX 7 Agricultural productivity gaps have been relatively stable or shrinking over the past decade Combining the gender gaps in agricultural productivity presented above with equivalently calculated gaps using the first wave of the Nigeria General Household Survey—which was conducted in 2010/11—suggests that unconditional gender gaps in agricultural productivity have remained relatively stable over the last decade, while conditional gender gaps have fallen, particularly in the North. Specifically, unconditional gender gaps in 2010/11 were estimated at 27% in the North and 21% in the South, relative to the 35% and 25% in 2018/19 shown in figure 7, panel a. As shown in figure 7, panel b below, the conditional gender gaps estimated in 2010/11 were 43% in the North and 28% in the South, relative to 23% and 22%, respectively, in 2018/19. Figure 7 Unconditional and conditional gender gaps in agricultural productivity by region a. Unconditional gaps have risen while... b. Conditional gaps have fallen 60 60 43% Unconditional gender gap (%) Conditional gender gap (%) 35% 40 40 28% 27% 25% 23% 22% 21% 20 20 0 0 2011 2018 2011 2018 2011 2018 2011 2018 North South North South Table 3 summarizes the decomposition analysis of the while variables shown in green moderate the gap. The agricultural productivity gap, listing the statistically analysis suggests three key factors associated with significant drivers from the full range of individual-, the productivity gap between male-managed and fe- household-, and plot-level characteristics included in male-managed plots: input use, crop choice, and the the analysis. Variables shown in orange drive the gap composition of labor used on the plot. 32 TABLE 3 Significant drivers of the agricultural productivity gender gap Significant factors Levels Returns Manager and household characteristics Dependency ratio - Land characteristics and agricultural practices Rented plot - Crop choice - + Fertilizer - Herbicide - - Own labor supply + Male household labor + Hired male labor - Note: The table presents only the statistically significant variables. The full list of individual-, household-, and plot-level variables included in the analysis is presented in appendix 2b. Variables shown in orange drive the gap while variables shown in green moderate the gap. 33 roots and tuber crops, holding back their agricultural productivity relative to men farmers: nationally, wom- en are 38 percentage points more likely to farm roots and tubers, and 19 percentage points less likely than Women use fewer inputs men to farm cereals; in the North, women are less like- ly to farm cereals and legumes and more likely to farm Agricultural inputs—such as fertilizer—can significant- horticulture and root and tuber crops; women in the ly improve farmer yields. Gender differences in the South are more likely to farm root and tuber crops. A use of these agricultural inputs are a key driver of the recent review noted that gender differences in crop gender productivity gaps in Nigeria. On average, dou- choice may stem from gender differences in risk pref- bling the quantities of fertilizer and herbicide used erences and skills, or norms around certain cash crops on a plot yields increases in agricultural productivity being perceived as “male crops.”26,VI of 6 percent and 18 percent, respectively. Nationally, male farmers use over 8 times more fertilizer and 50 percent more herbicide per hectare than their female counterparts, suggesting that equalizing input use could significantly increase female productivity. Given the high returns to fertilizer and herbicide, the lower use of these inputs is constraining female plot manag- Women use less productive labor er productivity. There are important regional gender differences in input use: women farmers in the North Farmers can draw on three sources of labor for their use significantly less fertilizer than their male counter- plot: their own labor, household labor, and hired labor. parts while women and men farmers in the South use Women plot managers do not appear to have trou- comparable amounts. In contrast, women farmers in ble accessing labor: women plot managers work more the South use less herbicide than men farmers in the than men plot managers and use more household and South, while women farmers in the North use slightly hired labor. However, the male labor used by female more than the men farmers. plot managers is significantly less productive than the male labor used by male farmers: hired male la- bor is less productive for women than men nationally, driven by hired male labor in the North, while male household labor is less productive for women plot managers in the South. Several mechanisms may drive the labor productivity gap: female plot managers may Women farm less valuable crops not have time to effectively supervise workers; male laborers working for a woman supervisor may exert Crop choice represents one of the most fundamen- lower effort; and women may lack resources to hire tal decisions farmers make, and the farming of dif- more productive workers.VII,27 ferent crops is a key driver of the productivity gap between female-managed plots and male-managed VI Women may opt not to grow cash crops based on the perception that they are riskier crops due to the large upfront investments plots. There are consistent value differences in both needed to produce at scale and the exposure to price fluctuations the North and the South across common crop types: in the market. Meanwhile, skills might also influence women’s decisions around cash crop farming: in Malawi, increases in women on average, households farming yams attain signifi- farmers’ noncognitive ability are linked with higher production of cantly higher values per hectare than those farming cash crops. VII The World Bank’s Africa Gender Innovation Lab is currently other roots and tuber crops. In both the North and conducting a high-frequency agricultural labor survey in Nigeria the South, women consistently farm the less valuable to examine these issues. 34 BOX 8  Differential participation in agricultural extension is a regional driver of the gender gap in the North Extension services can improve farmer productivity by teaching or demonstrating locality-specific best practices. However, only 3 percent of women plot managers in the North participated in extension services, relative to 20 percent of men. This is despite high returns to extension services in the North: plot managers that participated in extension activities within the last year had 18% higher productivity. The low extension activity participation rates for women may be attributable to several factors: (1) current extension services may focus on crops pre- dominantly farmed by men; (2) extension activities may be run by men with cultural norms precluding women participating; or (3) current outreach activities may target predominantly male social networks. Current Policy Landscape Several strategic documents at the federal level tack- Three key programs driving the gender strategy of the le constraints facing female farmers. Nigeria’s Agricul- APP are the Women and Youth Empowerment Pro- tural Promotion Policy (APP) formulates explicitly the gram (WYEP), the Graduate Unemployed Youth and objectives of reducing the gender bias in land alloca- Women Agro-Preneur Support program (GUYS), and tion and titling and expanding technical training and the Youth Employment in Agriculture Program (YEAP). access to financial services for women. In addition, the With the aims of removing barriers to finance, knowl- Federal Ministry of Agriculture’s Gender Policy in Ag- edge, and agribusiness information, the programs pro- riculture highlights the following activities in its policy vide in-kind grant, training, and business support to framework to address constraints facing women farm- women. In addition, the Agro-Processing Productivity ers: communication to equalize access to land and to Enhancement and Livelihood Improvement Support publicize existing credit facilities, facilitating access (APPEALS) project is evaluating the gender disaggre- to mechanization, making inputs distribution more gated impact of layering interpersonal and intraper- gender sensitive, promoting the production of small sonal socioemotional skills on top of business grant livestock, promoting agribusiness, developing women provision and technical training on agricultural pro- farmers’ business skills, disseminating information on ductivity, agribusiness survival, employment, and ag- market access, and extension services through Farmer ribusiness assets. Field Schools. 35 Self-Employment Self-employment is the most common income-generating activity among women in Nigeria with almost half of income-earning women spending most of their economically productive time in entrepreneurship-related activities. As with agriculture, female firm managers earn lower profits than their male counterparts. This section characterizes the profit gap between male- and female-managed enterprises and examines the differences in resources and returns to these resources that drive the gap. Who is a Business Manager? Decomposition of the Entrepreneurship Profit Gap This report defines business managers as individuals who manage or are most familiar with an income-gen- Self-employed women in Nigeria earn significant- erating activity. This broad definition encompasses ly lower profits from their enterprises than self-em- both formal and informal enterprises operated on an ployed men (figure 8). Nationally, women earn 66 own-account basis, or with hired or family labor. percent lower profits than men. The gender gap in profit increases to 71 percent after controlling for indi- This section uses data from the 2018/19 Nigeria Gen- vidual, household, and enterprise characteristics. The eral Household Survey. The survey data include 3,046 increase in the gender gap after controlling for these business managers, of whom 52 percent are women. characteristics suggests that self-employed women Female business managers have similar individual-lev- generally have characteristics associated with higher el characteristics as their self-employed male coun- profits. The analysis suggests significant regional var- terparts: almost 80 percent of both groups are married iation in the size of the gender gaps: self-employed and both live with around 3.8 other adults. Self-em- women in the North and in the South earn 56 percent ployed women are, on average, 0.7 years younger and less and 79 percent less than their male counterparts, have 0.7 fewer years of education. respectively. The gaps are larger in urban areas than rural areas. 36 FIGURE 8 Gender gaps in self-employment profits 66% National 71% 70% North 56% 63% South 79% 69% Urban 80% 64% Rural 68% 0 20 40 60 80 100 Gender gap (percent) Unconditional Conditional Source: Nigeria General Household Survey data. Note: The unconditional gender gap controls for plot size. 37 Table 4 summarizes the decomposition analysis of the between men’s and women’s firms. Transitioning to self-employment profit gap, listing the statistically sig- selling to traders or businesses further up the supply nificant drivers from the full range of individual- and chain may necessitate selling higher-value goods, a household-level characteristics included in the anal- wider range of items, or larger quantities. ysis. Variables shown in orange widen the gap while variables shown in green moderate the gap. The anal- ysis indicates two key factors associated with the prof- Current Policy Landscape it gap between male-managed and female-managed enterprises: physical capital and target market. The National Directorate of Employment (NDE) pro- vides no-fee vocational skills development, entrepre- neurship development, and agricultural skills training to both men and women across Nigeria, thereby elim- inating barriers to financial access. NDE is currently evaluating a gender disaggregated impact of provid- ing soft skills, internships, and apprenticeship on top of regular vocational and entrepreneurship trainings to beneficiaries under the Skills for Jobs program. The Women operate less-capitalized firms Small and Medium Enterprises Development Agen- cy of Nigeria (SMEDAN) operates the national poli- Self-employed women operate firms with significantly cy on micro, small, and medium enterprises (MSME) less capital than firms operated by men: on average, and has been promoting MSME operations across the the value of the equipment owned by women-operat- country through business process support operations, ed firms is only 16 percent of the value for firms operat- research and development and ICT (information and ed by men. More-capitalized firms have higher profits: communication technology), market and finance sup- nationally, doubling the capital stock of firms is asso- ports, and targeted enterprise development. The pol- ciated with a 4 percent increase in profits, suggesting icy provides for conscious and targeted development that equalizing the capital stocks of female-owned support to women in MSMEs. Moreover, discounted enterprises to equivalent levels of male-owned enter- rate lending and uncollateralized credit for women prises would increase their profits by 25 percent. are being explored through third-party financial in- stitutions. For example, partnering with Access Bank in Nigeria, Nigeria Development Bank is piloting cash- flow-based lending to women-owned MSMEs across the country. Women almost exclusively sell to final consumers Over 95 percent of female entrepreneurs sell to final consumers despite these firms generating 46 percent lower profits than firms that sell to traders or small businesses. While most male entrepreneurs (90 per- cent) also sell to final consumers, the higher rate at which women are operating these less-profitable firms represents a key driver of the profitability gap 38 TABLE 4 Significant drivers of the self-employment profit gap Factors Levels Returns Manager and household characteristics Polygamous household + Enterprise characteristics Physical capital stock - Sells products to final consumers - Sells products to traders and small businesses + Access to internet - Number of enterprises - Note: The table presents only the statistically significant variables. The full list of individual- and household-level variables included in the analysis is presented in appendix 2C. Variables shown in orange drive the gap while variables shown in green moderate the gap. 39 Wages Earning a wage is the least common income-generating activity among working women in Nigeria, with only 12 per- cent of women primarily earning wages. Public-sector wages are determined based on a gender-blind salary scale and while the Nigerian constitution prohibits gender discrimination, the data suggest private-sector firms differen- tially compensate male employees. This section characterizes the wage gap between male and female employees and examines the differences in firm characteristics and returns to these characteristics that drive the gap. Who is an Employee? Decomposition of the Wage Gap This report defines employees as individuals who re- Nationally, female wage earners in Nigeria earn 22 per- port working for someone outside the household for cent less than male wage earners (figure 9). The gap payment in cash or in kind. increases slightly to 28 percent after controlling for in- dividual-, sector-, and employer-level characteristics. This section uses data from the 2018/19 Nigeria Gen- The gender gap is smaller in the public sector, with a eral Household Survey. The survey data include 1,087 5 percent difference between female and male wage employees, of whom 34 percent are women. Female earners, which jumps to 27 percent in the conditional wage earners have similar individual-level charac- specification. It is worth noting that the survey data teristics as male wage earners: both have around do not include individual salary grades or levels for 12 years of education and live with an average of 4 individuals in the public sector and civil servant hir- adults. Wage-earning women are 2 years younger ing data suggest comparable proportions of men and than wage-earning men and while around 25 percent women rise to the different salary grade levels.28 The of both groups are single, almost four times as many larger gap in the conditional specification indicates women are divorced or widowed as men. that women may, on average, need to be more quali- fied than men to reach a given salary grade. Within the private sector, women earn an average of 40 percent less than their male counterparts, which falls to 24 percent after controlling for individual-, sector-, and employer-level characteristics. 40 FIGURE 9 Gender gap in wages 22% National 28% 5% Public 27% 40% Private 24% 0 20 40 60 Gender gap (percent) Unconditional Conditional Source: Nigeria General Household Survey data. Note: The unconditional gender gap controls for plot size. 41 Table 5 summarizes the decomposition analysis of the gender gap in wages. Variables shown in orange drive the gap while variables shown in green moderate the gap. The analysis indicates that the main driver of the wage gap between men and women is the sectoral Women are relatively more likely to work in the differences where women consistently work in low- public sector er-paid sectors. The analysis also shows that the gap is partially mitigated by the fact that, on average, wom- Women are 11 percentage points more likely to work en wage earners are more highly educated than male in government, which decreases the wage gap; while wage earners, although women have a lower return to both men and women working in the public sec- their education. We present findings from the survey tor benefit from employer-specific wage premiums, data on sectoral differences for wage earners below women also benefit from the lower gender gap stem- and examine the broader topic of occupational segre- ming from an official, gender-neutral salary structure. gation—the unequal distribution of men and women The wage benefit for government work is significant: across sectors, with women tending to work in less lu- women working in government earn 38 percent more crative sectors—in part III. than women with private-sector employers compared to an 11 percent public-sector work premium for men. Current Policy Landscape The Nigeria Labor Act provides the general rules of en- gagement between employers and employees of both Women work in less-remunerated sectors public and formalized private sectors in the country, and emphasizes fairness and justice in recruitment, Female wage earners are more than twice as likely to employment, contracts, and wages. It seeks to protect work in education, relative to male wage earners. This women and other vulnerable categories in the labor stark sectoral segregation is a significant driver of the market. The National Employment Policy (NEP) of wage gender gap because education is also the low- 2017 aims to provide a creative environment for pro- est-paying sector: education workers earn 28 percent ductive and employment-intensive growth in Nigeria less than other wage workers, after controlling for oth- and seeks to promote gender equality in employment er individual and occupation characteristics. Diversify- by eliminating constraints to participation of women ing women’s wage work toward relatively higher-paid in the workforce. The NEP highlights that the Govern- sectors—such as manufacturing, agricultural process- ment shall eliminate open or disguised discrimination ing, and service work—could shrink the wage gap and against women workers in recruitment, remuneration, promote substantial economic growth. In addition, promotion, and training. The policy direction here recent research has demonstrated that decreased could be complemented by closing gender gaps in occupational segregation can also generate substan- productivity, wages, and vertical and horizontal occu- tial growth by improving firm and worker productivity pational segregation. through a more efficient allocation of workers across sectors.VIII,29 VIII We discuss this point in more detail in the Occupational Segregation Deep Dive section in part III of this report. 42 TABLE 5 Significant drivers of the wage gap Factors Levels Returns Manager and household characteristics Married - Widowed + - Education + - Job characteristics Public sector + Industry: education - Note: The table presents only the statistically significant variables. The full list of individual- and household-level variables included in the analysis is presented in appendix 2D. Variables shown in orange drive the gap while variables shown in green moderate the gap. 43 Part III Deep Dives The analysis presented in part II documents large gen- ing secondary analysis in addition to the quantitative der gaps in earnings in Nigeria: 33 percent in agricul- analysis in this report. tural productivity, 66 percent in firm profits, and 22 percent in wages. In agriculture, the gap is driven by The first deep dive describes the land tenure and ti- crop choice, use of inputs, and the composition of la- tling institutional environment, how it shapes wom- bor a woman plot manager uses. In entrepreneurship, en’s use and ownership of land, and why land tenure is differences in profits stem from lower endowment in important for increasing earnings. Although land size capital and position in the value chain. Women wage and ownership did not appear to be a significant driv- workers earn less than men due to their sector of em- ing factor of the gender gap in agricultural productiv- ployment. ity, difficulty in measuring those variables accurately could cause this result. In the second deep dive, we The identified constraints interact with each other and delve into women’s livestock ownership and use. The with preferences regarding occupation, time use, and household survey used in part II does not capture family responsibilities. These preferences are in turn individual holdings or production of livestock, but shaped by contextual factors such as social norms rather measures them at the household level. Thus, and institutions. Contextual factors are not adequate- exploring the role that limited livestock access may ly captured by quantitative data or analyses; however, play as a constraint to women’s economic empower- they are key to understanding the barriers that wom- ment is key. Finally, the issue of occupational segre- en face in their economic trajectory. gation is a salient cross-cutting factor within each of the three sectors we analyzed in part II. For instance, In the following sections, we perform deep dives to women are more likely to farm lower-value crops, op- investigate the role of three factors that may shape erate in less-profitable positions in value chains, and the gender gap even though they are not necessarily work in less-remunerated jobs in wage employment. captured by the analysis presented in part II. In the In the third deep dive, therefore, we explore the issue preparation of this report, factors that limit women’s of occupational segregation more closely: its underly- economic empowerment were discussed in multiple ing causes and evidence on interventions to relax this rounds of consultations with government agencies constraint. at the federal and state level in Nigeria. Three salient factors emerged from those consultations as requir- 45 DEEP DIVE I future investments.34 Evidence from the region shows the importance of secure use and ownershipIX of land for agricultural productivity: in Ghana, for example, weaker land tenure security for women led to produc- tivity losses on their plots, while in Rwanda, formaliza- Deep Dive I: tion of women’s land rights through titling programs increased their investments in soil conservation.35 Land Tenure and Titling Without proper land rights, however, it is difficult for women to achieve high agricultural productivity or leverage the financing necessary to start productive enterprises.36 Women’s constraints in owning and con- Women have lower levels of land trolling land hinder women’s economic progress in Ni- ownership than men in Nigeria geria, given that land ownership can provide women with the security and stability to optimize production In Nigeria and throughout the developing world, wom- and unlock financial resources.37 en are less likely to own or use land than men—a ma- jor driver of gender inequality and gaps in economic The ways in which land use is regulated and property outcomes.30 Although women represent between 60 rights are defined have additional implications that and 79 percent of Nigeria’s rural labor force, men are extend far beyond the agricultural sphere.38 Wom- five times more likely than women to own land.31 The en’s land ownership has been linked to increased Food and Agriculture Organization (FAO) delineates household bargaining power, higher protection from two useful indicators of women’s land ownership: the domestic violence, better child nutrition, and low- distribution of landholders by sex and the percentage er exposure to human immunodeficiency virus and of men and women who own land. In both indicators, acquired immune deficiency syndrome (HIV/AIDS).39 women fare worse than men in Nigeria.32 Seventy per- These findings signal that land rights matter not only cent of plots are owned by individual men while only for agricultural productivity, but also for sustainable 8 percent are owned by individual women, and the re- development and gender equality as a whole in Ni- maining jointly owned plots do not necessarily entail geria.40 Nigeria’s Agricultural Promotion Policy 2016– equal rights between men and women. Meanwhile, 2020 and the National Livestock Transformation Plan only 5 percent of women in Nigeria own land relative 2019–2028 recognize the importance of women’s land to 23 percent of men. Less than 2 percent of women rights to agricultural productivity and broader devel- own at least one plot of land solely, relative to nearly opment outcomes, highlighting policies that reduce 17 percent of men.33 gender bias in land allocation and titling processes as a priority area for reform. The Nigeria National Gender Action Plan also identifies land as a major constraint for women farmers and proposes activities to facili- Why does women’s land tenure tate women’s access to land for agriculture on a sea- security matter? sonal basis. Land is an asset that can generate income via agricul- tural and livestock activities, rental, or sale; store and IX We refer to the rights to access, manage, rent, transform, and accumulate wealth; and serve as collateral to facilitate transfer land as ownership. 46 DEEP DIVE I How do women use and own land In Nigeria, the vast majority of land is acquired via fam- ily allocations or through the market: 71 percent of in Nigeria? male-managed plots and 69 percent of female-man- aged plots are acquired through family inheritance, while only 7 and 2 percent of male- and female-man- Statutory and customary laws aged plots, respectively, are acquired through pur- chase.46 An examination of various systems indicates Gender disparities in land tenure exist due to both that, by and large, women’s access to land depends statutory and customary law. The 1978 Land Use Act on their relationships with men through marriage, in- vested all land within each state to the governor and heritance, or borrowing.47 Women’s rights to land may local governments, introducing instability and un- often be considered “secondary”: since no formal certainty into the land market. Currently, in many re- concept of property co-ownership exists in traditional spects, customary law prevails in dictating the infra- Nigerian culture, all substantial property assumedly structure of the land market in Nigeria. belongs to the male head of household, and land is almost exclusively registered in men’s names.48 While In 1978, The Government of Nigeria introduced the all members of the landowning family, including wom- Land Use Act (LUA) into the Constitution, vesting all en, are typically entitled to use a portion of the prop- land situated within each state to the governor and erty for agriculture, this use is often limited to a short delegating land control, management, and allocation period of time and precludes making any long-term of urban and rural areas to the governor and local improvements to the land.49 The predominance of the governments, respectively.41 Little understood by the customary system has also created a complicated majority of Nigerians, the LUA introduced several in- land rights regime as systems vary widely across and efficiencies and uncertainties into the land market.42 within ethnic groups.50 The system outlined by LUA has not been fully im- plemented or realized: less than 3 percent of land in Nigeria is formally registered via a certificate of occu- Family allocations: marriage pancy.43 Though the LUA was established to make land accessible to all Nigerians and remove confusion and In Nigeria, marriage is a main pathway to women’s discrimination in land titling, it has failed to achieve land use, though women’s access to land often lasts either objective. Ultimately, instead of eliminating in- only so long as the marriage itself.51 While the Marriage formal land markets, the LUA has made room for more Act of 1990 stipulates that a woman is entitled to at informal transactions. As such, Nigeria’s land market least one-third of her husband’s estate in the case of predominantly functions on customary rather than death, this only applies to women who were married statutory practices.44 Though the LUA itself does not under statutory law and only if there is a will.52 Un- stipulate any discrimination on the basis of gender, its der sharia law, a widow has the right to inherit a share failed implementation and subsequent segue to in- of family property—although her share is typically formality open the door to discrimination and gender smaller than those received by male family members.53 inequality via customary practices.45 Women married under customary law face different outcomes: in cases of widowhood, they often have no right to inheritance, while in cases of divorce, they face threats of dispossession.54 47 DEEP DIVE I Family allocations: inheritance and register it in their names. However, tradition of- ten prevails and, out of compliance to norms, women Women may gain access to use and own land through may not register land in their names but rather in the inheritance, though the chiefly patrilineal inheritance names of their male family members.64 system in several of Nigeria’s regions often excludes women from the right to inherit land. By and large, traditional beliefs and practices, customary laws, and some statutory laws dictate that daughters cannot in- Toward more equal land rights: herit land when a father dies.55 The Southern states, Land reforms and women’s for example, largely follow primogeniture in which the first son inherits his father’s entire estate.56 While education may improve women’s women are generally excluded from inheritance in land tenure security and the South-South, South-East, and the Middle belt re- gions, the North is an exception: in this region, wom- empowerment en have land ownership rights through inheritance, as Sharia law entitles women to a small share of the Land reforms inheritance, or half of what their brothers inherit.57 However, women frequently relinquish the inherited Expanding women’s access to and ownership over land to male family members due to social pressures.58 land tenure requires changes in Nigeria’s statutory in- Women may also forgo their inheritance because they frastructure.65 However, given its entrenchment in the lack awareness of their rights.59 1999 constitution, the LUA is difficult to amend. In its nearly forty years of operation, the LUA has not been reviewed once.66 In 2007, President Musa Yar’adua Market: purchasing introduced a seven-point agenda that included land reforms, and in 2009, he founded a Presidential Tech- Although most land in Nigeria is acquired through in- nical Committee on Land Reform.67 The Committee heritance or gift, women can also obtain land own- sought to develop guidance for the government on ership rights through purchase.60 Intrahousehold dy- registering landholders and proposed an approach for namics and capital constraints may complicate this systematic land titling registration (SLTR).68 In its first process. To purchase land, in some traditional soci- year, the Committee developed a sensitization pro- eties, married women need a letter of authorization gram for the country to promote attitudinal changes from their husbands. While not a legal requirement, toward land tenure and an understanding of the eco- this customary practice creates an additional hur- nomic opportunities of modern land titling and regis- dle for women to purchase land.61 Women often lack tration; developed a study of technical requirements capital to buy land and, even if they do, it may be in for determining a nationwide cadaster; developed a the possession of husbands or male relatives.62 While capacity building program for collecting, processing, loans may provide a viable pathway toward ownership, and structuring spatial data into a useful, nationwide obtaining them requires collateral and sometimes the cadaster; and undertook pilot projects in urban and same written permission from a husband.63 In urban rural governments of six states.69 centers, customary laws in land transactions and own- ership are laxer, and women may buy a parcel of land 48 DEEP DIVE I Kogi, Kaduna, Jigawa, Anambra, Ondo, and Kano States en who were not formally married (e.g., common law undertook efforts to reform and develop new land ad- unions) would not be required to produce a marriage ministration systems for SLTR under the GEMS3 pro- certificate to benefit from the program.73 gram (Growth and Employment in States) in a collab- oration between the Committee, the World Bank, the Findings from an ongoing study in Uganda also suggest United Kingdom Department for International Devel- that small incentives can help nudge households to opment (DFID), and state governments. While GEMS3 cotitle land in the names of both spouses.74 The study successfully established modern electronic land reg- indicates that providing a subsidized land title con- istration systems, raised revenue, and addressed land ditional on registering the wife’s name as a co-owner conflicts in four pilot states, it still did not meet its increased demand for cotitling by 50 percent without targets for registering parcels due to political and im- adversely affecting the overall demand for titling. Of- plementation challenges. SLTR also did not remedy fering an educational video also raised the demand the fact that women lack equal access to land regis- for cotitling by 25 percent. Both the conditional sub- tration and administration, as well as the benefits of sidy and informational video show how policy mak- registration.70 Lessons from the SLTR pilot states in ers can strengthen women’s land tenure by leveraging Nigeria suggest that recording secondary land rights simple, low-cost interventions to boost joint titling (including long-term leases and use rights for work, within land formalization programs. income, and household food security) could benefit women in land transactions with the state or private Some emerging evidence from Ethiopia and India sug- sector, given the predominance of customary tenure gests that broader legal and institutional reforms can and the fact that women disproportionately lack pri- also play a role in fostering greater land rights for wom- mary land rights.71 However, little evidence exists on en. In Ethiopia, for example, the 2000 Revised Family effectiveness and impacts of interventions aiming to Code sought to increase women’s legal rights to as- formalize secondary land rights. sets by requiring the equal division of assets between spouses upon divorce. Combined with the 2003 Land The evidence to date on land formalization programs Registration Act, which aimed to equalize men’s and in Sub-Saharan Africa suggests that, when carefully women’s inheritance and property rights, these legal designed and implemented, they can help women changes shifted perceptions and norms around the secure and benefit from their land rights.72 An impact division of assets upon divorce. In 1997, about 40 per- evaluation of Rwanda’s pilot land tenure regulariza- cent of women in Ethiopia expected that land would tion program, for example, shows that gender-sen- be divided equally between spouses during a no-fault sitive land formalization interventions can improve divorce, and by 2009, more than 80 percent of wom- women’s land tenure security and induce more effi- en reported expectations of equitable division of land cient investments in land. The program systematically upon divorce.75 Moreover, in India, evidence on the demarcated land parcels, registered joint ownership equal inheritance reform under the Hindu Succession of land for husbands and wives, and provided certifi- Act suggests some positive impacts on girls’ likelihood cates of land rights. Results from the assessment find to inherit land and educational attainment; however, that female-headed households who participated in these gains were not sufficient to close the gender gap the program increased their investment in land by 19 in land inheritance.76 percent. In the national scale-up of the pilot, the pro- gram design was further adapted to ensure that wom- 49 DEEP DIVE I In Nigeria, recent efforts to strengthen women’s land ty through interventions such as ranching options in- rights through legal reform have concentrated on pro- itiatives. Other programmatic efforts could focus on tecting widows’ use and ownership of land, given that fostering networks and group formation to strengthen widows are often evicted from their marital homes a common female voice for land rights, as well as in- following the death of a husband. A recent federal law, cluding programs that educate rural women on how to the Violence Against Persons (Prohibition) (VAPP) Act, access land within existing agricultural extension ser- seeks to protect women from harmful widowhood vices.79 Integrating education on land ownership and practices by criminally sanctioning them; yet, imple- titling into extension services could promote synergy mentation, enforcement, and awareness of the law between existing extension work and other agencies has reportedly been poor thus far.77 At the state level, to advance women’s empowerment and productivity.80 the VAPP Act has been passed in Edo, Federal Capital Territory (FCT), Kaduna, and Ogun States. In addition While sparse evidence exists on the impact of these to the federal and state-level VAPP Act, some states programs on women’s use and ownership of land, fur- have already enacted other laws to protect widows, ther research may seek to address this question. It is including the State of Anambra’s Malpractices Against important to identify to what extent such programs, in Widows and Widowers (Prohibition) Law (2005), which addition to structural modifications to Nigeria’s stat- prohibits the extended family from forcefully dis- utory system, improve women’s economic empower- placing a widow from her marital home or removing ment, which largely hinges upon the ability to acquire property that she acquired during the lifetime of her or purchase land in a way that systematically and sta- deceased husband. Cross River, Ekiti, Lagos, Osun, and bly recognizes rights. Oyo States have adopted similar domestic and gen- der-based violence laws, but much more could be done at the state and federal levels to enshrine and enforce protections for widows.78 Education and empowerment Strengthening women’s communities and educating women on opportunities to access land may also help improve women’s land tenure security. The Nigeria National Gender Action Plan proposes to grant land use rights to women in rural communities via local government councils to improve their ability to use and derive income from the land. The Nigeria Gen- der Policy in Agriculture includes targets to sensitize women and men at the community level on existing land laws to promote understanding about the impor- tance of women’s equal ownership and use of land. In addition, the National Livestock Transformation Plan recommends improving women’s land tenure securi- 50 DEEP DIVE II Social norms and customs play a strong role in per- petuating these disparities, often restricting women’s entitlements to livestock and instead automatical- ly assuming men’s ownership and property rights.86 Among the Fulani community, which holds a large proportion of the nation’s cattle herds, men assume Deep Dive II: de facto rights to all household animals.87 Such dispar- ities in asset ownership can severely undercut wom- Livestock en’s economic potential.88 A marked gender divide in Why does women’s ownership of ownership characterizes livestock livestock matter? production in Nigeria In Nigeria, the livestock sector makes up 6 to 8 per- Across Nigeria, and throughout the developing world, cent of the national GDP and approximately 13 million a marked gender divide characterizes livestock own- households have farm animals.89 Livestock is a par- ership. Though nationally representative sex-disaggre- ticularly useful and versatile asset, especially for poor gated data on livestock ownership are limited, availa- households: it can store wealth, serve as collateral, and ble data from the Nigeria General Household Survey buffer against shocks.90 Livestock is also highly profit- indicate that male-headed households in Nigeria own able: it has high reproductive potential, can be com- more animals than female-headed households.81 Fur- mercialized either as a whole or through its byprod- thermore, women in Nigeria more commonly own and ucts, and can provide complimentary inputs for crop control smaller and less valuable livestock (i.e., goats, production owing to its draught power and soil organ- sheep, poultry) relative to men, who own and control ic matter.91 Given its economic significance, livestock larger and more valuable livestock (i.e., cows, calves, can play an important role in shifting decision-making bulls, oxen).82 power between members of a household.92 Local studies corroborate these disparities in live- In addition to owning less livestock, women also own stock ownership. Research in Akwa Ibom State, South- different types of farm animals, which has conse- ern Nigeria, finds that in the majority of households, quences on the total value of their assets. In analyz- women owned poultry, men owned goats and sheep, ing asset growth over a twenty-year horizon in Kaduna and joint ownership of livestock was uncommon.83 A State, Dillon and Quiñones (2010) find that women’s study in Nasarawa State in Northern Nigeria indicates assets grew at a slower rate than men’s, a phenome- that women owned mostly chicken and only a small non that the authors partly accredit to women’s and proportion of large animals.84 In Abia State in South- men’s differential livestock holdings.93 Men who pri- eastern Nigeria, evidence suggests that 71.2 percent marily held livestock experienced large increases in of women have chickens, 16.2 percent have goats, and their asset values due to livestock reproduction, while 6.9 percent have sheep.85 women’s assets increased only marginally in value over time. Considering that different levels of assets 51 DEEP DIVE II accumulate at different rates, these inequalities can Women lack adequate financial resources to multiply over time, driving long-term wealth and pov- purchase livestock erty dynamics and perpetuating economic, social, and empowerment gaps between men and women.94 Women’s ownership of livestock is hindered by wom- en’s limited access to capital, which is critical for pur- Nigeria’s National Livestock Transformation Plan chasing and caring for livestock via feed, veterinary (NLTP) 2019–2028 acknowledges the central role of services, and other supplies.96 Only 45 percent of the livestock sector in achieving the country’s eco- women farmers in Nigeria acquired livestock through nomic and development objectives. However, the market means, and women more commonly obtain barriers facing women’s ownership and production of livestock through nonmarket means (i.e., inheritance livestock hinder the progress of the sector as a whole: or gifts) relative to men.97 This occurs not because the Food and Agriculture Organization (FAO) iden- women necessarily face barriers to markets, but rather tifies discrimination against women among the top because they face barriers to capital.98 three constraints for livestock production in Nigeria.95 Women have little to no decision-making power over earnings and resource allocation What are the barriers to women’s participation in profitable Though women are less likely to own larger and more valuable livestock, they still take part in man- livestock activities? aging them: broadly speaking, women often assume tasks like collecting and cutting feed, retrieving wa- Women’s ownership and production of livestock de- ter, and cleaning pens.99 Studies across both Northern pends on a number of factors, including financial and Southern Nigeria indicate that women generally resources, decision-making power, and use of exten- contribute more than men to managing vulnerable sion services. The NLTP affirms that key gender issues animals, feeding animals, cleaning barns, performing in the livestock sector include women’s lower levels dairy-related activities, transporting farm manure, and of capital and access to credit; cultural stereotypes selling milk.100 Likewise, women are engaged in ac- which discourage women from ranching; and power tivities such as selection of livestock and poultry for dynamics which restrict women to the role of labor- breeding, and treatment of sick animals.101 Among the er, as opposed to manager, within the sector. Similarly, settled Fulani in Nigeria, women process, manage, and Nigeria’s Gender Action Plan recognizes that gender market milk.102 norms around division of labor and value chain ac- tivities tend to affect women by restricting them to Typically, however, the degree of decision-making less remunerative value chains or activities along the power that a woman has with respect to the income chain. The following sections outline the evidence on generated by livestock production is not commensu- several underlying constraints to women’s participa- rate with the time she allocates to the income-gen- tion in the livestock sector, as well as potential solu- erating activities.103 Despite women’s contributions tions for addressing important gender disparities. to livestock management, they have little to no de- cision-making power over the allocation of labor, productive resources, or earnings from livestock pro- 52 DEEP DIVE II duction due to unequal power relations within the Closing gender gaps in the household.104 livestock sector: Emerging evidence suggests that cash and Women lack access to relevant and gender-sensitive livestock extension services livestock transfer programs can support women’s ownership and Extension services play an important role in dissemi- nating information on effective livestock techniques production of livestock and markets, reducing asymmetries and transaction To address these constraints, the NLTP proposes a costs, and making livestock activities more profitable number of activities targeted specifically to women, and effective.105 However, women face greater barri- including provision of access to finance for women; ers to extension services and information than men.106 land tenure system reforms to improve women’s ac- While women are offered training in livestock produc- cess to land; agricultural extension services custom- tion in the North East, North West, and North Central ized and targeted to women; credits, land acquisition, zones, women’s access to training is limited in other and vocational skills training to increase women’s par- regions.107 Women who have access to extension ser- ticipation in ranching; and strengthening women and vices may not participate in them, while women who girls’ rights, especially as it relates to gender-based vi- participate in them may not reap benefits compara- olence (GBV). Moreover, Nigeria’s Gender Action Plan ble to those of their male counterparts. One study recommends activities to train farmers on value chain cites poor targeting of women as a challenge to ag- diversification, which could include providing train- ricultural extension in Nigeria, while another argues ing to women farmers on more lucrative and typically that extension agents should put more emphasis on male-dominated value chains such as ranching. We training rural women in livestock and poultry man- review here the emerging evidence of cash and live- agement strategies.108 stock transfer programs’ impact on women’s owner- ship and production of livestock. The Nigeria Gender Policy in Agriculture (NGPA) 2016 is aimed at addressing the gender gap in access to ex- tension services, particularly with its fifth objective, Financial resources could help foster “mainstreaming gender in extension services and im- empowerment within livestock production proving gender responsiveness in the delivery of agri- cultural services.” Under this objective, the NGPA pro- Expanding women’s access to capital could open up poses targeting the following areas for policy action: opportunities to purchase livestock and inputs like increasing the number of female extension agents, feed and veterinary services, leading to greater au- easing women’s time constraints due to domestic re- tonomy, decision-making power, and profitability. sponsibilities, and improving adult literacy. However, Findings from a study in Kebbi State, which delivered adoption and implementation of the federal policy at unconditional cash transfers to primary female de- the state level is of critical importance, given that the cision-makers in 1,200 households, show promising state is responsible for the execution and delivery of results. The households that received cash transfers extension services. had one and a half times more animal stock and 30 percent greater household asset value than house- 53 DEEP DIVE II holds that did not receive the transfers. Moreover, Outside of Africa, an assessment of a livestock transfer the female recipients were 14 percent more likely to and training program targeting women in Nepal found be economically active and earned 80 percent more that in the short run, beneficiaries’ financial inclusion profits from their businesses.109 However, there was no and empowerment increased significantly. Further- conclusive evidence on the effects of the transfers on more, randomized experiments in Bangladesh reveal women’s decision-making power or women’s individ- that a large-scale ultra-poor program boosted wom- ual ownership of livestock. Further research may seek en’s earnings, and the asset transfer program, which to answer such questions. provided mainly livestock and targeted women, led to significant increases in longer-term household land and livestock holdings, as well as enhanced savings Livestock transfers for the ultra-poor and sustained earnings outcomes.113 However, the pro- or graduation programs can increase women’s gram may have also diminished women’s property earnings and livestock holdings rights over household resources, mobility, and deci- sion-making power.114 Overall, more research is need- Interventions like the National Fadama development ed to understand the isolated impact of various com- project, and other humanitarian programs in the ponents of these programs on women’s access to and North-Eastern states of Nigeria, for example, provide earnings from livestock, given that implementing live- livestock to rural households and to women in par- stock programs without considering context-specific ticular throughout Nigeria. In Sub-Saharan Africa and gender dynamics may compromise benefits or even other regions of the developing world, many programs negatively affect women recipients.115 for the ultra-poor contain elements of livestock trans- fer in combination with training, temporary cash or in- kind grants, savings support, and health and life skills information, though no study to date has identified the particular effect of livestock transfers within such programs.110 For example, a study attempting to cap- ture the household-level effects of a livestock transfer program in Rwanda was unable to identify the caus- al impact.111 Still, some promising evidence is emerg- ing: in Ghana, a program targeting women combined a productive asset transfer (livestock), consumption support, technical skills training, home visits, health education and services, and savings support through deposit collection services. Households of benefi- ciaries experienced a significant boost in livestock revenue, earning 50 percent more than a comparison group, in addition to a 91 percent increase in nonfarm income one year after the program ended.112 54 DEEP DIVE III As described in the wage decomposition section, women wage earners consistently work in lower-paid sectors. Women in Nigeria are more than twice as likely to work in education than men, despite jobs in educa- tion paying significantly lower wages. Similarly, women are significantly less likely to work in manufacturing, despite the sector yielding higher wages than working Deep Dive III: in either service or education jobs. Another example of occupational segregation are the systematic gen- Occupational der differences in crop choice and livestock owner- ship. Women are about half as likely to plant cereals Segregation and more than twice as likely to farm roots and tubers, vegetables, and melons, despite higher returns to ce- reals. Similarly, women in Nigeria generally own and control smaller, less valuable livestock, such as chick- Relative to men, women in Nigeria ens, while men own and control larger, more valuable tend to concentrate in both lower- livestock, such as cows and oxen. profit sectors and lower-profit This deep dive outlines evidence on, implications of, positions within sectors and potential solutions to address occupational seg- regation in Nigeria. As evidence from rigorous impact Occupational segregation is the unequal distribution evaluations, which yields important insights on effec- of men and women both across sectors, typically tive policies, is limited in the sphere of occupational resulting in a high concentration of women in low- segregation in Nigeria, this section will occasionally paid or low-profit sectors, and within sectors, lead- draw on relevant regional and international findings. ing women to occupy lower-ranked and less lucra- tive positions within sectors.116 Data from the General Household Survey reveal widespread occupational segregation in Nigeria both across and within sectors, Why does occupational including in self-employment, wage work, and agricul- ture. For example, self-employed women in Nigeria segregation matter? are less than half as likely to work in a service sector This sectoral imbalance perpetuates the gender earn- and over 60 percent more likely to work in a trading ings gap. In the decomposition analysis of part II, we sector. Similarly, women are 4 percentage points more find that occupational segregation accounts for a sig- likely than men to sell to final consumers, indicating a nificant portion of the gender wage and agricultural higher concentration in downstream enterprise activ- productivity gap. As mentioned in the deep dive on ities. This sectoral and value chain segregation yields livestock, the different types of livestock owned by lower self-employment profits for women, as average women and men lead to long-term gaps in the value profits in trading are lower than average profits in ser- of their respective assets. vices, and profits from selling to final consumers are significantly lower than higher value chain operations. 55 DEEP DIVE III Occupational segregation also has further-reach- External influences from teachers, parents, and class- ing consequences in terms of overall productivi- mates further diminish girls’ motivation to learn a non- ty and earnings in the country. The misallocation of traditional trade. For example, differential treatment high-ability women into low-return occupations may of boys and girls in science, technology, engineering, reduce economic growth through suboptimal alloca- and mathematics (STEM) subjects in school due to an tion of labor across the sectors of the economy and overarching belief that STEM subjects are more “male” segments of the value chain. For instance, decreased than “female” decreases girls’ confidence and perpet- race and gender segregation in the United States’ la- uates the notion that girls lack the ability and the grit bor market is estimated to be responsible for 20–40 to succeed in STEM, a male-dominated area. percent of the increase in per capita output between 1960 and 2010.117 Meanwhile, a study using a model to As in many other countries, women in Nigeria spend quantify the effects of gender gaps in the labor market a disproportionate amount of time on domestic re- on average income by region finds substantial costs sponsibilities compared to men.121 Unequal sharing associated with these gaps: it estimates that regional of childcare and domestic work between female and long-run GDP losses range from 10.1 percent in Cen- male household members impedes women’s ability tral Asia, to 37.8 percent in the Middle East and North to work long hours and limits their choice of occupa- Africa, with losses in Sub-Saharan Africa amounting to tion. For example, women are more likely to choose 12.0 percent.118 jobs that allow part-time work or more flexibility to fulfill their care and domestic responsibilities.122 The interaction of endowments, self-selection, time Why is there occupational constraints, and external influences means that there segregation? are multiple levels and stages in a woman’s occupa- tional trajectory when an intervention may be help- Occupational segregation stems from girls and wom- ful in influencing her choice of sector. However, more en self-selecting into gender-typical occupation- research is needed to determine the relative impor- al paths, as well as social pressure that discourages tance of these factors and to identify effective inter- them—explicitly or implicitly—from choosing to pur- ventions that broaden women’s occupational choice. sue a non-gender-typical path. The self-selection and external, societal influence intersect with endowment differences, such as lower skills or capital, to narrow women’s occupational choices. Women and girls may self-select out of male-domi- nated trades due to perceptions that they are diffi- cult, uninteresting, unfeminine, or unappealing in the opportunities they offer.119 A girl’s decision to work in a predominantly male occupation—for example, as a mechanic—may be seen as weakening her identity as a girl and making her less feminine.120 56 DEEP DIVE III Expanding women’s occupational An impact evaluation of a joint World Bank and Gov- ernment of Nigeria program has demonstrated that choices: Improving access to role training programs can positively influence women’s models, social networks, and sectoral choice and ambitions, potentially filling en- couragement gaps from parents and educators. The information for women and girls, ACCESS (Assessment of Core Competency for Em- as well as protecting them from ployability in the Service Sector) program provided recent university graduates with 85 hours of training gender-based violence play a role to equip them with sufficient communication, com- puter, and cognitive skills to work in Nigeria’s ICT sec- tor. Two years after the program ended, participants Education and training programs were 26 percent more likely to work in the ICT sector than nonparticipants. The training was particularly ef- Girls in Nigeria are less likely to complete a basic ed- fective for female participants with a preprogram bias ucation despite having an education policy that guar- against thinking of women as professionals: yielding antees free access to six years of primary school and an impact three times as large for women with pro- three years of secondary school. In 2011, for example, male bias. Ultimately, ACCESS suggests that training the ratio of female to male students in primary school for specific sectors can shift employment and sectoral and secondary school was 0.91 and 0.89, respectively.123 outcomes, providing promising grounds for interven- The lower educational attainment of girls limits their tions supporting women in crossing over to nontradi- access to jobs and occupations that require high- tional sectors.128 er levels of human capital, constraining their choice even before entering the labor market. Role models The girls who do attend school typically pursue STEM education at lower rates than boys. Gender gaps in in- Role models play an important part in transferring terest in STEM subjects emerge during primary school knowledge, facilitating networks, and shifting percep- despite science and math both representing compul- tions about the breadth of women’s opportunities in sory components of the primary school curriculum.124 the workplace and economy, as well as the returns to Studies in Nigeria show that boys are given more time jobs in traditionally male sectors. Increasing girls’ ex- for a task in a science classroom and more opportuni- posure to successful men and women in male-dom- ties to ask and answer questions, use equipment and inated sectors strengthens the extent to which they learning materials, and lead groups than girls.125 Dispar- can see themselves in such sectors and boosts their ities in STEM engagement and participation in Nigeria motivation to pursue similar careers.129 tend to widen as students transition to higher levels of education, resulting in a phenomenon termed “the Evidence from India indicates that women role mod- leaky pipeline,” or the increasing absence of women els in the classroom, community, and workplace have in STEM as education level increases.126 Data from 12 positive effects on shifting norms on what is “men’s states in Nigeria from 1998 to 2002 indicate that the work” or not, and, in turn, improving girls’ and women’s ratio of men to women enrolled in STEM university labor market outcomes.130 For example, the presence courses remained fairly constant at 3:1.127 of a female leader in a village significantly increased 57 DEEP DIVE III girls’ aspirations and parents’ aspirations for their sons during job search, they do not for daughters.135 In daughters. Moreover, evidence from the United States Malawi, men are less likely to refer women for jobs, shows that female students performed better on im- regardless of how qualified and numerous the women plicit attitude scores for math when studying under a in their networks are.136 female professor.131 However, when women do have access to networks in However, female representation in the classroom,X male-dominated spheres, they may be more likely to community, and workplace is low in Nigeria. In the cross over. Studies in Uganda and Ethiopia find that classroom, fewer women than men choose to teach early exposure to a male role model plays a key role STEM subjects at the postprimary level.132 In the com- in a woman’s decision to enter a male-dominated sec- munity and workplace, women held 7.5 percent of tor: Ugandan women with a male role model are 12 to national assembly seats, 5.6 percent of state assem- 22 percent more likely to cross over to predominantly bly seats, and 4.4 of local government chairperson male sectors.137 Qualitative research indicates that a positions in 2015, while only 37.4 percent of lawyers father’s occupation may influence a woman’s decision and 26.2 percent of judges were women.133 This lack to enter a male-dominated sector, and a study from of representation limits not only the extent to which Ethiopia finds that women who take technical and girls can see themselves in certain paths, but also the vocational education and training do so largely due knowledge they have to pursue those paths. to their social networks, which provide them not only with exposure to the field but also information on its earning potential.138 Social networks Social networks play a critical role in occupational Lack of information outcomes, as they help women to identify and find opportunities, understand earning potential, build Exposure to role models and networks may provide relationships, and increase social capital. In the la- women with accurate information on the returns to bor market, imperfect information often forces em- work in traditionally male sectors and thus boost ployers to rely heavily on social networks and word preferences to enter these sectors. In Ethiopia, wom- of mouth to fill positions.134 Women are often more en in female-dominated sectors incorrectly believed network-constrained than men, given lower levels of that they earned the same or higher than their female education, gender norms regarding contact with the counterparts in male-dominated sectors.139 Research opposite sex, and limited mobility. There is limited re- in Kenya has demonstrated that providing women with search on how networks may support or limit wom- information on expected earnings in male-dominated en’s job prospects. For example, while fathers in South sectors can increase interest in pursuing a male-dom- Africa serve as useful network connections for their inated trade.140 A study in Uganda found that accurate information can address misconceptions of earnings in traditionally male and female sectors and assist X Female representation in the classroom is low generally and for STEM subjects in particular as well. World Bank (2019) notes that women in making better-informed sectoral decisions.141 48 percent of primary education teachers in Nigeria were women Information-based interventions are a promising path in 2010 and Ekine and Abay (2013) note that women teachers are often concentrated at the primary school level, suggesting an for encouraging women to choose nontraditional sec- even lower proportion of women educators at the secondary and tors. university level. 58 DEEP DIVE III Safety and gender-based violence (GBV) Gender-based violence (GBV) can affect women’s occupational outcomes through several channels in- cluding violence in the workplace, on the way to the workplace, or because employment-induced changes in empowerment can lead to increased intimate-part- ner violence. While nationally representative statistics on the prevalence of violence in the workplace do not yet exist, smaller-scale surveys indicate that women’s experiences with and perceived risk of harassment by men at work influence their occupational preferenc- es and outcomes.142 Over two-thirds of young female apprentices interviewed in Ibadan, Nigeria reported experiencing physical violence and 40 percent cit- ed their instructor as the perpetrator.143 Qualitative research in the states of Katsina, Edo, and Taraba re- vealed that women in male-dominated occupations (i.e., quarry work, shoe and bag making, farming, and cloth dyeing) are especially vulnerable to harassment and violence. Some female workers suggested that men may feel threatened by successful women and subsequently turn to harassment to disparage and discourage them from participating.144 These dynam- ics not only directly affect victims but also create an overarching environment perceived as unwelcom- ing to women. In fact, qualitative assessments reveal that young women entering job training programs in Liberia and Rwanda prefer self-employment or fe- male-dominated industries due to the threat of sex- ual harassment.145 It is important to note that concerns for safety may also be used as pretext to restrict women’s freedom. A recent review notes that “concern for women’s safety […] is often partly real and partly the expression of a patriarchal norm” and can be used to limit women’s interactions with men that are not unsafe but deemed inappropriate by traditional norms.146 59 Part IV What Are the Costs of Economic Gender Gaps in Nigeria? When women are less productive on their farms and earn less from their businesses or wages, the Nigerian economy misses out on the valuable contributions of nearly half of its labor force. As highlighted in the preceding sections, a number of gender-specific constraints underlie gaps in economic Costing the gender outcomes between men and women in Nigeria. These constraints not only prevent Nigerian women from gap in agriculture, reaching their full potential, but also represent a vast drain on the economy. When women are less produc- entrepreneurship, and tive on their farms and earn less from their businesses or wages, the Nigerian economy misses out on the val- wage employment uable contributions of nearly half of its labor force. This To cost the gender gap in agricultural productivity, we section estimates Nigeria’s forgone income in terms of calculate the total agricultural output that would be gross domestic product (GDP) from the gender gaps in obtained in the absence of the gap, that is, if women agriculture, entrepreneurship, and wage employment produced as much per hectare as men do. This fig- in Nigeria. The following subsections summarize the ure is then extrapolated at the macroeconomic level methodology used to estimate the cost of the gaps using available figures on total arable land in Nigeria, and present the results. A detailed description of the the share of crop production in the national agricul- methodology, steps, and assumptions is provided in tural GDP, and the share of agriculture in total GDP, to appendix 3. come up with a figure of forgone earnings due to the gap. Lastly, we apply a GDP multiplier to that figure to account for the existence of multiplier effect between sectors of the economy, and estimate the total cost in GDP due to the gap. To cost the gender gap in self-employment, we esti- mate the total national value of profits in the absence of the gap by multiplying the number of women self-employed in Nigeria by the average level of profit obtained by self-employed men. The difference be- tween this figure and the total aggregate profit in the presence of the gap represents the amount of forgone 61 earnings due to the gender gap. We then apply GDP Results multipliers, defined for seven subsectors of the econ- omy, to estimate a total cost in GDP caused by the The total estimated forgone earnings based on gen- gender gap. The GDP cost of the gender gap in wage der gaps in agricultural productivity, firm profits, and income is analogously obtained, by comparing the wage earnings amount to US$9.3 billion, or 2.3 per- total estimated wages with and without the gap and cent of overall GDP. This is a lower bound estimate, applying subsector multipliers. which corresponds to the amount of additional earn- ings generated if every woman’s productivity, profits, It is important to note that this partial-equilibrium or wage level was equalized with those of men. Using approach abstracts from subsequent changes in the GDP multipliers, which take into account potential input, output, and labor markets. For example, in the spillover effects across sectors of the economy, the short run, a large increase in agricultural input use by costs could actually be as high as 5.8 percent of over- women farmers may drive up the cost of those in- all GDP, or US$22.9 billion. puts. Similarly, a large increase in agricultural output driven by increased productivity could depress ag- Each component of the economic gender gaps and its ricultural prices. While these short-run equilibrium corresponding cost contributes to the overall amount effects would likely ease over time, they suggest that of forgone GDP as indicated in figure 10. the partial equilibrium approach may overestimate the potential GDP gains from closing the gender gap. Although the costs of closing these gaps—in terms of However, recent research (discussed in the deep dive investments or reallocation of resources—could be section) highlights the large and sustained dynamic sizable, the US$9.3 billion of forgone GDP is immense, impacts on growth that result from more efficiently signaling that there is an important economic oppor- allocating labor across the economy, likely overshad- tunity to be gained from closing the gender gaps. Yet, owing impacts in the input and output markets.147 to narrow these gaps, policy makers and their devel- opment partners need not only a detailed account of the key drivers of these gaps and the economic costs associated with them, but also rigorous evidence on effective policy options. The following section for- mulates policy recommendations for Nigerian policy makers to address the underlying constraints facing women farmers, entrepreneurs, and workers. 62 FIGURE 10 Cost of the gender gaps $22.9 billion (5.8% of GDP) The unconditional gender gap in agricultural pro- ductivity is estimated to be 30 percent. The total forgone earnings from the unconditional gender gap equate to 2.7 percent of agricultural GDP and 0.6 percent of overall GDP (or about US$2.3 billion). Using GDP multipliers, closing the gap in agriculture could represent a total estimated in- 8.1 crease of up to 2.0 percent of GDP (or approxi- mately US$8.1 billion). The cost of the unconditional gender gap in prof- its between men- and women-owned firms (66 Billion USD percent) in terms of total GDP similarly equates to 1.6 percent of total GDP (or about US$6.2 bil- lion). Using GDP multipliers, closing the gap in firm $9.3 billion profits could represent a total estimated increase (2.3% of GDP) of up to 3.3 percent of GDP (or approximately US$13.2 billion). 2.3 13.2 The cost of an unconditional gender gap in wages of 22 percent equates to 0.2 percent of total GDP 6.2 (about US$780 million). Using GDP multipliers, closing the gap in wages could represent a total estimated increase of up to 0.4 percent of GDP (or approximately US$1.6 billion). 1.6 0.8 At least With multipliers Agricultural productivity gap Self-employement profits gap Wage earnings gap Source: Nigeria General Household Survey data. 63 Part V Policy Priorities for Closing Gender Gaps in Economic Outcomes in Nigeria Without equal access to health, education, and reduced fertility, gender equality will likely be out of reach. However, the interventions proposed in part V are likely to contribute to closing existing gaps in economic outcomes, including for women later in the life cycle, which can have snowball effects on fertility, investment in the human capital of children, and overall economic growth. As Nigeria faces the immediate challenge of stimulat- ic gains for Nigerian women. It is important to note ing economic recovery, it also has the opportunity to that the policy options laid out in this section must address the sizable gender gaps that undermine wom- be complementary to a comprehensive approach to en’s economic empowerment and hinder inclusive closing gaps in human capital. Without equal access economic growth. As part II shows, women in Nige- to health, education, and reduced fertility, gender ria earn significantly lower incomes than men across equality will likely be out of reach. However, the inter- a range of economic activities, constraining national ventions proposed in part V are likely to contribute to income. Women in agriculture, self-employment, and closing existing gaps in economic outcomes, includ- wage employment earn 30 percent, 66 percent, and 22 ing for women later in the life cycle, which can have percent less, respectively, than men in those sectors. snowball effects on fertility, investment in the human These gender gaps in productivity, profits, and wages capital of children, and overall economic growth. burden the economy: the analysis in part IV indicates that closing these gender gaps could yield an increase The evidence review that follows (and is summarized in GDP of up to 6 percent. in table 6) includes both experimental and quasi-ex- perimental studies if they used a reliable method of This report identifies the key constraints limiting wom- identifying a counterfactual. While we prioritized re- en’s earnings in each sector: in agriculture, they are sults of studies conducted in Nigeria, we also highlight crop choice, access to inputs, and access to hired labor relevant research from other countries in Sub-Saharan for women farmers; in self-employment, they are capi- Africa. Notably, this section highlights that there is a tal stock and position in the value chain; and in wage dearth of rigorous evidence on what works to close employment, it is sectoral segregation for female gender gaps in Nigeria. It is important to note, how- wage earners. Each of these six constraints informs a ever, that there are large gaps in the global evidence policy priority that is backed by research and could base on effective gender interventions as well. There- help policy makers narrow gender gaps in earnings. fore, an opportunity exists in Nigeria to generate new This section reviews the existing rigorous evidence on evidence that would not only expand the knowledge how to address the six key drivers of gender gaps in ag- base at the national level, but would also help ad- ricultural productivity, firm profits, and wage earnings, vance the evidence-based policy debates on gender as identified in part II. An additional cross-cutting pol- equality in Africa and globally. icy area addresses the care and time constraints that women face and that limit their participation in the To conclude, this section presents broader lessons on labor market. Adopting an evidence-driven approach policies and programs that can help close gender gaps to policy and program design in each of these seven and offers guidance to Nigerian leaders on incorporat- priority areas can help ease these constraints, close ing gender innovation in policy making, programming, the economic gender gaps, and accelerate econom- and research. 65 TABLE 6 Note: The categories referenced in the “State of the evidence” column in table 6 are defined as follows: credible indicates that more than one impact evaluation from Sub-Saharan Africa demonstrates consistent, positive impacts of an intervention; emerging indicates that just one impact evaluation (from Sub-Saharan Africa or another developing context) shows positive impacts or multiple impact evaluations show mixed or not exclusively positive results; frontier indicates that there are no impact evaluations showing strong positive impacts, but other non experimental What works to close gender gaps in economic outcomes in Nigeria? evidence suggests that the intervention could address the given constraint; and not promising indicates that at least one impact evaluation shows no or negative impacts of an intervention. A review of the evidence Policy priority/ Policy Main Evidence State of the Policy priority/ Policy Main Evidence State of the constraint option conclusions from Nigeria evidence constraint option conclusions from Nigeria evidence Providing farmers with subsidized, improved seed varieties Subsidies for In-kind grants to female entrepreneurs can increase can increase crop yields and incomes for both women and improved seed Yes Emerging business profits, in some cases more effectively than cash men farmers, but may not necessarily result in gains for In-kind grants No Frontier varieties grants, given the pressure women face in redistributing cash women without specific efforts to target them. to household expenses. Engaging men to Engaging men and women through couples’ trainings can change norms encourage women's adoption and production of more No Frontier Large cash grants—provided to growth-oriented female around gendered valuable crops, typically farmed by men. entrepreneurs in the context of business plan competitions, crops Large grants Yes Credible for example—increase the likelihood of firm survival and Unlocking firm boost sales and profits of women-owned firms. Availability of Health insurance combined with weather index insurance owners’ access to Promoting women insurance products has the potential to increase women’s willingness to switch growth capital farmers’ choice of No Frontier higher-value crops suited to women to more valuable crops by decreasing the risk farmers they face. Secure savings mechanisms enable female Secure savings microentrepreneurs to set aside earnings and increase No Credible mechanisms investment in their businesses. Trainings to boost women’s socioemotional skills could Socioemotional skills facilitate their adoption of cash crops, given the strong No Frontier training correlation between noncognitive skills and take-up of more valuable crops, particularly in patriarchal societies. Programming that merges skills training and social Social network network building can increase earnings among female No Emerging building entrepreneurs, although the impacts of these programs tend to dissipate after they end. Employing more female extension service agents can increase women farmers’ use of productivity-enhancing Female extension technologies, since women in certain settings are more No Emerging service agents likely to adopt new technologies from female extension agents. Promoting women’s Psychology-based entrepreneurial training increases engagement in greater Psychology-based firm profits for both male and female entrepreneurs at Promoting women No Credible value addition business training significantly higher rates than traditional training, with farmers’ choice of women registering even greater gains than men. higher-value crops Local language voice and video agricultural extension and enhancing women Digital technology for messaging can increase women farmer’s participation No Emerging farmers’ use of farm agricultural extension in agricultural decision-making, boost their cultivation inputs of cash crops, and improve production outcomes. Job training in Job training in professional, male-dominated sectors can higher-return male- increase the likelihood of women entering the sector while Yes Emerging dominated sectors also changing gender-biased perceptions of the sector. Provision of subsidized fertilizer improves women Subsidies for inputs/ farmers’ fertilizer use and increases agricultural output, Yes Credible fertilizer especially when women farmers are specifically targeted. Enhancing women farmers’ use Access to accurate information regarding earnings in Information about male-dominated sectors can increase women’s interest in of farm inputs No Emerging earnings entering the sector, but may not sustain their participation in the sector, potentially due to challenging gender norms. Subsidies to While mechanization of agriculture has been shown to Decreasing mechanize farm benefit women relatively more than men, gender roles No Emerging labor can constrain women’s access to and use of technology. occupational segregation Access to subsidized daycare can help mothers shift from Childcare services lower-wage jobs with more flexible hours to higher-paying No Emerging jobs with fixed hours. Facilitating access Provision of cash transfers to households with children to farm labor and under five can increase spending on hired labor, increase mechanization Cash transfers No Emerging agricultural output, and decrease the time women spend on farm activities. Programs that Interventions aiming to expand women’s social networks build networks, or introduce them to role models and mentors could have No Frontier role modelling, and positive effects on shifting gender norms around women’s mentorship sectoral choice. Microcredit does not serve as sufficient capital to significantly impact women’s business outcomes; however, Microcredit Yes Not promising it has demonstrated other important impacts on women’s vulnerability, empowerment, and labor force participation. Enrolling young children in preschool can significantly decrease the number of hours caregivers, mostly women, Childcare services spend on childcare and increase the likelihood of women No Credible working outside of the home as well as increasing the Mesocredit/mid- Mid-sized loans for growth-oriented women-owned firms likelihood of women engaging in higher-paid work. No Emerging Unlocking firm sized loans can accelerate business growth. owners’ access to growth capital Easing women’s Gender norms and behavior training for men and young time constraints Engaging men Cash transfers and grants delivered within holistic couples can increase men’s participation in housework, to participate in No Emerging Cash transfers productive packages boost ultra-poor women’s business No Credible although women’s time spent doing housework does not housework start-ups and revenue. necessarily decrease as a result. 66 67 adequate in targeting and addressing the specific needs of women farmers.152 The evidence on exten- sion interventions that work for women farmers is ex- tremely scarce. One study in Mozambique shows that women farmers were more likely to learn and adopt a Policy Priority 1: Promoting Women Farmers’ technology when a female extension agent was added Choice of Higher-Value Crops to the extension system.153 Another study from Malawi suggests that although women extension workers learn In both the North and the South, women consistently information about new agricultural technologies just as farm less valuable roots and tuber crops, holding back well as their male counterparts, they were not as suc- their agricultural productivity relative to men farmers’. cessful at teaching or convincing others to adopt a new In Nigeria, the existing evidence on interventions that technology. However, this disparity disappeared when work to increase women’s adoption of higher-yield farmers had more accurate information about the skills crops is extremely thin. One experimental study finds of the female workers, suggesting that gender discrim- that subsidies for improved seeds aimed at boosting ination in the perception of skills is underlying the ef- rice production had positive impacts on household fect.154 More recent innovations using digital technolo- welfare outcomes, but these were significantly larg- gy for extension messaging are also promising policy er for male-headed households than female-headed options, as detailed in policy priority 2. households.148 Other studies combining input pro- vision with information suggest that this may be an More research is needed on other frontier areas that effective way to promote adoption. In Benin, an in- facilitate women’s adoption of more profitable crops. tervention providing information and inputs to grow For example, gender differences in crop choice may Nerica, a new rice variety, led to higher yields and in- also stem from gender differences in risk preferences, come for women farmers.149 Studies from Côte d’Ivo- skills, or norms around certain cash crops being per- ire and the Gambia find that the provision of seeds ceived as “male crops.”155 An ongoing study in Ugan- and information on Nerica had an equal impact on daXIV shows that engaging men through a coopera- adoption for female and male farmers.XI, 150 In Uganda, tion-based couples’ training and encouraging them the provision of resources through subsidized input to transfer or register new out-grower contracts to packages along with extension services was sufficient their wives led to an increase in women’s participa- to induce the majority of women farmers to cultivate tion in cash crop value chains.156 This suggests that and consume a biofortified orange-fleshed sweet po- couples-based interventions may be a promising ap- tato crop.XII, 151 proach to support norm changes around crop choice. Skills might also influence women’s decisions around While information provision may be effective at en- cash crop farming: in Malawi, women farmers’ higher couraging women to switch to higher-value crops noncognitive abilities are linked with higher produc- when combined with inputs, the lower impact of tion of cash crops.157 agricultural extension services on women farmers is well documented.XIII Extension services are often in- Developing women farmers’ socioemotional skills, XI It is not clear, however, whether the studies were sufficiently through psychology-based trainings for example, may powered to detect gender differences in impact. be another promising avenue for research, although XII While the orange-fleshed sweet potato is not considered a “cash no evidence to date exists on the impact of such in- crop,” this study highlights that women’s choice of crops can be shifted through inputs and information provision. XIII Although extension services did not stand out as a priority driver of the gender gap in agriculture at the national level, we found where the gap in access to extension services is as high as 17 that it was a significant driver of the gap in the North (see box percentage points. 8). Enhancing women’s access to extension services should be XIV Impacts on economic outcomes including productivity and considered a policy priority for the North of Nigeria specifically, earnings are forthcoming. 68 terventions on women farmer’s crop choice. Women grams where women farmers have not been explicitly may also opt not to grow cash crops based on the per- targeted indicate less promising results.162 ception that they are riskier crops due to the large up- front investments needed to produce at scale and the Since subsidies are difficult to scale back once start- exposure to price fluctuations in the market. Further ed, deepening the knowledge on alternative, more testing of insurance products that address both covar- nimble policy options for enhancing women’s use of iate and idiosyncratic shocks (e.g., related to maternal farm inputs is critical. Agricultural extension services health risks or childcare) could help inform wheth- that are targeted and designed to meet the needs of er insurance products adapted to women’s specific women farmers have the potential to help ease wom- needs could decrease the risks they face and encour- en farmers’ access to inputs (see policy priority 1 for a age them to switch to more valuable crops.158 summary of the emerging evidence on agricultural ex- tension services). Digital technology to diffuse agricul- tural extension information—about input prices and available suppliers, for example—could be a promis- ing option. Some evidence suggests that digital tech- nology has the ability to increase farmers’ access to information and services. Educational voice messages in local languages for farmers in northern Ghana have Policy Priority 2: Enhancing Women led to increases in farmers’ yields up to 55 percent.163 Farmers’ Use of Farm Inputs In Uganda, video-enabled agricultural extension mes- saging increased women farmers’ participation in ag- Women farmers’ productivity is limited by their lower ricultural decision-making, adoption of practices and use of inputs—particularly fertilizer and herbicides— inputs, and improved their production outcomes.164 on the plot. Lifting barriers to access and use of inputs Improved access to mobile technology, paired with is crucial to closing the gap in agricultural productivity. an education program, led to women planting more Evidence on what works to increase women’s use of cash crops and households diversifying their crops in inputs is still emerging, with only one study provid- Niger.165 Rigorous evidence on impacts of such policy ing evidence on an intervention that increases ferti- programs for women in Nigeria is still needed. lizer use in Nigeria. A quasi-experimental evaluation of the Growth Enhancement Support Scheme—which included subsidized fertilizer—found broad positive impacts, including increased fertilizer use, output, in- come, and expenditures.159 Impacts on income and the rate of diversification were higher for female-headed households. Policy Priority 3: Facilitating Access to Farm Since a critical underlying constraint to women’s ac- Labor and Mechanization cess to inputs is financing to purchase inputs, a num- ber of development initiatives offer seasonal financing While women farmers appear to use as much labor as to women farmers or provide them with subsidized their male counterparts, their earnings are stifled by inputs.160 An impact evaluation in Mali suggests that the lower productivity of the labor they use, namely providing women with free fertilizer improved their the male labor (hired or family, depending on the re- use of fertilizer and complementary inputs, including gion). The evidence on programs that increase wom- hired labor, and led to an increase in agricultural out- en farmers’ access to high-quality agricultural labor put.161 However, impacts from fertilizer subsidy pro- is sparse, and there is currently none from Nigeria. 69 Financial constraints may prevent women farmers from hiring more productive labor.166 In Zambia, an evaluation of cash transfers given to households with children under 5 showed that spending on hired la- bor increased by four times.167 Input provision could Policy Priority 4: Unlocking Firm Owners’ also be a promising intervention for enhancing wom- Access to Growth Capital en farmers’ labor use. 168 The aforementioned study of fertilizer distribution in Mali found that women who Nigerian women entrepreneurs are constrained by received the fertilizer increased their use of other their lower capital endowment: they operate firms complementary inputs, including hired labor. More with significantly less capital than firms operated by research is needed to understand the effectiveness men. This lowers their productivity but also limits their of interventions that promote women’s use of hired ability to purchase inputs, invest in new activities, and labor and that increase the productivity of hired labor. move up higher in the value chain. Several types of in- terventions may be considered to unlock women’s ac- Access to machinery that decreases time and labor cess to capital: facilitating access to loans, cash trans- requirements on the farm could greatly aid women in fers, and cash grants, and increasing women’s savings filling the productivity gap between themselves and by providing secure savings mechanisms. male farmers, yet women farmers often lack financ- ing to purchase valuable tools. One solution would be A study from Nigeria finds positive impacts from ac- to provide financial assistance through cash vouchers cess to small amounts of credit: an impact evaluation or in-kind transfers for the purchase of machinery. A of a rural microcredit scheme in Ekiti state found pos- study in Zambia found that women benefit more from itive impacts on labor supply, income, savings, and ex- mechanization than men by saving time which can penditure.XV, 172 Meanwhile, a recent World Bank report, be used in off-farm work.169 Mixed evidence was pro- which draws on rich survey data from 14 countries in duced from a study on mechanizing forage chopping Sub-Saharan Africa and impact evaluation evidence in Tanzania. The traditionally female role was adopt- throughout the region, suggests that microcredit ed by men as it was mechanized, leaving women with schemes, which have shown positive impacts on rural more available time, yet also a sense of dependence women’s household income in Nigeria, have had lim- on the head of household’s willingness to perform or ited effects on women’s business outcomes across the pay for the work.170 Social norms around gendered use region, as they typically provide insufficient funding to and ownership of machinery and the heavy upfront stimulate firm investment and growth.173 Still, emerg- investment required to purchase machinery may con- ing global evidence indicates that targeting specific strain women’s ability to purchase time-saving tech- subsets of entrepreneurs (e.g., experienced entrepre- nologies such as tractors.171 Hiring or leasing services neurs) for microfinance under certain conditions (e.g., may be a promising policy alternative. However, more with flexible credit terms) can improve women’s earn- research is needed to measure the effectiveness of such innovations on women’s adoption of time-sav- XV There are several ongoing studies that will contribute to the ing technologies and, ultimately, the impact on their small evidence base on what works for women entrepreneurs in Nigeria. First, an ongoing study in Ekiti is using a randomized productivity. design to measure the returns to microcredit for women microentrepreneurs and will also test whether there are complementarities between microcredit and digital banking services (Alzua and Olajide 2018). Second, an ongoing study is comparing the impacts of business consulting services to those of business training services on small and medium-sized enterprise outcomes (Anderson, Kaul, and McKenzie 2019). Finally, there is also an ongoing evaluation designed to measure the impacts of different packages of the Nigerian Government’s Skills for Jobs program (Okunogbe 2019). 70 ings. Growth-oriented firms requiring larger loans are and sales, and employment, with an increase of over particularly credit constrained as microfinance does 20 percentage points in the likelihood of a firm hav- not address their needs, but they are too small for ing ten or more workers.179 Effects on women’s busi- commercial banks. A quasi-experimental study from ness start-up were particularly strong, while effects Ethiopia found that mesoloans offered to growth-ori- on other business outcomes were similar for women ented women entrepreneurs had a significant impact and men. Positive impacts of large business grants are on accelerating their business growth and boosting confirmed by another impact evaluation from Ethio- employment levels.174 Innovative approaches to finan- pia, Tanzania, and Zambia.180 cial products, such as psychometric testing and non- asset-based forms of collateral provide a potential Providing women with secure savings mechanisms, in- avenue for facilitating access to larger business loans cluding through mobile platforms, is another avenue for women-owned firms—though more research is for unlocking women’s access to growth capital.181 A needed to further validate the effectiveness of these study from Kenya shows that among entrepreneurs interventions. Notably, the World Bank Group’s Nige- offered a bank account, women market vendors were ria Women Entrepreneurship Finance Initiative (We- more likely to sign up for a formal savings account Fi) is working with financial intermediaries to expand than male bicycle taxi drivers. The women that signed access to finance for women-owned firms by piloting up increased their savings and made more productive and evaluating innovative products for women entre- investments into their businesses.182 In Malawi, add- preneurs, such as loans requiring little or no collateral. ing access to business bank accounts to support firm formalization led to significant increases in women’s Cash transfers provided within safety net programs use of business bank accounts and insurance, which have the potential to alleviate the capital constraint of in turn led to large impacts on their sales and prof- extremely poor women. In Northern Nigeria, an uncon- its.183 Another study from Tanzania found that a pro- ditional cash transfer of approximately US$350 trans- gram promoting the registration of a mobile savings ferred over fifteen months had a positive and signifi- account among women entrepreneurs increased their cant effect on the likelihood of rural women starting savings and secondary business creation.184 a nonfarm business.175 In Niger, a comprehensive pro- ductive package delivered to ultra-poor women that included a cash grant had a positive impact on non- agricultural business start-ups and revenue.176 On the other hand, stand-alone small cash grants may have little effect on the outcomes of businesses owned by women. Evidence from Ghana suggests, however, that in-kind grants may have a positive impact on the prof- Policy Priority 5: Promoting Women’s its of women microentrepreneurs, when targeted at Engagement in Greater Value Addition more successful businesses.177 Women entrepreneurs in Nigeria are significantly Targeting growth-oriented entrepreneurs for large more likely to sell to the final consumer than to trad- cash grants through business plan competitions can ers or small businesses. Their relatively lower position ease women’s constraints to accessing capital.178 A in the value chain drives their profits down compared recent study found large returns to winning Nigeria’s to male business owners. YouWiN! business competition: random assignment of winners among highly ranked applicants demon- There is limited evidence in Nigeria on effective in- strated that the program’s grants of approximately terventions to help women engage in greater value US$50,000 increased firm entry, firm survival, profits addition or sell more expansive ranges or quantities 71 of products to improve their position in the value segregation which were also found to drive the gap chain. Yet, global evidence suggests that certain de- in earnings in those respective sectors. Recommenda- mand-side interventions (focusing on skills, access to tions laid out in policy priority 1 and policy priority 5 capital, intrahousehold allocation of time, safety and also contribute to reducing occupational segregation mobility, and networks and role models), in addition in agriculture and the private sector, respectively. Sup- to addressing supply-side constraints, hold promise porting women’s transition into higher-productivity, in developing country contexts.185 For example, an often male-dominated sectors is critical not only for impact evaluation of a psychology-based entrepre- closing gender gaps in earnings, but also for acceler- neurial mindset training implemented in Togo found ating structural transformation, and in turn, economic that the training not only boosted profits for women growth. microentrepreneurs by 40 percent, but also induced them to be more innovative, introduce new products, One study in Nigeria assesses the impact of a program take out more and larger loans, and make larger in- that aimed to decrease occupational segregation. vestments in their firms.186 A study from Ethiopia con- The program sought to equip recent university grad- firmed the positive impacts of mindset-oriented busi- uates with the communication, computer, and cog- ness trainings on the performance of women-owned nitive skills demanded by Nigeria’s ICT sector.190 Two micro- and small enterprises, but found that delivery years after the program ended, the training program quality is crucial to success.187 Meanwhile, some glob- increased the average likelihood of graduates work- al evidence suggests that combining skills trainings ing in an ICT job by 26 percent, with equal gains by with social network building for women can improve gender. For women who at baseline were implicitly outcomes through social support, given the critical biased against associating women with profession- importance of networks in developing soft skills, in- al attributes, the likelihood of switching into the ICT creasing aspirations, and building business support sector due to the program was more than three times contacts.188 One impact evaluation in Ethiopia found as large as that of unbiased women. This suggests that a mentoring intervention for business owners en- that job training in male-dominated sectors may help hanced their business practices, but that effects on women overcome internalized biases that hamper overall business outcomes were muted.189 mobility into higher-earnings sectors. Research from Kenya demonstrates that providing accurate information about the earnings in various trades shifted women toward training in male-domi- nated trades. However, this information alone was not sufficient to sustain engagement in male-dominated sectors. These findings suggest that providing informa- Policy Priority 6: Decreasing Occupational tion may need to be complemented with other inter- Segregation ventions to promote women’s sustained participation in male-dominated industries.191 Enhancing women’s Occupational segregation, that is, the concentration access to appropriate role models, social networks, of men and women in certain sectors, drives the gap and mentors, and decreasing the risk of GBV and har- in earnings for women employees. The deep dive on assment in the workplace are other promising poli- occupational segregation in part III lays out the mul- cy options; however, the evidence base on their im- tiple factors leading to women being concentrated in pact on women’s occupational choice is thin. Access lower-earning sectors. It is worth noting that women to childcare may also help expand the universe of farmers’ crop choice and women entrepreneurs’ posi- available occupations for women. A study from Ken- tion in the value chain are other forms of occupational ya showed that access to subsidized daycare helped 72 mothers shift from jobs with more flexible hours such haviors. In the Eastern Democratic Republic of Con- as retail or laundry to jobs with fixed hours, such as go, a 16-week training for men significantly increased those in the services sector or government-sponsored their participation in housework.196 Another study employment.192 from Rwanda finds that a training for young couples on norms and behaviors also increased men’s share of housework, although it did not decrease the time women spent on household chores.197 More research is needed to identify cost-effective interventions that engage men in sharing the burden of domestic re- sponsibilities. Policy Priority 7: Easing Women’s Time As mentioned earlier in the report, fertility rates in Constraints Nigeria can reach 7 or more children per woman in some of the Northern states. Policy options that di- Women farmers, entrepreneurs, and wage earners are rectly address high fertility are beyond the scope of constrained by their family and domestic responsibil- this report. However, it is worth noting that decreas- ities: the higher burden of childcare and housework ing the desired number of children and increasing the that falls on women translates into fewer or poten- use of contraceptives is likely to have transformative tially less productive working hours, if they are multi- effects on women’s participation in the labor market tasking by combining work with looking after children, and on their productivity, which could in turn trigger for example. Interventions such as providing childcare a virtuous cycle creating the conditions for a demo- services or promoting men’s involvement in house- graphic dividend. hold duties have the potential to increase the time women spend working, as well as their productivity, in the presence of constraints to hiring productive labor. Broader Lessons on Policies and Programs to Close Gender Gaps Childcare interventions are one option for which re- gional evidence is emerging. A randomized controlled Transforming gender norms can help unlock access trial of an early childhood development program in to more profitable activities and crops. Norms that in- rural Mozambique demonstrated that preschools al- fluence what is expected of women and men in soci- lowed caregivers (primarily mothers) to save 15 hours ety underlie many of the constraints identified in this of childcare duties per week and resulted in higher report, from occupational segregation in the labor labor force participation for the caregivers.193 A qua- market to crop choice and position in the value chain si-experimental study in southern Togo concluded for entrepreneurs. The interventions proposed in the that enrolling children three to five years of age in pre- policy menu outlined above can overcome some of school resulted in women being 37 percent more like- the restrictive norms that limit women’s economic ac- ly to work outside the home.194 In Kenyan urban slums, tivities by either changing the norm itself or circum- women who were given vouchers for daycare were 8.5 venting the norm. In the long run, transforming norms percentage points more likely to be employed than will be key to helping women move into more prof- those who were not given vouchers.195 itable activities and increasing the efficiency of the economy as a whole. Whether interventions should Another option for alleviating women’s time con- attempt to directly change norms or whether wom- straints is by aiming to get men to increase their share en’s economic empowerment can in itself lead to of housework. There is credible evidence on programs normative transformation is an open question for re- that engage men to change gender norms and be- search. In either case, understanding the set of norms 73 that constrain women’s economic activities, access to have yielded promising effects in terms of boosting land, socioemotional skills and behavior, access to ed- livestock revenue.202 ucation, and intrahousehold dynamics, to cite only a few relevant areas, is key to designing programs that Women face multiple reinforcing constraints that af- aim to close gender gaps. fect their strategic decisions regarding their choice of crops and activities, their investment in capital and Developing women’s skills can facilitate adoption of labor, and their preferences. Considering comple- more profitable crops and help them transition into mentarities between interventions is key to unlock- more lucrative sectors. Investing in girls’ education ing their potential. For example, unlocking women’s and in developing women’s socioemotional skills and formal ownership of land can help build their asset growth mindset through psychology-based trainings base, which could also work to increase their credit- could empower them to choose activities that help worthiness. them earn a higher income. Adolescent girl empower- ment programs that provide life skills and vocational or business skills training, often combined with safe Gender Innovation in Policy Making, spaces, have yielded positive impacts on young wom- Programming, and Research en’s economic empowerment. These programs also improve beneficiaries’ control over their bodies and Despite the vibrant role of Nigerian women in agri- intimate relationships, leading to delayed childbear- culture, entrepreneurship, and the labor market, a ing.198 Such programs may be particularly relevant in number of constraints still hold back women, and in the Northern geopolitical zones of Nigeria, where ad- turn, the Nigerian economy. Nigerian women contin- olescent marriage and childbearing is more prevalent. ue to earn much less than Nigerian men for the same amount of work. Women also have lower access to Strengthening women’s land rights and their use and and use of key inputs that could unlock their produc- control over livestock should also be considered as tivity and are confined to the less remunerative sec- priority policy areas to close gender gaps in earn- tors of the economy. While this report takes stock of ings in Nigeria. Although not captured by the quan- the evidence base on what works to empower wom- titative analysis conducted in this report, these areas en farmers, firm owners, and workers, it finds there appeared as salient factors in earnings inequality be- is more work to be done to generate new evidence tween women and men. The evidence to date on land about effective intervention across contexts and at formalization programs in Sub-Saharan Africa sug- scale in Nigeria. Moreover, translating this knowledge gests that, when carefully designed and implemented, into significant progress toward closing the gender they can help women secure and benefit from their gaps will require a profound change in the approach land rights.199 Small incentives and information inter- to programming for gender equality. ventions are low-cost programming options that can boost joint titling within land formalization programs.200 Innovation in policy and programming will be a central An unconditional cash transfer program from Nigeria facet of this approach. Generating innovative gender has yielded significant increases in animal stock hold- programming ideas will entail exploring and designing ings.201 In Sub-Saharan Africa and other regions of the interventions that are not “business as usual” and can developing world, many programs for the ultra-poor effectively address the factors that are keeping wom- contain elements of livestock transfer in combina- en from performing as well as men in key economic tion with training. Programs targeting women that sectors. While these innovative programs may be too combine a productive asset transfer (livestock), con- risky to implement at scale initially, piloting them and sumption support, technical skills training, home vis- experimenting with various delivery modalities could its, health education and services, and savings support help identify which implementation options are most 74 effective. Conducting rigorous impact evaluations to scale up effective pilots and scale back resources on understand the effects of these interventions will be ineffective programs. needed to further build the evidence base to inform scale-up efforts of the government and its develop- Furthermore, more strategic integration of gender ment partners. Notably, new knowledge produced equality as a key objective is crucial to the formulation by Nigeria has the potential to move the emerging of federal and state-level policy documents. Several regional and global evidence base forward on what national policy documents such as the Agricultural works to address the constraints to women’s econom- Promotion Policy, the National Livestock Transforma- ic empowerment. tion Plan, and the Gender Policy in Agriculture, already explicitly mention women as a target group for pro- A key barrier for policy makers is a lack of evidence gramming or recognize the constraints that they face. on policies and programs proven to effectively ease The Economic and Recovery Growth Plan, although the constraints. Despite Nigeria’s innovative and active it includes specific programming recommendations policy space, and the wide range of programs targeting aimed at improving women’s economic empower- the constraints facing women farmers, entrepreneurs, ment, does not specifically state gender inequality as and employees, this report highlights a conspicuous a constraint to economic growth. The next iteration shortage of evidence on the effectiveness of program- of the overarching national economic strategy could ming to improve women’s economic empowerment incorporate reducing gender inequality as an explicit in Nigeria. One of the main lessons of this report is the objective. This would reaffirm the government’s com- need for increased investment in rigorous research to mitment to empowering Nigerian women and build- show what works, and increased investment to scale ing policy momentum to address one of the key barri- up interventions for which there is evidence of impact. ers to economic growth. Recognizing that not all issues can be addressed with However, embedding gender equality in national strat- limited resources, this report helps identify priority egy documents is just one step in the chain of policy constraints that play an important role in equalizing actions needed to realize the promise of closing gen- earnings between women and men in Nigeria. To ad- der gaps in Nigeria. Delivering on these strategies will dress these gender constraints effectively, knowing require innovative thinking to design new gender pro- which interventions are likely to have the largest im- grams to test, pilot, and bring to scale, should they be pact for women is critical. Rigorous evidence of their proven effective. It will also involve harnessing the rig- cost-effectiveness is also a key component of making orous evidence from impact evaluations and comple- decisions on how to allocate scarce resources. In this mentary studies in ongoing and future policy debates regard, impact evaluations can serve as a critical tool to ensure sustained and coordinated action. Nigeri- in expanding the knowledge of cost-effective inter- an policy makers are already making demonstrable ventions. strides toward narrowing gender gaps. Equipped with additional evidence on the most effective and cost-ef- Well-designed evaluations can measure both pro- fective programming, they can accelerate progress gram impacts and costs. Costing policies is essential toward reaching their inclusive development and to operationalizing policy documents, preventing the economic recovery targets while advancing women’s prescription of overly ambitious policies that cannot empowerment and providing valuable lessons for the realistically be fully implemented. Impact evaluations rest of the world. can help policy makers assess whether programs are achieving intended goals, such as improving agricul- tural productivity or incomes, and at what cost. This evidence can then inform subsequent decisions to 75 Appendices 76 Appendix 1: Correlates of Labor Supply Women: Men: Women: Any Variables Conditional Men: Any work Conditional work hours worked hours worked Age 0.035*** 0.506** 0.032*** 1.170*** (0.005) (0.256) (0.005) (0.350) Age squared -0.000*** -0.007** -0.000*** -0.014*** (0.000) (0.003) (0.000) (0.004) Household head 0.146*** 3.523* 0.219*** 3.115 (0.038) (2.064) (0.036) (2.115) Married 0.144*** -0.665 0.080** -1.798 (0.033) (1.864) (0.034) (2.665) Polygamous 0.126*** 1.730 0.051 -4.836 (0.041) (2.337) (0.040) (3.681) Divorced/Widowed 0.104** -2.934 -0.104 -1.964 (0.044) (2.974) (0.082) (4.020) Did not complete primary 0.083* 0.266 -0.033 0.943 (0.044) (2.330) (0.052) (2.496) Completed primary 0.055* 1.969 -0.021 5.752*** (0.032) (1.288) (0.032) (2.020) Junior school certificate 0.014 -1.472 -0.132*** 2.742 (0.038) (1.901) (0.041) (2.159) Senior school certificate 0.107*** 4.542*** 0.031 6.332*** (0.033) (1.582) (0.028) (2.112) Postsecondary certificate 0.081* 5.525** -0.060* 5.173** (0.044) (2.264) (0.032) (2.260) University or college degree -0.010 3.629 -0.108*** 2.894 (0.049) (2.591) (0.040) (2.485) Quranic and integrated Quranic education -0.024 5.556*** 0.067* 13.008*** (0.043) (1.562) (0.035) (3.869) Wealth index -0.004 1.119*** -0.009** 0.754*** (0.004) (0.240) (0.004) (0.274) Household size 0.002 -0.180 -0.001 0.061 (0.004) (0.185) (0.002) (0.170) Dependency ratio (elderly and children under 5) 0.041* -0.706 0.004 1.596 (0.024) (1.245) (0.018) (1.310) Zone 1. North Central -0.090* -9.296*** -0.016 -8.133*** (0.049) (2.607) (0.028) (2.281) Table continues next page 77 Correlates of Labor Supply (Continued) Women: Men: Women: Any Variables Conditional Men: Any work Conditional work hours worked hours worked Zone 2. North East -0.250*** -9.692** -0.124*** -6.260* (0.049) (3.867) (0.037) (3.441) Zone 3. North West -0.175*** -8.712*** -0.047 -5.258** (0.041) (2.137) (0.032) (2.596) Zone 4. South East -0.008 -7.811*** -0.050* -9.229*** (0.037) (2.358) (0.030) (2.743) Zone 5. South South 0.080** -4.327** 0.002 -7.133*** (0.035) (2.161) (0.028) (1.875) Constant -0.199** 27.919*** 0.059 14.427** (0.090) (4.597) (0.093) (6.274) Observations 5,740 3,198 5,227 3,565 R-squared 0.172 0.118 0.260 0.089 Adjusted R-squared 0.169 0.112 0.257 0.0836 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 78 Appendix 2A: Kitagawa-Oaxaca-Blinder Decomposition Methods This appendix provides a technical overview of the an outcome variable (such as agricultural productivity quantitative methods used in this report. These regres- or self-employment profits) between two groups (such sion-based methods build on the seminal papers of as men and women) into two components: a compo- Kitagawa (1955), Oaxaca (1973), and Blinder (1973).XVI, 203 nent reflecting differences in levels of observable char- It is important to note that decomposition methods acteristics between the two groups, known as explained are based on correlations and cannot be interpret- or composition effect, and a component reflecting dif- ed as direct estimates of underlying causal parame- ferences in returns to the observable characteristics, ters: they can, however, help document the relative which is known as the unexplained or structural ef- quantitative importance of factors accounting for an fect. The analysis presented throughout this diagnostic observed difference in outcomes, suggesting priority compares women in productive activities (women farm topics for further analysis and policy intervention. The managers, business managers, or employees) to men in main purpose of the Kitagawa-Oaxaca-Blinder (KOB) equivalent productive activities (men farm managers, decomposition is to partition the difference in means of business managers, or employees). The KOB decomposition approach assumes that the outcome variable, Y, is linearly related to a series of covari- ates, X, and that the error term v is conditionally independent of X represented by the regression equation: ΣX K Ygi = βg0 + ik βgk + vgi, g = A,B k=1 where Ygi represents outcome Y for individual i in group g {A,B}, βg0 represents the conditional mean for group g, X are the covariates, and E(vgi|Xi) = 0. The overall difference in average outcomes between the two k groups can be represented as Δˆμ = y - y . 0 B A Combining these two equations and noting that E(vgi|X )=0 , yields i Σ ΣX K K Δ ˆ + ˆμ = β ˆ -β XBk β ˆ - ˆ 0 B0 Bk A0 Ak βAk. k=1 k=1 Additional details are presented in the Handbook of XVI Labor Economics, Volume 4, part A, pages 1–102, Chapter 1: Decomposition Methods in Economics (Fortin, Lemieux, and Firp 2011) and the report Levelling the Field (O’Sullivan et al. 2014). 79 K Adding and subtracting Σk=1 XAkβBk yields Σ Σ ΣX K K K Δ ˆ + ˆμ = β ˆ -β XBk β ˆ - ˆ + XAk β ˆ 0 B0 Bk A0 Ak Bk βAk, k=1 k=1 k=1 which can also be written as Σ Σ (X K K Δ ˆ -β ˆμ = (β ˆ )+ ˆ -β XBk(β ˆ )+ ˆ . XAk)β 0 B0 A0 Bk Ak Bk - Ak k=1 k=1 This formulation clearly shows the two underlying components: the unexplained/structural effect and the ex- plained/composition effect: ΣX K ˆ -β ˆμ(Explained) = (β ˆ ˆ ˆ Δ S B0 A0)+ Bk(βBk - βAk) k=1 Σ (X K Δˆμ(Explained) = XAk) βˆ , X Bk - Ak k=1 ˆ where βg0 and βgk(k=1, ... , K) are the estimated intercept and slope coefficients, respectively, of the regression models for the two groups. The split of the overall decomposition Δ ˆμ into the two subcomponents—the Δ ˆμ unexplained/structural effect 0 0 ˆμ and ΔX explained/composition effect—is referred to as the aggregate decomposition. Further decomposing the subcomponents, Δ ˆμ and Δˆμ, into the respective contributions of each covariate, Δμ and Δμ for (k=1, ... , K), S X S,k X,k is called the detailed decomposition. Under a linear form, the detailed decomposition can be calculated for the kth covariate as ˆμ = X (βˆ ˆ Δ S,k Bk Bk - βAk) Δ ˆ ˆμ = (X - X ) β X,k Bk Bk Ak. 80 Appendix 2B: Agriculture Balance table: Agriculture (1) (2) t-test Female plot Male plot difference manager manager Variables N Mean/SE N Mean/SE (1)−(2) Agriculture productivity (winsorized) 1060 323567.16 4862 314017.53 9549.63 [19027.07] [9718.59] Agriculture productivity (IHS) 1060 12.56 4862 12.61 -0.05 [0.06] [0.02] Area plot cultivated (IHS) 1060 0.22 4862 0.52 -0.29*** [0.01] [0.01] Age 1060 51.99 4862 47.62 4.37*** [0.65] [0.28] Age squared 1060 2921.38 4862 2477.08 444.30*** [67.87] [28.32] Relationship with household head: head 1060 0.65 4862 0.98 -0.34*** [0.02] [0.00] Relationship with household head: spouse 1060 0.28 4862 0.00 0.28*** [0.02] [0.00] Relationship with household head: son/daughter 1060 0.01 4862 0.01 0.00 [0.00] [0.00] Other relative 1060 0.06 4862 0.00 0.05*** [0.01] [0.00] Married 1060 0.23 4862 0.62 -0.39*** [0.02] [0.01] Polygamous 1060 0.11 4862 0.32 -0.21*** [0.02] [0.01] Divorced 1060 0.07 4862 0.01 0.07*** [0.01] [0.00] Widowed 1060 0.55 4862 0.01 0.54*** [0.02] [0.00] Did not complete primary 1060 0.15 4862 0.06 0.09*** [0.01] [0.00] Completed primary 1060 0.21 4862 0.19 0.01 [0.02] [0.01] Table continues next page 81 Balance table: Agriculture (Continued) (1) (2) t-test Female plot Male plot difference manager manager Variables N Mean/SE N Mean/SE (1)−(2) Junior school certificate 1060 0.09 4862 0.05 0.04** [0.01] [0.00] Senior school certificate 1060 0.09 4862 0.17 -0.08*** [0.01] [0.01] Postsecondary certificate 1060 0.04 4862 0.06 -0.02* [0.01] [0.00] University or college degree 1060 0.02 4862 0.05 -0.03*** [0.01] [0.00] Quranic and integrated Quranic education 1060 0.03 4862 0.18 -0.14*** [0.01] [0.01] Work on wage salary 1060 0.06 4862 0.11 -0.05*** [0.01] [0.01] Work on nonfarm enterprise 1060 0.20 4862 0.35 -0.14*** [0.02] [0.01] Stopped activities because of illness in the last 1060 0.26 4862 0.14 0.12*** 4 weeks [0.02] [0.01] Extension training individual (planting) 1060 0.07 4862 0.18 -0.11*** [0.01] [0.01] Household size 1060 5.71 4862 7.60 -1.89*** [0.16] [0.08] Dependency ratio (elderly and children under 11) 1060 0.61 4862 0.90 -0.30*** [0.03] [0.02] Area plot cultivated (IHS) 1060 0.22 4862 0.52 -0.29*** [0.01] [0.01] Area plot cultivated squared (IHS) 1060 0.13 4862 0.50 -0.37*** [0.01] [0.02] Plot was irrigated 1060 0.01 4862 0.03 -0.02*** [0.00] [0.00] Plot squatted (free) 1060 0.14 4862 0.10 0.04** [0.02] [0.01] Plot is rented 1060 0.14 4862 0.12 0.02 [0.01] [0.01] Plot has legal title 1060 0.03 4862 0.07 -0.04*** [0.01] [0.00] Quality of plot’s soil is good 1060 0.84 4862 0.85 -0.01 [0.02] [0.01] Uses animals for traction on plot 1060 0.06 4862 0.27 -0.21*** [0.01] [0.01] Uses machinery on the plot 1060 0.06 4862 0.11 -0.05*** [0.01] [0.01] Table continues next page 82 Balance table: Agriculture (Continued) (1) (2) t-test Female plot Male plot difference manager manager Variables N Mean/SE N Mean/SE (1)−(2) Crops: cereals 1060 0.51 4862 0.70 -0.19*** [0.02] [0.01] Crops: vegetables and melons 1060 0.17 4862 0.07 0.10*** [0.01] [0.00] Crops: oilseed 1060 0.12 4862 0.19 -0.07*** [0.01] [0.01] Crops: legumes 1060 0.05 4862 0.18 -0.14*** [0.01] [0.01] Crops: others 1060 0.06 4862 0.04 0.02 [0.01] [0.00] Crops: yams 1060 0.24 4862 0.11 0.13*** [0.02] [0.01] Crops: other roots and tubers 1060 0.60 4862 0.22 0.38*** [0.02] [0.01] Used purchased seed 1060 0.49 4862 0.49 0.00 [0.02] [0.01] IHS [fertilizer kilogram/hectare] 1060 1.91 4862 4.18 -2.27*** [0.13] [0.08] IHS [herbicide kilogram/hectare] 1060 0.73 4862 1.03 -0.31*** [0.06] [0.03] IHS [pesticide kilogram/hectare] 1060 0.15 4862 0.36 -0.21*** [0.04] [0.02] IHS [agricultural capital/hectare] 1060 -0.81 4862 0.26 -1.07*** [0.10] [0.04] IHS [plot manager labor days/hectare] 1060 4.74 4862 4.00 0.74*** [0.09] [0.03] IHS [hired female labor days/hectare] 1060 1.32 4862 0.72 0.60*** [0.09] [0.03] IHS [hired male labor days/hectare] 1060 2.77 4862 2.53 0.23* [0.11] [0.05] IHS [hired child labor days/hectare] 1060 0.23 4862 0.52 -0.29*** [0.04] [0.03] IHS [female family labor days/hectare] 1060 1.93 4862 2.00 -0.07 [0.11] [0.05] IHS [male family labor days/hectare] 1060 2.48 4862 1.55 0.93*** [0.11] [0.04] IHS [child family labor days/hectare] 1060 0.78 4862 0.97 -0.19** [0.08] [0.04] IHS [child family labor 11–15 year olds days/ 1060 1.34 4862 1.19 0.16 hectare] [0.09] [0.04] Table continues next page 83 Balance table: Agriculture (Continued) (1) (2) t-test Female plot Male plot difference manager manager Variables N Mean/SE N Mean/SE (1)−(2) Zone 1. North Central 1060 0.19 4862 0.23 -0.05** [0.02] [0.01] Zone 2. North East 1060 0.06 4862 0.16 -0.11*** [0.01] [0.01] Zone 3. North West 1060 0.06 4862 0.35 -0.29*** [0.01] [0.01] Zone 4. South East 1060 0.36 4862 0.10 0.26*** [0.02] [0.00] Zone 5. South South 1060 0.26 4862 0.08 0.18*** [0.02] [0.01] Zone 6. South West 1060 0.07 4862 0.07 -0.00 [0.02] [0.01] Wealth index 1052 -0.50 4822 -1.15 0.65*** [0.09] [0.04] Number of adults 1060 3.28 4862 3.93 -0.65*** [0.10] [0.04] Sample: full Note: The values displayed for t-tests are the differences in the means across the groups. SE refers to standard error. ***, **, and * indicate significance at the 1, 5, and 10 percent critical level. IHS indicates variables that have been transformed with the inverse hyperbolic sine transformation. Cross-section: Agricultural productivity Full North South Variables (1) (2) (3) Female member -0.176** -0.249** -0.240* (0.085) (0.105) (0.134) Age 0.001 -0.002 -0.002 (0.011) (0.011) (0.020) Age squared -0.000 0.000 -0.000 (0.000) (0.000) (0.000) Married 0.110 -0.091 0.413** (0.145) (0.156) (0.199) Polygamous 0.181 0.000 0.273 (0.153) (0.165) (0.290) Table continues next page 84 Cross-section: Agricultural productivity (Continued) Full North South Variables (1) (2) (3) Divorced 0.116 -0.302 0.484* (0.231) (0.202) (0.279) Widowed -0.096 -0.268 0.192 (0.185) (0.235) (0.261) Did not complete primary -0.181* -0.104 -0.380** (0.109) (0.111) (0.163) Completed primary -0.075 -0.000 -0.221 (0.074) (0.082) (0.143) Junior school certificate 0.039 0.004 -0.024 (0.093) (0.092) (0.161) Senior school certificate -0.091 -0.079 -0.260 (0.081) (0.095) (0.158) Postsecondary certificate -0.129 -0.227* -0.076 (0.109) (0.121) (0.219) University or college degree -0.063 -0.017 -0.262 (0.135) (0.157) (0.238) Quranic and integrated Quranic education -0.025 -0.005 0.821** (0.078) (0.076) (0.347) Stopped activities because of illness in the last 4 weeks 0.015 -0.020 -0.030 (0.060) (0.065) (0.094) Extension training individual (planting) 0.106* 0.168** -0.034 (0.063) (0.067) (0.118) Household size -0.009 -0.008 -0.011 (0.007) (0.007) (0.020) Dependency ratio (elderly and children under 11) -0.010 -0.003 -0.039 (0.031) (0.039) (0.059) Area plot cultivated (IHS) -3.318*** -2.778*** -4.996*** (0.296) (0.305) (0.533) Area plot cultivated squared (IHS) 1.454*** 1.109*** 2.567*** (0.177) (0.178) (0.316) Plot was irrigated 0.067 0.156 -0.181 (0.111) (0.122) (0.315) Plot squatted (free) -0.103 -0.004 -0.215* (0.079) (0.091) (0.121) Plot is rented 0.057 0.138 -0.019 (0.084) (0.086) (0.147) Plot has legal title 0.121 0.192** -0.095 (0.080) (0.077) (0.262) Quality of plot’s soil is good -0.020 -0.036 -0.006 (0.049) (0.053) (0.104) Uses animals for traction on plot 0.045 0.038 (0.061) (0.061) Table continues next page 85 Cross-section: Agricultural productivity (Continued) Full North South Variables (1) (2) (3) Uses machinery on the plot 0.095 -0.047 0.342 (0.095) (0.076) (0.216) Crops: cereals -0.172*** -0.037 -0.260*** (0.058) (0.069) (0.099) Crops: vegetables and melons 0.065 0.314** 0.026 (0.084) (0.126) (0.104) Crops: oilseed 0.084* 0.124** 0.138 (0.048) (0.051) (0.188) Crops: legumes 0.028 0.037 -0.038 (0.067) (0.066) (0.407) Crops: others 0.303** 0.166 0.409** (0.121) (0.150) (0.190) Crops: yams 0.894*** 1.029*** 0.841*** (0.083) (0.161) (0.092) Crops: other roots and tubers -0.495*** -0.285** -0.537*** (0.090) (0.128) (0.128) Used purchased seed -0.082 -0.039 -0.165** (0.052) (0.058) (0.081) IHS [fertilizer kilogram/hectare] 0.054*** 0.056*** 0.047*** (0.009) (0.010) (0.014) IHS [herbicide kilogram/hectare] 0.130*** 0.161*** 0.093*** (0.021) (0.024) (0.033) IHS [pesticide kilogram/hectare] 0.045 0.026 0.095* (0.030) (0.033) (0.057) IHS [agricultural capital/hectare] 0.006 -0.003 0.015 (0.014) (0.015) (0.022) IHS [plot manager labor days/hectare] 0.041** 0.001 0.076*** (0.019) (0.024) (0.024) IHS [hired female labor days/hectare] -0.024 0.019 -0.047** (0.015) (0.023) (0.020) IHS [hired male labor days/hectare] 0.044*** 0.048*** 0.044** (0.012) (0.013) (0.020) IHS [hired child labor days/hectare] 0.018 0.009 0.012 (0.014) (0.015) (0.040) IHS [female family labor days/hectare] -0.000 -0.004 0.002 (0.012) (0.017) (0.017) IHS [male family labor days/hectare] 0.037*** 0.013 0.070*** (0.013) (0.012) (0.019) IHS [child family labor days/hectare] 0.003 -0.008 0.025 (0.015) (0.015) (0.029) IHS [child family labor 11–15 year olds days/hectare] -0.011 0.003 -0.032 (0.013) (0.012) (0.022) Table continues next page 86 Cross-section: Agricultural productivity (Continued) Full North South Variables (1) (2) (3) Zone 1. North Central 0.084 (0.110) Zone 2. North East -0.064 -0.108 (0.095) (0.088) Zone 3. North West -0.033 (0.104) Zone 4. South East -0.067 (0.150) Zone 5. South South 0.160 0.154 (0.155) (0.122) Zone 6. South West 0.424*** 0.495** (0.156) (0.192) Constant 13.019*** 13.076*** 13.243*** (0.308) (0.325) (0.513) Observations 5,922 3,964 1,958 R-squared 0.360 0.406 0.363 Percentage gap -0.16 -0.23 -0.22 Fertilizer elasticity 0.056 0.057 0.056 Herbicide elasticity 0.178 0.201 0.196 Extension elasticity 0.109 0.180 -0.040 Note: Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 IHS indicates variables that have been transformed with the inverse hyperbolic sine transformation. Oaxaca: Agriculture Decomposition of the gender differential in agricultural productivity Full North South estimated estimated estimated coefficient coefficient coefficient [standard [standard [standard error] error] error] Overall Group 1 12.563*** 12.357*** 12.654*** [0.07] [0.11] [0.09] Group 2 12.610*** 12.564*** 12.743*** [0.04] [0.05] [0.07] Difference -0.047 -0.207* -0.088 [0.08] [0.11] [0.10] Table continues next page 87 Oaxaca: Agriculture (Continued) Full North South estimated estimated estimated coefficient coefficient coefficient Overall [standard [standard [standard error] error] error] Explained 0.130 0.043 0.151 [0.08] [0.12] [0.11] Unexplained -0.176** -0.249** -0.240* [0.08] [0.11] [0.13] Explained Age 0.004 -0.003 -0.005 [0.05] [0.02] [0.05] Age squared -0.018 0.002 -0.010 [0.04] [0.02] [0.04] Married -0.043 0.031 -0.221** [0.06] [0.05] [0.11] Polygamous -0.038 -0.000 -0.015 [0.03] [0.03] [0.02] Divorced 0.008 -0.015 0.030 [0.02] [0.01] [0.02] Widowed -0.052 -0.129 0.106 [0.10] [0.11] [0.14] Did not complete primary -0.016 0.003 -0.048** [0.01] [0.00] [0.02] Completed primary -0.001 -0.000 0.019 [0.00] [0.00] [0.01] Junior school certificate 0.001 -0.001 [0.00] [0.00] Senior school certificate 0.007 0.008 0.039 [0.01] [0.01] [0.03] Postsecondary certificate 0.002 -0.000 0.001 [0.00] [0.01] [0.00] University or college degree 0.002 0.013 [0.00] [0.01] Quranic and integrated Quranic education 0.004 0.001 -0.000 [0.01] [0.01] [0.00] Stopped activities because of illness in the last 4 weeks 0.002 -0.001 -0.004 [0.01] [0.00] [0.01] Extension training individual (planting) -0.011 -0.029** 0.001 [0.01] [0.01] [0.00] Household size 0.017 0.004 0.007 [0.01] [0.01] [0.01] Table continues next page 88 Oaxaca: Agriculture (Continued) Full North South estimated estimated estimated coefficient coefficient coefficient Overall [standard [standard [standard error] error] error] Dependency ratio (elderly and children under 11) 0.003 0.001 0.002 [0.01] [0.02] [0.00] Area plot cultivated (IHS) 0.968*** 0.541*** 0.807*** [0.12] [0.12] [0.19] Area plot cultivated squared (IHS) -0.535*** -0.333*** -0.505*** [0.08] [0.08] [0.12] Plot was irrigated -0.002 -0.004 0.002 [0.00] [0.00] [0.00] Plot squatted (free) -0.004 -0.000 0.004 [0.00] [0.01] [0.01] Plot is rented 0.001 0.007 0.002 [0.00] [0.01] [0.01] Plot has legal title -0.005 -0.009* 0.002 [0.00] [0.00] [0.00] Quality of plot’s soil is good -0.002 0.000 [0.00] [0.01] Uses animals for traction on plot -0.009 -0.006 [0.01] [0.01] Uses machinery on the plot -0.004 0.003 -0.012 [0.00] [0.00] [0.01] Crops: cereals 0.032*** 0.006 -0.007 [0.01] [0.01] [0.01] Crops: vegetables and melons 0.007 0.018 0.001 [0.01] [0.01] [0.00] Crops: oilseed -0.006 0.008 [0.00] [0.01] Crops: legumes -0.004 -0.004 -0.000 [0.01] [0.01] [0.00] Crops: others 0.006 0.001 -0.004 [0.00] [0.00] [0.01] Crops: yams 0.115*** 0.014 0.031 [0.02] [0.02] [0.02] Crops: tubers -0.188*** -0.019 -0.065*** [0.04] [0.01] [0.02] Used purchased seed -0.000 0.002 0.004 [0.00] [0.00] [0.01] IHS [fertilizer kilogram/hectare] -0.123*** -0.125*** 0.002 [0.03] [0.03] [0.01] Table continues next page 89 Oaxaca: Agriculture (Continued) Full North South estimated estimated estimated coefficient coefficient coefficient Overall [standard [standard [standard error] error] error] IHS [herbicide kilogram/hectare] -0.040*** 0.042 -0.020* [0.01] [0.03] [0.01] IHS [pesticide kilogram/hectare] -0.009 -0.007 -0.005 [0.01] [0.01] [0.01] IHS [agricultural capital/hectare] -0.007 0.002 -0.006 [0.02] [0.01] [0.01] IHS [plot manager labor days/hectare] 0.030* -0.000 0.049** [0.02] [0.01] [0.02] IHS [hired female labor days/hectare] -0.014 -0.001 -0.016 [0.01] [0.00] [0.01] IHS [hired male labor days/hectare] 0.010 0.012 -0.006 [0.01] [0.02] [0.01] IHS [hired child labor days/hectare] -0.005 -0.002 [0.00] [0.00] IHS [female family labor days/hectare] -0.001 -0.003 [0.00] [0.03] IHS [male family labor days/hectare] 0.035*** 0.013 0.075*** [0.01] [0.01] [0.02] IHS [child family labor days/hectare] -0.001 0.001 [0.00] [0.00] IHS [child family labor 11–15 days/hectare] -0.002 0.001 -0.009 [0.00] [0.00] [0.01] Zone 1. North Central -0.004 0.010 [0.01] [0.03] Zone 2. North East 0.007 0.003 [0.01] Zone 3. North West Zone 4. South East -0.017 [0.04] Zone 5. South South 0.029 0.008 [0.03] [0.01] Zone 6. South West -0.000 -0.091** [0.01] [0.04] Unexplained Age -0.184 2.475* -1.900 [1.14] [1.47] [1.54] Table continues next page 90 Oaxaca: Agriculture (Continued) Full North South estimated estimated estimated coefficient coefficient coefficient Overall [standard [standard [standard error] error] error] Age squared 0.036 -0.992 1.096 [0.56] [0.73] [0.75] Married 0.051 0.022 -0.103 [0.07] [0.10] [0.16] Polygamous 0.035 0.027 -0.014 [0.05] [0.12] [0.04] Divorced 0.004 -0.009 -0.018 [0.02] [0.02] [0.03] Widowed 0.115 -0.008 0.001 [0.12] [0.21] [0.15] Did not complete primary -0.026 0.028* -0.007 [0.02] [0.02] [0.04] Completed primary 0.065 0.054 0.100 [0.04] [0.05] [0.07] Junior school certificate 0.024 -0.013 0.041 [0.02] [0.02] [0.03] Senior school certificate 0.023 0.008 0.058 [0.02] [0.02] [0.05] Postsecondary certificate -0.001 -0.045 0.007 [0.01] [0.03] [0.02] University or college degree 0.002 -0.031 0.024 [0.01] [0.03] [0.02] Quranic and integrated Quranic education -0.002 0.010 [0.01] [0.03] Stopped activities because of illness in the last 4 weeks -0.003 -0.026 -0.034 [0.03] [0.04] [0.04] Extension training individual (planting) -0.003 0.011 0.007 [0.01] [0.02] [0.02] Household size 0.161 -0.010 0.272 [0.11] [0.16] [0.17] Dependency ratio (elderly and children under 11) -0.130*** -0.012 -0.172*** [0.05] [0.06] [0.07] Area plot cultivated (IHS) -0.403** -0.345 -0.141 [0.16] [0.28] [0.19] Area plot cultivated-squared (IHS) 0.176** 0.132 0.072 [0.07] [0.13] [0.08] Plot was irrigated -0.000 0.001 0.002 [0.00] [0.01] [0.01] Table continues next page 91 Oaxaca: Agriculture (Continued) Full North South estimated estimated estimated coefficient coefficient coefficient Overall [standard [standard [standard error] error] error] Plot squatted (free) 0.011 0.034 0.043 [0.03] [0.03] [0.04] Plot is rented -0.057** -0.012 -0.085** [0.02] [0.03] [0.03] Plot has legal title 0.013 0.024 0.029* [0.01] [0.02] [0.02] Quality of plot’s soil is good -0.024 0.007 0.044 [0.11] [0.14] [0.17] Uses animals for traction on plot 0.026 0.030 [0.02] [0.05] Uses machinery on the plot 0.051** -0.001 0.063** [0.02] [0.01] [0.03] Crops: cereals 0.122* 0.148 0.067 [0.07] [0.12] [0.09] Crops: vegetables and melons -0.025 -0.017 -0.014 [0.02] [0.02] [0.03] Crops: oilseed -0.027 -0.065 -0.006 [0.03] [0.04] [0.02] Crops: legumes 0.013 0.000 0.002 [0.01] [0.03] [0.01] Crops: others 0.027 -0.008 0.031 [0.02] [0.01] [0.04] Crops: yams 0.005 0.004 0.031 [0.03] [0.02] [0.05] Crops: tubers -0.018 -0.082** 0.146 [0.11] [0.04] [0.20] Used purchased seed -0.010 -0.119* 0.122 [0.06] [0.06] [0.08] IHS [fertilizer kilogram/hectare] -0.008 0.050 -0.011 [0.04] [0.06] [0.04] IHS [herbicide kilogram/hectare] -0.060** -0.006 -0.056* [0.03] [0.07] [0.03] IHS [pesticide kilogram/hectare] 0.017 -0.022 0.027 [0.02] [0.01] [0.02] IHS [agricultural capital/hectare] 0.011 -0.021 0.052 [0.02] [0.04] [0.03] IHS [plot manager labor days/hectare] 0.037 -0.270 0.093 [0.18] [0.17] [0.29] Table continues next page 92 Oaxaca: Agriculture (Continued) Full North South estimated estimated estimated coefficient coefficient coefficient Overall [standard [standard [standard error] error] error] IHS [hired female labor days/hectare] 0.036 -0.003 0.121** [0.04] [0.03] [0.06] IHS [hired male labor days/hectare] -0.096* -0.154 -0.070 [0.06] [0.10] [0.08] IHS [hired child labor days/hectare] -0.007 0.011 -0.005 [0.01] [0.02] [0.01] IHS [female family labor days/hectare] 0.010 0.071 -0.079 [0.05] [0.06] [0.10] IHS [male family labor days/hectare] -0.015 -0.058 -0.111** [0.04] [0.09] [0.05] IHS [child family labor days/hectare] 0.047 0.069 0.052 [0.03] [0.06] [0.03] IHS [child family labor 11–15 days/hectare] -0.068** 0.016 -0.048 [0.03] [0.06] [0.04] Zone 1. North Central -0.037 0.023 [0.06] [0.12] Zone 2. North East -0.009 0.013 [0.02] [0.02] Zone 3. North West -0.002 [0.05] Zone 4. South East -0.010 0.215** [0.15] [0.08] Zone 5. South South -0.039 0.159** [0.10] [0.07] Zone 6. South West -0.028 0.115 [0.03] [0.07] Constant -0.005 -1.189 -0.457 [0.64] [0.90] [0.95] N 5922 3964 1958 Note: ***, **, and * indicate significance at the 1, 5, and 10 percent critical level. IHS indicates variables that have been transformed with the inverse hyperbolic sine transformation. SE refers to standard error. 93 Appendix 2C: Self-Employment Balance table: Self-employment (1) (2) t-test Female Male difference Variables N manager N manager (1)−(2) Mean/SE Mean/SE Total profits (IHS) 1594 9.23 1446 10.30 -1.07*** [0.09] [0.11] Total profits (winsorized) 1594 14418.20 1446 36278.67 -21860.47*** [651.29] [1489.55] Age 1594 39.89 1446 40.75 -0.86 [0.52] [0.49] Age squared 1594 1776.71 1446 1855.69 -78.98 [49.51] [43.07] Married 1594 0.49 1446 0.58 -0.10*** [0.02] [0.02] Polygamous 1594 0.28 1446 0.20 0.09*** [0.02] [0.01] Divorced 1594 0.04 1446 0.01 0.03*** [0.01] [0.01] Widowed 1594 0.13 1446 0.01 0.12*** [0.01] [0.00] Never attended school 1594 0.27 1446 0.12 0.15*** [0.02] [0.01] Did not complete primary 1594 0.06 1446 0.04 0.02* [0.01] [0.01] Completed primary 1594 0.20 1446 0.20 -0.00 [0.01] [0.01] Junior school certificate 1594 0.07 1446 0.07 0.01 [0.01] [0.01] Senior school certificate 1594 0.22 1446 0.30 -0.08*** [0.02] [0.02] Postsecondary certificate 1594 0.04 1446 0.06 -0.02** [0.01] [0.01] Quranic and integrated Quranic education 1594 0.13 1446 0.17 -0.05** [0.01] [0.02] Stopped activities because of illness in the last 1594 0.12 1446 0.12 0.00 4 weeks [0.01] [0.01] Table continues next page 94 Balance table: Self-employment (Continued) (1) (2) t-test Female Male difference Variables N manager N manager (1)−(2) Mean/SE Mean/SE Access to internet 1594 0.17 1446 0.32 -0.14*** [0.01] [0.02] Relationship with household is other than 1594 0.82 1446 0.17 0.65*** household head [0.01] [0.01] Household size 1594 7.01 1446 7.06 -0.05 [0.15] [0.15] Number of adults 1594 3.84 1446 3.94 -0.10 [0.09] [0.09] Dependency ratio (elderly and children under 11) 1594 0.79 1446 0.79 -0.00 [0.03] [0.03] Use of formal credit 1594 0.01 1446 0.00 0.01** [0.00] [0.00] Use of informal credit 1594 0.01 1446 0.01 0.01 [0.01] [0.00] Value of your physical capital stock (tools, 1594 8.91 1446 10.76 -1.86*** equipment, etc.) (HIS) [0.14] [0.13] Value of your current stock of inputs or 1594 5.96 1446 5.75 0.21 supplies (IHS) [0.18] [0.21] Number of enterprises 1594 1.03 1446 1.03 -0.00 [0.01] [0.01] Enterprise officially registered with the 1594 0.02 1446 0.19 -0.17*** government [0.01] [0.01] Operated from home 1594 0.64 1446 0.24 0.39*** [0.02] [0.02] Industrial site 1594 0.01 1446 0.03 -0.02*** [0.00] [0.00] Traditional market 1594 0.13 1446 0.13 0.00 [0.01] [0.01] Roadside 1594 0.06 1446 0.14 -0.08*** [0.01] [0.01] Location of business: other 1594 0.09 1446 0.32 -0.23*** [0.01] [0.02] Household savings 1594 0.78 1446 0.78 -0.00 [0.01] [0.02] Formal loan 1594 0.01 1446 0.01 0.00 [0.00] [0.00] Informal loan 1594 0.26 1446 0.18 0.08*** [0.02] [0.01] Family income source (agriculture or other 1594 0.11 1446 0.18 -0.07*** business) [0.01] [0.01] Table continues next page 95 Balance table: Self-employment (Continued) (1) (2) t-test Female Male difference Variables N manager N manager (1)−(2) Mean/SE Mean/SE Relatives, friends, remittances 1594 0.25 1446 0.22 0.03 [0.02] [0.01] Other 1594 0.02 1446 0.04 -0.02** [0.01] [0.01] Sell product to final consumers 1594 0.96 1446 0.90 0.06*** [0.01] [0.01] Sell product to traders 1594 0.02 1446 0.04 -0.01 [0.01] [0.01] Sell product to small business 1594 0.01 1446 0.03 -0.02** [0.00] [0.01] Selling individual is missing 1594 0.00 1446 0.00 N/A [0.00] [0.00] Number of hired males 1594 0.01 1446 0.58 -0.57*** [0.00] [0.09] Number of hired females 1594 0.10 1446 0.06 0.04** [0.02] [0.01] Average number of hours worked by female 1594 162.91 1446 8.54 154.37*** household members [3.68] [1.69] Average number of hours worked by male 1594 2.21 1446 186.15 -183.94*** household members [0.37] [3.88] Average number of hours worked by child 1594 5.80 1446 1.88 3.92*** household members (<11) [1.34] [0.61] Average number of hours worked by child 1594 169.19 1446 195.90 -26.71*** household members (11–15) [3.83] [4.44] Industry: agriculture production 1594 0.02 1446 0.02 -0.00 [0.01] [0.00] Industry: manufacturing 1594 0.16 1446 0.12 0.04** [0.01] [0.01] Industry: services 1594 0.20 1446 0.43 -0.23*** [0.01] [0.02] Industry: trade 1594 0.62 1446 0.37 0.25*** [0.02] [0.02] Industry: other 1594 0.01 1446 0.06 -0.05*** [0.00] [0.01] Zone 1. North Central 1594 0.11 1446 0.10 0.01 [0.01] [0.01] Zone 2. North East 1594 0.09 1446 0.14 -0.05*** [0.01] [0.01] Zone 3. North West 1594 0.31 1446 0.38 -0.06** [0.02] [0.02] Table continues next page 96 Balance table: Self-employment (Continued) (1) (2) t-test Female Male difference Variables N manager N manager (1)−(2) Mean/SE Mean/SE Zone 4. South East 1594 0.13 1446 0.14 -0.01 [0.01] [0.01] Zone 5. South South 1594 0.16 1446 0.11 0.04*** [0.01] [0.01] Zone 6. South West 1594 0.20 1446 0.12 0.07*** [0.02] [0.01] Urban 1594 0.40 1446 0.40 0.00 [0.02] [0.02] Number of adults 1594 3.84 1446 3.94 -0.10 [0.09] [0.09] Sample: full Note: The value displayed for t-tests are the differences in the means across the groups. ***, **, and * indicate significance at the 1, 5, and 10 percent critical level. IHS indicates variables that have been transformed with the inverse hyperbolic sine transformation. Cross-section: Self-employment profits (1) (2) (3) Monthly Monthly Monthly Variables profits (IHS) profits (IHS) profits (IHS) Principal manager is female -1.134*** -0.730* -1.237 (0.431) (0.421) (0.799) Age 0.061** 0.116*** -0.042 (0.026) (0.034) (0.036) Age squared -0.001** -0.001*** (0.000) (0.000) Married 0.328 0.137 0.378 (0.302) (0.337) (0.410) Polygamous 0.780** 0.716* 0.514 (0.362) (0.399) (0.538) Divorced -0.349 0.173 -0.318 (0.722) (0.363) (0.978) Widowed 0.553 0.666 0.360 (0.578) (0.475) (0.781) Never attended school -0.540 -1.020** 0.033 (0.421) (0.460) (0.708) Did not complete primary -0.763 -1.287* -0.226 (0.535) (0.757) (0.731) Completed primary -0.560 -1.068** -0.276 (0.424) (0.478) (0.620) Table continues next page 97 Cross-section: Self-employment profits (Continued) (1) (2) (3) Monthly Monthly Monthly Variables profits (IHS) profits (IHS) profits (IHS) Junior school certificate -1.041** -1.165*** -0.935 (0.512) (0.439) (0.785) Senior school certificate -0.558 -1.114** -0.280 (0.422) (0.530) (0.584) Postsecondary certificate -1.006* -1.013* -1.255 (0.550) (0.519) (0.911) Quranic and integrated Quranic education -0.678 -1.155** (0.412) (0.462) Stopped activities because of illness in the last 4 weeks -0.184 -0.087 -0.403 (0.206) (0.207) (0.373) Access to internet -0.019 0.371 -0.469 (0.226) (0.262) (0.287) Relationship with household is other than household head 0.069 -0.184 0.123 (0.314) (0.291) (0.484) Household size -0.092 -0.101* -0.043 (0.058) (0.056) (0.115) Number of adults 0.116 0.120* 0.118 (0.080) (0.070) (0.178) Dependency ratio (elderly and children under 11) 0.279** 0.257** 0.332 (0.124) (0.112) (0.296) Use of formal credit -0.426 1.827*** -0.359 (1.243) (0.318) (1.597) Use of informal credit -0.565 0.603*** -1.697 (0.834) (0.220) (1.411) Value of your physical capital stock (tools, equipment, etc.) (IHS) 0.045** 0.054*** 0.010 (0.019) (0.019) (0.042) Value of your current stock of inputs or supplies (IHS) -0.008 -0.010 -0.005 (0.014) (0.019) (0.019) Number of enterprises 0.042 0.256 -0.701 (0.451) (0.471) (1.251) Enterprise officially registered with the government -0.325 -0.756 0.169 (0.368) (0.680) (0.325) Operated from home -0.433 0.198 -1.036*** (0.328) (0.483) (0.351) Industrial site 0.799* 0.980 0.391 (0.424) (0.676) (0.482) Traditional market 0.281 0.921* -0.346 (0.374) (0.476) (0.503) Roadside -0.046 0.361 -0.454 (0.382) (0.526) (0.499) Table continues next page 98 Cross-section: Self-employment profits (Continued) (1) (2) (3) Monthly Monthly Monthly Variables profits (IHS) profits (IHS) profits (IHS) Location of business: other 0.172 0.971 -0.541 (0.398) (0.631) (0.390) Household savings 0.284* 0.190 0.890*** (0.166) (0.171) (0.332) Formal loan 0.904* 1.741*** 0.811 (0.531) (0.440) (0.813) Informal loan -0.073 0.103 -0.062 (0.194) (0.201) (0.335) Family income source (agriculture or other business) -0.102 0.270 -0.994** (0.198) (0.183) (0.453) Relatives, friends, remittances 0.056 -0.118 0.425* (0.158) (0.190) (0.252) Other 0.219 0.222 0.362 (0.219) (0.237) (0.432) Sell product to final consumers -0.591** -0.904** -0.175 (0.265) (0.371) (0.406) Sell product to traders 0.224 -0.694 1.073** (0.361) (0.452) (0.503) Sell product to small business -0.020 -0.307 0.274 (0.279) (0.387) (0.545) Number of hired males 0.027 0.001 0.270*** (0.035) (0.024) (0.088) Number of hired females 0.190 0.289** 0.095 (0.121) (0.144) (0.163) Average number of hours worked by female household members 0.002 0.002 -0.001 (0.004) (0.005) (0.005) Average number of hours worked by male household members 0.001 0.003 -0.005 (0.003) (0.004) (0.005) Average number of hours worked by child household members (<11) -0.001 -0.002 0.004 (0.001) (0.001) (0.002) Average number of hours worked by child household members (11–15) -0.001 -0.001 0.004 (0.003) (0.004) (0.004) Industry: agriculture production -0.791 -0.160 -1.147 (1.120) (1.206) (2.132) Industry: manufacturing 0.288 0.469 0.293 (0.403) (0.337) (1.095) Industry: services 0.357 0.307 0.816 (0.374) (0.274) (1.003) Industry: trade 0.226 0.439 0.220 (0.396) (0.288) (1.043) Table continues next page 99 Cross-section: Self-employment profits (Continued) (1) (2) (3) Monthly Monthly Monthly Variables profits (IHS) profits (IHS) profits (IHS) Industry: other -0.464 -0.742 0.010 (0.939) (1.323) (1.182) Zone 1. North Central 0.282 (0.314) Zone 2. North East -0.320 (0.317) Zone 3. North West 0.455* 0.168 (0.272) (0.184) Zone 4. South East 0.154 0.105 (0.326) (0.319) Zone 5. South South 0.056 (0.400) Zone 6. South West 0.671** 0.711** (0.261) (0.301) Urban 0.252* 0.301** 0.183 (0.147) (0.132) (0.261) Constant 8.491*** 7.486*** 10.862*** (1.080) (0.983) (1.815) Observations 3,040 1,644 1,396 R-squared 0.097 0.182 0.101 Percentage gap -0.71 -0.560 -0.79 Capital stock 0.045 0.054 0.010 Sell to final consumers elasticity -0.465 -0.622 -0.226 Sell to traders elasticity 0.172 -0.549 1.57 Sell to small business elasticity -0.058 -0.317 0.134 Adjusted R-squared 0.0800 0.154 0.0655 Note: Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 IHS indicates variables that have been transformed with the inverse hyperbolic sine transformation. 100 Oaxaca: Self-employment Full North South Urban Rural estimated estimated estimated estimated estimated coefficient coefficient coefficient coefficient coefficient [standard [standard [standard [standard [standard error] error] error] error] error] Overall Group 1 9.228*** 9.044*** 9.426*** 9.464*** 9.072*** [0.09] [0.08] [0.16] [0.16] [0.10] Group 2 10.296*** 10.228*** 10.405*** 10.618*** 10.085*** [0.10] [0.14] [0.16] [0.13] [0.15] Difference -1.067*** -1.185*** -0.979*** -1.153*** -1.013*** [0.14] [0.16] [0.22] [0.22] [0.16] Explained 0.067 -0.454 0.258 0.227 0.028 [0.37] [0.36] [0.71] [0.55] [0.41] Unexplained -1.134*** -0.730* -1.237 -1.380** -1.041** [0.43] [0.40] [0.78] [0.62] [0.44] Explained Age -0.053 -0.313** -0.024 0.037 -0.141 [0.06] [0.14] [0.05] [0.06] [0.10] Age squared 0.054 0.336** 0.024 -0.042 0.143 [0.06] [0.14] [0.05] [0.06] [0.11] Married -0.032 -0.014 -0.054 -0.017 -0.024 [0.03] [0.03] [0.06] [0.02] [0.05] Polygamous 0.067** 0.140* 0.008 0.028 0.068 [0.03] [0.08] [0.01] [0.02] [0.06] Divorced -0.011 0.002 -0.015 -0.033 0.008 [0.02] [0.00] [0.05] [0.06] [0.01] Widowed 0.069 0.048 0.065 0.146 -0.020 [0.07] [0.03] [0.14] [0.09] [0.09] Never attended school -0.080 -0.249** 0.002 -0.040 0.028 [0.06] [0.12] [0.05] [0.03] [0.23] Did not complete primary -0.015 -0.028 -0.003 -0.026 -0.002 [0.01] [0.02] [0.01] [0.03] [0.02] Completed primary 0.001 0.017 0.004 -0.011 0.001 [0.01] [0.03] [0.01] [0.02] [0.01] Junior school certificate -0.006 0.025 -0.026 -0.019 -0.000 [0.01] [0.02] [0.03] [0.03] [0.00] Senior school certificate 0.045 0.149* 0.018 0.027 -0.023 [0.04] [0.08] [0.04] [0.04] [0.12] Postsecondary certificate 0.022 0.045* -0.006 0.024 0.004 [0.01] [0.03] [0.02] [0.02] [0.03] Quranic and integrated Quranic education 0.031 0.038 -0.005 0.005 [0.03] [0.05] [0.05] [0.08] Table continues next page 101 Oaxaca: Self-employment (Continued) Full North South Urban Rural estimated estimated estimated estimated estimated coefficient coefficient coefficient coefficient coefficient Overall [standard [standard [standard [standard [standard error] error] error] error] error] Stopped activities because of illness in the last -0.001 0.001 -0.011 0.007 -0.001 4 weeks [0.00] [0.00] [0.01] [0.02] [0.00] Access to internet 0.003 -0.062 0.063 0.037 0.002 [0.03] [0.04] [0.04] [0.06] [0.03] 0.044 -0.132 0.072 -0.035 0.031 Relationship with household is other than household head [0.20] [0.21] [0.28] [0.23] [0.24] Household size 0.004 -0.060 0.001 0.075 -0.023 [0.02] [0.04] [0.01] [0.08] [0.03] Number of adults -0.012 0.019 -0.009 -0.101 0.014 [0.02] [0.03] [0.02] [0.11] [0.02] Dependency ratio (elderly and children under 11) -0.001 0.009 0.009 0.039 -0.010 [0.01] [0.02] [0.02] [0.03] [0.01] Use of formal credit -0.004 0.003 -0.006 0.001 -0.004 [0.01] [0.00] [0.03] [0.00] [0.02] Use of informal credit -0.004 0.002 -0.018 0.005 -0.006 [0.01] [0.00] [0.02] [0.01] [0.01] -0.084** -0.121*** -0.017 -0.084 -0.069 Value of your physical capital stock (tools, equipment, etc.) (IHS) [0.04] [0.05] [0.07] [0.05] [0.05] Value of your current stock of inputs or -0.002 0.001 -0.004 -0.003 -0.000 supplies (IHS) [0.00] [0.00] [0.02] [0.01] [0.00] Number of enterprises -0.000 0.001 0.002 -0.020 -0.012 [0.00] [0.00] [0.01] [0.02] [0.01] Enterprise officially registered with the 0.055 0.117 -0.034 -0.056 0.110 government [0.06] [0.11] [0.06] [0.07] [0.09] Operated from home -0.171 0.119 -0.182*** -0.216* -0.067 [0.13] [0.29] [0.07] [0.13] [0.19] Industrial site -0.015 -0.016 -0.010 -0.021 -0.004 [0.01] [0.01] [0.01] [0.01] [0.01] Traditional market 0.001 -0.096* -0.049 0.003 [0.01] [0.05] [0.07] [0.02] Roadside 0.004 -0.041 0.021 0.002 -0.015 [0.03] [0.06] [0.02] [0.02] [0.05] Location of business: other -0.040 -0.267 0.108 0.005 -0.144 [0.09] [0.17] [0.08] [0.08] [0.15] Household savings -0.001 -0.006 -0.008 0.012 0.001 [0.01] [0.01] [0.02] [0.03] [0.00] Formal loan 0.001 -0.008 0.004 -0.006 0.001 [0.00] [0.01] [0.01] [0.01] [0.01] Informal loan -0.006 0.017 -0.000 -0.004 -0.009 [0.02] [0.03] [0.00] [0.02] [0.02] Table continues next page 102 Oaxaca: Self-employment (Continued) Full North South Urban Rural estimated estimated estimated estimated estimated coefficient coefficient coefficient coefficient coefficient Overall [standard [standard [standard [standard [standard error] error] error] error] error] Family income source (agriculture or other 0.007 -0.025 0.019 0.018 0.009 business) [0.01] [0.02] [0.02] [0.02] [0.02] Relatives, friends, remittances 0.001 -0.001 0.021 -0.000 0.006 [0.00] [0.00] [0.02] [0.00] [0.01] Other -0.004 -0.006 -0.005 -0.016 0.009 [0.00] [0.01] [0.01] [0.01] [0.01] Sell product to final consumers -0.038* -0.058** -0.012 -0.012 -0.049* [0.02] [0.03] [0.03] [0.03] [0.03] Sell product to traders -0.002 0.016 0.003 -0.011 0.001 [0.00] [0.01] [0.02] [0.02] [0.00] Sell product to small business 0.000 0.005 -0.010 -0.009 0.006 [0.01] [0.01] [0.02] [0.01] [0.01] Selling individual is missing Number of hired males -0.015 -0.000 -0.126*** -0.003 -0.009 [0.02] [0.01] [0.04] [0.03] [0.03] Number of hired females 0.007 0.022 -0.002 -0.002 0.010 [0.01] [0.01] [0.00] [0.01] [0.01] Average number of hours worked by female 0.357 0.282 -0.181 1.459 -0.040 household members [0.58] [0.65] [0.77] [1.78] [0.36] Average number of hours worked by male -0.155 -0.487 0.872 -1.023 0.196 household members [0.63] [0.75] [0.89] [1.95] [0.35] Average number of hours worked by child -0.004 -0.013 0.003 0.011 -0.010 household members (<11) [0.01] [0.01] [0.00] [0.01] [0.01] Average number of hours worked by child 0.014 0.039 -0.074 0.151 -0.074 household members (11–15) [0.09] [0.14] [0.09] [0.23] [0.06] Industry: agriculture production 0.000 0.001 -0.012 0.013 -0.001 [0.01] [0.01] [0.02] [0.02] [0.01] Industry: manufacturing 0.011 0.038 -0.003 0.061 -0.000 [0.02] [0.03] [0.01] [0.08] [0.00] Industry: services -0.083 -0.041 -0.301 -0.164 -0.026 [0.09] [0.04] [0.36] [0.26] [0.05] Industry: trade 0.056 0.048 0.096 0.122 0.023 [0.10] [0.03] [0.45] [0.24] [0.07] Industry: other 0.023 0.038 -0.001 -0.029 0.091 [0.05] [0.07] [0.06] [0.03] [0.08] Zone 1. North Central 0.004 -0.015 -0.008 [0.01] [0.02] [0.02] Zone 2. North East 0.017 0.068** [0.02] [0.03] Table continues next page 103 Oaxaca: Self-employment (Continued) Full North South Urban Rural estimated estimated estimated estimated estimated coefficient coefficient coefficient coefficient coefficient Overall [standard [standard [standard [standard [standard error] error] error] error] error] Zone 3. North West -0.029 -0.001 0.017 0.054 [0.02] [0.01] [0.02] [0.04] Zone 4. South East -0.002 0.066* 0.025 0.007 [0.01] [0.04] [0.03] [0.02] Zone 5. South South 0.002 -0.017 -0.012 -0.080** [0.02] [0.02] [0.01] [0.04] Zone 6. South West 0.050* -0.060 0.000 [0.03] [.] [0.05] [.] Urban 0.001 -0.006 -0.004 0.000 [0.01] [0.01] [0.01] [.] Unexplained Age 2.236 -0.432 7.374** 9.886*** -0.910 [2.35] [2.70] [3.67] [2.70] [3.02] Age squared -0.702 0.336 -2.991* -4.410*** 0.932 [1.06] [1.18] [1.70] [1.26] [1.31] Married 0.114 -0.090 0.238 0.822* -0.726** [0.35] [0.34] [0.55] [0.48] [0.33] Polygamous -0.042 -0.164 0.023 0.179 -0.551** [0.18] [0.34] [0.07] [0.14] [0.23] Divorced 0.050 -0.011 0.168 0.212 -0.020 [0.05] [0.01] [0.13] [0.16] [0.02] Widowed 0.030 -0.013 0.107 0.188** -0.111 [0.05] [0.03] [0.10] [0.09] [0.07] Never attended school 0.142 -0.061 0.061 -0.016 0.709 [0.24] [0.31] [0.14] [0.12] [0.79] Did not complete primary 0.061 0.060 -0.007 -0.100 0.203 [0.07] [0.06] [0.11] [0.06] [0.17] Completed primary 0.143 0.011 0.168 -0.093 0.507 [0.21] [0.11] [0.39] [0.15] [0.56] Junior school certificate 0.039 -0.008 0.033 -0.050 0.146 [0.09] [0.03] [0.18] [0.11] [0.17] Senior school certiicate 0.395 0.044 0.694 0.012 0.685 [0.27] [0.11] [0.54] [0.25] [0.51] Postsecondary certificate 0.037 0.026 0.026 0.005 0.053 [0.06] [0.03] [0.11] [0.06] [0.10] Quranic and integrated Quranic education 0.103 0.018 0.007 0.482 [0.15] [0.21] [0.09] [0.47] Stopped activities because of illness in the last -0.049 -0.009 -0.095 -0.141* 0.055 4 weeks [0.04] [0.05] [0.06] [0.08] [0.06] Table continues next page 104 Oaxaca: Self-employment (Continued) Full North South Urban Rural estimated estimated estimated estimated estimated coefficient coefficient coefficient coefficient coefficient Overall [standard [standard [standard [standard [standard error] error] error] error] error] Access to internet -0.276** -0.185* -0.395* -0.166 -0.229* [0.11] [0.10] [0.22] [0.20] [0.12] Relationship with household is other than 0.577 -0.268 0.820** 0.653* 0.574 household head [0.36] [0.41] [0.41] [0.36] [0.58] Household size 0.842 0.594 0.720 1.035 0.309 [0.82] [1.04] [1.05] [1.17] [0.85] Number of adults -0.459 -0.054 -0.627 -1.224 0.477 [0.64] [0.80] [0.96] [1.13] [0.60] Dependency ratio (elderly and children under 11) -0.155 -0.101 -0.256 -0.117 0.076 [0.22] [0.24] [0.33] [0.35] [0.24] Use of formal credit -0.001 -0.000 -0.002 0.002 -0.013 [0.01] [0.00] [0.01] [0.00] [0.01] Use of informal credit -0.015 -0.004 -0.023 0.009 -0.023 [0.01] [0.00] [0.02] [0.01] [0.02] Value of your physical capital stock (tools, 0.439 0.660** 0.178 0.372 0.616 equipment, etc.) (HIS) [0.33] [0.31] [0.73] [0.68] [0.40] Value of your current stock of inputs or 0.138 0.231 0.043 -0.109 0.235 supplies (IHS) [0.17] [0.18] [0.24] [0.19] [0.24] Number of enterprises -1.439* -1.935*** 0.091 -1.746 -1.239 [0.75] [0.62] [2.11] [1.23] [0.77] Enterprise officially registered with the 0.037 0.030 0.042 0.074 0.021 government [0.03] [0.02] [0.06] [0.06] [0.03] Operated from home -0.078 0.051 0.149 0.102 -0.165 [0.32] [0.67] [0.25] [0.23] [0.61] Industrial site -0.000 0.003 -0.017 -0.015 0.011 [0.01] [0.00] [0.03] [0.02] [0.01] Traditional market -0.043 -0.037 0.153* 0.034 -0.048 [0.08] [0.07] [0.08] [0.08] [0.15] Roadside 0.053 0.026 0.167** 0.057 0.061 [0.05] [0.04] [0.08] [0.07] [0.06] Location of business: other -0.087 0.034 -0.026 0.059 -0.092 [0.11] [0.08] [0.17] [0.13] [0.12] Household savings -0.138 -0.162 -0.618 -0.293 0.203 [0.25] [0.22] [0.59] [0.52] [0.26] Formal loan 0.003 0.001 0.018 0.003 0.016 [0.01] [0.00] [0.02] [0.01] [0.01] Informal loan 0.052 0.002 0.033 0.034 0.068 [0.08] [0.11] [0.08] [0.07] [0.11] Family income source (agriculture or other 0.029 -0.037 0.152 -0.044 0.160 business) [0.06] [0.06] [0.12] [0.08] [0.10] Table continues next page 105 Oaxaca: Self-employment (Continued) Full North South Urban Rural estimated estimated estimated estimated estimated coefficient coefficient coefficient coefficient coefficient Overall [standard [standard [standard [standard [standard error] error] error] error] error] Relatives, friends, remittances 0.014 0.032 -0.039 -0.000 0.084 [0.06] [0.07] [0.11] [0.12] [0.07] Other 0.007 0.003 -0.001 0.004 0.003 [0.02] [0.01] [0.04] [0.03] [0.02] Sell product to final consumers 0.775 0.741 1.412 0.950 1.073 [0.56] [0.73] [1.30] [1.08] [0.75] Sell product to traders 0.038* 0.010 0.067 0.052* 0.074* [0.02] [0.02] [0.06] [0.03] [0.04] Sell product to small business 0.026* 0.003 0.065* 0.043 0.032 [0.01] [0.01] [0.04] [0.03] [0.02] Number of hired males -0.002 0.005 -0.003 0.008 0.001 [0.01] [0.01] [0.02] [0.02] [0.02] Number of hired females -0.001 -0.021 0.026 -0.003 -0.012 [0.02] [0.02] [0.03] [0.02] [0.02] Average number of hours worked by female 0.092 0.173 0.223 -0.829 0.357 household members [0.25] [0.23] [0.71] [0.61] [0.33] Average number of hours worked by male 0.476 0.458 -1.102 1.098 0.849 household members [0.85] [0.93] [2.36] [2.33] [0.73] Average number of hours worked by child -0.001 -0.011 0.010 0.006 0.002 household members (<11) [0.01] [0.02] [0.01] [0.02] [0.02] Average number of hours worked by child -0.665 -0.696 0.791 -0.540 -1.272 household members (11–15) [1.04] [1.04] [2.89] [2.03] [0.99] Industry: agriculture production 0.042 0.011 0.071 0.046 0.032 [0.04] [0.04] [0.08] [0.04] [0.06] Industry: manufacturing 0.024 -0.002 0.004 0.221 0.079 [0.13] [0.11] [0.31] [0.27] [0.11] Industry: services -0.039 0.053 -0.254 0.074 0.134 [0.24] [0.18] [0.71] [0.49] [0.23] Industry: trade -0.172 -0.136 -0.585 0.472 -0.167 [0.40] [0.31] [1.33] [0.85] [0.38] Industry: other 0.027 0.045* -0.005 0.056 0.035 [0.03] [0.02] [0.09] [0.06] [0.03] Zone 1. North Central 0.096 0.018 0.015 -0.122** [0.06] [0.06] [0.06] [0.05] Zone 2. North East 0.140** 0.081 -0.245 -0.012 [0.07] [0.10] [0.18] [0.09] Zone 3. North West 0.169 -0.206 -0.289 -0.381 [0.15] [0.22] [0.19] [0.23] Table continues next page 106 Oaxaca: Self-employment (Continued) Full North South Urban Rural estimated estimated estimated estimated estimated coefficient coefficient coefficient coefficient coefficient Overall [standard [standard [standard [standard [standard error] error] error] error] error] Zone 4. South East 0.070 0.239 -0.090 -0.160** [0.05] [0.19] [0.15] [0.08] Zone 5. South South 0.074 0.287** -0.043 -0.117 [0.08] [0.13] [0.06] [0.12] Zone 6. South West 0.065 0.253 -0.266 -0.021 [0.07] [0.16] [0.22] [0.02] Urban 0.033 -0.149* 0.348 [0.11] [0.09] [0.23] Constant -4.456** 0.301 -9.443*** -7.343*** -4.004 [2.15] [2.07] [3.26] [2.48] [3.27] N 3040.000 1644.000 1396.000 1149.000 1891.000 Note: ***, **, and * indicate significance at the 1, 5, and 10 percent critical level. IHS indicates variables that have been transformed with the inverse hyperbolic sine transformation. Appendix 2D: Wages Balance table: Wages (1) (2) t-test Female plot Male plot difference manager manager Variables N Mean/SE N Mean/SE (1)−(2) Hourly wage (winsorized) 372 278.24 715 342.76 -64.52** [20.84] [20.16] Hourly wage (IHS) 372 5.92 715 6.17 -0.24*** [0.06] [0.05] Age 372 36.73 715 39.08 -2.35** [0.80] [0.57] Age squared 372 1473.71 715 1649.55 -175.83** [59.44] [45.55] Married 372 0.54 715 0.59 -0.04 [0.04] [0.03] Polygamous 372 0.11 715 0.14 -0.03 [0.02] [0.02] Divorced 372 0.04 715 0.02 0.02 [0.02] [0.01] Widowed 372 0.09 715 0.00 0.09*** [0.02] [0.00] Never attended school 372 0.03 715 0.03 0.00 [0.01] [0.01] Table continues next page 107 Balance table: Wages (Continued) (1) (2 ) t-test Female plot Male plot difference manager manager Variables N Mean/SE N Mean/SE (1)−(2) Did not complete primary 372 0.02 715 0.01 0.00 [0.01] [0.01] Completed primary 372 0.04 715 0.13 -0.08*** [0.01] [0.02] Junior school certificate 372 0.01 715 0.03 -0.02** [0.01] [0.01] Senior school certificate 372 0.28 715 0.34 -0.07 [0.03] [0.03] Postsecondary certificate 372 0.30 715 0.19 0.11*** [0.03] [0.02] University or college degree 372 0.30 715 0.23 0.06 [0.03] [0.02] Quranic and integrated Quranic education 372 0.03 715 0.03 -0.00 [0.01] [0.01] Nonhousehold head 372 0.84 715 0.20 0.64*** [0.03] [0.02] Household size 372 6.44 715 6.34 0.10 [0.30] [0.24] Dependency ratio (elderly and children under 11) 372 0.55 715 0.61 -0.06 [0.04] [0.03] Access to internet 372 0.61 715 0.58 0.04 [0.04] [0.03] Stopped activities because of illness in the last 372 0.11 715 0.07 0.04 4 weeks [0.03] [0.02] How many hours per week did you normally 372 34.99 715 39.38 -4.39*** work on this? [1.15] [1.16] Received any in-kind payment or allowance 372 0.14 715 0.18 -0.04 for work [0.02] [0.02] Number of coworkers 372 2.95 715 2.83 0.12 [0.08] [0.07] Government (state, local, federal) 372 0.47 715 0.36 0.11** [0.04] [0.03] Employer is other 372 0.04 715 0.06 -0.02 [0.01] [0.01] Enrolled in pension scheme 372 0.25 715 0.21 0.04 [0.03] [0.02] Has a written contract/agreement or letter of 372 0.66 715 0.39 0.27 appointment [0.15] [0.20] Employer provides health insurance coverage 372 0.06 715 -0.24 0.30 [0.54] [0.59] Table continues next page 108 Balance table: Wages (Continued) (1) (2 ) t-test Female plot Male plot difference manager manager Variables N Mean/SE N Mean/SE (1)−(2) Industry: agriculture production 372 0.06 715 0.07 -0.01 [0.02] [0.02] Industry: services 372 0.36 715 0.42 -0.06 [0.04] [0.03] Industry: education 372 0.42 715 0.18 0.24*** [0.04] [0.02] Industry: other 372 0.13 715 0.25 -0.12*** [0.02] [0.02] Occupation: managers, professionals 372 0.77 715 0.52 0.25*** [0.03] [0.03] Occupation: support, service, and sales 372 0.12 715 0.08 0.04 [0.02] [0.01] Occupation: skilled agriculture, crafts, and 372 0.05 715 0.19 -0.14*** machine operators [0.02] [0.02] Zone 1. North Central 372 0.14 715 0.16 -0.02 [0.02] [0.02] Zone 2. North East 372 0.06 715 0.12 -0.07** [0.02] [0.02] Zone 3. North West 372 0.11 715 0.16 -0.05* [0.02] [0.02] Zone 4. South East 372 0.24 715 0.12 0.12*** [0.03] [0.01] Zone 5. South South 372 0.26 715 0.24 0.02 [0.03] [0.03] Zone 6. South West 372 0.19 715 0.20 -0.01 [0.03] [0.02] Occupation: elementary occupations 372 0.07 715 0.22 -0.15*** [0.02] [0.02] Industry: manufacturing 372 0.03 715 0.08 -0.05** [0.01] [0.01] Number of adults 372 3.98 715 3.85 0.14 [0.16] [0.14] Sample: full Note: The value displayed for t-tests are the differences in the means across the groups. SE refers to standard error. ***, **, and * indicate significance at the 1, 5, and 10 percent critical level. IHS indicates variables that have been transformed with the inverse hyperbolic sine transformation. 109 Cross-section: Wages Full Public Public (1) (2) (3) hourly wage hourly wage hourly wage Variables (IHS) (IHS) (IHS) Female member -0.321*** -0.307** -0.264*** (0.090) (0.121) (0.095) Age 0.059*** 0.087** 0.056** (0.021) (0.035) (0.026) Age squared -0.001** -0.001* -0.001* (0.000) (0.000) (0.000) Married 0.192** -0.053 0.319*** (0.076) (0.143) (0.098) Polygamous 0.153 -0.070 0.283 (0.142) (0.173) (0.225) Divorced 0.445*** 0.381** 0.537*** (0.109) (0.162) (0.148) Widowed 0.268** -0.043 0.584** (0.134) (0.174) (0.254) Never attended school -0.600*** -0.806*** -0.718*** (0.200) (0.270) (0.214) Did not complete primary -0.528** -0.637** (0.261) (0.291) Completed primary -0.573*** -0.824*** -0.702*** (0.163) (0.190) (0.175) Senior school certificate -0.159 -0.519** -0.221 (0.158) (0.202) (0.159) Postsecondary certificate -0.052 -0.437** -0.077 (0.174) (0.183) (0.182) University or college degree 0.361** -0.138 0.393** (0.175) (0.204) (0.196) Quranic and integrated Quranic education -0.536* -1.250*** -0.475 (0.288) (0.285) (0.312) Nonhousehold head 0.099 0.233* -0.060 (0.101) (0.134) (0.113) Household size 0.017 0.009 0.040** (0.011) (0.010) (0.019) Dependency ratio (elderly and children under 11) -0.085 -0.101 -0.085 (0.064) (0.074) (0.089) Access to internet 0.153* 0.197*** 0.155 (0.081) (0.075) (0.099) Stopped activities because of illness in the last 4 weeks 0.104 -0.127* 0.229* (0.082) (0.073) (0.136) Table continues next page 110 Cross-section: Wages (Continued) Full Public Public (1) (2) (3) hourly wage hourly wage hourly wage Variables (IHS) (IHS) (IHS) How many hours per week did you normally work on this? 0.000 0.001 -0.002 (0.002) (0.003) (0.002) Received any in-kind payment or allowance for work 0.138* 0.233* 0.113 (0.082) (0.121) (0.079) Number of coworkers 0.086** 0.075* 0.069 (0.033) (0.045) (0.044) Government (state, local, federal) 0.197** (0.094) Employer is other -0.297*** (0.110) Enrolled in pension scheme 0.087 0.036 0.248 (0.082) (0.079) (0.151) Has a written contract/agreement or letter of appointment -0.001 0.005** -0.005 (0.002) (0.003) (0.004) Employer provides health insurance coverage 0.001 0.001 0.000 (0.002) (0.002) (0.002) Industry: agriculture production -0.044 -0.619** -0.201 (0.199) (0.306) (0.227) Industry: services -0.106 -0.985*** -0.004 (0.097) (0.148) (0.104) Industry: education -0.316*** -1.018*** -0.551*** (0.119) (0.164) (0.154) Industry: other 0.221** -0.602*** 0.221* (0.107) (0.189) (0.116) Occupation: managers, professionals -0.011 0.480*** -0.118 (0.130) (0.127) (0.157) Occupation: support, service, and sales -0.301** -0.045 -0.367** (0.134) (0.147) (0.168) Occupation: skilled agriculture, crafts, and machine operators 0.157 0.081 0.076 (0.122) (0.190) (0.146) Zone 1. North Central -0.227** -0.106 0.105 (0.113) (0.114) (0.200) Zone 2. North East -0.521*** -0.328** -0.237 (0.121) (0.129) (0.213) Zone 3. North West -0.091 0.096 (0.122) (0.129) Zone 4. South East 0.233 0.209 (0.154) (0.185) Zone 5. South South -0.053 0.131 0.186 (0.094) (0.130) (0.183) Table continues next page 111 Cross-section: Wages (Continued) Full Public Public (1) (2) (3) hourly wage hourly wage hourly wage Variables (IHS) (IHS) (IHS) Zone 6. South West -0.188 0.108 (0.123) (0.188) Junior school certificate -0.539* * (0.260) Constant 4.355*** 4.475*** 4.431*** (0.480) (0.774) (0.563) Observations 1,087 451 584 R-squared 0.493 0.526 0.452 Percentage gap -0.28 -0.27 -0.24 Education elasticity -0.276 -0.644 -0.430 Services elasticity -0.105 -0.631 -0.009 Government elasticity 0.213 0 0 Adjusted R-squared 0.474 0.483 0.415 Note: Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 IHS indicates variables that have been transformed with the inverse hyperbolic sine transformation. Oaxaca: Wages Decomposition of the gender differential in wages Full Public Private estimated estimated estimated coefficient coefficient coefficient Overall [standard [standard [standard error] error] error] Group 1 5.921*** 6.470*** 5.463*** [0.08] [0.10] [0.07] Group 2 6.166*** 6.511*** 5.977*** [0.05] [0.07] [0.06] Difference -0.245*** -0.041 -0.514*** [0.08] [0.10] [0.09] Explained 0.077 0.266** -0.250** [0.10] [0.13] [0.11] Unexplained -0.321*** -0.307*** -0.264*** [0.09] [0.12] [0.09] Explained Table continues next page 112 Oaxaca: Wages. (Continued) Full Public Private estimated estimated estimated coefficient coefficient coefficient [standard [standard [standard Overall error] error] error] Age -0.140* -0.174 -0.219* [0.08] [0.14] [0.12] Age squared 0.093 0.120 0.164 [0.06] [0.11] [0.10] Married -0.009 -0.001 -0.026 [0.01] [0.00] [0.02] Polygamous -0.004 0.009 0.002 [0.01] [0.02] [0.02] Divorced 0.011 0.021 -0.001 [0.01] [0.02] [0.01] Widowed 0.023* -0.006 0.028* [0.01] [0.02] [0.02] Never attended school -0.000 0.004 [0.01] [0.00] Did not complete primary -0.001 [0.00] Completed primary 0.047*** 0.019 0.008 [0.02] [0.02] [0.03] Junior school certificate -0.030* [0.02] Senior school certificate 0.011 0.008 -0.015 [0.01] [0.02] [0.03] Postsecondary certificate -0.006 -0.012 0.084 [0.02] [0.03] [0.05] University or college degree 0.023 -0.004 0.044 [0.02] [0.01] [0.04] Quranic and integrated Quranic education 0.000 0.023 0.003 [0.02] [0.02] [0.01] Nonhousehold head 0.063 0.149* -0.039 [0.06] [0.08] [0.07] Household size 0.002 -0.010 0.024 [0.01] [0.01] [0.03] Dependency ratio (elderly and children under 11) 0.005 0.022 -0.001 [0.01] [0.02] [0.01] Access to internet 0.006 -0.007 0.012 [0.01] [0.01] [0.01] Stopped activities because of illness in the last 4 weeks 0.004 -0.009 -0.004 [0.00] [0.01] [0.01] How many hours per week did you normally work on this? -0.001 -0.001 0.008 [0.01] [0.00] [0.01] Table continues next page 113 Oaxaca: Wages. (Continued) Full Public Private estimated estimated estimated coefficient coefficient coefficient [standard [standard [standard Overall error] error] error] Received any in-kind payment or allowance for work -0.005 -0.023 0.004 [0.01] [0.02] [0.01] Number of coworkers 0.011 0.001 -0.003 [0.01] [0.01] [0.01] Government (state, local, federal) 0.023 [0.01] Employer is other 0.005 [0.01] Enrolled in pension scheme 0.003 -0.006 [0.00] [0.01] Has a written contract/agreement or letter of appointment -0.000 0.003 -0.000 [0.00] [0.00] [0.00] Employer provides health insurance coverage Industry: agriculture production 0.002 0.001 [0.02] [0.01] Industry: services 0.007 0.050 0.000 [0.01] [0.05] [0.01] Industry: education -0.076** -0.031 -0.196*** [0.03] [0.06] [0.06] Industry: other -0.027* -0.008 -0.047* [0.01] [0.03] [0.03] Occupation: managers, professionals -0.003 0.036 -0.040 [0.03] [0.02] [0.05] Occupation: support, service, and sales -0.012 -0.001 -0.029 [0.01] [0.00] [0.02] Occupation: skilled agriculture, crafts, and machine operators -0.022 -0.004 -0.015 [0.02] [0.01] [0.03] Zone 1. North Central 0.004 -0.006 -0.001 [0.01] [0.01] [0.01] Zone 2. North East 0.035* 0.014 [0.02] [0.01] Zone 3. North West 0.005 -0.040* [0.01] [0.02] Zone 4. South East 0.093** 0.020 [0.04] [0.02] Zone 5. South South -0.001 0.031 0.004 [0.00] [0.04] [0.01] Zone 6. South West 0.001 0.011 0.001 [0.01] [0.02] [0.01] Table continues next page 114 Oaxaca: Wages. (Continued) Full Public Private estimated estimated estimated coefficient coefficient coefficient [standard [standard [standard Overall error] error] error] Unexplained Age -2.512* -2.504 -1.304 [1.37] [2.55] [1.97] Age squared 1.551** 1.692 0.718 [0.71] [1.34] [0.92] Married -0.202** -0.097 -0.162 [0.09] [0.17] [0.11] Polygamous -0.042 -0.028 -0.050 [0.03] [0.06] [0.04] Divorced -0.006 -0.007 -0.009 [0.01] [0.01] [0.01] Widowed -0.042*** 0.009 -0.045** [0.01] [0.02] [0.02] Never attended school 0.008 -0.004 -0.034 [0.01] [0.00] [0.02] Did not complete primary 0.006 0.004 -0.043 [0.01] [0.00] [0.03] Completed primary -0.029 0.007 -0.018 [0.03] [0.03] [0.03] Junior school certificate -0.024** -0.001 -0.006 [0.01] [0.01] [0.02] Senior school certificate 0.009 0.016 -0.614** [0.10] [0.06] [0.26] Postsecondary certificate 0.164* 0.130 -0.308** [0.09] [0.17] [0.14] University or college degree 0.168* 0.327 -0.231** [0.10] [0.22] [0.11] Quranic and integrated Quranic education 0.035 -0.009 0.010 [0.04] [0.02] [0.02] Nonhousehold head -0.026 -0.038 0.154 [0.11] [0.15] [0.18] Household size -0.002 -0.021 -0.065 [0.11] [0.15] [0.17] Dependency ratio (elderly and children under 11) 0.011 0.234*** -0.175** [0.05] [0.07] [0.08] Access to internet -0.028 -0.024 -0.061 [0.07] [0.11] [0.07] Stopped activities because of illness in the last 4 weeks 0.026 0.057** 0.024 [0.02] [0.02] [0.02] Table continues next page 115 Oaxaca: Wages. (Continued) Full Public Private estimated estimated estimated coefficient coefficient coefficient [standard [standard [standard Overall error] error] error] How many hours per week did you normally work on this? 0.095 0.026 0.010 [0.12] [0.15] [0.18] Received any in-kind payment or allowance for work -0.004 -0.056** 0.043* [0.02] [0.03] [0.03] Number of coworkers -0.031 0.253 -0.052 [0.17] [0.27] [0.22] Government (state, local, federal) 0.096* [0.05] Employer is other 0.006 [0.01] Enrolled in pension scheme 0.035 0.030 0.037** [0.03] [0.06] [0.02] Has a written contract/agreement or letter of appointment 0.001 0.128 -0.003 [0.00] [0.09] [0.00] Employer provides health insurance coverage -0.004 -0.002 [0.00] [0.00] Industry: agriculture production 0.033 0.016 0.093 [0.02] [0.02] [0.06] Industry: services -0.167** 0.175 -0.178** [0.07] [0.13] [0.08] Industry: education -0.103 0.205 -0.067 [0.08] [0.13] [0.09] Industry: other -0.058* 0.065 -0.058 [0.03] [0.05] [0.04] Occupation: managers, professionals -0.128 -0.196 0.077 [0.16] [0.19] [0.15] Occupation: support, service, and sales -0.013 -0.006 0.002 [0.02] [0.02] [0.04] Occupation: skilled agriculture, crafts, and machine operators -0.008 0.010 -0.014 [0.02] [0.01] [0.04] Zone 1. North Central 0.043 -0.026 0.066 [0.03] [0.04] [0.05] Zone 2. North East -0.014 -0.041* -0.014 [0.02] [0.02] [0.01] Zone 3. North West 0.008 -0.040 -0.027 [0.02] [0.03] [0.03] Zone 4. South East 0.043 -0.125** 0.095 [0.04] [0.06] [0.07] Zone 5. South South 0.023 -0.068 0.115 [0.04] [0.04] [0.10] Table continues next page 116 Oaxaca: Wages. (Continued) Full Public Private estimated estimated estimated coefficient coefficient coefficient [standard [standard [standard Overall error] error] error] Zone 6. South West 0.041 -0.045** 0.084 [0.03] [0.02] [0.09] Constant 0.717 -0.352 1.749 [0.82] [1.34] [1.30] N 1087.000 451.000 584.000 Note: ***, **, and * indicate significance at the 1, 5, and 10 percent critical level. IHS indicates variables that have been transformed with the inverse hyperbolic sine transformation. 117 Appendix 3: Technical Appendix Costing the gender gap in agricultural calculate the percent of additional agricultural output productivity from closing the gap. We define agricultural productivity as the average The final step includes computing the size of the gap monetary value of agricultural output produced per relative to total gross domestic product (GDP). To do unit of land managed (in hectares). While agricul- this, we need to know what fraction of agricultural GDP tural productivity is typically measured on a house- comes from crop production, and, in turn, the fraction hold-level basis, here we look at the plot level and of GDP that comes from agriculture. In Nigeria, crop identify whether the plot manager is a man or a wom- production represents 88 percent of agricultural GDP, an to measure the plot-level gender differences. The and agriculture represents 21 percent of total GDP. plot-level quantity of output is based on farm manag- The product of these two shares and of the percent in- ers’ estimates while the land area is based mostly from crease in output from closing the gender gap yields the GPS measures.XVII estimated increase in total earnings due to closing the gender gap in agricultural productivity. Furthermore, The difference between the average monetary val- because growth in the agricultural sector influences ues of output on land managed by men and women other sectors of the economy, the cost of the gender constitutes the unconditional gender gap in agricul- gap is likely higher than just the forgone agricultural tural productivity.XVIII We use available data on the GDP. To take this into account, we use an estimate of total number of hectares of arable land in Nigeria to the multiplier between agricultural sector growth and estimate the total harvested value of plots managed the rest of the economy obtained from a study by the by women and men—calculated separately based on African Development Bank (Alemu 2015).XIX, 204 each gender’s agricultural productivity and hectares of land managed. This yields the total harvested val- ue in the presence of the gender gap. Then, to calcu- Costing the gender gap in firm profits late the total harvested value in the absence of the gender gap, we apply men’s agricultural productivity To estimate the forgone GDP from the gender gap in to land managed by women and add it to the total firm profits, first we measure the difference in total output of land managed by men. The size of land man- monthly profits between male- and female-owned aged by women is obtained by using the fraction of businesses as the unconditional gender gap in profits. all plots managed by women and the average size of We apply this figure to the total number of self-em- plots managed by women. In Nigeria, women manage ployed men and women to calculate the total month- 16 percent of plots, and women’s plots are on average ly profits nationally in the presence and absence of 61 percent the size of men’s plots. Therefore, women the gender gap in profits. The World Bank reports that manage only 10 percent of agricultural land in Nige- of the 46 percent of women in the labor force, 87 per- ria. Using the two estimates of total harvested value in cent are self-employed. However, because the World the presence and absence of the gender gap, we can Bank statistic includes self-employed individuals em- ployed in farming or agricultural self-production, we use the General Household Survey (GHS) data used XVII Land area was based on GPS measures for 95 percent of the sample and plot manager estimates for 5 percent of the sample. XVIII In this analysis, the unconditional gender gap refers to the XIX The multipliers used in this analysis were extracted from African simple difference in averages between men and women in terms Development Bank (Alemu 2015) and were calculated based on a of agricultural productivity, firm sales, and wages. Social Accounting Matrix for Nigeria, as detailed by the author. 118 throughout this report to estimate the proportion of percent increase in annual profits from closing the individuals that are self-employed excluding those in gender gap, the proportion of self-employed women agriculture, to avoid double-counting earnings from in that subsector, and the sectoral GDP. We then apply farming.XX In Nigeria, 68 percent of women who are the multiplier between each subsector and the rest of self-employed are in sectors other than agriculture. the economy to calculate the total benefit added to GDP. Next, we calculate women’s total aggregate month- ly profits as the product of women’s average total monthly profits and the number of self-employed Costing the gender gap in wage earnings women entrepreneurs. We similarly calculate men’s total aggregate monthly profits. The sum of these two Following a similar approach to the one used for the figures constitutes the total monthly profits nation- gender gap in firm profits, we identify both the total ally in the presence of the gender gap. To calculate populations in wage employment and the proportion the total monthly profits nationally in the absence of of each of the subsectors that is salaried to determine the gender gap, we apply men’s average total monthly the forgone national income from the gender gaps in profits to both self-employed men and self-employed wages. First, we measure the difference in average to- women. We multiply these values by 12 to get the an- tal hourly wages between salaried men and women nual figures. as the unconditional gender gap in wages. This figure is applied to the total number of salaried men and Then, to estimate the impact of closing the gender women in the labor force to calculate the total hourly gaps in firm profits on national income, we identify wages nationally in the presence and absence of the the proportion of each subsector of the economy that gender gap. The World Bank cites that 46 percent of is involved in self-employment from the GHS data. We Nigerian women participate in the labor force and 13 categorize self-employed entrepreneurs by their sub- percent of women labor force participants are sala- sector of work (manufacturing, mining, construction, ried, indicating that approximately 6 percent of Nige- utility, transport, trade, or other services) using the cat- rian women or 3.3 million women are employed in the egories outlined in African Development Bank (2015): formal wage sector. 60.5 percent are employed in trade, 30.4 percent are employed in other services, and 9.1 percent are em- We calculate women’s total hourly wages nationally ployed in manufacturing.XXI Based on these shares, we as the product of women’s average total hourly wages estimate the increase in each subsector’s GDP due to and the number of women employed in the formal closing the gender gap in profits as the product of the wage sector. We similarly calculate men’s total hour- ly wages nationally. The sum of these two figures is XX The World Bank statistic is based on the International Labour the total hourly wages nationally in the presence of Organization definition, which refers to self-employed workers as workers who hold “jobs where the remuneration is directly the gender gap. To calculate the total hourly wages dependent upon the profits derived from the goods and services nationally in the absence of the gender gap, we ap- produced.” They fall into four subcategories, including employers, ply men’s average total hourly wages to both salaried own-account workers, members of producers’ cooperatives, and contributing family workers. Based on this definition, we separate men and salaried women. From these two figures, we those self-employed individuals working in agriculture from those estimate the percent of additional hourly wages from working in entrepreneurship for the purposes of the analysis in part IV. closing the gender wage gap. Using the average num- XXI Respondents in self-employment do not work in the following ber of hours worked in the last week (28.6 hours) and subsectors: mining, construction, utility, or transport. 119 the average number of working weeks per year (48 weeks), we convert these to annual figures.XXII Finally, we identify the proportion of each subsector of the economy involved in salaried work from the GHS data: 86.1 percent of the labor force employed in other services is salaried compared to 6.6 percent in trade, 4.3 percent in manufacturing, 1.2 percent in construction, 0.9 percent in utility, and 0.9 percent in transport.XXIII Based on these shares, we estimate the increase in each subsector’s GDP due to closing the gender gap in wages as the product of the percent in- crease in annual wages from closing the gender gap, the proportion of the subsector that is salaried, and the sectoral GDP. We then apply the multiplier be- tween each subsector and the rest of the economy to calculate the total increase in sectoral GDP and the total increase in overall GDP. XXII The average number of hours worked is the mean reported in the sample of analysis using the GHS data, and the average number of working weeks per year is the average that was reported at the national level. XXIII Respondents in wage employment do not work in mining. 120 Endnotes 1 World Bank. 2018. “Human Capital Index: Nigeria Brief.” World Bank, 10 United Nations Security Council. 2017. “Report of the Secretary-Gen- Washington, DC. eral on Children and Armed Conflict in Nigeria.” Vol (S/2017/304). 2 World Bank. World Development Indicator Database. 11 For example, in 2014, Boko Haram abducted 276 Chibok schoolgirls in Borno State, most of whom were Christian, and attempted to convert them to Islam (Inskeep 2019); Inskeep, Steve. 2019. “Nigerian School- 3 World Bank Group. 2019. Profiting from Parity: Unlocking the Po- girls’ Abduction Told in ‘Beneath the Tamarind Tree’.” National Public tential of Women’s Business in Africa. World Bank, Washington, DC. Radio, July 8, 2019. https://www.npr.org/2019/07/08/739416281/ni- gerian-schoolgirls-abduction-toldin-beneath-the-tamarind-tree. 4 UNESCO (United Nations Educational, Scientific and Cultur- al Organization). 2016. "Secondary School Enrollment Rates." 12 GCPEA (Global Coalition to Protect Education from Attack). 2018. “I http://tcg.uis.unesco.org/4-1-4-completion-rate-primary-low- Will Never Go Back to School: The Impact of Attacks on Education er-secondary-upper-secondary/. for Nigerian Women and Girls”; Inskeep, Steve. 2019. “Nigerian School- girls’ Abduction Told in ‘Beneath the Tamarind Tree’.” National Public 5 Aromolaran, Adebayo B. 2004. “Female Schooling, Non-Market Radio, July 8, 2019. https://www.npr.org/2019/07/08/739416281/ni- Productivity, and Labor Market Participation in Nigeria.” Center gerian-schoolgirls-abduction-told-in-beneath-the-tamarind-tree. Discussion Paper No. 879, Yale University, New Haven. 13 Bertoni, Eleonora, Michele Di Maio, Vasco Molini, and Roberto Nis- 6 Adamu, Zainab Inuwa. 2014. “Gender Inequality: Issues and Chal- tico. 2018. “Education Is Forbidden: The Effect of the Boko Haram lenges in Basic Education in Nigeria.” The Business and Manage- Conflict on Education in North-East Nigeria.” Working Paper 495, ment Review 5 (4): 112–117; UNICEF (United Nations Educational, Centre for Studies in Economics and Finance, Naples, Italy. Scientific and Cultural Organization), Nigeria Country Office. 2007. “Information Sheet: Girls’ Education.” 14 NPC (National Population Commission) [Nigeria] and ICF. 2019. Nigeria Demographic and Health Survey 2018. Abuja, Nigeria, and 7 Elder, Sara, and Koko Siaka Koné. 2014. “Labour Market Transitions Rockville, Maryland, USA: NPC and ICF. of Young Women and Men in Sub-Saharan Africa.” ILO Work4Youth Publication Series 9. 15 NPC (National Population Commission) [Nigeria] and ICF. 2019. Nigeria Demographic and Health Survey 2018. Abuja, Nigeria, and 8 Adamu, Zainab Inuwa. 2014. “Gender Inequality: Issues and Chal- Rockville, Maryland, USA: NPC and ICF. lenges in Basic Education in Nigeria.” The Business and Manage- ment Review 5 (4): 112–117; UNICEF (United Nations Educational, 16 Dieterich, Christine, Anni Huang, and Alun H. Thomas. 2016. “Wom- Scientific and Cultural Organization), Nigeria Country Office. 2007. en’s Opportunities and Challenges in Sub-Saharan African Job “Information Sheet: Girls’ Education”; Kritz, Mary, and Paulina Mak- Markets.” IMF Working Paper 16/118, International Monetary Fund, inwa-Adebusoye. 1999. “Determinants of Women’s Decision-Making Washington, DC. Authority in Nigeria: The Ethnic Dimension.” Sociological Forum 14(3): 399–424. 17 Onagoruwa, Adenike, and Quentin Wodon. 2017. “Child Marriage and Fertility in Nigeria.” HNPGP Knowledge Brief, World Bank, 9 UNHCR (United Nations Refugee Agency). 2019. “Nigeria Emergency, Washington, DC. May 31, 2019.” http://www.unhcr.org/en-us/nigeria-emergency.html. 121 18 Bloom, David, David Canning, Gunther Fink, and Jocelyn Finlay. 2009. 31 Aluko, Bioye Tajudeen, and Abdul-Rasheed Amidu. 2006. Women “Fertility, Female Labor Force Participation and the Demographic and Land Rights Reforms in Nigera. Promoting Land Administration Dividend.” Journal of Economic Growth 14: 79–101. and Good Governance, 5th FIG Regional Conference, Accra; British Council of Nigeria. 2012. “Gender in Nigeria Report 2012: Improving the Lives of Girls and Women in Nigeria.” Issues Policies Action 2. 19 Longwe, Abiba, Jereon Smits, and Eelke de Jong. 2013. “Number and Spacing of Children and Women’s Employment in Africa.” NiCE Working Paper 13–103. 32 FAO (Food and Agriculture Organization). 2018b.“The Gender Gap in Land Rights.” Policy Brief, FAO, Rome, Italy. 20 Osili, Una Okonkwo, and Bridget Terry Long. 2008. “Does Female Schooling Reduce Fertility? Evidence from Nigeria.” Journal of 33 Slavchevska, Vanya, Ana Paula De La O Campos, Chiara Brunelli, and Development Economics 87: 57–75. Cheryl Doss. 2016. “Beyond Ownership: Women’s and Men’s Land Rights in Sub-Saharan Africa.” Working Paper, Food and Agriculture Organization of the United Nations, Rome. 21 NPC (National Population Commission) [Nigeria] and ICF. 2019. Nigeria Demographic and Health Survey 2018. Abuja, Nigeria, and Rockville, Maryland, USA: NPC and ICF. 34 Binswanger, Hans P., and Mark R. Rosenzweig. 1986. “Behavioural and Material Determinants of Production Relations in Agriculture.” Journal of Development Studies 22 (3): 503–539. 22 World Bank. 2011. World Development Report 2012: Gender Equality and Development. Washington, DC: World Bank. 35 Goldstein, Markus, and Christopher Udry. 2008. “The Profits of Power: Land Rights and Agricultural Investment in Ghana.” Journal 23 Chakravarty, Subha, Smita Das, and Julia Vaillant. 2017. “Gender and of Political Economy 116 (6): 981–1022; Ali, Daniel Ayalew, Klaus Youth Employment in Sub-Saharan Africa: A Review of Constraints Deininger, and Markus Goldstein. 2014. “Environmental and Gender and Effective Interventions.” Policy Research Working Paper 8245, Impacts of Land Tenure Regularization in Africa: Pilot Evidence from World Bank, Washington, DC. Rwanda.” Journal of Development Economics 110: 262–75. 24 Osili, Una Okonkwo, and Bridget Terry Long. 2008. “Does Female 36 British Council of Nigeria. 2012. “Gender in Nigeria Report 2012: Schooling Reduce Fertility? Evidence from Nigeria.” Journal of Improving the Lives of Girls and Women in Nigeria.” Issues Policies Development Economics 87: 57–75; Akpa Onoja, Matthew, and Action 2. Osayomore Ikpotokin. 2012. “Modeling the Determinants of Fer- tility among Women of Childbearing Age in Nigeria: Analysis Using Generalized Linear Modeling Approach.” International Journal of 37 Chikwendu, D.O., and J.O. Arokoyo. 1997. “Women and Sustaina- Humanities and Social Sciences, 2 (18): 106–107. ble Agricultural Development in Nigeria.” Journal of Sustainable Agriculture 11 (1): 53–69; British Council of Nigeria. 2012. “Gender in Nigeria Report 2012: Improving the Lives of Girls and Women in 25 World Bank. World Development Indicator Database. Nigeria.” Issues Policies Action 2. 26 Buehren, Niklas. 2019. “Gender and Agriculture in Sub-Saharan 38 Deininger, Klaus, and Hans Binswanger. 1999. “The Evolution of Africa: Review of Constraints and Effective Interventions.” Africa the World Bank’s Land Policy: Principles, Experience, and Future Gender Innovation Lab, World Bank, Washington, DC. Challenges.” The World Bank Research Observer 14 (2): 247–276. 27 Buehren, Niklas. 2019. “Gender and Agriculture in Sub-Saharan 39 Deere, Carmen Diana, Abena D. Oduro, Hema Swaminathan, and Africa: Review of Constraints and Effective Interventions.” Africa Cheryl Doss. 2013. “Property Rights and the Gender Distribution Gender Innovation Lab, World Bank, Washington, DC. of Wealth in Ecuador, Ghana and India.” The Journal of Economic Inequality 11 (2): 249–65; Deininger, Klaus, Aparajita Goyal, and Hari 28 National Bureau of Statistics. 2019. “2018 Statistical Report on Nagarajan. 2013. “Women’s Inheritance Rights and Intergenerational Women and Men in Nigeria.” Transmission of Resources in India.” Journal of Human Resources 48 (1): 114–41; Panda, Pradeep, and Bina Agarwal. 2005. “Marital Violence, Human Development and Women’s Property Status in 29 Hsieh, C. T., E. Hurst, C. I. Jones, and P.J. Klenow. 2019. “The Allocation India.” World Development 33(5): 823–850; Allendorf, K. 2007. “Do of Talent and US Economic Growth.” Econometrica 87 (5): 1439–1474. Women’s Land Rights Promote Empowerment and Child Health in Nepal?” World Development 35 (11): 1975–1988; Strickland, R. S. 30 Meinzen-Dick, Ruth, Nancy Johnson, Agnes R. Quisumbing, Jemimah 2004. “To Have and to Hold: Women’s Property and Inheritance Njuki, Julia A. Behrman, Deobrah Rubin, Amber Peterman, and Rights in the Context of HIV/AIDS in Sub-Saharan Africa.” Working Elizabeth Waithanji. 2014. “The Gender Asset Gap and Its Impli- Paper, International Center for Research on Women. cations for Agricultural and Rural Development.” In In Gender in Agriculture: Closing the Knowledge Gap, edited by Agnes R. Qui- 40 Slavchevska, Vanya, Ana Paula De La O Campos, Chiara Brunelli, and sumbing, Ruth Meinzen-Dick, Terri L. Raney, Andre Croppenstedt, Cheryl Doss. 2016. “Beyond Ownership: Women’s and Men’s Land Julia A. Behrman, and Amber Peterman, 91–115. Rome: Food and Rights in Sub-Saharan Africa.” Working Paper, Food and Agriculture Agriculture Organization of the United Nations; British Council of Organization of the United Nations, Rome. Nigeria. 2012. “Gender in Nigeria Report 2012: Improving the Lives of Girls and Women in Nigeria.” Issues Policies Action 2; Agarwal, Bina. 1994. “Gender and Command over Property: A Critical Gap in 41 Mabogunje, Akin L. 2010. “Land Reform in Nigeria: Progress, Economic Analysis and Policy in South Asia.” World Development Problems & Prospects.” presented at Annual World Bank Con- 22 (10): 1455–1478. ference on Land Policy and Administration, Washington, 122 DC. http://siteresources.worldbank.org/EXTARD/Resourc- of Social Sciences 5 (26): 37 -43; National Bureau of Statistics. 2016. es/336681-1236436879081/5893311-1271205116054/mabogunje.pdf. “Nigeria - General Household Survey - Panel Wave 3 (Post Planting) 2015-2016, Third round.” 42 Adeniyi, Peter O. 2011. Improving Land Sector Governance in Nigeria: Implementation of the Land Governance Assess- 47 Aluko, Bioye Tajudeen, and Abdul-Rasheed Amidu. 2006. Women ment Framework . Washington, D.C.: World Bank Group. and Land Rights Reforms in Nigeria. Promoting Land Administration http://documents.worldbank.org/curated/ and Good Governance, 5th FIG Regional Conference, Accra; British en/68706150485597 3322/Improving-land-sector-govern- Council of Nigeria. 2012. “Gender in Nigeria Report 2012: Improving ance-in-Nigeria-implementation-of-the-land-governance-as- the Lives of Girls and Women in Nigeria.” Issues Policies Action 2. sessment-framework.; Ghebru, Hosaena, Hyacinth Edeh, Daniel Ali, Klaus Deininger, Austen Okuma, and Sileshi Woldeyohannes. 48 Hilhorst, Thea. 2000. “Women’s Land Rights: Current Developments 2014. “Tenure Security and Demand for Land Tenure Regularization in Sub-Saharan Africa.” Evolving Land Rights, Policy and Tenure in in Nigeria.” Working Paper 25, International Food Policy Research Africa, edited by C. Toulmin and J. Quan. 188–196. London: IIED with Institute, Washington DC; Mabogunje, Akin L. 2010. “Land Reform DFID and NRI; Tripp, Aili Mari. 2004. “Women’s Movements, Cus- in Nigeria: Progress, Problems & Prospects” presented at Annual tomary Law, and Land Rights in Africa: The Case of Uganda.” African World Bank Conference on Land Policy and Administration, Wash- Studies Quarterly 7 (4): 1–19; Whitehead, Ann, and Dzodi Tsikata. ington, DC. http://siteresources.worldbank.org/EXTARD/Resourc- 2003. “Policy Discourses on Women’s Land Rights in Sub-Saharan es/336681-1236436879081/5893311-1271205116054/mabogunje. Africa: The Implications of the Return to the Customary.” Journal pdf; Emeka, Nzeh, OS. Famobuwa, and Iroechenwu Chinemeze. of Agrarian Change 3 (1/2): 67–112; Iruonagbe, Charles Tunde. 2010. 2017. “Land Reform System and Its Implications on Agricultural “Women’s Land Rights and the Challenge of Patriarchy: Lessons from Production in Nigeria.” Universal Journal of Agricultural Research Ozalla Community, Edo State, Nigeria.” Gender and Behaviour 8 (1): 5 (6): 338–343; Otubu, Akintunde Kabir. 2013. “Land Reforms and 2603–2617; Slavchevska, Vanya, Ana Paula De La O Campos, Chiara Future of Land Use Act in Nigeria.” Nigerian Current Law Review. Brunelli, and Cheryl Doss. 2016. “Beyond Ownership: Women’s and Men’s Land Rights in Sub-Saharan Africa.” Working Paper, Food and 43 Ghebru, Hosaena, Hyacinth Edeh, Daniel Ali, Klaus Deininger, Austen Agriculture Organization of the United Nations, Rome; Ademola, A. Okuma, and Sileshi Woldeyohannes. 2014. “Tenure Security and 1994. “Women Farmers in Ondo State, Nigeria.” Journal of Anthro- Demand for Land Tenure Regularization in Nigeria.” Working Paper pological Research 50 (3): 311–326. 25, International Food Policy Research Institute, Washington, DC. 49 Famoriyo, S. 1984. “Administration of Land Allocation in Nigeria.” 44 Mabogunje, Akin L. 2010. “Land Reform in Nigeria: Progress, Land Use Policy 1 (3): 217–224; Aluko, Bioye Tajudeen, and Ab- Problems & Prospects” presented at Annual World Bank Con- dul-Rasheed Amidu. 2006. Women and Land Rights Reforms in ference on Land Policy and Administration, Washington, Nigeria. Promoting Land Administration and Good Governance, 5th DC. http://siteresources.worldbank.org/EXTARD/Resourc- FIG Regional Conference, Accra; Iruonagbe, Charles Tunde. 2010. es/336681-1236436879081/5893311-1271205116054/mabogunje.pdf; “Women’s Land Rights and the Challenge of Patriarchy: Lessons from Slavchevska, Vanya, Ana Paula De La O Campos, Chiara Brunelli, and Ozalla Community, Edo State, Nigeria.” Gender and Behaviour 8 Cheryl Doss. 2016. “Beyond Ownership: Women’s and Men’s Land (1): 2603–2617. Rights in Sub-Saharan Africa.” Working Paper, Food and Agriculture Organization of the United Nations, Rome; Adeniyi, Peter O. 2011. 50 Fabusoro, E., T. Matsumoto, and M. Taeb. 2008. “Land Rights Regimes Improving Land Sector Governance in Nigeria: Implementation in Southwest Nigeria: Implications for Land Access and Livelihoods of the Land Governance Assessment Framework. Washington, Security of Settled Fulani Agropastoralists.” Land Degradation & DC.: World Bank Group. http://documents.worldbank.org/cu- Development 19: 91–103; Aluko, Bioye Tajudeen, and Abdul-Rash- rated/en/687061504855973322/Improving-land-sector-govern- eed Amidu. 2006. Women and Land Rights Reforms in Nigeria. ance-in-Nigeria-implementation-of-the-land-governance-assess- Promoting Land Administration and Good Governance, 5th FIG ment-framework; Williams, Donald C. 1992. “Measuring the Impact Regional Conference, Accra; Mabogunje, Akin L. 2010. “Land Reform of Land Reform Policy in Nigeria.” Journal of Modern African Studies in Nigeria: Progress, Problems & Prospects” presented at Annual 30 (4): 587–608; Arua, Emea O., and Eugene C. Okorji. 1998. “Multi- World Bank Conference on Land Policy and Administration, Wash- dimensional Analysis of Land Tenure Systems in Eastern Nigeria.” ington, DC. http://siteresources.worldbank.org/EXTARD/Resourc- Land Reform Bulletin 1997/2. Rome: FAO. es/336681-1236436879081/5893311-1271205116054/mabogunje.pdf. 45 Adeniyi, Peter O. 2011. Improving Land Sector Governance in Nigeria: 51 Aluko, Bioye Tajudeen, and Abdul-Rasheed Amidu. 2006. Women Implementation of the Land Governance Assessment Framework. and Land Rights Reforms in Nigeria. Promoting Land Administra- Washington, DC. : World Bank Group. http://documents.worldbank. tion and Good Governance, 5th FIG Regional Conference, Accra; org/curated/en/687061504855973322/Improving-land-sector-gov- Iruonagbe, Charles Tunde. 2010. “Women’s Land Rights and the ernance-in-Nigeria-implementation-of-the-land-governance-as- Challenge of Patriarchy: Lessons from Ozalla Community, Edo State, sessment-framework. Nigeria.” Gender and Behaviour 8 (1): 2603–2617. 46 Aribigbola, Afolabi. 2007. “Urban Land Use Planning, Policies and 52 Ikejiofor, U. 2006. “Equity in Informal Land Delivery: Insights from Management in Sub-Saharan African Countries: Empirical Evidence Enugu, Nigeria.” Land Use Policy 23 (4): 448–459. from Akura, Nigeria.” Fourth Urban Research Symposium; Chikwen- du, D. O., and J. O. Arokoyo. 1997. “Women and Sustainable Agricul- tural Development in Nigeria.” Journal of Sustainable Agriculture 11 53 Namubiru-Mwaura, Evelyn. 2014. “Land Tenure and Gender: Ap- (1): 53 -69; Enwelu, I. A., U. L. Morah, M. U Dimelu, and C.I. Ezeano. proaches and Challenges for Strengthening Rural Women’s Land 2014. “Women Farmers' Access and Use of Land for Agriculture in Rights.” Women’s Voice, Agency, & Participation Research Series, Selected Communities of Anambra State.” Mediterranean Journal No. 6, World Bank, Washington, DC. 123 54 Small, J. 1997. “Women’s Land Rights: A Case Study from the Northern 62 Enwelu, I. A., U. L. Morah, M. U. Dimelu, and C. I. Ezeano. 2014. Transvaal.” Women, Land and Authority: Perspectives from South “Women Farmers’ Access and Use of Land for Agriculture in Selected Africa, edited by S. Meer. London: Oxfam; Iruonagbe, Charles Tunde. Communities of Anambra State.” Mediterranean Journal of Social 2010. “Women’s Land Rights and the Challenge of Patriarchy: Lessons Sciences 5 (26): 37–43. from Ozalla Community, Edo State, Nigeria.” Gender and Behaviour 8 (1): 2603–2617. 63 Okoroafor, C. U. and E. C. Nwadiaro. 2011. “Gender and Poverty in Nigeria: A Sociological Explanation.” European Journal of Social 55 Adeniyi, Peter O. 2011. Improving Land Sector Governance Sciences 18 (3). in Nigeria: Implementation of the Land Governance Assess- ment Framework; Washington, DC.: World Bank Group. http:/ 64 Adeniyi, Peter O. 2011. Improving Land Sector Governance in Nigeria: /documents.worldbank.org/curated/en/687061504855973322 Implementation of the Land Governance Assessment Framework. /Improving-land-sector-governance-in-Nigeria-implementa- Washington, DC.: World Bank Group. http://documents.worldbank. tion-of-the-land-governance-assessment-framework. FAO (Food org/curated/en/687061504855973322/Improving-land-sector-gov- and Agriculture Organization). 2018a. “National Gender Profile of ernance-in-Nigeria-implementation-of-the-land-governance-as- Agriculture and Rural Livelihoods – Nigeria.” Food and Agriculture sessment-framework. Organzation of the United Nations, Abuja; Iruonagbe, Charles Tunde. 2010. “Women’s Land Rights and the Challenge of Patriarchy: Lessons from Ozalla Community, Edo State, Nigeria.” Gender and Behaviour 65 Myers, Gregory Wilson. 1991. “Land and Power: The Impact of the 8(1): 2603–2617; Ikejiofor, U. 2006. “Equity in Informal Land Delivery: Land Use Act in Southwest Nigeria.” LTC Research Paper 108, Land Insights from Enugu, Nigeria.” Land Use Policy 23 (4): 448–459. Tenure Center, Madison, WI; Famoriyo, S. 1984. “Administration of Land Allocation in Nigeria.” Land Use Policy 1 (3): 217–224; Gheb- ru, Hosaena, Hyacinth Edeh, Daniel Ali, Klaus Deininger, Austen 56 Edu, O. K. 2016. “A Critical Analysis of the Laws of Inheritance in the Okuma, and Sileshi Woldeyohannes. 2014. “Tenure Security and Southern States of Nigeria.” Journal of African Law 60(1): 141–155; Demand for Land Tenure Regularization in Nigeria.” Working Paper Aluko, Bioye Tajudeen, and Abdul-Rasheed Amidu. 2006. Women 25, International Food Policy Research Institute, Washington DC; and Land Rights Reforms in Nigeria. Promoting Land Administration Adeniyi, Peter O. 2011. Improving Land Sector Governance in Nigeria: and Good Governance, 5th FIG Regional Conference, Accra. Implementation of the Land Governance Assessment Framework. Washington, DC. World Bank Group. http://documents.worldbank. 57 Adeniyi, Peter O. 2011. Improving Land Sector Governance in Nigeria: org/curated/en/687061504855973322/Improving-land-sector-gov- Implementation of the Land Governance Assessment Framework ernance-in-Nigeria-implementation-of-the-land-governance-as- Washington, D.C.: World Bank Group. http://documents.world- sessment-framework; Mabogunje, Akin L. 2010. “Land Reform in bank.org/curated/en/687061504855973322/ Improving-land-sec- Nigeria: Progress, Problems & Prospects.” presented at Annual tor-governance-in-Nigeria-implementation-of-the-land-govern- World Bank Conference on Land Policy and Administration, Wash- ance-assessment-framework; COHRE (Centre on Housing Rights ington, DC. http://siteresources.worldbank.org/EXTARD/Resourc- and Evictions). 2005. “A Place to Live: Women’s Inheritance Rights es/336681-1236436879081/5893311-1271205116054/mabogunje.pdf. in Africa.” Geneva: COHRE; British Council of Nigeria. 2012. “Gender in Nigeria Report 2012: Improving the Lives of Girls and Women in 66 Emeka, Nzeh, O. S. Famobuwa, and Iroechenwu Chinemeze. 2017. Nigeria.” Issues Policies Action 2. “Land Reform System and Its Implications on Agricultural Produc- tion in Nigeria.” Universal Journal of Agricultural Research 5 (6): 58 Slavchevska, Vanya, Ana Paula De La O Campos, Chiara Brunelli, and 338–343. Cheryl Doss. 2016. “Beyond Ownership: Women’s and Men’s Land Rights in Sub-Saharan Africa.” Working Paper, Food and Agriculture 67 United States Agency for International Development (USAID). Organization of the United Nations, Rome. 2010. Nigeria - Property Rights and Resource Governance. US- AID Country Profile; Olusola, Atilola. 2010. “Land Administra- 59 Namubiru-Mwaura, Evelyn. 2014. “Land Tenure and Gender: Ap- tion Reform Nigeria: Issues and Prospects.” FIG Congress: Fac- proaches and Challenges for Strengthening Rural Women’s Land ing the Challenges – Building the Capacity, Sydney. 11–16 April Rights.” Women’s Voice, Agency, & Participation Research Series, 2010; Mabogunje, Akin L. 2010. “Land Reform in Nigeria: Progress, No. 6, World Bank, Washington, DC; Slavchevska, Vanya, Ana Paula Problems & Prospects.” Presented at Annual World Bank Con- De La O Campos, Chiara Brunelli, and Cheryl Doss. 2016. “Beyond ference on Land Policy and Administration, Washington, DC. Ownership: Women’s and Men’s Land Rights in Sub-Saharan Africa.” http: / /siteresources.worldbank.or g/ E X TA R D / Resourc- Working Paper, Food and Agriculture Organization of the United es/336681-1236436879081/5893311-1271205116054/mabogunje.pdf. Nations, Rome; Iruoma, Kelechukwu. 2018. “Pushing for Stronger Laws to Protect Widows’ Rights in Nigeria,” Women’s Advancement 68 Resnick, Danielle, and Austen Okumo. 2016. “Institutional Limits to Deeply, April 25, 2018. Land Governance Reform: Federal-State Dynamics in Nigeria.” NSSP Working Paper 38. Washington, DC: International Food Policy Re- 60 Ghebru, Hosaena, Hyacinth Edeh, Daniel Ali, Klaus Deininger, Austen search Institute (IFPRI). http://ebrary.ifpri.org/cdm/ref/collection/ Okuma, and Sileshi Woldeyohannes. 2014. “Tenure Security and p15738coll2/id/130966. Demand for Land Tenure Regularization in Nigeria.” Working Paper 25, International Food Policy Research Institute, Washington, DC. 69 Mabogunje, Akin L. 2010. “Land Reform in Nigeria: Progress, Problems & Prospects.” presented at Annual World Bank Con- 61 Okoroafor, C. U., and E. C. Nwadiaro. 2011. “Gender and Poverty in ference on Land Policy and Administration, Washington, Nigeria: A Sociological Explanation.” European Journal of Social DC. http://siteresources.worldbank.org/EXTARD Resourc- Sciences 18 (3). es/336681-1236436879081/5893311-1271205116054/mabogunje.pdf. 124 70 World Bank. 2019. Nigeria: Review Land Acquisition Policies and file of Agriculture and Rural Livelihoods: Nigeria.” Country Gender Practice in Selected States. Washington, DC: World Bank. Assessments, FAO, Abuja. 71 World Bank. 2019. Nigeria: Review Land Acquisition Policies and 83 Edet, Glory E., and Nsikak-Abasi A. Etim. 2014. “Poverty and Gender Practice in Selected States. Washington, DC: World Bank. Inequality in Asset Ownership among Households in Southern Nigeria.” Journal of Studies in Social Sciences 7 (1): 82–95. 72 O’Sullivan, Michael. 2017. “Gender and Property Rights in Sub-Sa- haran Africa: A Review of Constraints and Effective Interventions.” 84 Ayoade, J. A., H. I. Ibrahim, and H. Y. Ibrahim. 2009. “Analysis of Wom- Policy Research Working Paper 8250, World Bank, Washington, DC; en Involvement in Livestock Production in Lafia Area of Nasarawa World Bank. 2019. “Africa’s Pulse, No. 20.” October. World Bank, State, Nigeria.” Livestock Research for Rural Development 21 (12). Washington, DC. 85 Olojede, J. C., and M. E. Njoku. 2007. “Involvement of Women in 73 Ali, Daniel Ayalew, Klaus Deininger, and Markus Goldstein. 2014. Livestock Production: A Means of Reducing Hunger and Malnu- “Environmental and Gender Impacts of Land Tenure Regularization trition in Ikwuano Local Government Area of Abia State, Nigeria.” in Africa: Pilot Evidence from Rwanda.” Journal of Development Agricultural Journal 2 (2): 231–235. Economics 110: 262–75. 86 Baba, Saadatu Umaru, and Dan Van der Horst. 2018. “Intrahousehold 74 Cherchi, Ludovica, Markus Goldstein, James Habyarimana, Joao Mon- Relations and Environmental Entitlements of Land and Livestock talvao, Michael O’Sullivan, Chris Udry, and Ariel Gruver. 2019. “Em- for Women in Rural Kano, Northern Nigeria.” Environments 5–2 (26); powering Women through Equal Land Rights: Experimental Evi- FAO (Food and Agriculture Organization). 2018. “National Gender dence from Rural Uganda.” Gender Innovation Lab Policy Brief 33, Profile of Agriculture and Rural Livelihoods: Nigeria.” Country Gender World Bank, Washington, DC. Assessments, FAO, Abuja; O’Sullivan, Michael. 2017. “Gender and Property Rights in Sub-Saharan Africa: A Review of Constraints and Effective Interventions.” Policy Research Working Paper 8250, World 75 Kumar, N., and A. R. Quisumbing. 2012. “Beyond ‘Death Do Us Part’: Bank, Washington, DC; Adisa, Rashid Solagberu. 2011. “Management The Long-Term Implications of Divorce Perceptions on Women’s of Farmer-Herdsmen Conflicts in North-Central Nigeria: Implications Well-Being and Child Schooling in Rural Ethiopia.” World Devel- for Collaboration between Agricultural Extension Service and Other opment 40 (12): 2478–2489. Stakeholders.” Journal of International Agricultural and Extension Education 18 (1): 60–72; Dahl, Gudrun. 1987. “Women in Pastoral 76 Deininger, K., A. Goyal, and H. Nagarajan. 2013. “Women’s Inheritance Production: Some Theoretical Notes on Roles and Resources.” Rights and Intergenerational Transmission of Resources in India.” Ethnos 52 (1–2): 246–279. Journal of Human Resources 48 (1): 114–41. 87 Abbass, Isah Mohammed. 2012. “No Retreat, No Surrender: 77 Iruoma, Kelechukwu. 2018. “Pushing for Stronger Laws to Protect Conflict for Survival Between Fulani Pastoralists and Farmers Widows Rights in Nigeria. ” Women’s Advancement Deeply. April in Northern Nigeria.” European Scientific Journal 8 (1): 331–346; 25, 2018. Iro, Ismail. N.d. “Fulani Herding System.” Accessed September 7, 2019. http://www.nigerdeltapeoplesworldcongress.org/arti- cles/fulani_herding_syst.pdf. http://unpub.wpb.tam.us.sitepro- 78 Bentina Alawari Mathiasa. 2015. “Widowhood Practice in Eastern tect.com/var/m_f/fa/fa2/22697/235214-fulani_herding_syst. Nigeria: A Comparative Study of Imo and Anambra States.” Sociology pdf. Study 5 (3): 223–231. 88 Galiani, Sebastian, and Ernesto Schargrodsky. 2011. “Land Property 79 Enwelu, I. A., U. L. Morah, M. U. Dimelu, and C. I. Ezeano. 2014. Rights and Resource Allocation.” Journal of Law and Economics “Women Farmers’ Access and Use of Land for Agriculture in Selected 54 (4): 329–345. Communities of Anambra State.” Mediterranean Journal of Social Sciences 5(26): 37–43. 89 Africa Sustainable Livestock 2050. 2018. Livestock Production Sys- tems Spotlight Nigeria. FAO (Food and Agriculture Organization), 80 Ghebru, Hosaena, Hyacinth Edeh, Daniel Ali, Klaus Deininger, Austen Rome. Okuma, and Sileshi Woldeyohannes. 2014. “Tenure Security and Demand for Land Tenure Regularization in Nigeria.” Working Paper 25, International Food Policy Research Institute, Washington, DC. 90 FAO (Food and Agriculture Organization). 2009. “The State of Food and Agriculture 2009.” Livestock in Balance. Rome: FAO. 81 Africa Sustainable Livestock 2050. 2018. Livestock Production Systems Spotlight Nigeria. FAO (Food and Agriculture Organiza- 91 Njuki, Jemimah, and Pascal Sanginga. 2013. “Gender and Livestock: tion), Rome; National Bureau of Statistics. 2016. “Nigeria – General Key Issues, Challenges and Opportunities.” In Women, Livestock Household Survey-Panel Wave 3 (Post Planting) 2015–2016, Third Ownership and Markets, edited by Jemimah Njuki and Pascal Round.” Sanginga, 1–8. Ottawa: International Development Research Centre; Sheahan, Megan, and Christopher B. Barrett. 2014. “Understanding the Agricultural Input Landscape in Sub-Saharan Africa: Recent 82 Oseni, Gbemisola, Markus Goldstein, and Amarachi Utah. 2013. Plot, Household, and Community-Level Evidence.” Policy Research “Gender Dimensions in Nigerian Agriculture.” Africa Region Gender Working Paper 7014, World Bank, Washington, DC. Practice Policy Brief Issue 5, World Bank, Washington, DC; FAO (Food and Agriculture Organization). 2018. “National Gender Pro- 125 92 World Bank. 2011. World Development Report 2012: Gender Equality 100 Ayoade, J. A., H. I. Ibrahim, and H. Y. Ibrahim. 2009. “Analysis of Wom- and Development. Washington, DC: World Bank; Sanginga, Pas- en Involvement in Livestock Production in Lafia Area of Nasarawa cal, Jemimah Njuki, and Elizabeth Waithanji. 2013. “Improving the State, Nigeria.” Livestock Research for Rural Development 21 (12). Design and Delivery of Gender Outcomes in Livestock Research and Development in Africa.” In Women, Livestock Ownership and 101 Farinde, A. J., and A. O. Ajayi. 2005. “Training Needs of Women Farm- Markets, edited by Jemimah Njuki and Pascal Sanginga, 129–142. ers in Livestock Production: Implications for Rural Development in Ottawa: International Development Research Centre. Oyo State of Nigeria.” Journal of Social Sciences 10 (3): 159–165. 93 Dillon, Andrew, and Esteban J. Quiñones. 2010. “Gender Differen- 102 Waters-Bayer, Ann. 1985. “Dairying by Settled Fulani Women in tiated Asset Dynamics in Northern Nigeria.” ESA Working Paper Central Nigeria and Some Implications for Dairy Development.” No. 11-06, IFPRI (International Food Policy Research Institute), Agricultural Administration Unit, Overseas Development Institute, Washington, DC. London; Nunow, A. A. 2000. Pastoralists and Markets. Research report 61. Leiden: African Studies; Dimelu, M. U., D. E. Salifu, A. I.En- 94 Sabates-Wheeler, R. 2006. “Asset Inequality and Agricultural Growth: welu, and E. M. Igbokwe. 2016. “Challenges of Herdsmen-Farmers’ How Are Patterns of Asset Inequality Established and Reproduced?” Conflict in Livestock Production in Nigeria: Experience of Pastoralists WDR Background Paper on Asset Inequality and Agricultural Pro- in Kogi State, Nigeria.” African Journal of Agricultural Research 128 ductivity, World Bank, Washington, DC; O’Sullivan, Michael. 2017. (8): 642–650. “Gender and Property Rights in Sub-Saharan Africa: A Review of Constraints and Effective Interventions.” Policy Research Working 103 Njuki, Jemimah, Samuel Mburu, and Paula Pimentel. 2013. “Livestock Paper 8250, World Bank, Washington, DC. Markets and Intra-Household Income Management.” In Women, Livestock Ownership and Markets, edited by Jemimah Njuki and 95 Africa Sustainable Livestock 2050. 2018. Livestock Production Sys- Pascal Sanginga, 60–78. Ottawa: International Development Re- tems Spotlight Nigeria. FAO (Food and Agriculture Organization), search Centre. Rome; FAO (Food and Agriculture Organization). 2018. “National Gender Profile of Agriculture and Rural Livelihoods: Nigeria.” Country 104 World Bank, FAO (Food and Agriculture Organization), and IFAD Gender Assessments, FAO, Abuja. (International Fund for Agricultural Development). 2009. Gender in Agriculture Sourcebook. Washington, DC: World Bank; Assan, 96 Africa Sustainable Livestock 2050. 2018. Livestock Production Sys- N. 2014. “Gender Disparities in Livestock Production and Their tems Spotlight Nigeria. FAO (Food and Agriculture Organization), Implication for Livestock Productivity in Africa.” Scientific Journal Rome. of Animal Science 3 (5): 126–138. 97 Olojede, J. C., and M. E. Njoku. 2007. “Involvement of Women in 105 Mburu, Samuel, Jemimah Njuki, and Juliet Kariuki. 2013. “Women’s Livestock Production: A Means of Reducing Hunger and Malnutrition Access to Livestock Information and Financial Services.” In Wom- in Ikwuano Local Government Area of Abia State, Nigeria.” Agricul- en, Livestock Ownership and Markets, edited by Jemimah Njuki tural Journal 2 (2): 231–235; Kristjanson, Patti, Ann Waters-Bayer, and Pascal Sanginga, 79–94. Ottawa: International Development Nancy Johnson, Annita Tipilda, Jemimah Njuki, Isabelle Baltenweck, Research Centre. Delia Grace, and Susan MacMillan. 2014. “Livestock and Women’s Livelihoods.” In Gender in Agriculture: Closing the Knowledge Gap, 106 Arokoyo, Tunji. N.d. “ICTs in the Transformation of Agricultural edited by Agnes R. Quisumbing, Ruth Meinzen-Dick, Terri L. Raney, Extension: The Case of Nigeria.” https: / /repositor y.ruforum. Andre Croppenstedt, Julia A. Behrman, and Amber Peterman, org/sites/default/files/Case_study_Nigeria.pdf. 209–234. Food and Agriculture Organization of the United Nations. 107 FAO (Food and Agriculture Organization). 2018. “National Gender 98 Kristjanson, Patti, Ann Waters-Bayer, Nancy Johnson, Annita Tip- Profile of Agriculture and Rural Livelihoods: Nigeria.” Country Gender ilda, Jemimah Njuki, Isabelle Baltenweck, Delia Grace, and Susan Assessments, FAO, Abuja. MacMillan. 2014. “Livestock and Women’s Livelihoods.” In Gender in Agriculture: Closing the Knowledge Gap, edited by Agnes R. Quisumbing, Ruth Meinzen-Dick, Terri L. Raney, Andre Croppen- 108 Chibuzo, Izuogu, and Atasie Chikerenma. 2015. “Redefining the stedt, Julia A. Behrman, and Amber Peterman, 209–234. Food and Nigerian Agricultural Extension System for Effective Agricultural Agriculture Organization of the United Nations. Transformation.” Developing Country Studies 5 (11): 11–16; Farinde, A. J., and A. O. Ajayi. 2005. “Training Needs of Women Farmers in Livestock Production: Implications for Rural Development in Oyo 99 UNECA (United Nations Economic Commission for Africa) and State of Nigeria.” Journal of Social Sciences 10 (3): 159–165. FAO (Food and Agriculture Organization). 1975. Women, Population and Rural Development in Africa. New York: United Nations; Wa- ters-Bayer, A., and B. Letty. 2010. “Promoting Gender Equality and 109 Bastian, Gautam, Markus Goldstein, and Sreelakshmi Papineni. Empowering Women through Livestock.” In The Role of Livestock 2017. “Are Cash Transfers Better Chunky or Smooth?: Evidence in Developing Communities: Enhancing Multifunctionality, edited from an Impact Evaluation of a Cash Transfer Program in Northern by Frans Swanepoel, Aldo Stroebel, and Siboniso Moyo, 31–50. Nigeria.” Gender Innovation Lab Policy Brief No. 21, World Bank, Bloemfontein: University of Free State (UFS) and The Technical Washington, DC. Centre for Agricultural and Rural Cooperation (CTA); Oji, I., and Ekumankama, O. 2002. “Women Involvement in Animal Agriculture 110 O’Sullivan, Michael. 2017. “Gender and Property Rights in Sub-Sa- in Afikpo South Area of Ebonyi State of Nigeria.” haran Africa: A Review of Constraints and Effective Interventions.” Policy Research Working Paper 8250, World Bank, Washington, DC. 126 111 Argent, Jonathan, Britta Augsburg, and Imran Rasul. 2014. “Livestock 122 Smita Das, Aphichoke Kotikula, and Eliana Carranza. 2018. Gen- Asset Transfers with and without Training: Evidence from Rwanda.” der-Based Employment Segregation: Understanding Causes and Journal of Economic Behavior & Organization 108 (December): Policy Interventions. World Bank, Washington, DC. 19–39. 123 British Council. 2014. Girls’ Education in Nigeria. Report 2014: Issues, 112 Banerjee, Abhijit, Esther Duflo, Nathanael Goldberg, Dean Karlan, Influencers and Actions. Robert Osei, William Parienté, Jeremy Shapiro, Bram Thuysbaert, and Christopher Udry. 2015. “A Multi-faceted Program Causes Lasting 124 Ekine, Adefunke O. 2010. “The Impact of Videotaped Instructional Progress for the Very Poor: Evidence from Six Countries.”  Science 348 Strategy on Pupils’ Interest and Achievement on Primary Science (6236). in Some Selected Schools in Oyo State.” PhD thesis, University of Ibadan. 113 Bandiera, Oriana, Robin Burgess, Narayan Das, Selim Gulesci, Imran Rasul, and Munshi Sulaiman. 2017. “Labor Markets and Poverty in 125 Herz, Barbara, and Gene B. Sterling. 2004. “What Works in Girls’ Village Economies.”  Quarterly Journal of Economics  132 (2): 811–70. Education: Evidence and Policies from the Developing World.” Council on Foreign Relations, New York. 114 Roy, Shalini, Jinnat Ara, Narayan Das, and Agnes R. Quisumbing. 2015. “‘Flypaper Effects’ in Transfers Targeted to Women: Evidence from 126 Brotman, Jennie S., and Felicia Moore Mensah. 2008. “Girls and Sci- BRAC’s ‘Targeting the Ultra Poor’ Program in Bangladesh.” Journal ence: A Review of Four Themes in the Science Education Literature.” of Development Economics 117 (November): 1–19. Journal of Research in Science Teaching 45 (9): 971–1002; Rathgeber, Eva M. 2009. “Women and Girls in Science and Technology: Increasing 115 Kristjanson, Patti, Ann Waters-Bayer, Nancy Johnson, Annita Tip- Opportunities in Education, Research and Employment.” Expert ilda, Jemimah Njuki, Isabelle Baltenweck, Delia Grace, and Susan paper, United Nations, Geneva. MacMillan. 2014. “Livestock and Women’s Livelihoods.” In Gender in Agriculture: Closing the Knowledge Gap, edited by Agnes R. 127 Federal Office of Statistics, 2003 abstracts. Quisumbing, Ruth Meinzen-Dick, Terri L. Raney, Andre Croppen- stedt, Julia A. Behrman, and Amber Peterman, 209–234. Food and Agriculture Organization of the United Nations. 128 Croke, Kevin, Markus Goldstein, and Alaka Holla. 2017. “Can Job Training Decrease Women’s Self-Defeating Biases? Experimental Evidence from Nigeria.” Gender Innovation Lab Policy Brief 28, 116 Das, Smita, Aphichoke Kotikula, and Eliana Carranza. 2019. “Gen- World Bank, Washington, DC. der-Based Employment Segregation: Understanding Causes and Policy Interventions.” Jobs Working Paper 26, World Bank, Wash- ington, DC. 129 Stout, Jane G., Nilanjana Dasgupta, Matthew Hunsinger, and Melissa A. McManus. 2011. “STEMing the Tide: Using Ingroup Experts to Inoculate Women’s Self-Concept in Science, Technology, Engineer- 117 Hsieh, C. T., E. Hurst, C. I. Jones, and P. J. Klenow. 2019. “The Allo- ing, and Mathematics (STEM).” Journal of Personality and Social cation of Talent and US Economic Growth.” Econometrica 87 (5): Psychology 100 (2): 255–270. 1439–1474. 130 Beaman, Lori, Esther Duflo, Rohini Pande, and Petia Topalova. 2012. 118 Cuberes, D., and M. Teignier. 2016. “Aggregate Effects of Gender Gaps “Female Leadership Raises Aspirations and Educational Attainment in the Labor Market: A Quantitative Estimate.” Journal of Human for Girls: A Policy Experiment in India.”  Science 355 (2012): 582–586. Capital 10 (1): 1–32. 131 Stout, Jane G., Nilanjana Dasgupta, Matthew Hunsinger, and Melissa 119 Brotman, Jennie S., and Felicia Moore Mensah. 2008. “Girls and A. McManus. 2011. “STEMing the Tide: Using Ingroup Experts to Science: A Review of Four Themes in the Science Education Liter- Inoculate Women’s Self-Concept in Science, Technology, Engineer- ature.” Journal of Research in Science Teaching 45 (9): 971–1002; ing, and Mathematics (STEM).” Journal of Personality and Social Scantlebury, Kathryn. 2012. “Still Part of the Conversation: Gender Psychology 100 (2): 255–270. Issues in Science Education.” In Second International Handbook of Science Education, edited by B. Fraser, K. Tobin, and C. McRobbie, 499–512. New York: Springer. doi: 10.1007/978-1-4020-9041-7_34. 132 Solarin, Tai. 2013. “University Student Enrollment per Department, 2011–2013.” Office of the Registrar, Ijagun, Ogun State, Nigeria; World Bank. 2010. “Primary Education, Teachers (% Female).” The World 120 Brotman, Jennie S., and Felicia Moore Mensah. 2008. “Girls and Bank Group. Accessed September 7, 2019. https://data.worldbank. Science: A Review of Four Themes in the Science Education Liter- org/indicator/SE.PRM.TCHR.FE.ZS?locations=NG; Ekine, Adefunke, ature.” Journal of Research in Science Teaching 45 (9): 971–1002; and Negar Ashtari Abay. 2013. “Enhancing Girls’ Participation in Sinnes, Astrid. 2006. “Three Approaches to Gender Equity in Science Science in Nigeria.” https://www.brookings.edu/wp-content/up- Education.” NORDINA 1 (6): 72–83. loads/2016/07/ekine_girls_education.pdf. 121 Nigeria National Bureau of Statistics (NBS). 2019. LSMS Integrated 133 National Bureau of Statistics. 2018. “Statistical Report on Women Surveys on Agriculture Nigeria General Household SurveyPanel, and Men in Nigeria: 2017.” Wave 4. NBS, Abuja, Nigeria. https://www.nigerianstat.gov.ng/pd- fuploads/LSMS_Integrated_Panel_Survey_Report.pdf. 134 Chakravarty, Subha, Smita Das, and Julia Vaillant. 2017. “Gender and Youth Employment in Sub-Saharan Africa: A Review of Constraints 127 and Effective Interventions.” Policy Research Working Paper 8245, 147 Hsieh, C. T., E. Hurst, C. I. Jones, and P. J. Klenow. 2019. “The Allo- World Bank, Washington, DC. cation of Talent and US Economic Growth.” Econometrica 87 (5): 1439–1474. 135 Magruder, Jeremy. 2010. “Intergenerational Networks, Unemploy- ment and Persistent Inequality in South Africa.” American Economic 148 Awotide, Bola Amoke, Aziz Karimov, Aliou Diagne, and Tebila Na- Journal: Applied Economics 2 (1): 62–85. kelse. 2013. “The Impact of Seed Vouchers on Poverty Reduction among Smallholder Farmers in Nigeria.” Agricultural Economics 44: 647–658. 136 Beaman, Lori, Esther Duflo, Rohini Pande, and Petia Topalova. 2012. “Female Leadership Raises Aspirations and Educational Attainment for Girls: A Policy Experiment in India.”  Science 355 (2012): 582–586. 149 Agboh-Noameshie, A. R., F. M. Kinkingninhoun-Medagbe, and A. Diagne. 2007. “Gendered Impact of NERICA Adoption on Farmers’ Production and Income in Central Benin.” In 2nd Conference of the 137 Campos, F., M. Goldstein, L. McGorman, A. M. Boudet, and O. Pim- African Association of Agricultural Economists, Accra, Ghana. hidzai. 2015. “Breaking the Metal Ceiling: Female Entrepreneurs Who Succeed in Male-Dominated Sectors in Uganda.” Policy Research Working Paper 7503, World Bank, Washington, DC. 150 Diagne, A. 2006. “Diffusion and Adoption of NERICA Rice Varieties in Côte d’Ivoire.” The Developing Economies 44 (2): 208–231; Dibba, L., A. Diagne, S. C. Fialor, and F. Nimoh. 2012. “Diffusion and Adoption 138 Alibhai, S., N. Buehren, and S. Papineni. 2015. “Female Entrepre- of New Rice Varieties for Africa (NERICA) in the Gambia.” African neurs Who Succeed in Male Dominated Sectors in Ethiopia.” Gen- Crop Science Journal 20 (1). der Innovation Lab Policy Brief 12, World Bank, Washington, DC; Buehren, Niklas, and Taylor Van Salisbury. 2017. “Female Enrollment in Male-Dominated Vocational Training Courses: Preferences and 151 Buehren, N., S. Chowdhury, M. Sulaiman, and S. Papineni. Forthcom- Prospects.” World Bank, Washington, DC. ing. “Growing, Eating, and Selling Orange-Fleshed Sweet Potatoes: Impact and Spillovers of Nutrition-Sensitive Agricultural Programs to Women Farmers in Uganda.” 139 Alibhai, S., N. Buehren, and S. Papineni. 2015. “Female Entrepreneurs Who Succeed in Male Dominated Sectors in Ethiopia.” Gender Innovation Lab Policy Brief 12, World Bank, Washington, DC. 152 Ragasa, C. 2014. “Improving Gender Responsiveness of Agricultural Extension.” In Gender in Agriculture, edited by A. R. Quisumbing, R. Meinzen-Dick, T. L. Raney, A. Croppenstedt, J. A. Behrman, and 140 Hicks, Joan Hamory, Michael Kremer, Isaac Mbiti, and Edward Miguel. A. Peterman, 411–430. Dordrecht: Springer Netherlands. 2011. “Vocational Education Voucher Delivery and Labor Market Returns: A Randomized Evaluation among Kenyan Youth.” Working paper. 153 Kondylis, Florence, Valerie Mueller, and Jessica Zhu. 2017. “Seeing Is Believing? Evidence from an Extension Network Experiment.”  Jour- nal of Development Economics 125 (C): 1–20. 141 Campos, F., M. Goldstein, L. McGorman, A. M. Boudet, and O. Pim- hidzai. 2015. “Breaking the Metal Ceiling: Female Entrepreneurs Who Succeed in Male-Dominated Sectors in Uganda.” Policy Research 154 BenYishay, Ariel, Maria Jones, Florence Kondylis, and Ahmed Mushfiq Working Paper 7503, World Bank, Washington, DC. Mobarak. 2016. “Are Gender Differences in Performance Innate or Socially Mediated?” Policy Research Working Paper 7689, World Bank, Washington, DC. 142 Chakravarty, Shubha, Smita Das, and Julia Vaillant. 2017. “Gender and Youth Employment in Sub-Saharan Africa: A Review of Constraints and Effective Interventions.” Policy Research Working Paper 8245, 155 Buehren, Niklas. 2019. “Gender and Agriculture in Sub-Saharan World Bank, Washington, DC. Africa: Review of Constraints and Effective Interventions.” Africa Gender Innovation Lab, World Bank, Washington, DC. 143 Fawole, Olufunmilayo I., Adedibu L. Aderonu, and Adeniran O. Fawole. 2005. “Intimate Partner Abuse: Wife Beating among Civil 156 Ambler, Kate, Kelly M. Jones, and Michael O’Sullivan. 2018. “What Servants in Idaban, Nigeria.” African Journal of Reproductive Health Is the Role of Men in Connecting Women to Cash Crop Markets? 9 (2): 54–64. Evidence from Uganda.” IFPRI Discussion Paper 1762, International Food Policy Research Institute, Washington, DC. 144 Qasim, Farwah, and Wumi Asubaro. 2019. “Gender Based Violence: An Analysis of the Implications for the Nigeria For Women Project.” 157 Montalvao, Joao, Michael Frese, Markus Goldstein, and Talip Kilic. World Bank, Washington, DC. 2017. “Soft Skills for Hard Constraints: Evidence from High-Achieving Female Farmers.” Policy Research Working Paper 8095, World Bank, Washington, DC. https://doi.org/10.1596/1813-9450-8095. 145 Chakravarty, Subha, Smita Das, and Julia Vaillant. 2017. “Gender and Youth Employment in Sub-Saharan Africa: A Review of Constraints and Effective Interventions.” Policy Research Working Paper 8245, 158 Delavallade, C., F. Dizon, R. V. Hill, and J. P. Petraud. 2015. “Managing World Bank, Washington, DC. Risk with Insurance and Savings: Experimental Evidence for Male and Female Farm Managers in the Sahel.” Policy Research Working Paper 7176, World Bank, Washington, DC. 146 Jayachandran, S. 2020. Social Norms as a Barrier to Women’s Em- ployment in Developing Countries. No. w27449. National Bureau of Economic Research. 159 Ogunniyi, Adebayo, Olagunju Kehinde Oluseyi, Ogundipe Adeyemi, Salman K. Kabir, and Francis Philips. 2017. “Scaling Up Agricultural 128 Innovation for Inclusive Livelihood and Productivity Outcomes in Targeting Women on Household Vulnerability and Empowerment: Sub-Saharan Africa.” Africa Development Review 29 (52): 121–134. Evidence from South West Nigeria.” Working Paper 2016-01, Part- nership for Economic Policy; Okunogbe, Oyebola. 2019. “Impact Evaluation of Skill for Jobs Training Program.” AEA RCT Registry. 160 Buehren, Niklas. 2019. “Gender and Agriculture in Sub-Saharan June 10. https://doi.org/10.1257/rct.3782-2.0; Anderson, Stephen, Africa: Review of Constraints and Effective Interventions.” Africa Rupali Kaul and David McKenzie. 2019. “Building Customer-Cen- Gender Innovation Lab, World Bank, Washington, DC. tric and Market Driven Firms Through Management Consulting: A Field Experiment with SMEs in Nigeria.” AEA RCT Registry. April 161 L. Beaman, D. Karlan, B. Thuysbaert, and C. Udry. 2013. “Profitability 06. https://doi.org/10.1257/rct.4045-2.0; Alzua, Maria and Damilola of Fertilizer: Experimental Evidence from Female Rice Farmers in Olajide. 2018. “Complementarities in microcredit and financial Mali.” American Economic Review 103 (25): 381–386. services interventions for financial inclusion and empowerment: evidence from randomized evaluation in South West Nigeria.” AEA RCT Registry. February 5. https://doi.org/10.1257/rct.2705-2.0. 162 R. W. Karamba. 2013. “Input Subsidies and Their Effect on Cropland Allocation, Agricultural Productivity, and Child Nutrition: Evidence from Malawi.” http://hdl.handle.net/1961/15104. 173 World Bank Group. 2019. Profiting from Parity: Unlocking the Po- tential of Women’s Business in Africa. World Bank, Washington, DC. 163 World Bank. 2017. "ICT in Agriculture (Updated Edition): Connecting Smallholders to Knowledge, Networks, and Institutions." June 2017, 174 Alibhai, A., N. Buehren, and S. Papineni. 2018. “Better Loans or Better 71–96. Borrowers? Impact of Meso-Credit on Female-Owned Enterpris- es in Ethiopia.” Policy Research Working Paper 8511, World Bank, Washington, DC. https://doi.org/10.1596/1813-9450-8511. 164 Lecoutere, Els, David J. Spielman, and Bjorn Van Campenhout. 2019. “Empowering Women with Digital Extension in Uganda: Effects of Information and Role Models.” 6th African Conference of Agricultural 175 Friedman, Jed, Goldstein Markus, Gonzalez Martinez Paula, and Economists, September 23–26, 2019, Abuja, Nigeria. Papineni Sreelakshmi. 2020. “Cash Is Queen: Local Economy Growth Effects of Unconditional Cash Transfers to Women.” Manuscript. 165 Aker, C., and C. Ksoll. 2016. “Can Mobile Phones Improve Agricultural Outcomes? Evidence from a Randomized Experiment in Niger.” 176 Bossuroy, Thomas, Markus Goldstein, Dean Karlan, Harounan Ka- Food Policy 60 (C): 44–51. zianga, William Parienté, Patrick Premand, Christopher Udry, Julia Vaillant, and Kelsey Wright. 2019. “Promoting Productive Inclusion and Resilience among the Poor: Multi-Country RCT of the Sahel 166 O’Sullivan, M., A. Rao, R. Banerjee, K. Gulati, and M. Vinez. 2014. Adaptive Social Protection.” AEA RCT Registry. August 29.  “Levelling the Field: Improving Opportunities for Women Farmers in Africa.” World Bank, Washington, DC. 177 Fafchamps, Marcel, David McKenzie, Simon Quinn, and Christopher Woodruff. 2014. “Microenterprise Growth and the Flypaper Effect: 167 Daidone, Silvio, Benjamin Davis, Joshua Dewbre, Mario González- Evidence from a Randomized Experiment in Ghana.” Journal of Flores, D. C. Usa, Sudhanshu Handa, David Seidenfeld, and Gelson Development Economics 106: 211–226. Tembo. N.d. “Zambia’s Child Grant Programme: 24-Month Impact Report on Productive Activities and Labour Allocation.” 178 World Bank Group. 2019. Profiting from Parity: Unlocking the Po- tential of Women’s Business in Africa. World Bank, Washington, DC. 168 L. Beaman, D. Karlan, B. Thuysbaert, and C. Udry. 2013. “Profitability of Fertilizer: Experimental Evidence from Female Rice Farmers in Mali.” American Economic Review 103 (25): 381–86. 179 McKenzie, David. 2017. “Identifying and Spurring High-Growth Entrepreneurship: Experimental Evidence from a Business Plan Competition.” American Economic Review 107 (8): 2278–2307. 169 Daum, Thomas, Filippo Capezzone, and Regina Birner. 2019. “Of Trackers and Tractors: Using a Smartphone App and Compositional Data Analysis to Explore the Link between Mechanization and 180 Fafchamps, Marcel, and Simon Quinn. 2017. “Aspire.” The Intra-Household Allocation of Time in Zambia.” ZEF-Discussion Journal of Development Studies 53 (10): 1615–1633. doi: Papers on Development Policy No. 278. 10.1080/00220388.2016.1251584. 170 Fischer, G., S. Wittich, G. Malima, G. Sikumba, B. Lukuyu, D. Ngunga, 181 World Bank Group. 2019. Profiting from Parity: Unlocking the Po- and J. Rugalabam. 2018. “Gender and Mechanization: Exploring the tential of Women’s Business in Africa. World Bank, Washington, DC. Sustainability of Mechanized Forage Chopping in Tanzania.” Journal of Rural Studies 64: 112–122. 182 Dupas, Pascaline, and Jonathan Robinson. “Savings Constraints and Microenterprise Development: Evidence from a Field Experiment 171 UN Women, UNDP (United Nations Development Programme), in Kenya.” American Economic Journal: Applied Economics 5, no. UNEP (United Nations Environment Programme), and the World 1 (January 1, 2013): 163–92. https://doi.org/10.1257/app.5.1.163. Bank Group. 2015. The Cost of the Gender Gap in Agricultural Productivity in Malawi, Tanzania, and Uganda. Washington, DC: 183 Campos, Francisco Moraes Leitao, Markus P. Goldstein, and David World Bank Group. J. Mckenzie. 2015. “Short-Term Impacts of Formalization Assistance and a Bank Information Session on Business Registration and Access 172 Ikenwilo, Divine, Damilola Olajide, Olufemi Obembe, Ngozi Ibeji, and to Finance in Malawi.”  Policy Research Working Paper 7183, World Rufus Akindola. 2016. “The Impact of a Rural Microcredit Scheme Bank, Washington, DC. 129 184 Bastian, Gautam, Iacopo Bianchi, Markus Goldstein, and Joao Mon- 194 Tabbert, F. 2009. “Child Care and Female Labor Force Participation talvao. 2018. “Short-Term Impacts of Improved Access to Mobile in Developing Countries: Quasi-Experimental Results from Togo.” Savings, with and without Business Training: Experimental Evidence Faculty of Economics and Management, Leibniz-University of Hannover. from Tanzania.” Working Paper 478, Center for Global Development: Washington, DC. 195 Clark, S., C. W. Kabiru, S. Laszlo, and S. Muthuri, 2019. “The Impact of Childcare on Poor Urban Women’s Econom- 185 Das, Smita, Aphichoke Kotikula, and Eliana Carranza. 2019. “Gen- ic Empowerment in Africa.” Demography 56 (4): 1247–1272. der-Based Employment Segregation: Understanding Causes and https://doi.org/10.1007/s13524-019-00793-3. Policy Interventions.” Jobs Working Paper No. 26, World Bank, Washington, DC. 196 Vaillant, Julia, Estelle Koussoubé, Danielle Roth, Rachael S. Pierotti, Mazeda Hossain, and Kathryn Falb. 2020. “Cluster-Randomized Con- 186 Campos, Francisco, Michael Frese, Markus Goldstein, Leonardo Iaco- trolled Trial to Reduce Intimate Partner Violence in North and South vone, Hillary Johnson, David McKenzie, and Mona Mensmann. 2017. Kivu, DRC: Engaging Men to Transform Inequitable Gender Attitudes “Teaching Personal Initiative Beats Traditional Training in Boosting Approach.” World Bank, Washington, DC. Small Business in West Africa.” Science 357 (6357): 1287–90. 197 Doyle, Kate, R. Levtov, G. Barker, G. Bastian, J. Bingenheimer, S. Kazimbaya, 187 Alibhai, Salman, Niklas Buehren, Michael Frese, Markus Goldstein, A. Nzabonimpa, J. Pulerwitz, F. Sayinzoga, V. Sharma, and D. Shattuck. Sreelakshmi Papineni, and Kathrin Wolf. 2019. “Full Esteem Ahead? 2018. “Gender-Transformative Bandebereho Couples’ Intervention Mindset-Oriented Business Training in Ethiopia.” Policy Research to Promote Male Engagement in Reproductive and Maternal Health Working Paper 8892, World Bank, Washington, DC. and Violence Prevention in Rwanda: Findings from a Randomized Controlled Trial.” PLoS One 13 (4). 188 Feigenberg, B., E. Field, and R. Pande. 2013. “The Economic Returns to Social Interaction: Experimental Evidence from Microfinance.” 198 World Bank. 2020.  GIL Top Policy Lessons on Empowering Adolescent The Review of Economic Studies 80 (4): 1459–1483; Field, E., S. Jay- Girls. Gender Innovation Lab, World Bank, Washington, DC. https:// achandran, R. Pande, and N. Rigol. 2015. “Friendship at Work: Can openknowledge.worldbank.org/handle/10986/33138. Peer Effects Catalyze Female Entrepreneurship?” National Bureau of Economic Research No. w21093; Austrian, K., and E. Muthengi. 199 O’Sullivan, Michael. 2017. “Gender and Property Rights in Sub-Saharan 2014. “Can Economic Assets Increase Girls’ Risk of Sexual Harass- Africa: A Review of Constraints and Effective Interventions.” Policy Re- ment? Evaluation Results from a Social, Health and Economic search Working Paper 8250, World Bank, Washington DC; World Bank. Asset-Building Intervention for Vulnerable Adolescent Girls in 2019. “Africa’s Pulse, No. 20” (October). World Bank, Washington, DC. Uganda.” Children and Youth Services Review 47: 168–175; Chong, A., M. Gonzalez-Navarro, D. Karlan, and M. Valdivia. 2013. “Effec- tiveness and Spillovers of Online Sex Education: Evidence from a 200 Cherchi, Ludovica, Markus Goldstein, James Habyarimana, Joao Montal- Randomized Evaluation in Colombian Public Schools.” National vao, Michael O’Sullivan, Chris Udry, and Ariel Gruver. 2019. “Empowering Bureau of Economic Research No. w18776. Women through Equal Land Rights: Experimental Evidence from Rural Uganda.” Gender Innovation Lab Policy Brief 33, World Bank, Washington, DC. 189 Mehrab Bakhtiar, Gautam Bastian, and Markus Goldstein. 2020. “Business Training and Mentoring: Experimental Evidence from Women-Owned Microenterprises in Ethiopia.” Draft paper. 201 Bastian, Gautam, Markus Goldstein, and Sreelakshmi Papineni. 2017. “Are Cash Transfers Better Chunky or Smooth?: Evidence from an Impact Evaluation of a Cash Transfer Program in Northern Nigeria.” Gender 190 Croke, Kevin, Markus Goldstein, and Alaka Holla. 2018. “Can Job Innovation Lab Policy Brief No. 21, World Bank, Washington, DC. Training Decrease Women’s Self-Defeating Biases? Experimental Evidence from Nigeria.” Other Operational Studies 30495, World Bank, Washington, DC. 202 Banerjee, Abhijit, Esther Duflo, Nathanael Goldberg, Dean Karlan, Robert Osei, William Parienté, Jeremy Shapiro, Bram Thuysbaert, and Christopher Udry. 2015. “A Multi-Faceted Program Causes Lasting 191 Hicks, Joan Hamory, Michael Kremer, Isaac Mbiti, and Edward Miguel. Progress for the Very Poor: Evidence from Six Countries.” Science 348 2011. “Vocational Education Voucher Delivery and Labor Market (6236). Returns: A Randomized Evaluation among Kenyan Youth.” Working paper. 203 Kitagawa, E. M. 1955. “Components of a Difference between Two Rates.” Journal of the American Statistical Association 50 (272), 1168–1194; 192 Clark, S., C. W. Kabiru, S. Laszlo, and S. Muthuri. 2019. “The Oaxaca, R. 1973. “Male-Female Wage Differentials inUrban Labor Mar- Impact of Childcare on Poor Urban Women’s Econom- kets.” International Economic Review 14 (3), 693–709; Blinder, A. S. ic Empowerment in Africa.” Demography 56 (4): 1247–1272. 1973. “Wage Discrimination: Reduced Form and Structural Estimates.” https://doi.org/10.1007/s13524-019-00793-3. Journal of Human Resources 8 (4), 436–455; Fortin, N., T. Lemieux, and S. Firpo. 2011. “Decomposition Methods in Economics.” In Handbook of 193 Martinez, S., S. Naudeau, and V. Pereira. 2012. “The Promise of Pre- Labor Economics, edited by O. Ashenfelter and D. Card, Vol. 4, 1–102. school in Africa: A Randomized Impact Evaluation of Early Child- Elsevier; O’Sullivan, M., A. Rao, R. Banerjee, K. Gulati, and M. Vinez. 2014. hood Development in Rural Mozambique.” World Bank and Save “Levelling the Field: Improving Opportunities for Women Farmers in the Children. Africa.” World Bank, Washington, DC. 204 Alemu, Z. G. 2015. “The Challenge of Job Creation in Nigeria.” Chief Economist Complex AEB 6 (8): 1–16. 130