ETHIOPIA GENDER DIAGNOSTIC REPORT PRIORITIES FOR PROMOTING EQUITY © 2019 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 of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Gender Innovation Policy Initiative for Ethiopia The Gender Innovation Policy Initiative for Ethiopia (GIPIE) is a country-level unit of the Africa Gender Innovation Lab (GIL) based in the World Bank Ethiopia Country Office. GIL conducts impact evaluations of development interventions and leads policy research to generate evidence on how to close gender gaps in earnings, productivity, assets, and agency. With these findings, GIL equips project teams and policy makers to design innovative and scalable interventions to address gender inequality. The GIL team is currently working on over 60 impact evaluations in nearly 30 countries with the aim of building an evidence base with lessons for the region. 2 ETHIOPIA GENDER DIAGNOSTIC REPORT PRIORITIES FOR PROMOTING EQUITY Table of Contents ACKNOWLEDGMENTS 6 INTRODUCTION 8 PART I: AN OVERVIEW OF WOMEN IN THE WORKFORCE 10 PART II: MEASURING THE GENDER GAPS 16 Agricultural Productivity 17 Business Sales 24 Wage Earnings 30 PART III: CONTEXTUAL FACTORS INFLUENCING THE GENDER GAPS 36 Labor Market Skills 37 Basic Literacy 37 Technical and Vocational Knowledge 38 Social Networks 40 Entrepreneurial, Noncognitive Skills 42 4 Social Norms 42 Marriage and Childbearing 43 Intrahousehold Dynamics 44 Asset Ownership 45 Internalized Beliefs 46 PART IV: GENDER AND INNOVATION IN PROGRAMMING 47 Policy Priorities 48 Gender Innovation 49 APPENDIXES 60 Appendix A: Sample Descriptive Statistics 61 Appendix B: Decomposition Methods 74 ENDNOTES 77 5 ACKNOWLEDGMENTS This report was prepared by a team of authors from Laketch Mickael, Senior Manager, Cross-cutting the World Bank’s Africa Gender Innovation Lab (GIL), Initiatives, Ethiopian Agricultural Transformation comprising of Niklas Buehren, Markus Goldstein, Agency; Aynie Habtamu, Senior Technical Expert Paula Gonzalez, Adiam Hagos Hailemicheal, Daniel for Inclusive Growth, Ethiopian Agricultural Kirkwood, Patricia Paskov, Michelle Poulin, Chandni Transformation Agency; Sahlu Mulu, Adviser Raja, and Andrew Tartar. Amy Copley and Zuzana to Minister, Ministry of Agriculture; Mestefakir Johansen provided editorial assistance, and Blossom Alebachew, Gender and Nutrition Specialist, (Blossoming.it) led the design and layout of the report. Ministry of Agriculture; Tsehaynesh Kidane, Gender and Nutrition Specialist, Ministry of Agriculture; The authors would like to gratefully acknowledge Tiruwork Tizazu, Gender and Cross-cutting Issues Bénédicte de la Brière and Tom Bundervoet, who Director, Ethiopian Chamber of Commerce; Kelifa peer reviewed the report. The team would also like Hussain, Deputy Director General, Federal Small and to express its deep gratitude to Michael O’Sullivan for Medium Scale Industries Development Agency; Tigist his valuable guidance. Digafe, Gender and Cross-cutting Issues Director, Federal Small and Medium Enterprise Development The Ethiopia Gender Diagnostic Report has Agency; Tewodros Asmare, Senior Gender Expert, benefited from the feedback of over 40 experts Federal Small and Medium Enterprise Development and policy makers. The team is especially grateful Agency; Taddesse Gedefa, Senior Gender Expert, to the participants of consultations and workshops Federal Small and Medium Enterprise Development held in December 2017, August 2018, and December Agency; Getachew Negash, Director General, Federal 2018: Dr. Fitsum Assefa, Minister, Planning and Technical and Vocational Education and Training Development Commission; Ahmed Shide, Minister, (TVET) Agency; Bizuneh Adugna, Adviser to State Ministry of Finance; Admasu Nebebe, State Minister, Minister of the TVET Agency, Federal TVET Agency; Ministry of Finance; Yalem Tsegay, Minister, Ministry Solomon Assefa, Deputy Director General, Directors of Women, Youth and Children’s Affairs; Semegn of Citizens’ Engagement, Research, Job Creation, Wubie, State Minister of Women’s Affairs, Ministry Project Management and Planning, Livelihoods of Women, Youth and Children’s Affairs; Dr. Teferi Development and Social Protection, Federal Urban Mequaninte, Director of Cross-cutting Initiatives, Development, Job Creation and Food Security Ethiopian Agricultural Transformation Agency; Agency; Bekele Mengistu, Deputy Director General, 6 Federal Urban Job Creation and Food Security Woldelissalsie, Planning and Finance Directorate Agency; Yohannes Letiso, Acting Adviser to the Director, National Planning Commission of Director General, Federal Urban Development, Job Ethiopia; Aster Sullamo, Senior Gender Expert, Creation and Food Security Agency; Tesfayenesh National Planning Commission of Ethiopia; Abebe Lemma, Gender and Cross-cutting Issues Director, Gebremedhin, Adviser to Minister, Ministry of Labor Federal Urban Development, Job Creation and and Social Affairs; Tsega Teka, Adviser to Minister, Food Security Agency; Etenesh Gebre, Gender Ministry of Trade and Industry; Eyerusalem Damte, and Cross-cutting Issues Expert, Federal Urban Gender Director, Ministry of Industry; Elsabeth Development, Job Creation and Food Security Beyene, Gender Director, Ministry of Trade; Azeb Agency; Fisseha Abera, Director of International Rezene, Director of Strategic Management, Ministry Financial Institutions Cooperation, Ministry of of Women, Youth and Children’s Affairs; and Seleshi Finance; Neteru Wondwossen, Director of Gender Tadessie, Director of Women’s Mobilization, Ministry and Cross-cutting Issues, Ministry of Finance; Martha of Women, Youth and Children’s Affairs. Hailemariam, Advisor to the Governor, National Bank of Ethiopia; Workabeba Bahru, Director for The team would also like to acknowledge the Human Resources, National Bank of Ethiopia; generous support of the Umbrella Facility for Gender Abate Mitiku, Director for Financial Inclusion and Equality (UFGE). The UFGE is a multi donor trust Communication, National Bank of Ethiopia; Kibre fund administered by the World Bank to advance Moges, Director for Legal Services, National Bank gender equality and women’s empowerment through of Ethiopia; Mohammod Abdela, Senior Expert, experimentation and knowledge creation to help National Bank of Ethiopia; Fillimon Rezene, Expert, governments and the private sector focus policy National Bank of Ethiopia; Abel Solomon, Senior and programs on scalable solutions with sustainable Communications Officer, National Bank of Ethiopia; outcomes. The UFGE is supported with generous Ambawork Mekonnen, Officer for Gender and Cross- contributions from Australia, Canada, Denmark, cutting Issues, National Bank of Ethiopia; Mawerdi Finland, Germany, Iceland, Latvia, the Netherlands, Abdurahman, Association of Ethiopian Microfinance Norway, Spain, Sweden, Switzerland, the United Institutions; Getachew Adem, Vice Commissioner, Kingdom, the United States, and the Bill and Melinda National Planning Commission of Ethiopia; Rebecca Gates Foundation. 7 INTRODUCTION Ethiopia has experienced remarkable economic firms account for only 44 percent of firms, they make success in recent years. In the past decade, its average up nearly 70 percent of failed businesses.5 annual growth rate far exceeded the regional average, at slightly over 10 percent relative to a regional 5 The unmet potential of women in the workforce is percent. Agriculture grew at 7 percent, services at 12 intrinsically linked to a lack of opportunities for women percent, and industry at 21 percent.1 in education, health, and human rights. Women are less literate, suffer from poorer health outcomes, and have Despite significant economic growth, however, fewer basic rights than men. These wide and pervasive women continue to face significant barriers in gender gaps hinder not only female livelihoods, but the workforce. Women experience high rates of also the potential for poverty alleviation and growth unemployment (50 percent), seasonal employment on a national level. (37 percent), and temporary employment (13 percent). Women are also less likely than men to 2 This report, the Ethiopia Gender Diagnostic, presents be paid for their work: over half of all women engaged evidence on the mechanisms underlying gender gaps in in the agricultural sector, for example, receive no the Ethiopian workforce. Using data from the 2011–2016 payment. Similar trends exist in other industries like Ethiopia Socioeconomic Surveys, this report provides small-scale manufacturing, where 58 percent of a detailed understanding of the constraints faced by female workers are unpaid family workers, relative to female farmers, entrepreneurs, and employees. To that 40 percent of male workers.3 end, the diagnostic makes four key contributions: Furthermore, for women active in the workforce, their First, this report provides an overview of the labor productivity lags behind that of men. Female farmers force in Ethiopia and identifies the factors that predict have lower rates of agricultural productivity than their whether, how much, and in what sector an individual male counterparts, and in entrepreneurship, female- works. Part I reveals that gender, among other factors, owned firms underperform those owned by men in serves as a strong predictor of workforce participation. an array of critical dimensions including profitability, A simple average indicates that women are 17 percent survival rate, average size, and growth trajectory.4 In less likely than men to participate in the labor force. a study of small- and medium-sized enterprises in This disparity widens to 29 percent when controlling Ethiopia, researchers found that while female-owned for demographic factors. 8 Women are also less likely than men to be paid for their work: over half of all women engaged in the agricultural sector, for example, receive no payment. Second, this report uses Oaxaca-Blinder Third, this report identifies the links between labor market decompositions to measure and account for gender skills, social norms, and gender gaps in the Ethiopian gaps in economic outcomes in agriculture, self- workforce. In doing so, Part III explores the mechanisms employment, and wage labor. Part II shows that women that hinder Ethiopian women from accessing or help fare worse off in agricultural productivity, business them to identify relevant resources and opportunities. revenues, and hourly wages than men due largely to differential access to productive resources like credit, Fourth, part IV provides policy makers with a menu of assets, inputs, and education. innovative programming examples. Virtually all of the innovations presented in part IV have the potential to be adapted to the Ethiopian context. BOX 1 Ethiopia Socioeconomic Surveys Implemented by the Central Statistical Agency The ESS was implemented in three waves: 2011–2012, (CSA) in collaboration with the World Bank’s 2013–2014, and 2015–2016. Data for the latter two Living Standards Measurement Study Integrated waves are nationally representative, allowing for a rich, Surveys on Agriculture (LSMS-ISA) team, the detailed analysis on the links between gender gaps in Ethiopia Socioeconomic Survey (ESS) produces the workforce and socioeconomic factors in Ethiopia. comprehensive, high-quality data on agriculture, For additional information on sampling, please refer to entrepreneurship, employment, interinstitutional appendix A1. Survey reports provided by LSMS, available collaboration, welfare indicators, and socioeconomic through http://microdata.worldbank.org/index.php/ characteristics across Ethiopia. catalog/2053, provide more granular details. 9 Part I An Overview of Women in the Workforce 10 BOX 2 Labor Supply in Ethiopia This report defines working-age individuals as Married individuals age 14 or above, in accordance with The Labor Proclamation of 2004, and working individuals as any working-age individual who spent one hour or more in the past week on agricultural activities; non 61% agricultural business activities; casual, part-time, or temporary labor; or any other work for which they received a wage or salary. 57% Part I uses the Ethiopia Socioeconomic Survey (ESS) third wave data (2015–2016), which represent 13,316 working-age individuals, of whom 52 percent are 0% 20% 40% 60% 80% 100% women. Some 53 percent of working-age people and 55 percent of working-age women work. Relative to working men, working women in our data have, on Widowed, Separated, or Divorced average, lower levels of education, are less likely to be widowed, separated, or divorced, and are less often the head of household. 4% Illiteracy Rate 20% 56% 35% 0% 20% 40% 60% 80% 100% Single Household Head 35% 24% 60% 23% 0% 20% 40% 60% 80% 100% Female Male 11 Gender serves as a strong predictor of workforce relevant: belonging to a household that has received participation in Ethiopia. A simple average indicates formal credit is associated with a higher likelihood that women in our data are 17 percent less likely than that an individual works. Women in households men to participate in the labor force.i This disparity that have received formal credit work more hours widens to 29 percent when taking into account other than men from such households. Fourth, household factors such as age, education, and household wealth. headship matters: household heads are more likely Among individuals active in the workforce, a gender to work than non-household heads, and female gap of 4.4 hours exists: while men work 31 hours per household heads work 3.2 more weekly hours, on week, on average, women work only 27. Table 1 displays average, than their male counterparts. Fifth, marital individual and household characteristics that, holding status is pertinent: married women work 4.2 fewer all else constant, predict whether an individual works hours per week than married men. and the extent to which he or she works.ii Factors highlighted in orange positively predict outcomes, Figures 1 and 2 display the breakdown of sectors in while factors highlighted in blue inversely predict which working individuals participate over time.iv outcomes.iii Factors that are not highlighted do not Based on the data, the majority work in agriculture, predict the outcome. followed by non farm enterprise, and wage employment.v Women make up roughly 40 percent of Table 1 reveals five key factors that strongly predict the agricultural sector and 30–40 percent of the wage workforce participation. First, age is critical: the sector. Meanwhile, women represent the majority of probability that a working-age individual works non farm enterprise operators, particularly in the increases with every additional year of life. This rural and small-town areas during 2011–2012. About trend is more prominent among women than men. 3 percent of women work in more than one sector, Second, education matters: women with a university with most of these individuals spending roughly half degree are 16 percent more likely to work than their of their time in non farm enterprise and the other half male counterparts. Third, accessing formal credit is in either the wage or agricultural sectors. i This section draws upon ESS third wave labor and time use data (2015–2016). ii Factors held constant in the remainder of part I include age, household headship, marital status, education level, child dependency ratio, whether the individual was ill or injured in the past month, household formal credit, household wealth, household size, and household geographic location. iii For example, in column 1, as an orange factor holds true and/or increases, the likelihood that an individual will work at least one hour increases, holding all else constant. Meanwhile, as a blue factor holds true and/or increases, the likelihood that an individual will work at least one hour decreases, holding all else constant. iv To calculate this, we use the total number of hours that a respondent reported working in each category, calculate the proportion of his or her time worked in each sector, then identify if he or she worked mostly in agriculture, entrepreneurship, or wage employment. If the person worked in equal proportions in two or more sectors, then he or she falls into the category “more than one sector.” v Agricultural activities include livestock- and fishing-related activities, either for sale or for household use. Non farm enterprise activities include all non agricultural or non fishing household businesses, big or small, whether for an entrepreneur or for the household. Wage employment is any work for a wage, salary, commission, or any payment in kind, excluding temporary employment. 12 TABLE 1 Factors That Explain If and How Much an Individual Works WORKED AT LEAST TOTAL NUMBER ONE HOUR OF HOURS (1) (2) FACTORS Female - Age (years) + + Female and age (years) + Age squared - Female and age squared Head of household + Female and head of household + Married + Female and married - Completed secondary - University degree - Female and university degree + Individual reported illness/injury - Female and reported illness/injury + Household received formal credit (ever) + - Female and household received formal credit (ever) + Wealth index - + Factors that positively predict outcomes Factors that inversely predict outcomes 13 FIGURE 1 Labor Participation by Sector 100% 90% 79% 80% 69% 70% 67% 60% 50% 40% 30% 27% 20% 17% 13% 11% 10% 5% 3% 3% 3% 4% 0% 2011–2012 2013–2014 2015–2016 Agriculture Non farm enterprise Wage More than one sector Note: To calculate this, we use the total number of hours that a respondent reported working in each category, calculate the proportion of his or her time worked in each sector, then identify if he or she worked mostly in agriculture, entrepreneurship, or wage employment. If the person worked in equal proportions in two or more sectors, then he or she falls into the category “more than one sector.” 14 FIGURE 2 Women’s and Men’s Labor Participation by Sector 2011–2012 55% 45% Overall 2013–2014 57% 43% 2015–2016 55% 45% 2011–2012 59% 41% Agriculture 2013–2014 59% 41% 2015–2016 57% 43% 2011–2012 33% 67% enterprise Non farm 2013–2014 42% 58% 2015–2016 46% 54% 2011–2012 72% 28% Wage 2013–2014 70% 30% 2015–2016 63% 37% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Female Male 15 Part II Measuring the Gender Gaps 16 Women active in the workforce still fare worse than percent of the population. The vast majority of men in agricultural productivity, business revenues, agricultural production, 90 percent, comes from and hourly wages in Ethiopia. To understand the factors the work of 12 million smallholder households.13 that help explain these gaps in economic outcomes, the Several key constraints, however, inhibit agricultural analysis in part II uses Oaxaca-Blinder decompositions productivity in Ethiopia and throughout Africa. to determine whether differential access to resources Population growth, climate variability, weak land like credit, assets, inputs, and education—or differences tenure systems, and low fertilizer use degrade arable in the returns to these resources—drive the gaps in land. 14 Poor transport infrastructure limits access women’s economic empowerment. to markets, information, and inputs. Inconsistent or low yields hinder smallholders’ ability to commercialize and participate in global supply AGRICULTURAL PRODUCTIVITY chains. Of particular relevance to smallholder farmers, limited access to credit through the formal In Ethiopia, roughly 40 percent of GDP comes from financial system inhibits their adoption of new the agricultural sector, which employs nearly 80 technologies and innovations. BOX 3 Background and Macroeconomic Context With a population of 105 million and a population to education, health, and water provision.8 More recently, growth rate of 2.85 percent, Ethiopia is the second foreign direct investment (FDI) has played an influential largest country on the African continent.6 Ethiopia’s role in national growth. In 2016, Ethiopia registered over average annual growth rate of slightly over 10 percent in $3 billion USD in FDI, an increase of 46 percent from the the past decade has far exceeded the regional average previous year. China, India, Turkey, and the European Union of 5 percent. Services grew at 12 percent, industry at 21 (EU) were the primary sources of Ethiopia’s 2016 FDI.9 percent, and agriculture at 7 percent.7 While Ethiopia remains predominantly agricultural, with 80 percent of Ethiopia’s sustained economic growth over the past its population living in rural areas, recent rapid economic decade has dramatically decreased poverty rates, which growth disproportionately favoring services and industry fell from 30 percent to 24 percent between 2011 and signals the advent of a demographic transition. 2016.10 Rapid expansion in the agricultural sector, which employs most of the population living below the poverty Ethiopia’s flourishing growth comes in no small part line, played a particularly critical role in reducing the from public investment, which increased from a mere poverty rate: each percent increase in agricultural output 5 percent of gross domestic product (GDP) in the early led to a decrease in poverty of nearly one percent.11 If 1990s to 19 percent in 2011, strengthening everything from Ethiopia continues on its current growth trajectory, it will power production, roads, railways, and industrial parks advance to middle income status by 2025.12 17 Though women make up more than 40 percent of the agricultural labor force and head approximately 25 percent of all farming households, they have less access to land and other factors of production than men. While all smallholder farmers face constraints to Addressing these challenges is a necessary step to fulfill productivity, female farmers encounter particularly ambitious targets such as those that were set in the acute challenges. Though women make up more than Growth and Transformation Plan II (GTP II), including 40 percent of the agricultural labor force and head achieving 8 percent growth in agriculture and allied approximately 25 percent of all farming households, sectors and increasing the percentage of rural women they have less access to land and other factors farmers who are benefiting from extension services of production than men.15 What is more, women to 30 percent. More research is needed to identify experience lower returns than men from a given level scalable solutions that can effectively address specific of resource expenditure.16 These lower returns point to barriers faced by women farmers. broader social norms, market failures, and institutional constraints that prevent resources from translating Part II provides an in-depth analysis of the factors into the same levels of agricultural productivity as they linked to lower agricultural productivity among female would for men. farmers in Ethiopia. 18 BOX 4 Who Is a Farm Manager? This report defines farm managers as individuals Married who have decision-making power over one or several land parcels, or a piece of land divided into fields. These decisions may include how or when to prepare land, sow crops, weed, harvest, process, 24% produce, or sell a surplus. This section uses the Ethiopia Socioeconomic 94% Survey (ESS) third wave data (2015–2016), which represent 2,907 farm managers, of whom 21 percent are women. Relative to male farm managers, female farm managers in our data are, on average, five years 0% 20% 40% 60% 80% 100% older; are more likely to be illiterate; have smaller households; and are less likely to be married. Widowed, Separated, or Divorced 74% Illiteracy Rate 4% 88% 59% 0% 20% 40% 60% 80% 100% Single 5 years older 2% 2 fewer household members 2% 0% 20% 40% 60% 80% 100% Female Male 19 BOX 5 How We Measure Agricultural Productivity Agricultural productivity is defined in this report as measured by farm managers’ estimates. Monetary the average value of agricultural output produced values are based on local and regional sales price per unit of land managed (in hectares). The analysis information to allow for comparability and land uses the total value of agricultural output per unit area is measured either by GPS devices or by farm of land, where the quantity of agricultural output is managers’ own estimates. BOX 6 Examining Differences in Levels of and Returns to Resources The Oaxaca-Blinder decomposition has been they may not achieve the same results. The structural employed extensively in the existing literature to effect thus refers to the portion of the gender gap isolate the factors contributing to gender gaps in that exists because of differences in the returns to agricultural productivity and wages, among other resources. For example, the structural effect captures outcomes.17 The decomposition delineates the gender the difference in agricultural output per hectare gap into two main components: the endowment for men versus women, given the same levels of effect and the structural effect. The endowment education, equivalent use of fertilizer, or equal effect captures the difference in levels of resources, amounts of credit. Discrimination, social norms, and such as education, fertilizer, or amount of credit that institutional constraints all perpetuate the structural women have relative to men. Policies and programs effect. In this report, the endowment effect may be may diminish the endowment effect by ensuring referred to as “levels,” while the structural effect equal access to and use of resources for men and may be referred to as “returns.” Appendix B provides women. However, even when men and women have additional technical details on the Oaxaca-Blinder access to the same quantity and quality of resources, decomposition. 20 Measuring the Gap Accounting for the Gap A simple average indicates that female farm managers Compared to male farm managers, female farm in our data produce 36 percent less per hectare than managers control smaller plots of land, cultivate fewer their male counterparts.vi This disparity, however, crops, use fewer inputs, and are less likely to access lessens to 6 percent when considering individual-, extension programs and formal credit. All of these household-, and plot-level characteristics (figure 3).vii factors are associated with women’s lower yields.viii The magnitude of the difference between the Table 2 displays the manager-, household-, and plot- conditional and unconditional averages signals that, in level characteristics that, holding all else constant, large part, the agricultural gender gap may stem from contribute to the gender gap in productivity via levels unequal levels of productive factors. and returns, respectively. Factors highlighted in orange widen the gender gap, while factors highlighted in blue narrow the gender gap. Factors that are not highlighted neither widen nor narrow the gap.ix FIGURE 3 Unconditional and Conditional Gender Gap in Agriculture Unconditional 36%*** Conditional 6% 0% 10% 20% 30% 40% Percentage Note: The symbols */**/*** denote statistical significance at the 10%, 5%, and 1% levels, respectively. vi Part II draws upon ESS third wave labor and time use data (2015–2016) and follows a similar approach to that in Aguilar et al. (2014), which employs data from the 2011–2012 Ethiopian Rural Socioeconomic Survey to compute the gender gap in agriculture for Ethiopia. vii The characteristics used to compute the conditional gender gap are 1) at the individual level: age, education, marital status, civil status, and religion; 2) at the household level: family labor, skills, household size, child dependency ratio, receipt of technical assistance, and household asset wealth index; and 3) at the plot level: ownership, log plot size, indicator variables for good soil quality, irrigation, remoteness of the plot, fertilizer and pesticide use, household land, hired labor, as well as percentages for cultivating cash crops, food crops, horticulture crops, and indicator variables for different agro-climatic conditions. See appendix A2 for descriptive statistics and details. viii The term extension programs in this context refers to the advisory services provided by agricultural extension workers. ix A much wider set of factors was tested in the decomposition but here we focus on only those found to significantly impact the gender gap. Appendix A2 provides additional information on the construction of the variables and descriptive statistics by gender for each of the factors that were taken into account during the gap analysis. 21 These findings highlight four key points: On the demand side, lack of access to agricultural education and complementary inputs often hinders 1. Women receive fewer extension services than men women from properly understanding and advocating for their extension needs. In Ethiopia, policies have Agricultural extension services are the primary recognized and prioritized the importance of closing platform through which smallholder farm managers the gender gap in access to extension services. The access information about new technologies and challenge now lies in identifying and effectively information. Female farm managers in our data, addressing constraints that keep women farmers from however, are less likely than their male counterparts benefiting from the existing extension system.xii to have attended an extension program, leading to a widening of the gender productivity gap.x Figure 4 2. Women access formal credit less than men displays the gender gap in extension attendance, which has increased over time. This gap implies that women Credit and other financial services can provide small- are, relative to men, increasingly less exposed to and scale farmers with the opportunity to improve farm aware of new techniques, farming knowledge, and productivity and transition from subsistence farming to management practices. Both supply- and demand- large-scale and commercial farming.20 In the short run, side factors may perpetuate the gap in extension credit can help farmers increase their purchasing power program attendance.18 On the supply side, top-down to acquire necessary production inputs and finance their agricultural extension models historically targeted operating expenses, while in the long run it can help farmers who were more likely to adopt technological farmers to make profitable investments.21 Female farm innovations, frequently leaving out female farmers.xi,19 managers in our data, however, are 9 percentage points FIGURE 4 Attendance in Extension Programs 38% 40% 35% 31% 30% Percentage 21% 22% 23% 20% 10% 0% 2011–2012 2013–2014 2015–2016 Female plot manager Male plot manager x This result is related to the mean comparison described in appendix A2. This sample includes all those farmers who had a positive productivity in the last harvest season and for whom we have information on individual, household, and plot characteristics. xi Because adopting new technologies requires large initial investments of time that most women did not have due to domestic duties, women were a less attractive target group for extension programs. xii The Women Development and Change Package recognizes that female farmers have restricted access to extension services and highlights a set of agricultural extension services that women should benefit from. The services cover a broad range of options including input use, labor-saving technologies, participation in horticulture, nutrition-dense crop production, irrigation soil management, and agro-processing. Moreover, the Gender Equality Strategy for the Agriculture Sector proposes to address these limitations through capacity building of staff on gender-sensitive planning, programming, and service delivery. 22 TABLE 2 Factors That Explain the Gender Gap in Agricultural Productivity FACTORS LEVELS RETURNS MANAGER AND HOUSEHOLD CHARACTERISTICS Farmer attended extension programs + Household ever received formal credit - Household size + - LAND CHARACTERISTICS AND AGRICULTURAL PRACTICES Total land managed - Total number of crops harvested + Pesticide, herbicides or fungicide (% of total) + + Oxen per hectare - Total hired labor use (days/ha) - Factors that widen the gender gap Factors that narrow the gender gap 23 less likely to live in a household with access to credit One reason why women, relative to men, may use less than male farm managers. fertilizer and other chemical inputs is that they are typically sold in large quantities, requiring a sizable One reason why women may have less access to formal upfront cost that cash-constrained women may struggle credit is that they are less likely to own and control to afford. Furthermore, women can be constrained in their physical assets that serve as collateral. Furthermore, mobility and limited in the transport options available to on average, women have lower levels of human and them. In remote rural areas, this affects access to fertilizer social capital which, in turn, can decrease their eligibility and, more generally, access to markets.23 Women farmers for formal credit. This puts women at a disadvantage: may also require lower quantities of chemical inputs due when credit is constrained, farmers are likely to use to their smaller average plot sizes. suboptimal levels of productive inputs, thereby limiting their productive capacity.22 BUSINESS SALES 3. Women manage less land and harvest a narrower range of crops In Ethiopia, 38.6 percent of the GDP comes from the informal sector, which includes self-employment. This Having more land and harvesting more diversified crops figure is consistent with the average of 38.4 percent for matter. Female farmer managers in our data, however, Sub-Saharan Africa and 38 percent for all low-income manage an average of only 0.6 hectares, compared countries.24 As in most countries in Sub-Saharan Africa, to an average of 1 hectare for male farm managers. self-employment opportunities lie largely in small Improving women’s access to land is a challenge that informal firms. The self-employed sector, on which this is highlighted in existing policy priorities in Ethiopia.xiii section focuses, is particularly important for women, Female farmers also grow a narrower range of crops who tend to have fewer opportunities in the wage sector. than their male counterparts. While the size of land managed reduces the gender gap in productivity, with In general, female business managers have less access smaller plots tending to be more efficient, the less to resources to grow and formalize their businesses, varied range of crops harvested widens the gender gap leading to smaller firms with lower profitability, in productivity. compared to male business managers. While many women enter informal self-employment as a last resort 4. Women access fewer production inputs with little intention to grow beyond a subsistence level, other women launch firms that have the potential to Modern agricultural inputs such as fertilizers, pesticides, grow into larger enterprises integrated within the formal herbicides, and fungicides were developed to increase sector. In practice, it is often difficult to distinguish agricultural productivity and protect farmers against between “necessity” entrepreneurs, who may have harvest fluctuations linked to pests, adverse weather, little entrepreneurial motivation beyond meeting and soil degradation, and thus mitigate crop losses. subsistence needs, and “opportunity” entrepreneurs, Female farm managers in our data use lower levels of who see an opportunity and make a decisive choice to these agricultural inputs (by about 2 percentage points) start a business with ambitions to thrive.25 It is important than their male counterparts, which limits productivity to keep these realities in mind when reading the results and may imply greater vulnerabilities to shock-induced on economic outcomes of self-employed women in variations in production. this section. xiii The Gender Equality Strategy for the Agriculture Sector emphasizes the need to support the revision and implementation of land-related policies in Ethiopia to increase the profitability and productivity of women in the agriculture sector. Some strategies and programs have been developed to address these needs. For instance, the Women Development and Change strategy has listed a set of interventions related to ownership, access, and use of land. Among these are, encouraging sharecropping where women lack the required labor to cultivate their land, ensuring women obtain fair sharecropping agreements, assigning plots to landless women, and making women aware of their land ownership rights. 24 BOX 7 Who Is a Business Manager? A business manager is an individual within a Married household in charge of the decisions regarding the earnings from an enterprise. This section uses the Ethiopia Socioeconomic Survey (ESS) third wave data (2015–2016), which represents 1,822 enterprises 40% and 1,600 principal managers, of which 40 percent are women. Relative to male managers, female managers in our data are, on average, less educated; 84% more likely to be household head; less likely to be married; and have smaller households. Male and female managers are, on average, the same age. 0% 20% 40% 60% 80% 100% Some 57 percent of principal managers work by themselves and 43 percent have a co-manager. Finally, 34 percent of female managers and 14 percent Widowed, Separated, or Divorced of male managers work in manufacturing, while 31 percent of female managers and 38 percent of male managers work in trade (figure 5). 49% Illiteracy Rate 4% 0% 20% 40% 60% 80% 100% 58% 34% Single 11% Same age 1 fewer household member 12% 0% 20% 40% 60% 80% 100% Female Male 25 FIGURE 5 Women’s and Men’s Enterprises by Sector 5% Agriculture production 10% 34% Manufacturing 14% 33% Services 38% 31% Trade 38% 1% Other 8% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Female Male Note: The calculations of shares by sector use data on enterprises at the manager level. For managers with multiple businesses in different sectors, more than one sector is recorded. As a result, the sums of the shares from each sector do not equal 100 percent. 26 Measuring the Gender Gap level factors, including the costs of operating the business, indicating that the gender gap may stem The simple difference in averages indicates that largely from differences in levels of resources female business managers’ sales are nearly 79 (figure 6).xv percent less than those of male managers.xiv This disparity, however, lessens to 24 percent when We use sales as a proxy for earnings given that considering individual-, household-, and enterprise- business profits are not directly reported in the data. FIGURE 6 Unconditional and Conditional Gender Gap in Entrepreneurship Unconditional 79%*** Conditional 24%** 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Percentage Note: The symbols */**/*** denote statistical significance at the 10%, 5%, and 1% levels, respectively. xiv This section draws upon ESS third wave labor and time use data (2015–2016). Because profits are not directly reported in the ESS, this analysis uses sales as a proxy for earnings, while controlling for costs incurred. xv The characteristics used to compute the conditional gender gap are 1) at the individual level: age, education, civil status, and religion; 2) at the household level: household size, child dependency ratio, and wealth index; and 3) at enterprise level: years of operation, number of hired workers, number of managers in firm, formal credit, logarithm of total costs, source of capital, main client, industries, and geographical location. See appendix A3 for descriptive statistics and details. The decomposition results focus on only those variables found to significantly impact the gender gap. 27 Accounting for the Gap extra income, the manager may decide to shift her time toward domestic activities and chores. Alleviating time Compared to male managers, female managers constraints stemming from domestic duties might be a spend less time on business activities; hire less labor; promising policy option to allow women entrepreneurs are less likely to have a business license; and access to allocate more time towards their businesses.xvi less formal credit. All of these factors appear to be Finally, women’s fewer working hours could simply linked to women’s lower business revenues. Table 3 reflect how work may become less safe after certain displays the individual-, household-, and enterprise- hours, and women decide not to work during those level characteristics that, holding all else constant, hours to avoid getting harassed or harmed. contribute to the gender gap in business sales via levels and returns, respectively. Factors highlighted in 2. Women hire less labor than men orange widen the gender gap, while factors highlighted in blue narrow the gender gap. Factors that are not Many businesses operate in labor-intensive sectors in highlighted neither widen nor narrow the gap. These which labor holds a high premium. Our data indicate findings highlight four key points: that, relative to male managers, female managers hire 0.3 fewer employees, and each additional employee is 1. Women spend less time on business activities associated with a 3.9 percent increase in monthly sales. than men Women may hire less labor due to a number of factors. For instance, women may have less entrepreneurial The quantity of time that self-employed managers training or weaker business support networks and dedicate to their firms is closely correlated with thus face more challenges in growing enterprises and business sales, especially in the context of small managing a larger body of employees.26 enterprises. In our data, each additional hour per week spent by a manager on his/her business is associated 3. Women are less likely to have a business license with a 1 percent increase in monthly sales. On average, self-employed women spend 17 hours per week on In our data, business licenses are positively correlated self-employment activities, compared to their male with entrepreneurial sales. This might be because counterparts’ 23 hours. licenses facilitate access to cheaper capital, larger markets, and additional business opportunities, for Examples of why self-employed women might work example, through government contracts. While the fewer hours relate to domestic responsibilities, share of managers with licensed enterprises has enterprise type, and gender-specific risks in increased over the years, the unconditional gender entrepreneurship. Since women often take on gap has not decreased (figure 7). Indeed, female- the majority of the domestic responsibilities, they managed enterprises are less likely to have a business necessarily have less available time for business- license than those managed by men. In our data, only related activities. For this and a number of other 15 percent of female managers operate an enterprise reasons, female-managed enterprises are infrequently with a business license, compared to 37 percent of the main source of household income. As such, once a male managers, and this difference widens the self- female-managed business achieves its desired level of employment sales gap. xvi In order to alleviate women’s time constraints, the Women’s Development and Change Package proposes labor-saving household technologies to ease the burden of household chores and the establishment of child care facilities to reduce the time women have to spend on childcare. 28 TABLE 3 Factors That Explain the Gender Gap in Self-Employment FACTORS LEVELS RETURNS MANAGER CHARACTERISTICS Head of household is the manager + Total number of hours spent on business + - HOUSEHOLD CHARACTERISTICS Wealth index + Household size + ENTERPRISE CHARACTERISTICS Total costs (log) + Enterprise of manager has a license + Number of hired workers + Formal credit - Amount borrowed for this enterprise + Factors that widen the gender gap Factors that narrow the gender gap 29 FIGURE 7 Share of Women and Men Managers with a Registered Business 40% 40% 34% 30% Percentage 20% 20% 17% 12% 10% 6% 0% 2011–2012 2013–2014 2015–2016 Female manager Male manager 4. Women access formal credit less than men improve firm performance by promoting business growth and boosting employment levels.28 Access to formal credit can ease capital constraints and thus spur firm growth, as well as protect WAGE EARNINGS businesses from economic shocks.27 Male managers in our data are 3.7 percent more likely to take out loans than female managers. What is more, in terms Until recently, wages in Ethiopia remained exceedingly of the size of the loans, male managers borrow about low: in 2009, the average wage in Ethiopia was only 50 percent more than female managers. one-third of the Sub-Saharan African average and less than one-half of the global average for low-income What really matters for entrepreneurs is not whether economies. In 2012, the monthly average real income they borrow money, but rather how much money they was ETB 421.70 (USD 23.40): less than USD 1.25 per borrow: while managers with credit in our data have day.29 Low levels of productivity and investment likely lower sales than non-credit-borrowing managers, contributed to stunted wage growth.30 credit-borrowing managers’ sales increase as their quantity borrowed increases. In our data, male employees are more likely to work in agricultural production and construction industries, In general, however, women face more barriers in while female employees are more likely to work in accessing formal credit, often failing to meet minimum the manufacturing and education sectors (figure 8). eligibility criteria such as collateral requirements. When While no significant gender differences exist in they do access credit, they tend to receive smaller loans, participation in the services and trade sectors, there which are often too small to meaningfully invest in their are, however, significant gender differences in the firms, thus widening the gender gap. A study from the wages that men and women receive for their work. World Bank’s Women Entrepreneurship Development Women are less likely to receive wages than men. Project (WEDP) in Ethiopia suggests that offering larger When they do, female wages are, on average, less loans to growth-oriented women entrepreneurs can than male wages. 30 BOX 8 Who Is an Employee? In our analysis, an employee is an individual Married who reported at least one paid job over the last 12 months. This section uses the Ethiopia Socioeconomic Survey (ESS) third wave data (2015–2016), which include 1,347 workers, of whom 45% 37 percent are women. Relative to male employees, female employees in 65% our data are, on average, 4 years younger than male workers; less educated; living in smaller households; and less likely to be married. Male employees are more likely to work in agricultural production and 0% 20% 40% 60% 80% 100% construction industries, while female employees are more likely to work in manufacturing and education (figure 8). Widowed, Separated, or Divorced 22% University Degree 6% 12% 20% 0% 20% 40% 60% 80% 100% Single 4 years younger 33% 1 fewer household member 29% 0% 20% 40% 60% 80% 100% Female Male 31 FIGURE 8 Women’s and Men’s Wage Employment by Sector 6% Agriculture production 10% 37% Services 36% 12% Manufacturing 8% 6% Construction 13% 5% Trade 4% 19% Education 15% 24% Other 22% 0% 10% 20% 30% 40% Female Male 32 Measuring the Gender Gap considering individual-, household-, and job-level characteristics (figure 9).xviii The marginal difference A simple difference in averages across both formal and between the conditional and unconditional average informal sectors indicates that female employees in suggests that, in large part, the gender wage gap our data earn 44 percent less per hour than their male stems from unobservable characteristics like aptitude, counterparts.xvii This disparity drops to 36 percent when motivation, and formal versus informal sector work. FIGURE 9 Unconditional and Conditional Gender Gap in Wages Unconditional 44%*** Conditional 36%*** 0% 10% 20% 30% 40% 50% Percentage Note: The symbols */**/*** denote statistical significance at the 10%, 5%, and 1% levels, respectively. xvii This section draws upon ESS third wave labor and time use data (2015–2016). xviii More precisely, the characteristics used to compute the conditional gender gap are: 1) at the individual level: whether employee is head of household, age, education, civil status, religion, and whether worker reported to be ill or injured; 2) at the household level: household size, child dependency ratio, and wealth index; and 3) at the job level: time spent on agricultural and non farm enterprise activities, type of employment, and industry of employment. See appendix A4 for descriptive statistics and details. The decomposition results focus on only those variables found to significantly impact the gender gap. 33 Gender differences in education partly explain the gender wage gap. Secondary and post- secondary education help individuals develop more advanced skills to garner higher wages: employees in our data who hold a bachelor’s or graduate degree, have, on average, a 50 percent higher hourly wage relative to individuals who only completed secondary education, and a 20 percent higher wage than those who only completed their primary education. Accounting for the Gap As before, factors that are not highlighted neither widen nor narrow the gap.xix These findings highlight Compared to male employees, female employees are two key points: younger; less likely to be married; more likely to hold a diploma or certificate as the highest degree; and less 1. Gender differences in education, age, and marital likely to hold a university degree. All of these factors status explain part of the wage gap seem to contribute to women’s lower hourly wage. In contrast to the agriculture and self-employment Gender differences in education partly explain the gaps, however, few factors other than demographics gender wage gap. Secondary and post secondary explicitly explain the gender wage gap. As such, education help individuals develop more advanced skills a large share of the gap stems from unobserved to garner higher wages: employees in our data who hold characteristics that were not included in the a bachelor’s or graduate degree, have, on average, a 50 analysis—either because they are difficult to measure percent higher hourly wage relative to individuals who or because it is unknown which characteristics only completed secondary education, and a 20 percent should be measured. Table 4 displays the individual-, higher wage than those who only completed their household-, and job-level characteristics that, primary education. While more women have diplomas holding all else constant, relate to the gender wage as their highest degree (22 percent of women compared gap via levels and returns, respectively. Factors to 12 percent of men), fewer women have university highlighted in orange widen the gender gap, while degrees as their highest degree (12 percent of women factors highlighted in blue narrow the gender gap. and 20 percent of men have at least a bachelor’s degree). xix A much wider set of factors was tested in the decomposition, but here we focus on only those factors found to significantly impact the gender gap. See appendix A4 for descriptive statistics and details. 34 Age and marital status also explain part of the gender 2. Role of unobservable factors in gender wage gap wage gap. Age is positively associated with wages. This trend contributes significantly to the gender wage Characteristics that are unobservable in the ESS data gap as male employees in our data are, on average, that is used in this analysis—such as motivation, or five years older than female employees. Marriage inter-personal communication skills—may explain is also positively associated with wages. This trend, some portion of the gender gap. These characteristics too, contributes significantly to the gender wage gap. are unaccounted for in this analysis because they are Some 65 percent of the male employees are married, difficult to measure and/or identify. A significant share compared to 45 percent of the female employees. of the gender wage gap may also possibly arise due to Being widowed or divorced is positively associated with gender-based discrimination, whether intentional or earnings. Given that only 6 percent of male employees subconscious, or because women in Ethiopia are more are widowed or divorced, compared to 22 percent of likely to work in the informal sector, where wages are female employees, this narrows the gender gap. usually lower than those in the formal sector.31 TABLE 4 Factors That Explain the Gender Gap in Wage Earnings FACTORS LEVELS RETURNS EMPLOYEE CHARACTERISTICS Age (years) + Age squared - Married + Widowed or divorced - Diploma or certificate - University degree + Factors that widen the gender gap Factors that narrow the gender gap 35 Part III Contextual Factors Influencing the Gender Gaps 36 The gender gap analysis in this report presents clear opportunities, and a perpetuation of disparities in skill evidence that the differences in access and returns sets, job opportunities, and wages. to resources result in poorer economic outcomes for Narrowing the gender gap in labor market skills women in Ethiopia: lower agricultural productivity (36%), requires consideration of literacy, technical and business sales (79%), and wage income (44%) compared vocational knowledge, informal learning channels, and to men. In agriculture, lower access to and usage of entrepreneurial and noncognitive skills. This section agricultural extension services, agricultural inputs, and seeks to explore these areas further. formal credit, as well as lower crop diversity appear to drive the gender productivity gap. Meanwhile, in self- employment, differential business revenues between Basic Literacy men and women stem from differences in time spent Formal education equips young girls with literacy on business activities, access to hired labor and credit, and business licensing, whereas in wage employment, Literacy is an important cornerstone of workforce demographic factors and education help explain some skill sets: it facilitates technical and vocational skill (but not all) of the gender gap in income. development and predicts participation in a broader range of businesses, including those in more profitable, In addition to the key drivers identified in the report, high-value-added sectors.35 Some 42 percent of there are other important factors that may affect Ethiopian women are literate, relative to 69 percent the gender gaps in economic outcomes. Notably, of men.36 gender gaps in agriculture, entrepreneurship, and wage employment do not exist independently of Most individuals learn literacy skills through the one another, but rather stem from and are linked formal education system, for which attendance has through an underlying set of contextual factors and increased in the past two decades, especially for girls social norms. The sections in part III explore these and women. Between 2000 and 2016, the share of links further, focusing on the role of labor market women who had never attended school dropped skills and social norms in influencing gender gaps in from 77 percent to 49 percent. Similarly, the gender the Ethiopian workforce. In doing so, these sections parity index, or the ratio of female to male primary highlight the particular mechanisms that hinder school attendance, increased from 73 percent to 99 Ethiopian women from accessing or help them to percent.37 The effects of increased attendance on identify relevant resources and opportunities. literacy are beginning to show for younger cohorts, with the current gender literacy gap for individuals LABOR MARKET SKILLS ages 15–24 standing at only 3 percent. Attempts to decompose gender differentials in labor While attendance rates and subsequent literacy rates market outcomes in Ethiopia suggest that between for younger cohorts have improved dramatically 20 and 39 percent of the gender wage gap for formal over the past decade, low schooling quality and sector workers stems from differences in education, educational outcomes still present a critical challenge. experience, and training.32 Women’s limited labor A non-negligible percentage of students enrolled in market skills pigeonhole them into jobs concentrated primary school may still exit the education system in low-profitability sectors with more women working without attaining basic skills, posing particular risks for in informal wage employment than men.33 Furthermore, young women, who drop out of school more rapidly 37 percent of women report seasonal employment than their male counterparts.38 Indeed, the gender and 13 percent report occasional employment.34 These parity index in attendance drops with age, standing at trends of sporadic employment likely lead to less 0.98 for the second cycle of primary school, 0.88 for on-the-job training, fewer professional development secondary school, and 0.76 for preparatory school.39 37 Ensuring gender sensitivity in schools boosts The gender gap in adult education starkly diverges attendance and learning from that in primary, secondary, and even tertiary school, since women are likely to experience more Formal and non formal education may fail to offer time and social constraints with age relative to men. girls and women a safe, accessible, and gender- Enrollment rates in adult education programs reflect sensitive environment. In some cases, they can face this reality: while 70 percent of illiterate men are safety concerns in reaching education due to violence enrolled in adult education programs, only around experienced on the way to, from, and at schools and 40 percent of illiterate women are enrolled in such learning centers. These security risks may fuel women’s programs.43 In order to facilitate women’s involvement preferences to stay at home in order to avoid abuse in continuing education programs, it may be effective and violence. Furthermore, education centers often to offer financial incentives to offset time costs do not meet girls, and women’s sanitary needs due to associated with attendance and transit. a lack of latrines and clean water. Technical and Vocational Knowledge Initiatives from the World Bank, United Nations, and other multilateral organizations focus on reducing the Mid-level and urban workers can access formal distance to schools and learning centers, facilitating technical education through TVET and ATVET safe transportation, hiring more female instructors and instructors who are well-attuned to the needs The gender gap in technical knowledge can lead women of female students, and ushering boys and men into to self-sort into less technical positions. Accordingly, discussions on cultural and societal practices.40 women have a much smaller presence in technical positions in the formal wage sector, accounting for roughly 30 percent of professional workers. Formal Adult and non formal education gives older cohorts technical and vocational education and training (TVET) the tools to build literacy skills and agricultural technical and vocational education While younger cohorts’ gender literacy gaps are and training (ATVET) address these skill gaps, seeking decreasing, overall gender literacy gaps remain wide to create mid-level workers for the formal wage and due to a persistent share of older illiterate cohorts. agricultural sectors. These programs cater to individuals Adult and non formal education programs cater to who passed grade 10 and received sufficient scores on this generation of women. In particular, they target the national grade 10 exam. illiterate individuals, individuals with some schooling, and dropouts of the formal education system with TVET offers pre-employment training, additional training the goal of putting them on a path toward formal for employed workers, and remedial training for workers education, literacy, and skill training. Out-of-school outside of the wage labor force. Individuals can participate children ages 7–14 may enroll in Alternative Basic in TVET at various stages in their formal education.xx Education, a three-year school equivalency program In rural areas, over 25 ATVET colleges provide agricultural that covers material from the first four years of primary skills training to both technical graduates entering the school.41 Children and adults ages 15 and up may enroll labor market as extension workers, and to smallholder in Adult Literacy Training to learn literacy skills. Finally, farmers with limited formal education. Extension any adult may enroll in community skills training workers, or development agents, complete three years of centers (CSTCs) to learn basic literacy, numeracy, and training and are then assigned to farmer training centers other entrepreneurship and trade-related skills via in rural localities. There, they provide information to and regionally governed centers.42 conduct training sessions for farmers. xx Students who do not pass the nationally administered exam in grade 10 and cannot continue to preparatory school may continue to government TVET colleges if they receive sufficient scores on the grade 10 exam. Students who do not receive sufficient scores on the nationally administered exam in grade 12 to attend university can enter TVET colleges. Students can also enter the TVET track at a higher level from university. 38 Female enrollment rates in TVET and ATVET are Training programs for traditionally female jobs thus high, though room remains for more programmatic reinforce a “gender-stereotyped distribution of skills” in inclusivity a landscape in which women make up only 20 percent of science and engineering professionals, 21 percent of The constraints that limit women’s participation business services agents, and 30 percent of information in formal education seem to fade with respect to and communication technologies (ICT) technicians.47 TVET and ATVET. Women’s enrollment rates in these programs are nearly at parity with men’s: of the 237,877 Social networks, however, can help to broaden women’s students enrolled in the Government of Ethiopia’s training and employment decisions. Research by the TVET program in the 2012–2013 academic year, over World Bank Gender Innovation Lab indicates that 50 percent of students were female.44 However, it women who take TVET courses in traditionally male- is also important to continue encouraging female dominated fields like furniture making, manufacturing, enrollment in technical and vocational training, for or electricity do so largely due to their existing social example, by facilitating easier and safer transportation networks in the given field and their consequent to education centers, ensuring gender-sensitive exposure to the field and its earning potential.48 spaces, or offering financial incentives for attendance. As such, occupational segregation results just as much from a woman’s social capital and exposure as it To break down the barriers between female farmers does from gendered social norms on what is “men’s and extension services, the Women Development work” or not. Facilitating social networks for women and Change Package highlights a set of agricultural and increasing the presence of female role models extension services from which women may benefit. and instructors in higher-level education can shift The services cover a wide range of issues including occupational gender norms, creating pathways for input use, labor-saving technologies, participation women into more lucrative industries and professions. in horticulture, nutrient-dense crop production, irrigation, soil management, and agro-processing. The ICT-based methods jump the physical hurdles to strategy furthermore proposes to educate staff on education and facilitate learning gender-sensitive planning, programming, and service delivery in order to better serve female farmers. Education offered outside of brick-and-mortar settings Finally, it suggests integrating relevant information, can significantly impact women’s economic outcomes. technology, and trainings into extension packages in Programs that operate via print, radio, or television, for order to respond effectively to women’s needs. example, do not require physical attendance, thus unlocking opportunities for women to learn without Social networks can nudge women into more lucrative constraints like transportation costs, distance, or technical training threats of violence in transit. These methods will likely grow in conjunction with digital radio transmission While women’s enrollment rates in technical training in many parts of the developing world; they are programs are nearly at parity with men’s, women highlighted as a particularly low-cost intervention in tend to enroll in shorter-term technical training challenging educational environments.49 programs that provide skills for less lucrative, more traditionally female positions in commerce, textiles, ICT-based methods like mobile phone-based and hospitality.45 Meanwhile, men more often enroll applications may also effectively facilitate knowledge, in longer-term programs that provide training in though not without shortcomings. Internet and mobile science, technology, engineering, and math (STEM) phone penetration remain low in Ethiopia at about 12 fields or construction. Such fields have greater and 43 percent, respectively.50 Furthermore, women, potential for higher incomes and executive functions.46 and rural women in particular, are not likely to have 39 In fostering social networks, it is important to actively consider and address the factors that often limit women’s capacity to access and leverage networks: mobility constraints, time poverty, and cultural norms that may restrict female and male interactions. the same access to these technologies as men due to a source of information for agricultural input use as limited resources and social norms.51 While 46 percent are formal public sources.56 Thus, a well-established of male-owned businesses used mobile phones social network gives individuals a community for business purposes, only 3 percent of female- through which they can share information and owned businesses did.52 As such, any attempts to resources. In fostering social networks, it is important engage the population with ICT-based mobile phone to actively consider and address the factors that methods should operate within the framework of low often limit women’s capacity to access and leverage broadband penetration and limited access to smart networks: mobility constraints, time poverty, and phones, particularly among women. cultural norms that may restrict female and male interactions. Social Networks Formal female professional networks foster woman- Low-level and rural workers acquire technical to-woman knowledge transfer knowledge through social networks and experience Formally organized female professional networks Informal learning channels like social networks facilitate social norm change, encourage relationships and experience play an important and often with and exposure to female role models, and foster underestimated role in disseminating technical, woman-to-woman knowledge transfer. In doing so, entrepreneurial, and noncognitive skills. In these networks expose young women to new industries entrepreneurship, previous experience is equally and occupations and build powerful formal social important as, if not more important than, formal capital that women may otherwise lack. In urban areas, education in determining productivity levels.53 In such networks may connect women who are engaged in agriculture, experience and learning by doing have formal wage labor and/or are female business owners. In positive effects on technical efficiency, especially in rural areas, existing women’s cooperative organizations traditional environments.54 can provide a mode of technical knowledge transfer In addition to experience, social networks matter. In and learning that women may otherwise be unable Ethiopia, where 74 percent of women and 62 percent to achieve independently. Importantly, these types of of men lack access to print, television, or radio, social female organizations can operate under existing social networks are an efficient way to spread information norms that often restrict women’s interactions with and technical knowledge in a relatively quick manner. 55 men outside of their immediate family and thus affect Research finds that neighbors are just as important women’s networking opportunities. 40 Social networks improve outcomes for entrepreneurs Further, social networks provide entrepreneurs with knowledge of relevant macro-level trends and market Social networks play a particularly important role conditions.58 Of specific importance to women, social in entrepreneurial endeavors. For example, social networks help entrepreneurs to secure more favorable networks often help entrepreneurs to identify input and output prices for goods. In a qualitative business and labor market opportunities, as well as study in Southern Ethiopia, female farmers reported form reliable business relationships with potential experiencing extortion and intimidation from male clients, partners, and suppliers. Approximately 40–50 buyers when selling their crops at the market.59 percent of firms surveyed in Ethiopia indicated that Such cases indicate that women may receive below they had made contact with new suppliers through market value for their output, pay above market value for business acquaintances.57 Entrepreneurs in both small their inputs, and have lower overall levels of allocative and large firms most frequently find their business efficiency and productivity. Strengthening women’s partners through business acquaintances. social networks may provide greater knowledge of and access to fair input and output prices for goods and, in turn, greater entrepreneurial success. BOX 9 What is a Crossover? Female enterprises in male-dominated sectors, over. When parents and husbands support women or “crossovers,” often perform better than those entrepreneurs, however, they are more likely to cross in traditionally female-dominated sectors. When over into men’s work.61 In general, crossover firms are women cross over into male-dominated sectors more likely to report having started their business due in Uganda, their firms are almost three times more to an opportunity provided to them by their husband. profitable than firms owned by women in traditionally These findings lend support to build policies to help female sectors and equally profitable as firms owned women entrepreneurs enter sectors that are male- by men in male-dominated sectors.60 dominated, and to involve men as a means to create opportunities for women. Developing strategies to Access to information and support from family members nudge and support women entrepreneurs to move into influence women’s decisions to cross over. When more profitable sectors can help to reach goals such as women lack information about the earnings potential those set out in the GTP II target of increasing the share in male-dominated sectors, they are less likely to cross of manufacturing industry in GDP to 8 percent. 41 Entrepreneurial, Noncognitive Skills SOCIAL NORMS Entrepreneurial and noncognitive skills boost Social norms are shared beliefs or informal rules success about which behaviors are appropriate, typical, or desirable in a particular social group.66 By providing a Entrepreneurial and noncognitive skills like social group with a common understanding of goals dependability and persistence improve labor market and values, norms lead to behavioral patterns. While outcomes, especially in low-skilled labor markets, norms do not dictate behavior, they certainly influence where such skills are valued equally as much as, if the likelihood of particular behaviors by establishing not more than, cognitive skills.62 Meanwhile, skills expectations of rewards and approval or, conversely, like extroversion, a desire to accomplish, risk-taking sanctions and disapproval. behavior, and overestimation of one’s capabilities are key psychological determinants of entrepreneurial Gender norms, in particular, stem from and give root activity at the individual level.63 These characteristics to the belief that men and women are and should be are particularly important as potential areas for different in behavior, aspirations, status, and economic programming in the context of Sub-Saharan Africa, activity. Norms influence everything from educational where roughly half of women in the non agricultural investments early on in life, to factors later in life like labor force are entrepreneurs.xxi,64 the timing and dynamics of marriage, childbearing, and household tasks. This section explores the current The formulation of these skills starts at a young age. state of four key domains in which gender norms A number of studies of noncognitive skills in Sub- influence women’s economic outcomes: marriage Saharan Africa reveal that girls have the same levels and childbearing; intrahousehold dynamics; asset of confidence, motivational skills, and aspirations ownership; and internalized beliefs. as boys. However, in one particular study, Ethiopian girls scored lower in self-efficacy and agency than Link between norms and behaviors their male counterparts at ages 12 and 15, with girls Norms do not dictate behavior, but rather guide it in more frequently reporting that family members two primary ways: “make all the decisions about how [the girls] spend [their] time.”65 By increasing girls’ and women’s sense 1. People conform to social norms to achieve social of agency over their own time, decisions, and lives, recognition or to avoid social sanction. policies can further boost female aspirations and optimize labor market outcomes. 2. Norms shape preferences by affecting individuals’ self-conceptions and perceptions of the options available to them. xxi Entrepreneurs here are defined as individuals who own their own businesses. 42 BOX 10 What are Gender Norms? Gender norms are informal rules that define which Gender category beliefs are beliefs about shared behaviors are typical and desirable for men and characteristics of men or women, as well as women. They are informed by different types of differences between men and women: “men are like common beliefs:67 X; women are like Y” or “men are good at X; women are good at Y.” Gender role beliefs are ideas that associate some tasks and relational dynamics with men and others Gender status beliefs are hierarchical distinctions that with women. attribute greater competence and authority to men. Marriage and Childbearing changed the legal age of marriage to 18. By 2005, five regions and two charter cities had implemented the Delaying marriage promotes educational and law. In these regions, the average age at first marriage economic advancement for women increased by 0.26 years for women aged 15–19 years and by 0.13 years for women aged 20–24 years. The increased Marriage in Ethiopia occurs early in life, with the median marriage age helped improve participation in the age at first marriage for women standing as one of the labor market, particularly for young women. In the five lowest on the continent: 17.1 years.68 Among women regions, labor force participation rose by 15–24 percent aged 30–34, 27.3 percent had married by age 15.69 more than in regions that had not yet implemented Early marriages can result from both social norms and the law.70 Furthermore, the law significantly increased economic pressures. Norms that emphasize women’s single women’s participation in paid work, year-round role as mothers rather than providers may motivate employment, work outside the household, and work in girls to move into adulthood through marriage and higher-than-average education occupations. The 2000 motherhood rather than through education and Revised Family Code represents a powerful signal from employment. Other norms emphasizing virginity the government about women’s status in marriage. Such may encourage marriage in adolescence. Economic socially salient signals have the power to shift norms. pressures often motivate marriages too, leading The marriage age reform now applies across all regions parents to arrange their daughters’ marriages in order and cities in Ethiopia, giving women greater incentives to to escape poverty at home. Women who marry early pursue economic opportunities. are more likely to drop out of school earlier and less likely to spend time acquiring valuable skills for Increasing access to education shifts fertility economic success. preferences Delaying marriage may result in better educational and In Ethiopia, 13 percent of adolescent girls aged 15–19 are economic outcomes for women in Ethiopia. In attempts mothers or are pregnant with their first child, while 38 to improve women’s ability to earn, work, and thrive percent of women give birth by age 18. Most women give outside of the home, the 2000 Revised Family Code birth within 2.1 years of marriage, placing the median 43 age for a mother’s first birth at 19.2 years. Women who Bank found that female business leaders experience give birth in their teenage years are more likely to drop intense “work overload” attributed to their “inability out of school, and throughout their childbearing years, to say no, the nature of their company and their women continue to grapple with decisions related to work… and the imbalance of their responsibility and fertility, motherhood, and the labor market.71 their required working hours.”74 Women experience an increasing trade-off between career and family as Increasing access to education, however, shifts fertility they enter roles with higher pay and responsibility, preferences. In Ethiopia, universal primary education in part deterring women from aspiring to particular policies, implemented in the mid-1990s, essentially occupations or positions. eliminated school fees for all grades in primary school. Subsequently, women’s schooling increased Affordable and accessible child care, especially in and women’s ideal family sizes decreased: a one-year urban settings, could ease the constraints that women increase in schooling was associated with a 0.34 drop in face when seeking employment outside of the home. women’s ideal family size.72 As such, norms surrounding In Kenya, subsidized early child care (ECC) for mothers women’s education can influence women’s economic living in a Nairobi slum led mothers to feel more eager activity in both direct and indirect ways. Education can to send their children to ECC centers. Women who improve skill development and lead to reductions in received subsidized ECC were 17 percent more likely fertility expectations, in turn increasing the amount of to be employed than those who did not receive it.75 time that women can contribute to the labor market. As more accessible child care could reduce gender inequalities in Africa, further research on this topic While research indicates that increased education in Addis Ababa and other urban settings in Ethiopia changes fertility preferences, evidence does not yet would be of great value. show that it delays first pregnancies. Between 2000 and 2011, the proportion of women with some primary Intrahousehold Dynamics education in Ethiopia rose from 34 to 67 percent, or about three percentage points each year.73 However, Labor-saving household technologies and increasing this large rise in education did not result in a large shift the participation of men and boys in housework can in the mother’s age at first birth, as it did in other Sub- ease the female burden of household chores Saharan African countries. Further research is needed In Ethiopia, like in other parts of Sub-Saharan to understand this trend. Perhaps, in Ethiopia, the age Africa, gender norms delegate to women and girls of first birth has not yet caught up with increases in the majority of domestic work, including child education, or possibly, most women in Ethiopia still do rearing, cleaning, food preparation, wood and water not complete primary school despite receiving some collection, and food production. The Young Lives primary education. time-use survey indicates that Ethiopian women Affordable, accessible child care opens up workforce ages 18–19 spend 4.1 hours per day on domestic tasks opportunities for women compared to 1.5 hours for boys of the same age.76 These domestic responsibilities impede women’s Women may avoid jobs or self-employment opportunities to study, develop professional opportunities that curtail the time they can devote to experience and skills, run a business, or engage in caring for family members and the household. Such paid work. Sixteen percent of girls drop out of school choices impact their lifetime earnings and contribute to look after siblings and 12 percent of girls drop out to the gender gaps in wages and profits. In Ethiopia, of school due to family issues.77 Time and care taking a small, qualitative study of major companies like constraints increase with age for women. Older girls Ethiopian Airlines, Ethio Telecom, and NIB International carry the greatest burden. 44 To address these constraints, the Women Development marriage.83 Greater decision-making authority can and Change Package proposes labor-saving household positively influence women’s control over productive technologies to ease the burden of household chores assets and investments in economic activities. On and the establishment of child care facilities to reduce the other hand, divorce can lead to higher numbers the time women have to spend on child care. of female-headed households, which are typically disadvantaged in terms of asset ownership and labor Interventions for gender-based violence could supply. Policies to support women’s economic activity broaden women’s opportunities to productively will need to account for the disadvantages faced by contribute to their communities women in female-headed households. Violence against women impacts a woman’s physical Asset Ownership and mental health and affects her ability to engage in everyday activities and the workforce. Fear of violence The 2000 Revised Family Code increases women’s can also curtail women’s willingness to pursue certain legal rights to assets economic activities, especially activities uncommon for women. According to 2016 Ethiopia Demographic Assets such as land and business equipment serve not and Health Survey data, one-third of ever-married only as crucial inputs, but also as potential collateral women in Ethiopia had experienced spousal violence for credit. Half of all women in Ethiopia own a house and 27 percent had experienced spousal violence in part or in full, while forty percent of women own within the past year. Sixty-three percent of women and land. Of women who own land, half report having their 28 percent of men agreed that a husband is justified in name on a title or deed.84 However, relative to men and beating his wife if she burns the food, goes out without male-headed households, women and female-headed telling him, neglects the children, or refuses to have households fare worse in land and asset ownership. Male- sex with him. One in ten women reported having headed households own 2.2 hectares of land, on average, experienced sexual violence.78 compared to 1.7 hectares for female-headed households. The 2000 Revised Family Code addresses these Divorce rights grant women more bargaining power inequalities by increasing women’s legal rights to Divorce is common throughout Sub-Saharan Africa, assets. While the previous Family Code granted contributing to union dissolution more than widowhood permission to married women to control assets or does.79 In Ethiopia, the divorce rate among ever-married pursue a profession, it failed to offer protection to women is just above 30 percent, a relatively high unmarried or widowed women. The 2000 Revised percentage in comparison to other Sub-Saharan African Family Code better protects women by granting equal countries.80 Almost half of all marriages in Ethiopia end rights to spouses during the duration, conclusion, and within 30 years and two-thirds of women who seek dissolution of marriage, requiring equal asset division divorce do so within the first five years of marriage.81 between the husband and wife upon divorce. When men seek divorce, they do so with arguments that the wife is unable to conceive. When women seek Bridging gender gaps in land registration improves divorce they sometimes state that they have married women’s tenure security too young or that they perceive their spouses to be unsupportive of employment agendas.82 Compared to female-headed households, male-headed households have larger plot sizes, a larger proportion of Divorce affects women’s economic participation in a cultivable land, and a larger fraction of registered land.85 number of ways. On the one hand, a woman’s ability Women in male-headed households are very rarely to leave an unsupportive spouse or an unhealthy primary land managers, though the reverse is not the marriage can increase her bargaining power in case for men in female-headed households. 45 The Gender Equality Strategy for the Agriculture Sector found for the allocation of the house and livestock. urges policy makers to revise and implement land- Moreover, women and men who were aware of the related policies in Ethiopia to boost female farmers’ land registration were more likely to report there profitability and productivity. The Women Development should be a more equitable division of livestock and and Change strategy lays out a set of interventions to land upon divorce.88 address gaps in ownership, access, and use of land. Key recommendations include encouraging sharecropping Internalized Beliefs where women lack the required labor to cultivate their land, ensuring women obtain fair sharecropping Affirmative action can combat internalized beliefs agreements, assigning plots to landless women, and and stereotypes making women aware of their land ownership rights. Women’s and men’s subjective assessments of Furthermore, the 2003 Land Registration Act sought to their own capabilities contribute to gender gaps. grant equal inheritance and property rights to women. In experimental games, when exposed to the It facilitated land registration at a low cost, issuing one- concept of male superiority, men’s self-perceived page certificates with the household head’s names competence exceeds that of women. Gender gaps (the husband for married households), neighbors’ in self-assessed ability are especially prominent names, and details about the size, location, and land for tasks typically performed by only one gender quality of farm plots. Some certificates included or tasks for which either men or women have a maps and photos of the husband and wife, making perceived natural advantage. However, when gender it more difficult for husbands to sell or rent out land was said to be irrelevant to the task, men and women without their wives’ consent. To ensure transparency, showed no difference in self-perceived competence. land certificates were issued after public registration. Even in samples of established business owners, Female heads of household (widows, divorced, and women maintain lower levels of self-assessed single women) also received certificates in their entrepreneurial capabilities compared to men.89 names for land in their possession. Furthermore, the Such gender differences in self-confidence or self- Ethiopian land certification scheme required that land assessed skills can contribute to differences in administration committees at the kebele level, or the entrepreneurial investments and outcomes. smallest administrative unit in Ethiopia, include at least one female member.86 The presence of female Stereotypes and gender bias also contribute to members in the land administration committees gender participation gaps. A study of U.S. major encouraged female-headed households to participate metropolitan symphony orchestras finds that in land certification. concealing the identities of the performers auditioning significantly boosted women’s chances Overall, Ethiopia’s land registration process increased of succeeding in selection.90 Other recent research of tenure security for women.87 A study across 15 villages tech companies in the United States shows that hiring in Ethiopia found that, combined with the Family practices are gendered: recruiters use “geek culture Code revisions, the 2003 Land Registration Act shifted references” and overt gender stereotypes, resulting perceptions and social norms related to the division in women having a difficult time making it through of assets upon divorce. While in 1997, only about recruitment.91 Bearing these challenges in mind, the 40 percent of women perceived that land would Women Development and Change Strategy and the be divided equally between the husband and wife National Employment Policy and Strategy promote upon a no-fault divorce, this percentage increased to affirmative action to ensure a certain proportion of more than 80 percent by 2009. Similar patterns were women participate in programs and projects.92 46 Part IV Gender and Innovation in Programming 47 Despite Ethiopia’s remarkable economic progress license, could help reduce the gender gaps in over the past decade, current gender gaps in productivity and earnings. agricultural productivity, entrepreneurship, and wages reveal that challenges remain to realizing the • Improving women’s access to credit, and full potential of women’s economic empowerment. especially to mid-sized lending products, when Evidence from this report suggests that an array of they often lack access to enough collateral to be contextual factors and social norms holds women considered “creditworthy”. Two key approaches back, including time poverty, limited labor market could work in making women more creditworthy: skills and technical knowledge, and a lack of gender- the first is building women’s assets (e.g. land, sensitive spaces, social networks, and mentors. These through co-titling in land registration), and the challenges manifest themselves in a wide range of second is developing innovative lending products factors identified in the decomposition analysis, that use nontraditional, non-asset-based forms such as more limited access to agricultural extension of collateral. These types of interventions could services, less intensive use of agricultural inputs, enable women to access credit at levels that higher costs for their businesses, and fewer hours would enhance their productivity and earnings in spent running their businesses, among others. agriculture and self-employment. To accelerate the country’s progress in achieving • Promoting educational opportunities, as well inclusive growth and equality for women, government as job skills development through vocational leaders and their development partners should and technical training. Giving women and girls consider the following policy responses—informed the opportunities to advance their education, by the report’s comprehensive analysis on the drivers develop their skills, and possibly cross over into and the underlying causes of the gender gaps—as more lucrative male-dominated sectors could priority areas where the potential to close gender help narrow the gender gap in employment and gaps in economic empowerment is greatest. earnings. POLICY PRIORITIES • Tackling gender norms and institutional constraints that limit women’s economic • Expanding access to customized agricultural empowerment. Entrenched gender norms and extension services for female farmers. Agricultural institutional barriers underlie many of the key extension services could play a considerable role in drivers of the gender gaps identified in the report. closing the gender gap in agricultural productivity Recent legal changes, however, have already if they target women on a larger scale. They could shown promise in beginning to shift norms in also tailor their interventions to focus on the marriage, childbearing, and asset ownership. factors widening the gender gap in productivity, such as use of agricultural inputs. • Alleviating time constraints for women by providing services and interventions to reduce • Increasing women’s access to key inputs in the time burden posed by household duties. Such agriculture and entrepreneurship. For female interventions could make more time available for farmers, boosting women’s access to fertilizer women to dedicate to other productive activities and pesticides, and for female entrepreneurs, and contribute to closing the gender gap in increasing access to hired labor and a business employment and earnings. 48 GENDER INNOVATION spread knowledge among women in rural areas, drawing on insights from behavioral science to To address the sizable gender gaps in economic reduce gender-based violence, or developing new outcomes and achieve Ethiopia’s inclusive growth partnership models such as a social impact bond tied targets will require a fundamental rethinking to employment outcomes of young men and women.94 of policies and programming aiming to reduce gender inequality. The intersection of gender and The section that follows presents a collection of innovation in programming can play a significant previously implemented innovations that can role in addressing the factors that limit women provide examples of potential programming options economically. in Ethiopia. Virtually all of the innovations have the potential to be adapted to the Ethiopian context, Incorporating gendered perspectives into all aspects with Ethiopian policy makers best-positioned to of a policy or a development project—from initial determine the appropriateness of a given intervention. design and planning to monitoring, evaluation, and While some stories are illustrative or exemplary, learning—can help meet multiple objectives and others have the real potential for replication and boost overall development outcomes, including scalability from local to regional and national those related to women’s economic empowerment, targets. Moving forward, further evidence on impacts, education, gender-based violence, and aggregate feasibility, and costs could help support policy makers economic growth.93 Innovation, which includes but who are considering adapting these solutions to is not limited to technological advances, can lead to specific contexts or taking them to scale. This report progress in achieving multiple development goals sets the stage for further design, experimentation, at once by driving economic growth and addressing and evaluation of innovative solutions in gender social challenges. Innovation may consist of, for programming to increase their effectiveness and example, using virtual peer-support networks to maximize impact on women’s empowerment. 49 Gender Innovation Menu The innovations presented below are sourced principally from partners of the International Development Innovation Alliance’s Gender and Innovation Working Group (IDIA/GIWG), which is composed of senior gender and innovation representatives from the world’s major donor and development institutions, including Australian Aid, the Bill and Melinda Gates Foundation, Global Affairs Canada, Global Innovation Fund, Grand Challenges Canada, Results for Development, Swedish International Development Cooperation Agency (Sida), The Rockefeller Foundation, # Case study Description Agricultural knowledge and use of inputs 1 Farm.Ink Farm.Ink has developed an automated chat-bot application, which helps connect farmers looking for advice and input, and selling agricultural products. The chat-bot links farmers to various social media platforms where they can connect with peers, solicit advice, and receive tips related to agricultural timing and crop growth. It also helps farmers make connections to potential buyers in East Africa. Implementing partners are using some of the latest technology and user-centered design methods to engage and connect with farmers who already have access to smartphones and mobile financial services. 2 Reel The entrepreneurial founder of Reel Gardening, Claire Reid, developed a method of properly Gardening spacing seeds by placing them in strips of newspaper fused with the right combination of flour and fertilizer. The proper spacing and depth of the seed within the paper envelope also varied depending on the type of seed being utilized. The biodegradable innovation is now available online and will soon be implemented in communities in South Africa. 3 Precision Precision Agriculture for Development (PAD) utilizes precision agriculture technologies Agriculture from developed countries including mobile soil analysis, satellite and drone photographs, for Deve- and weather prediction models to disseminate personalized agricultural advice via mobile lopment phone and other ICT-based extensions. In Ethiopia, PAD partners with the Agricultural Transformation Agency on the existing voice-based mobile advisory services. 4 Orange- The research and development of the orange-fleshed sweet potato (OFSP) and related Fleshed research and application of effective cultivation techniques have together been heralded by Sweet many as one of the most successful examples of micronutrient and vitamin bio-fortification Potato efforts in agriculture and development. 5 BioEnsure The technical innovation ‘BioEnsure’ are seeds resistant to extreme climates. The program teaches women in small, rural farming villages how to apply the seed treatment protocol and start their own businesses as agricultural distributors of the product. The product is resilient to drought and heat- waves. The program sells the product directly to these women who treat the seeds and sell them within their respective communities. 50 Department for International Development (DFID), UNICEF (United Nations International Children’s Emergency Fund), United Nations Development Programme (UNDP), United States Agency for International Development (USAID), and World Bank Group. The IDIA/GIWG works at the nexus of gender and innovation in an attempt to accelerate efforts toward meeting the United Nations’ Sustainable Development Goals (SDGs). The advantage of drawing innovation case studies from these partners lies in the fact that they represent both breadth and depth, across diverse institutional settings and varied geographies. Countries of Implementing Gender component/constraints addressed operation partner Constraints addressed Kenya Farm.Ink + Agricultural knowledge + Access to extension + Limited professional networks and role models (application has capacity to create a stronger network among female farmers) Limitations • Must recognize gendered division of labor in agriculture • Gender divide in access to ICTs may limit women’s access to these resources • Limited mobile penetration in Ethiopia Constraints addressed South Africa Reel Gardening + Agricultural knowledge + Limited job opportunities (manufacturing these strips creates jobs for previously unemployed men and women) Constraints addressed Ethiopia, India, Precision Agriculture + Agricultural knowledge Pakistan, Kenya, for Development + Access to extension Rwanda, Ecuador, Bangladesh Limitations • Must recognize gendered division of labor in agriculture • Gender divide in access to ICTs may limit women’s access to these resources • Given the limited mobile penetration, PAD works exclusively through voice- based mobile advisory in Ethiopia. Note: In other countries it operates using SMS and internet. Constraints addressed N/A CGIAR (International + Nutritional (vitamin A) deficiency that disproportionately impacts Potato Center) women and children Constraints addressed United States BioEnsure + Impact of drought on crops + Limited job opportunities Limitations • Not in operation outside of the United States • Price point is unclear 51 # Case study Description Financial products 6 Caregiver Caregiver is a financial innovation in the sense that it is a new design and delivery model of an existing category of financial products (insurance). Caregiver is cash based and covers “hidden” costs of illness. It is also a social innovation in the sense that it has effects beyond the insured individual, which extends to the broader family. 7 M-PESA M-PESA is a mobile phone-based money transfer, financing, and micro-financing service that saves beneficiaries time (waiting at banks), money (interest rates associated with cash transfers), and improves security (reducing cash-related crime). The innovation has scaled at an incredible pace and is now estimated to serve approximately 200,000 households. An estimated 25% of the country’s gross national product (GNP) flows through M-PESA. Entrepreneurial employment 8 Targeting Bandhan-Konnagar Targeting the Hardcore- Poor (THP) program utilizes an innovative, the Hard- 24-month graduation model to support extremely poor women on the pathway out of core- Poor poverty through entrepreneurship. Research showed that Bandhan-Konnagar’s innovative approach promoted positive gains in income, livelihoods, and health, which persisted even one year after the program concluded. There was a four-fold return in income spent on ultra- poor households, and participants ended the training program with an approximate one- quarter increase in their personal purchasing power. 9 Interacti- Babajob created an Interactive Voice Response (IVR) application and platform that permits lower- ve Voice literate or illiterate job- seekers to access and utilize the service, using a basic mobile phone. Response Clients can use the platform to create profiles (called ‘digital resumes’), search listings for better- (IVR) Em- quality jobs (higher pay, closer to home), apply to positions of interest to them, and have access to ployment updates about relevant job opportunities. Program 10 Living Living Goods employs and trains local people as community health promoters (CHPs) to sell Goods goods and life-saving medical supplies door-to-door at affordable prices. Living Goods equips mostly saleswomen with entrepreneurial skills, while simultaneously improving health outcomes in their communities. 11 U-Report The U-Report tool is UNICEF’s free digital youth engagement tool. It is used by approximately five million people around the world and implemented in over thirty-five different countries. The U-Report platform gives youth a place to voice opinions, pose and answer questions, and address youth concerns through online and SMS-based mobile features that include games, polls, surveys, and anonymous peer counseling services. 52 Countries of Implementing Gender component/constraints addressed operation partner Constraints addressed Jordan Women’s World + Shocks to household income Banking Limitations • Gender divide in access to ICTs may limit women’s access to these resources • Limited mobile penetration in Ethiopia Constraints addressed Kenya, Tanzania, Safaricom + Household consumption and savings patterns Afghanistan, + Costs associated with financial transactions South Africa, India, + Time poverty from time spent in travel to physical financial institution Romania, Albania Limitations • Gender divide in access to ICTs may limit women’s access to these resources • Limited mobile penetration in Ethiopia Constraints addressed India Bandhan-Konnagar + Identifying extremely poor women (program uses an innovative Participatory Rural Appraisal approach to identify and cater their programs to the most marginalized women) Limitations • Successful in the Indian context but may have different results in Sub- Saharan Africa Constraints addressed India Babajob + Limited job opportunities Limitations • Program has not yet been explicitly adapted to women but potential exists • Participants need access to a mobile phone; limited mobile phone penetration in Ethiopia Constraints addressed Uganda, Kenya, Living Goods + Limited job opportunities Myanmar, Zambia + Healthcare Constraints addressed N/A UNICEF + Gender divide in ICTs (gender equality is at the heart of program’s mandate and to increase the involvement of more girl ‘U-Reporters’, the program has created and promoted multimedia content and key events that speak directly to the concerns and needs of adolescent girls) Limitations • Gender divide in access to ICTs may limit women’s access to these resources • Limited mobile penetration in Ethiopia 53 # Case study Description Networking and mentorship 12 Let Girls Let Girls Map is one campaign of YouthMappers, a broader effort to involve and link youth Map locally, regionally, and globally, to empower them around identifying and prioritizing solutions to their development challenges. Let Girls Map connects and empowers girls around identifying educational resources that directly serve girls and women. 13 Rural Rural Women Striding Forward is a 2.5-year initiative that funded 22 rural women’s groups Women in Burkina Faso, Kenya, and Uganda, working on sustainable agriculture and the promotion Striding of women’s rights. These grassroots grantees trained women in agricultural techniques, Forward promoted participation of thousands of women and men in women’s rights activities, and increased women’s access to networking opportunities. 14 Meri Men- Meri Mentorship & Leadership Program is an initiative that is currently being designed and torship & implemented to match recent female university graduates from Addis Ababa University with Leadership older, more experienced women to create a network of mentees and mentors. It aims to Program provide young women with networks, information, and career guidance. Gender-sensitive spaces and adolescent girls 15 Biruh Tesfa Biruh Tesfa is a program that aims to reach out-of-school adolescent girls with female mentors and non formal education and life skills training. Adult female mentors go house to house to contact eligible girls. It is managed by the Population Council alongside the Ethiopia Ministry of Women, Children, and Youth Affairs. 16 ZanaAfrica ZanaAfrica delivers reproductive health education and sanitary pads through local Group organizations in order to increase school attendance and confidence among adolescent girls. 17 EkoLakay, These innovations provide clientele with household lavatories, which permits a private space to Enviro Loo, utilize the toilet. All wastes are safely treated and transformed into nutrient- rich compost that Sanivation, can be safely applied to support agriculture, agroforestry, and reforestation efforts. The lack of Afrisol access and proximity to toilets or latrines means that women and girls must travel considerable Energy distances to reach suitable sanitation facilities. This reality places women and girls in dangerous situations—the correlation between gender-based violence (GBV) and lack of access to safe sanitation services is well- documented worldwide. 54 Countries of Implementing Gender component/constraints addressed operation partner Constraints addressed Global Youth Mappers + Agricultural knowledge (this enables the mapping of water and natural resources and classification of land types for agricultural purposes) + Limited professional networks and role models (creates network among girls and young women) Limitations • Requires access to ICTs and internet, which is limited in Ethiopia Constraints addressed Burkina Faso, Global Fund for + Agricultural knowledge Kenya, Uganda Women + Use of inputs (finances women to purchase inputs if they cannot afford them) + Limited professional networks and role models (creates information sharing and networking among female farmers) + Land rights and certification (grantees inform women about their rights and encourage activism) Constraints addressed Ethiopia Earuyan Solutions + Limited professional networks and role models + Lack of exposure or information on career paths Limitations • Program is in its early stages and commenced in October 2018 with planned uptake of 30 girls per quarter Constraints addressed Ethiopia Population + Relatively low participation in formal school, non formal education, Council, Ethiopian and literacy programs among adolescent girls government + Vulnerable girls (disabled, child domestic workers, child sex workers) often confined to the home Constraints addressed Kenya, United States ZanaAfrica + School drop-out rate among adolescent girls Foundation Constraints addressed Haiti (EkoLakay), Sustainable + Need for gender-sensitive spaces that include private toilets Kenya (Sanivation, Organic Integrated + School drop-out rate among adolescent girls Afrisol Energy), Livelihoods (SOIL) + Gender-based violence South Africa + Limited job opportunities (program manages waste production through (Enviro Loo) weekly waste collection service that employs female managers) + Time poverty from collecting fuel and water 55 # Case study Description Time poverty and travel outside the home 18 Solar Sister The Solar Sister project addresses energy poverty by working to empower women with new economic opportunities. The organization funnels innovative, clean energy technologies through a woman-centered direct sales network to bring both light and economic opportunities to remote communities in rural Africa. 19 Green Heat Green Heat is a technological innovation—an anaerobic digester that transforms biodegradable biomass (waste, leaves, manure) into a clean burning fuel. The Green Heat innovation reduces reliance on traditional wood-fuel stoves. 20 eWaterpay The eWaterPay innovation system helps customers lower their water management costs by up to 50% and use savings to increase water access points in both rural and urban settings. 21 Eco-Fuel The Eco-Fuel Africa program encourages and instructs rural farmers, unemployed women, and Africa young people to transform locally-sourced farm and municipal and agricultural wastes into clean-cooking fuel briquettes. 22 XPRIZE The Skysource/Skywater Alliance was the winning entry to the XPRIZE Water Prize, an for Water innovation challenge to apply science and technology to find a way to extract water from Abundance the air. Skysource/Skywater Alliance won the competition for producing at least 2,000 liters Winner: of water per day from the atmosphere at a cost of no more than 2 cents per liter with only Skysource/ renewable energy. Skywater Alliance 23 M-KOPA M-KOPA installs home solar systems, using a socially innovative approach: a ‘pay as you go’ mobile phone payment system for energy. It has helped provide approximately one million people in Kenya, Uganda, and Tanzania access to green, safe, and affordable energy options for lighting and mobile phone charging in their own homes. 56 Countries of Implementing Gender component/constraints addressed operation partner Constraints addressed Nigeria, Tanzania, SolarSister + Time poverty from collecting fuel and water Uganda + Limited job opportunities (organization creates sales opportunities for women) Constraints addressed Uganda Green Heat + Time poverty from collecting fuel and water (the digester conserves water and reduces trips to gather water and firewood) + Negative respiratory health effects from traditional wood-fuel stoves Constraints addressed Gambia, Tanzania eWaterPay + Time poverty from collecting fuel and water (increased access to water near or at the home) + Gender-based violence (reduction in violence that takes place during travel to and from the home) Limitations • Utilizes mobile money to pay for set amount of water; limited mobile phone penetration in Ethiopia Constraints addressed Uganda Development + Time poverty from collecting fuel and water (charcoal briquettes Innovation Ventures producing hydraulic jacks reduce or eliminate need to gather wood-fuel outside home) + Gender-based violence + Youth unemployment Constraints addressed N/A XPRIZE + Time poverty from collecting fuel and water + Limited water in areas affected by drought Limitations • Skysource/Skywater was recently selected as the winner and there have been limited deployments of the invention thus far Constraints addressed Kenya, Tanzania, International Center + Time poverty from collecting fuel and water (increases time Uganda for Research on available to girls and women for productive activities through rural Women electrification) + Unemployment (provides jobs in management, installation, maintenance, small enterprise, and more) Limitations • Utilizes mobile money to provide energy on credit; limited mobile phone penetration in Ethiopia 57 # Case study Description General 24 mSTAR The Mobile Solutions Technical Assistance and Research (mSTAR) program advances mobile solutions and attempts to close the gender gaps that prevent or delay access to and adoption of mobile technology. The mSTAR program supports wide, coordinated action along a value chain of market stakeholders, including governments, donors, nongovernmental organizations (NGOs), mobile service providers, and their customers. The mSTAR program is designed to catalyze game-changing interventions to support, improve, and scale access to mobile money, mobile access, and mobile data collection and dissemination. 25 Digital Li- The Digital Livelihoods: Youth and the Future of Work at Scale project seeks to scale up velihoods: established financial and economic programming, social and technological innovations, and Youth and empowerment and leadership development skills training to assist 200,000 young women the Future and men. It employs a comprehensive gender equality strategy based on realities of the local of Work at contexts and builds on lessons- learned from previous Digital Opportunity Trust projects. Scale 26 Design UNICEF widely employs a ‘human-centered design’ social innovation approach that for Girls, incorporates young girls’ and boys’ perspectives at every step of the design process for by Girls. solutions to address development challenges. An innovative ‘human-centered design’ process Period. is being used at UNICEF Pakistan to design more gender-equitable program solutions targeting gender issues, such as menstrual health and hygiene management (MHM). 58 Countries of Implementing Gender component/constraints addressed operation partner Constraints addressed Bangladesh, FHI 360 + Gender divide in access to ICTs India, Liberia, + Functionality of other mobile-based innovations Mozambique Constraints addressed Ghana, Kenya, Digital Opportunity + Gender divide in access to ICTs Malawi, Morocco, Trust + Entrepreneurial training Rwanda, Senegal, Tanzania, Zambia Constraints addressed N/A UNICEF Pakistan + Lack of participation and agency among girls in development projects 59 Appendixes 60 Appendix A: Sample Descriptive Statistics APPENDIX A1. ETHIOPIA SOCIOECONOMIC SURVEY (ESS) SAMPLE URBAN Total Rural Small town Large town Panel A. ESS sample wave 1 (2011–2012) Number of enumeration areas 333 290 43 N/A Number of households 3,969 SHARE BY REGION Tigray 23% 10% 9% N/A Afar 10% 3% 5% N/A Amhara 56% 21% 26% N/A Oromiya 54% 19% 26% N/A Somali 17% 7% 7% N/A Benishangul-Gumuz 7% 3% 2% N/A SNNP 58% 26% 23% N/A Gambella 7% 3% 2% N/A Harari 3% 3% 0% N/A Dire Dawa 3% 3% 0% N/A 61 APPENDIX A1. ETHIOPIA SOCIOECONOMIC SURVEY (ESS) SAMPLE Panel B. ESS sample wave 2 (2013–2014) and wave 3 (2015–2016) Number of enumeration areas 433 290 43 100 Number of households 5,469 SHARE BY REGION Tigray 11% 10% 9% 15% Afar 3% 3% 5% 1% Amhara 20% 21% 23% 15% Oromiya 20% 19% 23% 20% Somali 6% 7% 7% 3% Benishangul-Gumuz 3% 3% 2% 0% SNNP 23% 26% 23% 15% Gambella 3% 3% 2% 1% Harari 3% 3% 2% 3% Dire Dawa 4% 3% 2% 7% Addis Ababa 5% N/A N/A 20% 62 APPENDIX A2. GENDER GAP IN AGRICULTURAL PRODUCTIVITYi AGRICULTURAL GAP (1) (2) (1) - (2) Variable Male Female Difference MAIN OUTCOME Self-reported productivity (logarithm) 8.5 8.1 0.359*** INDIVIDUAL CHARACTERISTICS Age (years) 46.8 51.4 -4.541*** Relationship with head of household Head of household 98% 96% 0.022*** Spouse of the head of household 0% 3% -0.026*** Son/daughter of the head of 2% 0% 0.012** household Other relative 0% 1% -0.008*** Civil status Married 94% 24% 0.704*** Widowed/divorced 4% 74% -0.701*** Single 2% 2% -0.003 Education Illiteracy rate 59% 88% -0.295*** Read and write (no school at all) 13% 4% 0.092*** Completed primary 22% 5% 0.169*** i Values in appendixes A2, A3, and A4 are rounded to the nearest tenth. 63 Variable Male Female Difference Completed secondary 4% 2% 0.023*** Diploma or certificate 1% 0% 0.005 University degree 1% 0% 0.005 Individual disabilityii 11% 17% -0.063*** Hours in the last seven days spent on Agricultural activities 15.9 9.7 6.220*** Non agricultural household business 2.1 1.8 0.371 Casual, part-time, or temporary labor 1.3 0.9 0.441 Wage, salary, any payment (excluding 1.4 0.5 0.983*** temporary) Unpaid apprenticeship 0.0 0.0 0.024 HOUSEHOLD CHARACTERISTICS Household ever received formal credit 26% 17% 0.090*** Household size 5.8 3.8 1.925*** Child dependency ratio (under 10)iii 0.7 0.6 0.141*** Wealth indexiv -0.3 -0.5 0.196** MANAGER CHARACTERISTICS (AGRICULTURE) Attended extension programs 44% 33% 0.106*** Distance of the household to closest 67.1 64.4 2.694 market (km) ii Individual disability: Manager/worker reported to be absent from usual activity due to this illness/injury at some point during the last month. iii The dependency ratio definition used throughout the report is number of children below age 10 over number of individuals above age 10 in a household. iv Wealth index was built using principal component analysis. The variables included in the analysis are a set of variables that capture living standards such as household ownership of durable assets (e.g., TV), dwelling ownership, infrastructure, and housing characteristics (source of water, sanitation facility, source of energy, number of room per capita). 64 Variable Male Female Difference More than half of the household 3% 3% 0.003 production sold Manager has non agriculture labor 17% 28% -0.107*** income Total land managed (hectares) 1.0 0.5 0.408*** Number of fields managed 11.2 8.5 2.783*** Number of plots managed 3.6 2.6 0.992*** MANAGER’S PLOT/FIELD CHARACTERISTICS Land under certificate (% of total)v 46% 46% -0.002 Manager's plot occupation: rented (% 4% 2% 0.020*** of fields) Intercropping (% of fields)vi 16% 16% -0.003 Average plot slope 12.2 11.5 0.670* Plot distance to household 1.2 0.6 0.609 MANAGER’S AGRICULTURAL NON LABOR INPUT USE Fields that use irrigation (% of total) 4% 3% 0.003 Fields that use fertilizer (% of total) 35% 36% -0.007 Fields that use organic fertilizer (% of 1% 2% -0.009** total) Fields that use pesticide, herbicide, or 10% 8% 0.018** fungicide (% of total) Fields that use improved seeds (% of 4% 4% 0.006 total) Chemical fertilizer used per hectare 117.8 97.7 20.14 (kg/ha) Oxen per hectare 1.4 1.0 0.329 v Formal certification is created using the question Does your household have a certificate for this [PARCEL]? vi Intercropping refers to planting more than one crop on a single field (the alternative to intercropping is having a pure-stand field). vii Agricultural index is created using principal component analysis and dummies of holding of the following resources: sickle, axe, pickaxe, traditional plough, modern plough, water pump, and agricultural livestock availability. 65 Variable Male Female Difference Agriculture implemented access indexvii 0.4 -0.2 0.631*** MANAGER’S AGRICULTURAL LABOR INPUT USE Total number of hours worked by 234.8 154.1 80.752*** household members (hours/ha) Total hired labor use (days/ha) 13.9 5.4 8.569 Total exchange labor use (days/ha) 15.9 34.0 -18.093*** CROP CHOICE Total number of crops harvested 5.4 4.4 0.926*** Food crop (% of total crops harvested) 62% 59% 0.023 Cash crop (% of total crops harvested) 14% 12% 0.018 Horticulture (% of total crops 6% 7% -0.007 harvested) Nuts, peas, and beans (% of total crops 9% 9% 0.003 harvested) Spice and oil seed (% of total crops 2% 2% 0.002 harvested) Other (% of total crops harvested) 7% 10% 0.031 Crop damage 85% 82% 0.029* GEOGRAPHICAL LOCATION Rural 100% 100% 0 Tropic-warm/semiarid 5% 4% 0.01 Tropic-cool/semiarid 31% 33% -0.019 Tropic-cool/subhumid 43% 38% 0.050** Tropic-cool/humid 18% 21% -0.033* Observations 2,289 618 2,907 66 APPENDIX A3. GENDER GAP IN SALES FROM SELF - EMPLOYMENT SELF - EMPLOYMENT GAP (1) (2) (1) - (2) Variable Male Female Difference MAIN OUTCOME Total monthly sales (logarithm) 8.26 7.47 0.791*** SECONDARY VARIABLES Average monthly operating costs (log) Wages 1.2 0.6 0.682*** Purchase of goods for sale 4.2 3.3 0.838*** Raw materials 3.1 3.8 -0.757*** Transportation 2.9 2.0 0.881*** Other costs 3.1 2.2 0.906*** Total costs 7.2 6.7 0.528*** INDIVIDUAL CHARACTERISTICS Age (years) 39.4 39.7 -0.3 Relationship with head of household Head of household 84% 56% 0.286*** Spouse of the head of household 3% 32% -0.288*** Son/daughter of the head of 10% 10% 0.007 household Other relative 3% 3% -0.003 67 Variable Male Female Difference Civil status Married 83% 40% 0.421*** Widowed/divorced 5% 48% -0.431*** Single 13% 12% N/A Education Illiteracy rate 30% 58% 0.28*** Read and write (no school at all) 11% 9% 0.028* Completed primary 33% 19% 0.142*** Completed secondary 19% 11% 0.077*** Diploma or certificate 4% 2% 0.023** University degree 2% 1% 0.013** Individual disability 9% 12% -0.035** Hours in the last seven days spent on Agricultural activities 7.2 5.5 1.662** Non agricultural household business 22.7 16.8 5.998*** Casual, part-time, or temporary labor 2.8 1.4 1.383** Wage, salary, any payment (excluding 1.9 1.4 0.49 temporary) Unpaid apprenticeship 0.3 0.0 0.287 HOUSEHOLD CHARACTERISTICS Household ever received formal credit 28% 24% 0.040* Household size 5.3 4.6 0.703*** 68 Variable Male Female Difference Child dependency ratio (under 10) 0.6 0.5 0.090*** Wealth index 0.11 -0.39 0.499*** GEOGRAPHICAL LOCATION Rural 51% 53% -0.025 Small town 16% 13% 0.03 Medium and large town 33% 34% -0.005 MANAGER OF ENTERPRISE CHARACTERISTICSviii Number of enterprises under principal 1.2 1.1 0.085*** manager Work with another manager 66% 9% 0.571*** Number of years that enterprise has 15.5 15.4 0.153 been operating Enterprise has a license 0.4 0.1 0.228*** Are the activities of this enterprise 0.3 0.3 0.066*** seasonal? Total number of hired workers under 0.6 0.3 0.321** manager Formal credit 0.1 0.0 0.037*** Amount borrowed for enterprises 1,422.0 628.6 793.419*** under manager Amount repaid on loans for 1,181.3 406.9 774.421*** enterprises under manager Client of manager is: local consumers, 97.2% 99.5% -0.023*** market, traders, and cooperative Client of manager is: NGO or 3.9% 0.6% 0.033*** governments Client of manager is: other 3.4% 2.5% 0.009 viii When a manager manages more than one enterprise we use the following: Mean of the number of years that enterprises have been operating, at least one enterprise has a license, at least one of the enterprises has seasonal activity, at least one enterprise acquired a credit. 69 Variable Male Female Difference WORK INDUSTRY (OF THE EMPLOYEE) Agriculture production 10% 5% 0.050*** Services 38% 33% 0.055** Manufacturing 14% 34% -0.201*** Trade 38% 31% 0.072*** Other 8% 1% 0.062*** Observations 961 639 1,600 70 APPENDIX A4. GENDER GAP IN WAGE EARNINGS WAGE GAP (1) (2) (1) - (2) Variable Male Female Difference MAIN OUTCOME Total hourly wage (logarithm) 3.4 2.9 0.438*** INDIVIDUAL CHARACTERISTICS Age (years) 36.5 31.7 4.751*** Relationship with head of household Head of household 69% 38% 0.307*** Spouse of the head of household 4% 31% -0.270*** Son/daughter of the head of 19% 18% 0.014 household Other relative 8% 13% -0.051*** Civil status Married 65% 45% 0.191*** Widowed/divorced 6% 22% -0.158*** Single 29% 33% -0.033 Education Illiteracy rate 11% 13% -0.023 Read and write (no school at all) 6% 5% 0.01 Completed primary 24% 18% 0.066*** 71 Variable Male Female Difference Completed secondary 28% 32% -0.038 Diploma or certificate 13% 22% -0.096*** University degree 20% 12% 0.081*** Individual disability 8% 12% -0.044*** Hours in the last seven days spent on Agricultural activities 2.5 1.0 1.584*** Non agricultural household business 2.3 2.0 0.377 Casual, part-time, or temporary labor 1.6 0.9 0.724* Wage, salary, any payment (excluding 28.3 28.6 -0.284 temporary) Unpaid apprenticeship 0.3 0.5 -0.117 HOUSEHOLD CHARACTERISTICS Household ever received formal credit 20% 18% 0.029 Household size 5.2 4.7 0.456*** Child dependency ratio (under 10) 0.4 0.4 0.067** Wealth index 2.9 3.8 -0.901*** GEOGRAPHICAL LOCATION Rural 27% 17% 0.101*** Small town 13% 12% 0.007 Medium and large town 61% 71% -0.108*** 72 Variable Male Female Difference WORKER/EMPLOYEE CHARACTERISTICS Did casual labor in the last 12 months? 8% 2% 0.060*** Employee worked for other 13% 13% 0.003 households for free? Employee has a secondary job 3% 2% 0.007 Employer is private company or 48% 44% 0.034 individual Employer is government 47% 51% -0.036 Employer is church/religious 3% 3% 0.006 organization Was employed as temporary labor by the Productive Safety Net Program 2% 0% 0.015** (PSNP)? WORK INDUSTRY (OF THE EMPLOYEE) Agricultural production 10% 6% 0.039** Services 36% 37% -0.013 Mining 1% 0% 0.010* Manufacturing 8% 12% -0.034** Construction 13% 6% 0.077*** Trade 4% 5% -0.006 Education 15% 19% -0.044** Other 22% 24% -0.025 Observations 1,347 843 504 73 Appendix B: Decomposition Methods The main purpose of decomposition methods is to The standard assumption used in these decompositions partition the overall difference of a given distribution is that the outcome variable Y is linearly related to the between two groups, group A and group Bi.In this covariates, X, and that the error term is conditionally diagnosis, group A will be women (women farm independent of X. managers, business managers, or employees) and group B will be men (men farm managers, business managers, and employees). Where and is the vector of covariates. The overall difference in average outcomes between groups B and A is which means where i This appendix is based mainly on the analysis presented in the Handbook of 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). 74 And are the estimated effect (explained by differences in covariates) are intercept and slope coefficients, respectively, of the identified with the aggregate decomposition. The regression models for groups A and B. detailed decomposition involves subdividing both components into the respective contributions of each covariate, and for The overall decomposition and its two components structural effect (unexplained) and composition Oaxaca-Blinder-decomposition of mean productivity differentials where E[v_g |X]=0. Letting D_B=1 be an indicator of group B (men plot managers) and taking the expectations over X, the overall mean productivity gap can be written as Where Adding and subtracting the average counterfactual productivity that group B (women) would have obtained under the productivity structure of group A (men), the expression becomes: Replacing the expected value of the covariates by the sample averages the decomposition is estimated as The first term in the equation is the structure effect (unexplained, or the part due to discrimination) while the second term is the composition effect . In practice, we computed it by plugging in the sample means and the OLS estimates in the above formula. 75 Interpretation of the Oaxaca-Blinder decomposition in the current context The  Oaxaca-Blinder decomposition  is a statistical method that explains the difference in the  means of a dependent variable between two groups by decompo- sing this gap into two parts: The Endowment Effect (Explained) The Structural Effect (Unexplained) Explains the differences Captures the return to resources. between men and women in The differences in what is obtained terms of factors of production from a given amount of a factor of such as years of experience, production, e.g., the difference in total inputs, or access to credit. productivity that the men obtain It refers to the differences in the compared with women who quantities or levels of resources have exactly the same years of used in plots by male managers experience or who use the same compared to women managers. total amount of inputs. In other words, this is the portion Even when men and women of the gender gap attributable to farm managers have access to the quantity or level of resources the same quantities of resources, that can be reduced by ensuring they do not achieve the same that women receive the resources results. Providing women they lack, relative to men. farmers with the same resources will not necessarily reduce the structural portion of the gender gap. Policies need to address broader issues of constraints faced by women managers. In other words, this part captures a discrimination component and the unobservable variables. 76 ENDNOTES 1. Ethiopian Agricultural Transformation Agency. 2017. 8. D. Rodrik. 2016. “The Return of Public Investment.” “Ethiopian Agriculture and Strategies for Growth.” https: / /www.project-syndicate.org/commentar y/ https://www.innovasjonnorge.no/globalassets/afrika/ public-infrastructure-investment-sustained-growth- mirafe-g-marcos---ata---pdf.pdf. by-dani-rodrik-2016-01. 2. CSA (Central Statistical Agency) Ethiopia and ICF. 2017. 9. CIA. 2017. “The World Factbook: Ethiopia,”. “Ethiopia Demographic and Health Survey 2016.” Addis Ababa, Ethiopia: CSA and ICF. 10. Government of Ethiopia. 2017. “Ethiopia’s Progress Towards Eradicating Poverty: An Interim Report on 3. United Nations Entity for Gender Equality and the 2015/16 Poverty Analysis Study.” Addis Ababa, Ethiopia: Empowerment of Women. 2014. “Preliminary Gender National Planning Commission. Profile of Ethiopia.” Addis Ababa, Ethiopia: UN Women. 11. World Bank. 2016. “Ethiopia: Priorities for Ending 4. A. Aguilar, E. Carranza, M. Goldstein, T. Kilic, and G. Extreme Poverty and Promoting Shared Prosperity.” Oseni. 2014. “Decomposition of Gender Differentials in Systematic Country Diagnostic Report. Washington, DC: Agricultural Productivity in Ethiopia.” Policy Research World Bank. Working Paper. Washington, DC: World Bank. 12. World Bank. 2015. “With Continued Rapid Growth, 5. E. Bekele and Z. Worku. 2008. “Women Entrepreneurship Ethiopia Is Poised to Become a Middle-Income Country in Micro, Small and Medium Enterprises: The Case of by 2025.” http://www.worldbank.org/en/country/ Ethiopia.” Journal of International Women’s Studies ethiopia/publication/ethiopia-great-run-grow th- 10(2): 3–19. acceleration-how-to-pace-it. 6. CIA (Central Intelligence Agency). 2017. “The World 13. FAO (Food and Agriculture Organization). 2014. “Ethiopia Factbook: Ethiopia.” https://www.cia.gov/library/ Country Programming Framework.” Addis Ababa: Office publications/the-world-factbook/geos/et.html. of the FAO Representative to Ethiopia. 7. Ethiopian Agricultural Transformation Agency. 2017. 14. World Bank. 2018. “Resilient Livelihoods and Landscape “Ethiopian Agriculture and Strategies for Growth.” Management Project, Project Appraisal Document.” https://www.innovasjonnorge.no/globalassets/afrika/ Washington, DC: World Bank. mirafe-g-marcos---ata---pdf.pdf. 77 15. FAO. 2014. 22. N. Adamon and A. M. Mukasa. 2017. “Credit Constraints and Farm Productivity: Micro-Level Evidence 16. M. O’Sullivan, A . Rao, R. Banerjee, K . Gulati, and from Smallholder Farmers in Ethiopia.” African M. Vinez. 2014. “Leveling the Field: Improving Development Bank Group, No. 247; A. De Janvry, M. Opportunities for Women Farmers in Africa.” Fafchamps, and E. Sadoulet. 1991. “Peasant Household Washington, DC: World Bank. Behaviour with Missing Markets: Some Paradoxes Explained.” The Economic Journal 101(409): 1400–1417; 17. A . Aguilar, E. Carranza, M. Goldstein, T. Kilic, and G. I. Singh, L. Squire, and J. Strauss, eds. 1986. “Agricultural Oseni. 2014. “Decomposition of Gender Differentials Household Models: Extensions, Applications, and in Agricultural Productivity in Ethiopia”; A .P. de la Policy.” Baltimore, Maryland: Johns Hopkins University Campos, K . A . Covarrubias, and A . Prieto Patron. 2015. Press; J. E. Stiglitz and A. Weiss. 1981. “Credit Rationing “How Does the Choice of the Gender Indicator Affect in Markets with Imperfect Information.” American the Analysis of Gender Differences in Agricultural Economic Review 7 1: 393–410. Productivity? Evidence from Uganda.” World Development 77: 17–33; T. Kilic, A . Palacios-Lopez, 23. World Bank. 2009a. “Ethiopia: Diversifying the Rural and M. Goldstein. 2013. “Caught in a Productivity Economy: An Assessment of the Investment Climate Trap,” World Development 70: 416–463; G. Oseni, P. for Small and Informal Enterprises.” Washington, DC: Corral, M. Goldstein, and P. Winters. 2014. “Explaining World Bank. Gender Differentials in Agricultural Production in Nigeria.” World Bank Policy Research Working Paper. 24. IMF (International Monetary Fund). 2013. “Selected Washington, DC: World Bank. Issues Paper on the Federal Democratic Republic of Ethiopia.” Washington, DC: IMF. 18. J. R. Anderson and G. Feder. 2007. “Agricultural Extension, Chapter 44.” Handbook of Agricultural Economics 3: 25. M. Hallward-Driemeier. 2013. “Enterprising Women: 2343–2378; T. Wossen, T. Berger, T. Mequaninte, and B. Expanding Economic Opportunities in Africa.” Alamirew. 2013. “Social Network Effects on the Adoption Washington, DC: World Bank. of Sustainable Natural Resource Management Practices in Ethiopia.” International Journal of Sustainable 26. S. J. Davis, J. C. Haltiwanger, R. S. Jarmin, J. Lerner, and J. Development & World Ecology. 20: 477–483; T. Wossen, T. Miranda. 2012. “Private Equity and Employment.” NBER Berger, and S. Di Falco. 2015. “Social Capital, Risk Preference Working Paper No. 17399; I. Drine and M. Grach. 2010. and Adoption of Improved Farm Land Management “Supporting Women Entrepreneurs In Tunisia.” WIDER Practices in Ethiopia.” Agricultural Economics. 46: 81–97. Working Paper 2010/100. Helsinki: UNU-WIDER; B. M. Kitching and A. Woldie. 2004. “Female Entrepreneur 19. A.R. Quisumbing and L. Pandolfelli. 2010. “Promising in Transactional Economies: A Comparative Study of Approaches to Address the Needs of Poor Female Business Women in Nigeria and China, Proceedings from Farmers: Resources, Constraints, and Interventions.” Hawaii International Conference on Business, Hawaii.” World Development 38(4): 581–592. 27. S. De Mel, D. McKenzie, and C. Woodruff. 2007. “Returns 20. A. N. Mukasa, A. M. Simpasa, and A.O. Salami. 2017. to Capital in Microenterprises: Evidence from a Field “Credit Constraints and Farm Productivity: Microlevel Experiment.” The Quarterly Journal of Economics 124(1): Evidence from Smallholder Farmers in Ethiopia.” ADB 423; S. De Mel, D. McKenzie, and C. Woodruff. 2008. Working Paper Series No. 247. Abidjan, Côte d’Ivoire: “Are Women More Credit Constrained? Experimental African Development Bank. Evidence on Gender and Microenterprise Returns.” American Economic Journal: Applied Economics 1(3): 21. J. Conning and C. Udry. 2007. “Rural Financial Markets 1–32; D. Karlan and J. Morduch. 2009. “Chapter 71 - Access in Developing Countries, Chapter 56.” Handbook of to Finance.” Handbook of Development Economics 5, Agricultural Economics 3: 2857–2908. 4703–4784; D. McKenzie. 2012. “Beyond Baseline and Follow-Up: The Case for More T in Experiments.” Journal of Development Economics 99(2): 210–221. 78 28. S. Alibhai, N. Buehren and S. Papineni. 2018. “Better Loans 41. Ministry of Education. 2008. “National Report on the or Better Borrowers? Impact of Meso-Credit on Female- Development and State of the Art of Adult Learning and Owned Enterprises in Ethiopia.” World Bank Policy Education (ALE).” Addis Ababa: Ethiopia. Research Working Paper. Washington, DC: World Bank. 42. Ministry of Education. 2008. “National Report on the 29. T. Ferede and S. Kebede. 2015. “Economic Growth and Development and State of the Art of Adult Learning and Employment Patterns, Dominant Sector, and Firm Education (ALE).” Addis Ababa: Ethiopia. Profiles in Ethiopia: Opportunities, Challenges and Prospects.” Swiss Programme for Research on Global 43. UN Women. 2014. “Preliminary Gender Profile of Issues for Development. R4D Working Paper 2015/2. Ethiopia.” 30. World Bank. 2009b. “Ethiopia: Toward the Competitive 44. UN Women. 2014. “Preliminary Gender Profile of Frontier: Strategies for Improving Ethiopia’s Investment Ethiopia.” Climate.” Washington, DC: World Bank. 45. UN Women. 2014. “Preliminary Gender Profile of 31. J. S. Arbache, A. Kolev, and E. Filipiak, eds. 2010. “Gender Ethiopia. ” Disparities in Africa’s Labor Market.” Washington, DC: World Bank. 46. UN Women. 2014. “Preliminary Gender Profile of Ethiopia.” 32. J. S. Arbache, A. Kolev, and E. Filipiak, eds. 2010. “Gender Disparities in Africa’s Labor Market.” Washington, DC: 47. N. Kabeer. 2012. “Women’s Economic Empowerment World Bank. and Inclusive Growth: Labour Markets and Enterprise Development.” SIG Working Paper; H. Beyene. 2015. 33. Arbache, Kolev, and Filipiak. 2010. “Gender Disparities in “National Assessment: Ethiopia: Gender Equality and Africa’s Labor Market,”; World Bank. 2009a. the Knowledge Society.” Los Angeles, CA: Women in Global Science and Technology and Organization for 34. Central Statistical Agency. 2017. “Ethiopia Demographic Women in Science for the Developing World. and Health Survey 2016”. 48. N. Buehren and T. Van Salisbury. 2017. “Female 35. M. Hallward-Driemeier. 2013. “Enterprising Women: Enrollment in Male-Dominated Vocational Training Expanding Economic Opportunities in Africa.” Courses: Preferences and Prospects.” . Washington, DC: World Bank. 49. Y. Kim, T. Kelly, and S. Raja. 2010. “Building Broadband: 36. Central Statistical Agency. 2017. “Ethiopia Demographic Strategies for the Developing World.” Washington, DC: and Health Survey 2016.” World Bank. 37. Central Statistical Agency. 2017. “Ethiopia Demographic 50. Freedom House. 2018. “Freedom on the Net 2017: and Health Survey 2016.” Ethiopia Country Profile.” https://freedomhouse.org/ report/freedom-net/2017/ethiopia. 38. M. Amare and S. Asfaw. 2012. “Poverty Reduction Impact of Food Aid in Rural Ethiopia.” Journal of Development 51. N. Buehren and T. Van Salisbury. 2017. “Female Effectiveness 4(2): 235–256. Enrollment in Male-Dominated Vocational Training Courses: Preferences and Prospects.” 39. UN Women. 2014. “Preliminary Gender Profile of Ethiopia.” 52. Global Information Society Watch. 2013. “Women’s Rights, Gender, and ICTs.” The Hague, Netherlands: Association 40. N. Buehren and T. Van Salisbury. 2017. “Female for Progressive Communications and Humanist Institute Enrollment in Male-Dominated Vocational Training for Cooperation with Developing Countries. Courses: Preferences and Prospects.” Washington, DC: World Bank. 79 53. Hallward-Driemeier. 2013. “Enterprising Women: Behavioral Approach.” Journal of Economic Literature Expanding Economic Opportunities in Africa.” 39(4): 1137–1176; G. J. Duncan and R. Dunifon. 1998. “Soft- Skills and Long-Run Labor Market Success.” Research 54. A. D. Foster and M. R. Rosenzweig. 1995. “Learning by in Labor Economics 17: 123–149; J.J. Heckman and Y. Doing and Learning from Others: Human Capital and Rubinstein. 2001. “Noncognitive Skills: Lessons from Technical Change in Agriculture.” Journal of Political the GED Testing Program.” American Economic Review Economy 103(6): 1176–1209. 91(2): 145–149; J.J. Heckman. 2005. “Lessons from the Technology of Skill Formation.” Annals of the New York 55. Central Statistical Agency. 2017. “Ethiopia Demographic Academy of Sciences 1038(1): 179–200; J.J. Heckman, J. and Health Survey 2016.” Stixrud, and S. Urzua. 2006. “The Effects of Cognitive and Noncognitive Abilities on Labor Market Outcomes 56. Foster and Rosenzweig. 1995. “Learning by Doing and and Social Behavior.” Journal of Labor Economics Learning from Others: Human Capital and Technical 24(3): 411–482; J.J. Heckman and T. D. Kautz. 2012. Change in Agriculture.” “Hard Evidence on Soft Skills.” NBER Working Paper; B. Roberts, R. Shiner, N. Kuncel, and L.R. Goldberg. 2007. 57. World Bank. 2009b. “Ethiopia: Toward the Competitive “The Power of Personality: The Comparative Validity of Frontier: Strategies for Improving Ethiopia’s Investment Personality Traits, Socioeconomic Status, and Cognitive Climate.” Ability for Predicting Important Life Outcomes.” Perspectives on Psychological Science 2(4): 313–345; 58. World Bank. 2009b. “Ethiopia: Toward the Competitive F.L. Schmidt and J. Hunter. 2004. “General Mental Frontier: Strategies for Improving Ethiopia’s investment Ability in the World of Work: Occupational Attainment Climate.”; E. B. Geleta, P. Elabor-Idemudia, C. Henry, and Job Performance.” Journal of Personality and and N. Reggassa. 2017. “The Challenges of Empowering Social Psychology 86(1): 162–173. Women: The Experience of Pulse Innovation Project in Southern Ethiopia.” SAGE Open. 63. L.W. Busenitz and J.B. Barney. 1997. “Differences between Entrepreneurs and Managers in Large Organizations: 59. World Bank. 2009b. “Ethiopia: Toward the Competitive Biases and Heuristics in Strategic Decision-Making.” Frontier: Strategies for Improving Ethiopia’s investment Journal of Business Venturing 12(1): 9–30; A. Cuervo. Climate.” 2005. “Individual and Environmental Determinants of Entrepreneurship.” The International Entrepreneurship 60. A. P. de la O Campos, K. A. Covarrubias, and A. Prieto and Management Journal 1(3): 293–311; W. McClelland. Patron. 2016. “How Does the Choice of the Gender 1968. “The Process of Effecting Change.” George Indicator Affect the Analysis of Gender Differences Washington University., Alexandria, Virginia: Human in Agricultural Productivity? Evidence from Uganda.” Resources Research Office; S.A. Shane. 2003. A World Development 77(C): 17–33. General Theory of Entrepreneurship: The Individual- Opportunity Nexus Cheltenham: Edward Elgar. 61. S. Alibhai, N. Buehren, and S. Papineni. 2015. “Female Entrepreneurs Who Succeed in Male-Dominated 64. F. Campos, R. Coleman, A. Conconi, A. Donald, M. Gassier, Sectors in Ethiopia.” Gender Innovation Lab Policy Brief M. Goldstein, Z. Hernandez, J. Mikulski, A. Milazzo, M. No. 12. Washington, DC: World Bank. Paryavi, R. Pierotti, M. O’Sullivan, and J. Vaillant. 2019. “Profiting from Parity: Unlocking the Potential of Women’s 62. D. Almond and J. Currie. 2011. “Human Capital Businesses in Africa.” Washington, DC: World Bank. Development before Age Five.” NBER Working Paper; L. Borghans, H. Meijers, and B. Weel. 2008. “The Role of 65. S. Dercon and A. Singh. 2011. “From Nutrition to Noncognitive Skills in Explaining Cognitive Test Scores.” Aspirations and Self-Efficacy: Gender Bias over Time Economic Inquiry 46(1): 2–12; S. Bowles, H. Gintis, and among Children in Four Countries.” Young Lives Working M. Osborne. 2001. “The Determinants of Earnings: A Paper. Oxford, UK: Young Lives. 80 66. E.L. Paluck and L. Ball. 2010. “Social Norms Marketing to 76. A. Pankhurst, G. Crivello, and A. Tiumelissan. 2016. Reduce Gender-Based Violence.” IRC Policy Briefcase. “Children’s Work in Family and Community Contexts: Examples from Young Lives Ethiopia.” Oxford, UK: 67. E.A. Cech. 2013. “Ideological Wage Inequalities? The Young Lives. Technical/Social Dualism and the Gender Wage Gap in Engineering.” Social Forces 91(4): 1147–1182; C.L. 77. M. Frost and C. Rolleston. 2013. “Improving Education Ridgeway and S.J. Correll. 2004. “Unpacking the Gender Quality, Equity and Access: A Report on Findings from System: A Theoretical Perspective on Cultural Beliefs in the Young Lives School Survey (Round 1) in Ethiopia.” Social Relations.” Gender & Society 18(4): 510–531. Oxford, UK: Young Lives. 68. J. Bongaarts, B. S. Mensch, and A. K. Blanc. 2017. 78. Central Statistical Agency. 2017. “Ethiopia Demographic “Trends in the Age at Reproductive Transitions in the and Health Survey 2016.” Developing world: The Role of Education.” A Journal of Demography 79(2): 139–154; S. Clark, A. Koski, and E. 79. S. Clark and S. Brauner-Otto. 2015. “Divorce in Sub- Smith-Greenaway. 2017. “Recent Trends in Premarital Saharan Africa: Are Unions Becoming Less Stable?” Fertility across Sub-Saharan Africa.” Studies in Family Population and Development Review 41(4): 583–605. Planning 48(1): 3–22. 80. CSA (Central Statistical Agency) Ethiopia and ICF. 2012. 69. Central Statistical Agency. 2017. “Ethiopia Demographic “Ethiopia Demographic and Health Survey 2011.” Addis and Health Survey 2016.” Ababa, Ethiopia: CSA and ICF. 70. M. Hallward-Driemeier and O. Gajigo. 2015. 81. D. Tilson and U. Larsen. 2000. “Divorce in Ethiopia: The “Strengthening Economic Rights and Women’s Impact of Early Marriage and Childlessness.” Journal of Occupational Choice: The Impact of Reforming Ethiopia’s Biosocial Science 32(3): 355–372. Family Law.” World Development 70(C): 260–273. 82. T. Ketema, G. Bastian, O. Gras, Z. Abro, K. Manchester, and 71. Central Statistical Agency. 2017. “Ethiopia Demographic E. Carranza. 2015. “Ethiopia Women Agribusiness Leaders and Health Survey 2016.” Network Impact Evaluation: Baseline Survey Report.” Washington, DC: World Bank. Unpublished data. 72. J. A. Behrman. 2015. “The Effect of Increased Primary schooling on Adult Women’s HIV Status in Malawi 83. B. Stevenson and J. Wolfers. 2006. “Bargaining in the and Uganda: Universal Primary Education as a Natural Shadow of the Law: Divorce Laws and Family Distress.” Experiment.” Social Science & Medicine 127: 108–15. The Quarterly Journal of Economics 121(1): 267–288. 73. Bongaarts, Mensch, and Blanc 2017. “Trends in the Age at 84. Central Statistical Agency. 2017. “Ethiopia Demographic Reproductive Transitions in the Developing World: The and Health Survey 2016.” Role of Education.” 85. CSA (Central Statistical Agency) Ethiopia and ICF. 2010. 74. B. Bahriu and A. B. Megistu. 2018. “The Challenges “Ethiopia Demographic and Health Survey 2009.” Addis of Women Leaders of Business Organizations in Ababa, Ethiopia: CSA and ICF. Addis Ababa, Ethiopia, in Balancing Work-Family Responsibilities.” Journal of International Women’s 86. S.T. Holden, K. Deininger, and H. Ghebru. 2011. “Tenure Studies 19(2): 140–158. Insecurity, Gender, Low-Cost Land Certification and Land Rental Market Participation in Ethiopia.” Journal 75. S. Clark, M. De Almada, C. W. Kabiru, S. Muthuri, and M. of Development Studies 47: 31–47. Wanjohi. 2018. “Balancing Paid Work and Child Care in a Slum of Nairobi, Kenya: The Case for Centre-Based 87. N. Kumar and A. R. Quisumbing. 2012. “Beyond ‘Death Child Care.” Journal of Family Studies 1–19. Do Us Part’: The Long-Term Implications of Divorce 81 Perceptions on Women’s Well-Being and Child 91. A. Wynn and S. Correll. 2018. “Puncturing the Pipeline: Do Schooling in Rural Ethiopia.” World Development 40(12): Technology Companies Alienate Women in Recruiting 2478–2489. Sessions?” Social Studies of Science 48(1): 149–164. 88. N. Kumar and A. R. Quisumbing. 2012. “Beyond ‘Death 92. MoWCA (Ministry of Women’s and Children’s Affairs). Do Us Part’: The Long-Term Implications of Divorce 2017. “Women Development and Change Package”. Perceptions on Women’s Well-Being and Child Ministry of Women’s and Children’s Affairs; MoLSA Schooling in Rural Ethiopia.” World Development 40(12): (Ministry of Labor and Social Affairs). 2009. “National 2478–2489. Employment Policy and Strategy“, Ministry of Labor and Social Affairs. 89. S. Thébaud. 2010. “Masculinity, Bargaining and Breadwinning: Understanding Men’s Housework in the 93. UN Women. 2014. “Preliminary Gender Profile of Cultural Context of Paid Work.” SAGE Open. Ethiopia.” 90. C. Goldin and C. Rouse. 2000. “The Impact of ‘Blind’ 94. UNDP (United Nations Development Programme). 2017. Auditions on Female Musicians.” The American “Spark, Scale, Sustain: Innovation for the Sustainable Economic Review 90(4): 715–741. Development Goals.” New York City, NY: UNDP. 82