TOWARDS A M O R E ABL E D E VEL O P M EN T EQUI T © 2023 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 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. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http:// creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Malaeb, B.; Duplantier, A., Maarraoui, G., Fluet, E. (2023). Towards a More Equitable Development: Constraints and Opportunities of Women in Djibouti. Washington DC. © World Bank. Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by the World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. Design: Florencia Micheltorena Cover page: Photo credit: Shutterstock Acknowledgements The report was prepared by Bilal Malaeb, Anne Duplantier and Giorgio Maarraoui of the Poverty and Equity team and Eloise Fluet of the Social Development team at the World Bank. The team acknowledges the contribution of Monica Vidili, and Agathe Marie Christien to the GBV analysis in the report, as well as Sarah Houssein, Farida Aboulmagd and Elissar Tatum Harati who contributed to the policy directions. The team wishes to thank Johannes Hoogeveen, Jonna Lundvall, Mirjam Kalle, Federica Marzo, Maria Eugenia Genoni, Bridget Crumpton, and colleagues from the World Bank Djibouti team for their guidance in preparing the report. The team is also grateful to the Ministry of Women and Family in Djibouti and the Gender Observatory for sharing background documents that helped shape this report. Table of Contents EXECUTIVE SUMMARY 7 Key Findings 9 Key messages 13 INTRODUCTION 15 CHAPTER 1. INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS 19 Women’s legal status 20 National strategies to promote gender equality 26 CHAPTER 2. ENDOWMENTS 30 Education 31 Health 43 CHAPTER 3. ECONOMIC OPPORTUNITIES 49 Overview of the labor market 50 Characteristics of labor force participation 52 Characteristics of unemployment 55 Characteristics of employment 61 Female entrepreneurship and access to the world of finance 69 Poverty 70 CHAPTER 4. VOICE AND AGENCY 71 Gender based violence prevalence and drivers 72 Intra-household decision making 77 Participation in political life and access to decision-making spheres 78 CHAPTER 5. GENDERED COVID-19 IMPACTS 80 CHAPTER 6. CONCLUSION AND POLICY RECOMMENDATIONS 83 Closing gaps 84 Creating opportunities 85 Bolstering agency 86 BIBLIOGRAPHY 88 ANNEXES 92 Annex 1. International and regional human rights treaties 93 Annex 2. Non-exhaustive gender relevant national laws 94 Annex 3. Legal gaps in criminal and personal status laws 95 TABLE OF FIGURES Figure E.1 Labor market characteristics of women (with at least secondary education) 11 Figure 0.1 Conceptual framework – Understanding the status of women in Djibouti 17 Figure 1.1 Timeline of the legal framework changes for women’s situation in Djibouti 21 Figure 1.2 Global index and indicators for Djibouti and comparable countries 25 Figure 1.3 Timeline of the national strategies for women’s situation in Djibouti 26 Figure 2.1 Ever attended school by region and age groups 32 Figure 2.2 Literacy rate by poverty status and age groups 32 Figure 2.3 Educational attainment by area and age groups 33 Figure 2.4 Proportion of non-educated by region 34 Figure 2.5 Educational mobility by poverty status and age groups 35 Figure 2.6 Literacy rate and educational attainment in Djibouti-ville 35 Figure 2.7 Percentage of girls among students in public schools by location in 2019 35 Figure 2.8 Attendance and literacy rate by region and age groups 36 Figure 2.9 Current attendance and educational attainment 37 Figure 2.10 Attendance and stopping school age by area 39 Figure 2.11 Acceptance rate at exams in public and private schools in 2019 (percent) 41 Figure 2.12 Acceptance rate at exams in public schools by region (percent) 41 Figure 2.13 Girls at the primary education level in comparable countries (percent) 42 Figure 2.14 Girls at the secondary education level in comparable countries (percent) 43 Figure 2.15 Need and use of health services in the last 30 days per region 44 Figure 2.16 Possession of pregnancy and vaccination booklet by region 46 Figure 2.17 Distance and transport to main services 48 Figure 2.18 Distance and transport to health center by area 48 Figure 3.1 Summary of status of the active age population by gender and year 51 Figure 3.2 Labor force participation by region and area 53 Figure 3.3 Labor force participation by years of education and age 53 Figure 3.4 Labor force participation and unemployment in comparable countries 55 Figure 3.5 Unemployment rate by region and poverty status 56 Figure 3.6 Unemployment rate by education level and age groups 57 Figure 3.7 NEET by poverty status and educational attainment 58 Figure 3.8 Type of unemployment by age groups and poverty 58 Figure 3.9 Length of unemployment by education level 59 Figure 3.10 Reason for not searching for a job by education level 59 Figure 3.11 Labor force participation and unemployment 57 Figure 3.12 Sector and type of unemployment 58 Figure 3.13 Sector of employment by area and educational attainment 63 Figure 3.14 Type of employment by poverty status and educational attainment 63 Figure 3.15 Contracts of waged workers by age groups and educational attainment 65 Figure 3.16 Social security coverage of waged workers by educational attainment and poverty status 65 Figure 3.17 Firm size and branches of employed by education level 66 Figure 3.18 Labor force participation and unemployment rate by age group 67 Figure 3.19 Sector and type of employment 65 Figure 4.1 FGM/C prevalence by region (2012-2019) 76 Figure 4.2 Decision makers on food expenditure 78 Figure 4.3 Decision makers on equipment expenditure and marriage 78 Figure 4.4 Proportion of seats held by women in national parliament 79 TABLE OF TABLES Table E.1 Summary of policy recommendations 13 Table 2.1 Gender parity ratio in public and private schools in 2019 38 Table 3.1 National and ILO definition of labor market participation 51 Table 3.2 Comparison between national and ILO definitions on labor market outcomes (percent) 51 Table 3.3 Determinants of labor force participation among the working-age (15+) population 54 Table 4.1 IPV prevalence 73 Table 4.2 Reasons for not seeking help 73 Table 4.3 Prevalence of early marriage 75 Table 4.4 Causes of early/forced marriages in urban and rural areas 75 Table 4.5 FGM/C prevalence at the national level (2002-2019) 76 Table 4.6 FGM/C prevalence by area 76 TABLES OF BOXES Box 1 Women, Business and the Law 24 Box 2 Changes over time in education outcomes 36 Box 3 Access to health and education services 47 Box 4 Deep Dive: Djibouti-ville and economic opportunities 60 Box 5 Changes over time in economic opportunities (EDAM 2012/2017) 64 Acronyms ACHPR – African Charter on Human & Peoples’ Rights MASS – Ministry of Social Affairs and Solidarity ACHPRRWA – African Charter on Human and MENA – Middle East and North Africa Peoples’ Rights on the Rights of the Women in Africa MICS – Multiple Indicator Cluster Surveys ACRWC – African Charter on the Rights and Welfare MPI – Multidimensional Poverty Index of the Child MSMEs – Micro, Small, and Medium Enterprises AFD – Agence Française de Développement NEET – Not in Employment, Education, or Training BEF – Brevet d’Études Fondamentales NER – Net Enrollment Rate CEDAW – Convention on the Elimination of All Forms OECD – Organization of Economic Co-operation and of Discrimination Against Women Development CEIO – Cellule d’Écoute, d’Information et OFPRA – Office Français de Protection des Réfugiés d’Orientation et Apatrides CRC – Convention on the Rights of the Child p.p. – percentage point(s) CTOCIDTP – Convention Against Torture & PAPFAM – Pan Arab Project for Family Health Other Cruel, Inhuman or Degrading Treatment or PPP – Purchasing Power Parity Punishment PSEA – Sexual Exploitation and Abuse Prevention DHS – Demographic and Health Survey SCAPE – Stratégie de Croissance Accélérée et de la DISED – Direction de la Statistique et des Études Promotion de l’Emploi Démographiques SDGs – Sustainable Development Goals EDAM – Enquête Djiboutienne Auprès des Ménages SSE – Social and Solidarity Economy EDESIC – Enquête Djiboutienne sur l’Emploi, le UNDP – United Nations Development Programme Secteur Informel et la Consommation des ménages UNESCO – United Nations Educational, Scientific and ENPH – L’Enquête Nationale de Prévalence des Cultural Organization Handicaps UNFD – Union Nationale des Femmes de Djibouti EDIM – Enquête Djiboutienne à Indicateurs Multiples UNGA – United Nation General Assembly FGM/C – Female Genital Mutilation/Cutting UNFPA – United Nations Fund for Population GBV – Gender Based Violence Activities GER – Gross Enrollment Rate UNHCR – United Nations High Commissioner for ICCPR – International Covenant on Civil & Political Refugees Rights UNICEF – United Nations International Children’s ICESCR – International Covenant on Economic, Social Emergency Fund & Cultural Rights USAID – U.S. Agency for International Development ICT – Information and communication technologies WBL – Women, Business, and the Law IFAD – International Fund for Agricultural WDI – World Development Indicators Development WDR – World Development Report ILO – International Labor Organization WED – Women’s Entrepreneurship Development INSD – Institut National de la Statistique de Djibouti WFP – World Food Programme IPV – Intimate Partner Violence WHO – World Health Organization KAP – Knowledge, Attitudes and Practices DJIBOUTI GENDER ASSESSMENT EXECUTIVE SUMMARY Photo credit: Shutterstock 7 DJIBOUTI GENDER ASSESSMENT EXECUTIVE SUMMARY D jibouti has experienced a record of strong economic growth and has made considerable progress in various developmental indicators in recent years. Due to its strategic geographical location and its regional importance, Djibouti made times higher than it is in urban areas, and deprivations in sanitation, access to clean water, electricity, and decent housing are significantly higher in rural areas. Women in Djibouti are at a disadvantage in education, work, access to finance, and are prone to experience important gains from trade which largely propelled gender-based violence. its economic growth. Real GDP growth has averaged around 5 percent in the last decade, bringing per Djibouti has experienced dynamic social change capita income from 2,386 USD in 2013 to 3,246 reflected in the considerable improvement over the USD in 2020.1 This increase in income is reflected in years in women’s rights and responsibilities as well improvements in other indicators of welfare. During as their access to opportunities, but the patriarchal the same period, life expectancy for both men and societal structure and the disconnect between women has increased from around 62 to 67 years; constitutional and legal frameworks continue to under-5 child mortality decreased from 77 to 56 per impede women. Progress has been made in key 1000 births; and enrollment in primary education areas, such as participation in political life with 26 increased by 12 percent. percent of parliamentary seats reserved for women and about a fifth of the current ministerial positions Despite this important progress, major challenges held by women. Progress was also made regarding remain as Djibouti’s development has not been female genital mutilation and cutting in terms of equitable nor inclusive, including from a gender legislation. At the same time societal structures perspective. Despite the relatively high per capita continue to privilege men over women giving them income, in 2017, around 21 percent of the population disproportionate control over resources and power. were estimated to live under the national poverty For instance, men have better access to credit and line; and 17 percent in extreme poverty under the financial services, land ownership, and business international poverty line of 1.90 dollars per person opportunities. Certainly, changes have been made at per day. In addition, half of the economically active the legal and institutional level and the constitution population was unemployed. The urban-rural divide appears to be progressive since it stipulates that and disparities between residents of the capital men and women are equal before the law. However, city and the outer regions remained an important some components of the Civil Code hinder women’s challenge. Indeed, outer regions are three times opportunities, notably through unequal inheritance poorer than Djibouti city, rural poverty is around four and divorce rights between men and women. 1  In constant 2015 USD. Growth declined substantially in 2020 due to the COVID-19 pandemic, but remained positive. 8 DJIBOUTI GENDER ASSESSMENT EXECUTIVE SUMMARY This gender assessment provides a comprehensive against FGM/C. However, the report highlights view of the status of women in Djibouti along several shortcomings in the legal framework and its three dimensions: endowments in human capital, implementation. Specifically, these relate to parental economic opportunities, as well as voice and leave, divorce and custody, inheritance, and forced agency, which encompasses issues such as gender- and early marriage – all of which have economic based violence, decision making, and political and social consequences on the status of women participation, among others. The report follows the in Djibouti. The Family Code, for instance, restricts framework of the 2012 World Development Report women’s inheritance and divorce rights. In addition, (WDR) and conceptualizes women’s empowerment as the enforcement and implementation of laws relating an interaction between formal and informal institutions, to FGM/C have lagged, marital rape continues to be as well as markets and households, that affect women’s neither prohibited nor criminalized, and the Family human capital endowments, economic opportunities, Code does not adequately protect women from and agency. As such, the report takes stock of existing intimate partner violence (IPV). legal frameworks, policies, and national strategies for women, analyzes existing data on women’s education In the education dimension, despite an increase and health, and assesses their labor market outcomes. in school attendance in recent years (for both It also examines social issues such as gender-based males and females) and a narrowing gender gap violence, influence over decision-making, and in children’s education, girls are still less likely to participation in political life. The analysis relies on attend school than boys and female illiteracy is existing studies as well as primary and secondary very high. The younger the cohort, the smaller the data sources, such as the household budget survey gender gap in school attendance. Further, compared (EDAM24 2017) and World Development Indicators, to women in older cohorts, those in younger cohorts among others. The gender assessment concludes with have higher rates of upward mobility with respect a set of policy recommendations on how to improve to their fathers’ education. Literacy and educational development outcomes for women in Djibouti. attainment exhibit an analogous trend. In spite of these positive developments, there are notable regional disparities. For instance, women living in Tadjourah Key Findings are nearly twice less likely than those in Djibouti- ville to have ever attended school. In comparison Equality between men and women is guaranteed to neighboring countries and countries with similar by the Djiboutian constitution and laws, but a socio-economic indicators, Djibouti fares worse than disconnect exists between the constitutional Cabo Verde and São Tomé and Príncipe as well as guarantee, the family code3 and policy the average of Middle East and North Africa (MENA) implementation. Laws such as the Family Code and region in terms of girls’ primary school enrollment but Labor Code ensure women’s labor rights and access performs better than its neighboring country Eritrea. to education for all. The legal framework imposes At the secondary education level, the gap between minimum quota for the representation of women Djibouti and comparable countries is even higher. in political life and institutionalizes the fight against the practice of female genital mutilation or cutting Gender differences in health appear minimal (FGM/C) and other kinds of gender-based violence but maternal mortality is higher than in many (GBV). Several national strategies and measures comparable countries. For children, no gender promoting gender equality have been adopted by difference is observed in the area of vaccination the Djiboutian government since the 1990s. They and girls’ mortality rate is lower than that of boys. focus on various priorities such as education and Girls from 6 months to 5 years of age are less likely children, economic empowerment, and the fight than boys to experience stunted growth, acute 2  Short for “Enquête Djiboutienne Auprès des ménages“. 3  It should be noted that an update of the Family Code is currently underway by the Ministry of Women and the Family in collaboration with the Ministry of Justice. The new version should be finalized and adopted during 2023. 9 DJIBOUTI GENDER ASSESSMENT EXECUTIVE SUMMARY malnutrition and ponderous insufficiency. The gender those who completed primary education but did not gap in HIV prevalence has closed during the last reach secondary education, which is the case for 22 20 years. However, women living in rural areas are percent of working age women. Among those with less likely than men to receive health services when this level of education, only 31 percent are active in needed. Djibouti has one of the highest ratios in the the labor force and 21 percent of those active ones are world for maternal mortality and the highest in the employed. Women with at least a secondary level of region and across economically comparable nations. education (11 percent of working age women) are the Furthermore, at the infant age, the mortality rate most likely to be active (45 percent). Less than half of of Djiboutian girls is one of the highest among the economically active women with secondary education neighboring countries and is much higher than the are employed (46 percent). average of the MENA region. Social norms and the general economic landscape There exist notable gender differences in labor likely discourage women from participating in market outcomes among Djiboutians aged 15 and the labor force, however, even among those who over, with women’s labor force participation being are active in the labor force, the market is only able half that of men. The share of women in employment to absorb a small proportion of women. Across age or training (11 percent) is roughly a third that of males groups, the youth (15-24 years old) have the highest (35 percent). Women are more likely than men to be unemployment rate but the lowest gender gap, unemployed and 55 percent of women are neither suggesting that young men and women face many in education nor in the labor force, while this is the of the same constraints in the labor market. Most case for only 27 percent of men. The labor force working women are in the informal sector. In fact, participation of women is very low (33 percent), and nearly all women without education work in the men are almost twice as likely to participate in the labor informal sector or as self-employed. As such, women force. This considerable gender gap has increased are less likely than men to have a contract and a social from 2012 to 2017. Even among educated women, security coverage. Relative to comparable countries, labor force participation is low. However, the gender women’s labor force participation4 in Djibouti is similar gap in labor force participation decreases with years to that of Cabo Verde and higher than neighboring of education from from 30 percentage points (p.p.) of countries Somalia and Yemen, and nearly twice difference among the non-educated to less than 10 as high as the MENA regional average. However, p.p. difference among the most educated. Women’s Djiboutian women’s participation in the labor force labor force participation is the lowest in Obock and lags far behind neighboring countries such as Eritrea Tadjourah (where women are also the least likely to and Ethiopia (according to figures from the WDI be educated, and poverty levels among the highest in database). In terms of unemployment, Djiboutian the country) and the highest in Djibouti-ville and Arta. women are more likely to be unemployed than those in neighboring countries Eritrea and Ethiopia, but the Half of all economically active women are female unemployment rate is lower than the average unemployed and an overwhelming majority of for the MENA region, Cape Verde and São Tomé and those who are working is engaged in the informal Príncipe. sector or in self-employment. In fact, a very low percentage (11 percent) of all women aged 15 and Access to finance is identified as the biggest more are working. The gender gap has decreased hindrance to female entrepreneurship. Around 58 between 2012 and 2017 by 5 percentage points, but percent of women-owned businesses and micro, small, remains high at 25 percent. The unemployment rate and medium enterprises (MSMEs) report that lack of of women is the highest in Tadjourah and Arta, where finance is among the greatest challenges they face. education and labor force participation are also the Compared to their male counterparts, they have lower lowest. Women’s unemployment is the highest for rates of owning bank accounts (54 percent compared 4  According to the World Bank’s World Development Indicators (WDI) database for the year 2017, a harmonized database that allows comparisons at the global level (between countries). 10 DJIBOUTI GENDER ASSESSMENT EXECUTIVE SUMMARY Figure E.1 Labor market characteristics of women (with at least secondary education) Djiboutian Women with at least a secondary education Employed Unemployed Inactive Source: Authors’ calculations based on EDAM 2017. to 71 percent), using loans (20 percent compared to No significant differences are observed between 25 percent), long-term loans (8 percent compared to women and men both in terms of monetary and 15 percent), and rely more on informal mechanisms multidimensional poverty. In 2017, 21 percent of the such as the tontine system and borrowing from family population in Djibouti was below the extreme poverty members (53 percent compared to 33 percent). line. While men and women are almost equally likely to be living in extremely poor households (22 and 21 Containment measures associated to the COVID-19 percent, respectively), rural women are much more outbreak had varied labor impacts depending likely to be considered extremely poor than urban on the gender of the breadwinner. According to ones (63 percent compared to 14 percent). The rapid phone surveys of households’ breadwinners Multidimensional Poverty Index (MPI), which combines in Djibouti, men were more likely than women to monetary poverty and indicators on education and be working as a result of the pandemic. Among the access to basic services and infrastructure, shows breadwinners who worked less, females were less that 28 percent of the population is estimated to live likely than males to receive a full paycheck during the in a multidimensionally poor household. The MPI is early stages of the COVID-19 pandemic. In March/ higher among women than men for those aged 15- April 2021, households with a female breadwinner 24 and 25-39, while the inverse is observed for the were more likely than their male-headed counterparts older age-groups. Women are slightly more likely to have a poor food consumption score. In addition, than men to live in a household deprived of improved a relatively larger share of households with a female sanitation. In addition, women are more likely to live in breadwinner could not access healthcare. Evidence a household with no individuals aged 13 and over with suggests that the COVID-19 pandemic has also primary education than men (32 percent compared exacerbated existing vulnerabilities in Djibouti, to 29 percent). However, men and women have on especially the GBV risks such as Intimate Partner average the same probability to be deprived of Violence (IPV). access to electricity and improved drinking water (34 11 DJIBOUTI GENDER ASSESSMENT EXECUTIVE SUMMARY percent and 7 percent respectively), as well as living in district of the capital – are the least likely to be literate a monetarily poor household. There is also no gender compared to the other districts, but the gender gap difference in the likelihood of living in a household is the highest in the 1st, 2nd and 3rd districts, the less where at least one school-aged child is not enrolled in poor districts. Women aged 15 and over from the school (18 percent). poorest district are half as likely to have completed at least secondary school than those from the richest Gender gaps across all dimensions are also observed districts. On the economic outcomes, the 4th district among the refugees and asylum seekers population also appears to lag behind others. Indeed, women in Djibouti. In terms of educational outcomes, from Balbala, which live in the 4th and 5th districts, refugee girls experience higher school drop-out than are the least likely to be active in the labor market and refugee boys. The gender gap in literacy is high with more likely to be unemployed when active. In addition, 30 percent of refugee women being literate compared women from the 4th district predominantly work in the to half of their male counterparts. In addition, female informal sector and have a lower likelihood to work refugees are almost twice as likely as males to have no in the public sector than women in other districts of education. As in the national population, girls aged 6 the capital. Moreover, the gender gaps in terms of months to 5 years are less prone to stunted growth, unemployment rate and sectors of employment are ponderous insufficiency and acute malnutrition than the highest in the 4th district. boys. However, the data shows that around 9 percent of refugee children aged 5-14 years old are working, While the Penal Code does not explicitly penalize whereas less than 1 percent of Djiboutian children domestic violence and marital rape, this appears to work. The limited job opportunities in refugee camps be widespread and largely underreported. Often, result in refugees migrating to the capital to work. IPV is a taboo topic and a private matter solved within Indeed, 21 percent and 38 percent of refugee women families without resorting to the formal legal system, and men aged 15 and more are working, lower than and women who experienced IPV often do not seek for their Djiboutian counterparts. This limited access for help. There is a lack of nationally representative to work increases food insecurity, especially for data on non-partner violence in Djibouti. Vulnerable female-headed households. This precarity is further populations, especially female migrants and refugees exacerbated by the limited access to credit. Vulnerable are at risk of non-partner violence, including sexual populations, such as refugees, are particularly prone violence, abuse and exploitation while on the move to GBV during the migration journey and upon arrival and upon arrival in Djibouti. Forced/early marriage in both refugee camps and urban settings. In refugee is common in Djibouti, especially in rural areas and camps for example, firewood collection points are family considerations and community traditions are usually distant, which increases the risk and exposure the main drivers of forced/early marriage. of girls to violence. Moreover, refugee women have been particularly affected by domestic violence. Despite the government of Djibouti having The nationally representative household budget criminalized FGM/C, it remains highly prevalent. survey does not represent the migrant groups and FGM/C is more prevalent among older women, undocumented persons and underrepresents the indicating that the practice is increasingly abandoned nomadic population. As such, this limits the analysis by younger generations. There are however major and conclusions of this report on women’s situation spatial disparities in FGM/C prevalence. While the within these population groups. practice has decreased in urban areas for women aged 15-year-old and above from 93 percent in 2006 Within the Djiboutian capital, gendered educational to 87 percent in 2019, FGM/C prevalence remains and economic outcomes vary across districts. the same for the same age group in rural areas at Djiboutians living in the capital represent around 80 96 percent. Moreover, the most mutilating forms of percent of the whole population. While in all districts FGM/C remain widespread in rural areas. Indeed, women aged 15 and over are less likely than men to FGM/C is strongly rooted in socio-cultural norms, be literate, those from the 4th district – the poorest religion, and tradition. 12 DJIBOUTI GENDER ASSESSMENT EXECUTIVE SUMMARY Key messages a more holistic approach to incentivize private sector development to create a market for skills, Overall, while Djibouti has made progress over for both men and women alike, given the current the last two decades to advance gender equality, dual labor market (comprising public and informal there remains serious shortcomings that require sectors) is unable to absorb the active labor force. policy attention. By custom and law, men continue Taking a progressive approach to family and civil to have more rights and higher status than women affairs, in line with the constitutional guarantee of and a pattern of gender roles exists that give men equality between men and women, is also essential predominance in public life, business, and politics. to give women equal footing in marriage, divorce Improving the implementation and enforcement and custody, as well as inheritance in order to of gender-relevant policies and laws is a priority. achieve their full potential. Based on the identified Furthermore, Djibouti must address gaps in public constraints and opportunities for women in Djibouti, service provision in education and health. Despite the report’s policy recommendations fall within three women being disadvantaged in the labor market, areas summarized in Table E.1: closing gaps, creating the analysis suggests that the labor market requires opportunities, and bolstering agency. Table E.1  Summary of policy recommendations Closing gaps Creating opportunities Bolstering Agency i) Closing gaps in i) Raising awareness among At the individual level: education outcomes and young girls (12 to 20 years i) Empowering civil society maternal mortality through old) who are out of school organizations to advocate for investments in literacy or have not attended school women’s rights and the ability programs and lifelong on employment and work to train on IPV in order to better education opportunities, opportunities, including non- inform, raise awareness and and improvement in access conventional ones support victims to schooling for girls, as well ii) Improving access of women ii) Gearing policy attention towards as safe and decent access to to productive inputs and credit regularization of the legal status contraception, and medical through national financial of marginalized women (e.g. and midwifery services for inclusion strategy and regulatory undocumented migrants) women environment for microfinance iii) Equipping women, and ii) Raising awareness on and non-conventional GBV survivors in particular, with the status of women in financing to develop female knowledge about their rights and the public, private and entrepreneurship legal procedures to access justice community spheres i) Reducing rural/urban Promoting a healthy and At the legal level: gaps in education and robust private sector with i) Strengthening the justice economic opportunities by equal opportunities for all, and system’s capacity to address GBV reducing poverty, equitably providing incentives for formal ii) Bridging the gaps between the distributing resources and businesses who provide social legal system and customary laws, investing in services in security and decent work for which hinder law enforcement in infrastructure employees relation to GBV ii) Decentralizing vocational iii) Raising awareness,sensitizing training programs and local communities and building revising them to meet labor societal and institutional capacity market needs to address GBV challenges 13 DJIBOUTI GENDER ASSESSMENT EXECUTIVE SUMMARY Table E.1  continued Closing gaps Creating opportunities Bolstering Agency Closing data gaps in gender i) Improving access to land, At the institutional level: disaggregated data on asset internet and information and i) Empowering government actors, ownership, access to finance communication technologies providing them with necessary and credit, normal and intra- (ICT) resources and capacity building household decision making ii) Reducing gaps in inheritance there are necessary to enforce laws information law which inhibit women from and implement policies against achieving their full potential GBV ii) Strengthen the leadership and commitment of women executives, parliamentarians or ministers so that they act more effectively in decision-making iii) Accelerating progress towards the implementation of the National Protocol 14 DJIBOUTI GENDER ASSESSMENT INTRODUCTION Photo credit: Shutterstock 15 DJIBOUTI GENDER ASSESSMENT INTRODUCTION D jibouti has enjoyed remarkable economic growth in the last decade, however this growth is not always inclusive and sustainable. In spite of growth averaging around 5 percent in last decade, income inequality in Djibouti A renewed focus on reducing gender disparities in Djibouti has the potential to contribute to more inclusive and sustainable growth. Djiboutian women are currently disadvantaged in key domains, and the COVID pandemic has likely worsened the is among the highest in the MENA region with a situation and exacerbated inequalities. Compared to Gini index of 41.6. Spatial inequality is also palpable, men, women lag in key educational and labor market where outer regions face three times the poverty rate outcomes. One out of two women over 10 years old in Djibouti-ville. In 2017, around 21 percent of the is not able to read and write compared to a rate of population were estimated to live under the national 34 percent among men (EDAM, 2017). Compared to poverty line and 17 percent in extreme poverty the MENA countries’ rate which stands at 75 percent, ($1.90 a day), and half of the active population was female literacy is particularly low in Djibouti (WDI unemployed (EDAM, 2017). The urban-rural divide database, United Nations Educational, Scientific in Djibouti also creates challenges in access to and Cultural Organization (UNESCO) estimates for opportunities and in promoting inclusive growth and the year 2017). Moreover, a third of the women aged gender equality. Developing the private sector and 15 years old and more completed primary education ensuring healthier, educated, and productive workers compared to about 50 percent for men. Such gaps are key development goals in Djibouti. As such, gender in terms of education may explain at least in part equality plays a crucial role in developing Djibouti’s the high unemployment and inactivity rates among workforce and unlocking inclusive sustainable growth women, particularly young women, who are more potentials, in addition to its inherent and instrumental likely than men to be in the informal sector when importance in the path to sustainable development. employed. The ratio of female to male labor force Increasing women’s access to finance, labor markets, participation is 0.39 at the national level, which is healthcare, schooling, and inheritance rights, as well significantly lower than the lower middle-income as access to justice and protection, are all important to countries’ average of 0.50 and the world average improve household welfare, growing the economy, and of 0.68 in 2017 (WDI, 2017). In addition, there are securing fundamental human rights. Intergenerational marked disparities in labor market outcomes across development, poverty reduction, and social mobility subgroups of the female population, such as by also rely fundamentally on women’s empowerment location and poverty status. The current COVID in light of women’s role and importance in improving pandemic is likely to have worsened such divides, health and educational outcomes for children. disproportionally affecting them. 16 DJIBOUTI GENDER ASSESSMENT INTRODUCTION Figure 0.1 Conceptual framework - Understanding the status of women in Djibouti R EQUALIT ND E Y po GE lic ies TIONS STITU MAL IN INFOR ECONOMIC OPPORTUNITIES TS HOUSEHOLDS RKE MA AGENCY ENDOWMENTS FOR MAL INST ITUT ION S GR OW TH Source: WDR On Gender and Development (2012). Gender equality is a core development objective, of women’s skills and talents negatively affects a smart development policy, and a core priority of productivity and poverty, and impact opportunities of the Government of Djibouti’s 2035 vision. Gender the future generations. Indeed, the literature5 shows equality is considered a key pathway to ensure lasting that women’s empowerment has positive effects on poverty reduction and shared prosperity, and a the health and education outcomes of their children. fundamental human right (Sustainable Development Goal (SDG) 5). Identifying the main gender gaps and This analysis relies on the analytical framework windows of opportunity that the country faces across developed in the World Bank 2012 WDR on Gender different domains contributes to a better-informed Equality and Development (Figure 0.1). The WDR policy design. As such, since the publication of the 2012 sees gender equality as resulting from the WDR on Gender and Development (2012), the World interaction of: (i) existing formal institutions, stemming Bank Strategy has been progressively geared towards directly from the government such as the legal and providing an increased attention and resources to regulatory frameworks, the education and health improve gender outcomes in countries, via support to services, and all other basic infrastructures provided both policy reforms and investment lending. Moreover, by the state, (ii) different economic markets such as the gender equality and economic development affect labor market, the financial system and the goods and each other simultaneously, as the misallocation services market, (iii) informal institutions which include 5  Haddad, Hoddinott, and Alderman 1997; Katz and Chamorro 2003; Duflo 2003; Thomas 1990; Hoddinott and Haddad 1995; Lundberg, Pollak, and Wales 1997; Quisumbing and Maluccio 2000; Attanasio and Lechene 2002; Rubalcava, Teruel, and Thomas 2009; Doss 2006; Schady and Rosero 2008; Thomas, Strauss, and Henriques 1990; Allendorf 2007; Andrabi, Das, and Khwaja 2008; Dumas and Lambert 2010; Kishor and Johnson 2004; Jeyaseelan and others 2007; Hindin, Kishor, and Ansara 2008; Koenig and others 2006; Martin and others 2002. 17 DJIBOUTI GENDER ASSESSMENT INTRODUCTION norms, beliefs and roles that shape social interactions, The section aims to identify key constraints to and (iv) households. Formal institutions set the shape an understanding of priority policy areas. framework of functioning of the different markets, • Voice and agency: The WDR (2012) defines the which in turn influence the norms and beliefs of the women’s agency as “their ability to make choices society over time. Thus, individuals and households that lead to desired outcomes”.6 While agency, will make choices in terms of education, health and including norms, voice and representation, and economic life influenced by their preferences, which are freedom from violence, are difficult to measure shaped by the informal institutions and the incentives empirically given the limited data, this section received from the markets. These choices are also of the report aims to highlight challenges and constrained by the interplay of markets, informal, opportunities women face in Djibouti in terms and formal institutions. Governments can implement of GBV, access to infrastructure, and household policies and reforms through formal institutions, and decision making. thus impacting markets and informal institutions, all of which could influence households’, employers’, and Several data constraints have limited this analysis, service providers’ choices in achieving greater gender which call for greater investments in data equality in education, health, and access to social and production and dissemination in general, and economic opportunities. This gender assessment aims gender-specific ones in particular. Indeed, recent to provide the key analytical underpinnings to inform and accessible micro-data allowing disaggregation the preparation of such activities and constitute a by gender on different topics are limited in Djibouti. reference document describing the status of women Availability of health data is one of the issues, as for along the key socio-economic dimensions: example there is no recent Demographic and Health Survey (DHS) or Multiple Indicator Cluster Survey • Legal and regulatory systems, as enablers, can (MICS) data for the country. When some data exists, play an important role in women’s social, economic, access is not always guaranteed. Thus, the report relies and political participation by promoting and at times on survey reports, instead of the raw data. protecting women’s rights, including rights related In addition, there are concerns on some of the data to financial inclusion, equal pay and treatment at quality. Indeed, finding high quality and reliable data work, family law issues, and child marriage. This that allows measuring the gender wage gap is one of section includes details on the legal framework, these particular challenges. Representativeness of the national strategies, and an analysis of stakeholders existing datasets at times are limited. For instance, involved in the institutional framework relevant for the household budget survey (EDAM4, 2017) excludes women’s empowerment. a portion of the resident population (e.g. migrants, • Endowments in education and health are key undocumented persons, certain nomadic groups, and for women’s human capital development. They refugees). promote better work opportunities, empower decision making, reduce the incidence of violence, This report is organized as follows: Chapter 1 and foster inter-generational mobility. This section describes the legal and regulatory systems affecting outlines the constraints and opportunities women women in Djibouti, Chapter 2 provides empirical face in these human capital dimensions and evidence on the endowment of women in human highlight the regional differences in outcomes. capital (namely, education and health), Chapter 3 • Women’s economic opportunities: This section describes women’s employment opportunities and includes a presentation of key indicators such as labor market characteristics, Chapter 4 addresses labor force participation, unemployment, youth voice and agency topics such as FGM/C and intra- in “not in employment, education or training” household decision making, Chapter 5 highlights (NEET), quality of employment, entrepreneurship impacts of COVID-19 pandemic and associated and assets, access to finance, poverty (monetary lockdown measures on women, and Chapter 6 and multidimensional) and their interactions with concludes with policy recommendations. other socio-economic characteristics of women. 6  WDR, 2012: 48. 18 1 DJIBOUTI GENDER ASSESSMENT CHAPTER INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS Photo credit: Shutterstock 19 DJIBOUTI GENDER ASSESSMENT INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS Women’s legal status high offices in the State allowed the entry of Djiboutian women into the Parliament. Further in 2018, this law E was amended to 25 percent (Law No. 219/AN/18/7th). quality between men and women is Finally, a decree asking women to be in 20 percent guaranteed by the Djiboutian constitution of senior state functions was set in 2008 (Decree No. and legal framework, but a disconnect exists 2008-0270/PR/MPF). with civil affair laws. In the constitution (1992), article 07 evokes the principle of equality between The legal framework on working rights guarantees women and men and Art. 01 and 03 guarantee equal equal access to employment with, in theory, the treatment of citizens regardless of gender. To push in same treatment, remuneration and allowances. this direction, a Grand Prize of the Head of State for Indeed, Art. 03 of the Labor Code forbids discrimination the advancement of women was created in 2000 to in employment based on gender (among other promote female human resources in socio-economic characteristics). This is reflected by the WBL index7 on and community development. The Djiboutian law workplace, which is 100. Another reason why Djibouti guarantees some rights for girls too. Education at scores 100 on workplace is that the law prohibits gender the primary and middle school levels is mandatory discrimination in the workplace and sexual harassment for all children since 2000 (section 4 of Law No. 96/ in employment with a criminal penalty. When it comes AN/00/4th). Moreover, Djibouti adopted in 2015 a law to pay, according to Art. 112 of the Labor code, women on the Legal Protection Code for Minors designed to are not allowed to work in jobs that are deemed protect and promote the rights of girls and boys (Law arduous in the same way a man can, and according to No. 95/AN/15/7th). In addition, Djiboutian women Art. 111, they cannot work in certain industrial jobs in can pass their nationality to their children (Law No. the same way a man can. On the other hand, Art. 137 79/AN/04/5th L) as the Nationality Code created in guarantees equal pay for work of equal value, for all 2004 states that there must be no discrimination with workers without regard to their sex. For these reasons, respect to acquiring citizenship. Djibouti scores 50 on the WBL pay index. However, as the data suggests, and as will be presented in In terms of participation of women in political life, subsequent sections of this report, women have fewer some steps have been done to encourage a better employment opportunities and appear to be paid less representativeness. In 2002, a quota law (Law No. 192/ than men for the same work. According to the United AN/02/4th) requiring at least 10 percent of either sex Nations Development Programme (UNDP) (2020), being representatives in elective functions and having women’s per capita income is almost half of men’s: in 7  The Women, Business and the Law (WBL) is an index representing gender equality stemming from the laws and regulations that affect women’s economic opportunities. See Box 1 for more details. 20 DJIBOUTI GENDER ASSESSMENT INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS Figure 1.1 Timeline of the legal framework changes 2019, women’s estimated gross national income per for women’s situation in Djibouti capita was $4,151 compared to $7,077 for men (in 2017 purchasing power parity (PPP)). Moreover, according to Anos Casero and Seshan (2006), the average monthly TIMELINE OF income of women was $169 compared to $354 for LEGAL FRAMEWORK men (using EDAM 1996). Art. 113 of the Labor Code ensures that every pregnant woman is entitled to 26 weeks maternity leave which begins seven weeks 1946 before the expected date of delivery and ends 19 Women can vote in elections weeks after the date of delivery, with a 14-week period of maternity leave fully paid (i.e. 98 days of mandatory 1986 paid maternity leave as reports UNDP 2020 in the last Women can stand for Human Development Report). Moreover, Djibouti elections acceded in 2020 the International Labor Organization 1992 (ILO) Maternity Protection Convention (2000). That The principle of equality said, although there is maternity leave, there is no between women and men parental leave and women are often made redundant is guaranteed by the constitution during or after pregnancy. (Organization of Economic 1995 Co-operation and Development (OECD), 2019) even FGM/C is illegal and punishable if Art. 114 of the Labor Code stipulates prohibition of dismissal in the event of pregnancy. In addition, fathers are entitled to 3 days of paid paternity leave (Art. 100 of 2000 Education is compulsory for the Labor Code), which brings the mother at the front of all children at the primary the childcare and less likely to pursue her work life easily. and middle school levels This drives the WBL index on parenthood down to 60. 2003 Djibouti has taken actions to fight the practice At least 10% of either sex as representatives in of FGM/C. In 1995, the National Assembly declared elective functions and in illegal and punishable the practice of FGM/C (through high offices 2004 a fine and up to 5 years of prison) and a few years later, Ability of children to a new law was adopted defining as guilty whoever is acquire citizenship from aware of FGM and did not inform authorities. In 2002, a 2006 either parent PAPFAM8 survey measured FGM/C prevalence for the Right to equal pay for work first time in the country and Djibouti is one of the very of equal value for all few states that measures the prevalence rate clinically workers without regard of their sex 2007 A listening, information, (OFPRA 9, 2017). Moreover, Djibouti established in and guidance unit for 2007 a listening, information, and guidance unit for 2008 violence against women is established violence against women. This includes a social, a 20% of women in senior legal, and a health section designed to help victims by state functions 2015 informing them about their rights and guiding them Legal protection for all to seek the appropriate institutions, services, and minors legal proceedings (Institut national de la statistique 2018 At least 25% of either sex de Djibouti10 (INSD) and Ministry of Women and the as representative in elective Family, 202011). functions and in high offices 2020 Legal prevention, protection, and care for 8  The Pan Arab Project for Family Health survey. women and children who 9  « Office Français de Protection des Réfugiés et Apatrides » in are victims of violence French. 10  National Institute of Statistics in Djibouti in English. 11  Rapport de l’enquête nationale sur les mutilations génitales féminines et les violences faites aux femmes. 21 DJIBOUTI GENDER ASSESSMENT INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS Despite legal progress in criminalizing FGM/C, law legal protection, and access to legal aid and referral enforcement has lagged. Djibouti’s Penal Code (Art. to judicial institutions. The law also sets specific 333, 1995) has criminalized FGM/C with up to five funds to establish “Centers for Integrated Care”13 to years of imprisonment and a fine of one million DJF provide shelters and emergency supports for women (approx. US$5,600). Anyone failing to file a report a and girls, especially minors at risk of forced marriage case of FGM/C to the authorities is punishable with and abuse. These centers have however not been up to one year of imprisonment and a fine of 100,000 established yet at the time of writing. It is important DJF (United Nations Fund for Population Activities to note, however, that the implementing decree for (UNFPA), 2017). In 2009, Law No. 55 on violence against this law has still not been adopted, which limits the women complemented Art. 333 by legally defining application of this new law. FGM/C and calling for penal sanctions. This legislation also allowed associations representing survivors who Although the Penal Code contains specific are unable to file a complaint to bring a claim forward provisions on rape and sexual assault, as well as on their behalf (Thomson Reuters Foundation, 2018). provisions against domestic violence, marital rape Despite this legal progress, the criminalization of is neither prohibited nor criminalized. The Penal FGM/C remains limited to date and FGM/C cases are Code defines rape as any act of sexual penetration of not sufficiently reported, prosecuted or punished.12 any kind committed by violence, coercion, threat or Indeed, the law is perceived by communities as an surprise (Art. 343). Aggravated cases are punishable attack on traditions, and families and villages are up to 20 years of imprisonment, including when lobbying to prevent prosecutions (OFPRA, 2017). In committed against vulnerable persons due to age, addition, access to legal administrations in rural areas illness, disability, pregnancy, abuse of authority, or remains limited and legal procedures are complex when committed with the use of a weapon, gang and represent significant financial implications for rape, or against a child under 15 years old (Art. 344). complainants. Women and girls would also not report When a person dies as a result of rape, or when FGM/C cases due to social stigma (Petit, 2013). rape is coupled with acts of barbarism or torture, the penalty may be imprisonment for life (Art. 345) The Government of Djibouti has recently made (OECD, 2019). Abortion that is not for therapeutic progress through enacting the 2020 law on purposes is illegal under the Penal Code, including prevention, protection and care for women and for survivors of rape or incest (UNDP, 2018). Sexual children survivors of violence, which can set the assault is also punishable with up to 20 years of ground rules for a comprehensive framework on imprisonment when accompanied by barbaric acts GBV prevention, protection and response (Law or torture (Art. 326). Nevertheless, these legislations no. 66/AN/719/8th). This law provides definitions offer protection for victims rather than criminalize of different forms of violence against women and the attacker in domestic violence situations. children, including sexual harassment and assault, economic violence, psychological and physical The Family Code (2002)14 also does not adequately violence. It also includes measures to guarantee the protect women from IPV and forced/early marriage confidentiality, security and safety of GBV survivors. and impedes women’s economic opportunities due Provisions on prevention include an early warning inequity in inheritance law. Art. 07 of the Family Code mechanism at the family level, especially related to states that marriage is not valid without the consent of GBV against children. Protection mechanisms include both husband and wife however a girl’s legal guardian 12  In 2014, an isolated case was brought to court when a mother and the person who undertook the FGM/C procedure were charged and received 6-month suspended sentences. No other cases can be identified. See Thomson Reuters Foundation (2018). Djibouti: The Law and FGM/C. https://www.28toomany.org/static/media/uploads/Law percent20Reports/djibouti_law_report_v1_(july_2018).pdf; United Nations International Children’s Emergency Fund (UNICEF) and UNFPA (2015). Rapport sur l’Évaluation de la mise en œuvre de la Stratégie Nationale pour l’Abandon Totale à Toutes Formes d’Excision. https://www.unicef.org/evaldatabase/files/Final5_report_ Evaluation_MGF_Djibouti_2015-001.pdf 13  « Centres de Prises en Charge Intégrées » in French. 14  It should be noted that an update of the Family Code is currently underway by the Ministry of Women and the Family in collaboration with the Ministry of Justice. The new version should be finalized and adopted during 2023. 22 DJIBOUTI GENDER ASSESSMENT INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS can give consent on her behalf. While Art. 13 of the (Law No. 133/AN/05/5thL.) as amended by Law No. Family Code sets the minimum age of marriage at 18 221/AN/17/8thL, defines sexual harassment at the years old, marriage below 18 years old is subject to workplace as the act of repeatedly exerting behaviors consent of the husband and the wife’s legal guardians. with a sexual connotation against a person, which The World Bank’s Women, Business and the Law15 harms the person’s dignity, humiliates, or creates an Report for Djibouti scores 20 for the marriage index. intimidating, hostile or offensive situation against him/ This is a low score, which is explained by the fact that her. The employer shall take all necessary measures the wife is bound by law to obey her husband and to prevent acts of sexual harassment and punish them cannot be the head of the family or household in the (Art. 4 of the Labor Code). Employees who carried out same way. than a man (Art. 31 of the 2002 Family Code). sexual harassment are liable to a disciplinary sanction In addition, women cannot obtain a divorce judgment (Ministry of Women and the Family, 2019). Art. 290 of under the same conditions as men and do not have the Labor Code also includes penalties for the coercion the same rights to remarry as men. Indeed, a woman of workers by threats or violence in the employment can ask for a divorce if she has a justification, such as context (UNDP 2018). While the Penal Code covers proof of injuries, and may have to pay damages to the some forms of sexual harassment through the offence husband, which is not the case for the husband (Family of “public insult” (Art. 430-431), comprehensive Code 2002, 39.1, 39.2 and 39.3). In addition, she must provisions on all forms of sexual harassment in both wait a minimum of three months after the divorce to public and private spaces are lacking in the current be able to remarry while the husband has no temporal legal framework. constraints (Arts. 42 and 43 of the 2002 Family Code). After a divorce, both parents have custody rights, and Djibouti’s legal system has not yet opted to classify the decision is ruled out by the judge based on the sexist crimes as feminicides. There are no specific child’s best interests. Nevertheless, a woman loses provisions in the Penal Code related to femicides custody of her children if she remarries unless it is in or so-called “honor crimes” (United Nations, 2019). the child’s best interests or if the new husband is within At the international level, Djibouti abstained in the the prohibited degree of relationship to the child (Art. adoption of United Nation General Assembly (UNGA) 70 and 71 of the 2002 Family Code). Similarly, Art. 31 Res 55/66 on the elimination of crimes against women of the Family Code states that the man is the head of committed in the name of honor.16 the family and that his wife must respect him and obey his requests (UNFPA, n.d). In IPV cases, there is also no The government of Djibouti has ratified major legal process to obtain a protection order from a court international and regional conventions on women’s and domestic violence is not criminalized. Moreover, rights and empowerment, which demonstrate under Sharia law applicable in Djibouti (which rules the government’s commitments to address GBV under the Family Code), a woman’s inheritance share and advance gender equality. These international is half that of a man’s. This is one reason why Djibouti treaties and conventions17 include the African Charter only scores 40 on assets in WBL. Other reasons include on Human and Peoples’ Rights (1981, Banjul Charter) the fact that surviving female and male spouses do not and the Protocol to the African Charter on Human have equal inheritance rights over assets and the law and Peoples’ Rights on the Rights of Women in does not provide for the valuation of non-monetary Africa (Maputo Protocol, 2003) seeking to eliminate contributions. Nevertheless, women and men have traditional harmful practices. In 1995, the country has equal ownership rights to any property as well as equal taken part of the Platform of Actions of Beijing, which administrative authority over assets during marriage. is the fourth world conference on women and that produced an important document of reference for The Labor Code has provisions on sexual advancing women’s rights. Djibouti is also a signatory harassment at the workplace. The Labor Code to the Convention on the Elimination of all forms of 15 https://wbl.worldbank.org/en/data/exploreeconomies/djibouti/2022 16  UN GA Res 55/66, 4 Dec. 2000, UN doc. A/RES/55/66. 17  More details about the ratification of international and regional frameworks can be found in Annex 1. 23 DJIBOUTI GENDER ASSESSMENT INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS Discrimination Against Women (CEDAW, originally gender equality on all levels, and ratified in 2015 the adopted in 1979 by the UN General Assembly) since 2030 SDGs from which the 5th goal tackles gender 1998, and has fully ratified it, unlike some other Arab equality. Moreover, the Djiboutian government states which made reservations upon ratification. adopted in 2018 the Francophonie Strategy from the That said, it has not signed the optional protocol XVII Francophone Summit which calls for promoting on the “International Convention on the Protection gender equality and ensuring that women and girls of the Rights of All Migrant Workers and Members are receiving their rights and are empowered. At the of their Families”. In 2003, the country adopted the regional level, Djibouti has been part of Chamikhat, Declaration of Cairo on the Elimination of FGM/C a regional network that conducts awareness and and the subsequent Declaration FGM/C +5 in 2008. advocacy campaigns in the sub-region about the More recently, Djibouti has adopted the “Agenda harmful impacts of FGM/C (Ministry of Women and 2063” of the African Union (2013) seeking to achieve the Family, 2019). BOX 1  Women, Business and the Law The Women, business and the law (WBL) index measures the laws and regulations that affect women’s economic opportunities. The global index is the average of 8 indicators that represent the milestones a woman might experience in her lifetime and affect her economic empowerment: mobility, workplace, pay, marriage, parenthood, entrepreneurship, assets, pension. Each indicator measures gender-based differences in the law and is scored over 100 – the highest possible score. Overall, Djibouti scores 68.1 on the 2021 WBL index, up from 53.1 in 2000. Compared to similar countries (Figure 1.2), this is higher than Yemen (26.9) and Somalia (46.9) but lower than Eritrea (69.4), Ethiopia (76.9), Cabo Verde and São Tomé and Príncipe (both at 86.3). Moreover, Djibouti is among the 10 countries that improved the most their global score since 2017. On mobility, workplace and entrepreneurship, Djibouti scores 100, meaning that there is no gender difference in the law in this area. For example, for mobility, a woman can choose to live where she wishes, travel outside her home, apply for a passport and travel outside the country the same way as a man does. Regarding the workplace, Djibouti’s score is 100, indicating that a woman can get a job the same way a man can, and that the law prohibits gender discrimination in the workplace and sexual harassment in employment with a criminal penalty. Finally, on entrepreneurship, the law in Djibouti prohibits any gender discrimination in credit access and ensures that a woman can sign a contract, register a business, and open a bank account in the same way a man can. However, on five other indicators, namely pension, parenthood, pay, assets and marriage, Djibouti fares less well. On pension, Djibouti scores 75, similar to Ethiopia and Cabo Verde, but lower than São Tomé and Príncipe. Although any absences from work due to childcare are accounted for in pension benefits, women and men do not retire with full pension benefits at the same age. On the dimension of parenthood, Djibouti scores 60, similar to Ethiopia and São Tomé and Príncipe, and higher than Somalia, Cabo Verde, Eritrea and Yemen. Indeed, the law prohibits the dismissal of pregnant workers and provides a paid leave for both mothers and fathers. However, there is no paid parental leave and the government does not administer all benefits of the maternity leave. 24 DJIBOUTI GENDER ASSESSMENT INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS BOX 1  continued Figure 1.2 Global index and indicators for Djibouti and comparable countries MOBILITY WORKPLACE PAY MARRIAGE PARENTHOOD ENTREPRENEUR ASSETS PENSION WBL 2022 SHIP SCORE 100 100 50 20 60 100 40 75 69.1 DJI 100 100 75 100 40 100 100 75 86.3 CPV 100 100 75 80 60 75 100 100 86.3 STP 100 100 75 60 20 75 100 25 69.4 ERI 100 100 25 80 60 75 100 75 76.9 ETH 75 50 50 20 40 75 40 25 46.9 SOM 25 25 25 0 0 75 40 25 25.9 YEM Source: Women, Business and the Law, 2021. Note: DJI = Djibouti, CPV = Capo Verde, STP = São Tomé and Príncipe, ERI = Eritrea, ETH = Ethiopia, SOM = Somalia, YEM = Yemen. In the domain of pay, Djibouti scores 50, similar to Somalia and higher than Ethiopia and Yemen, but lower than Cabo Verde, São Tomé and Príncipe, and Eritrea. Indeed, Djiboutian laws require equal pay for equally-valued work and women can work at night in the same way as men. Regarding the indicator on assets, Djibouti ranks the same as Somalia and Yemen but lower than all the other comparable countries and neighbors. In Djibouti, although women and men have equal ownership rights to any property as well as equal administrative authority over assets during marriage, daughters and sons do not have equal inheritance rights over their parents’ assets. Moreover, surviving female and male spouses do not have equal inheritance rights over assets and the law does not provide for the valuation of non-monetary contributions. Finally, the lowest score for Djibouti is the marriage indicator, at 20 as in Somalia. Even though this is higher than Yemen, it is lower than all the other comparable countries and neighbors. A wife has legal requirements to obey her husband and cannot be the head of household or family in the same way that a man can. In addition, the woman cannot obtain a divorce judgment under the same conditions as the man and does not have the same rights to remarry as a man. The only component in marriage in which Djibouti scores points is related to specific legislations addressing domestic violence. Nevertheless, these legislations offer protection for victims rather than criminalize the attacker in domestic violence situations. 25 DJIBOUTI GENDER ASSESSMENT INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS National strategies to promote gender Figure 1.3 Timeline of the national strategies for women’s situation in Djibouti equality Since the early 2000s, the Djiboutian government TIMELINE OF has adopted a series of strategies and policies that focus on or include specific targets for women’s NATIONAL STRATEGIES inclusion. First, the creation of the Ministerial Department for the Advancement of Women in 1999 – AND INSTITUTIONS transformed into the Ministry of Women and the Family 1991 - 1999 in 2016 - headed by a woman, gave the opportunity • Literacy campaigns • Creation of Ministerial for a woman to enter for the first time as a minister Department for Women into the government. Then an important first national (1999) 2002 strategy for the integration of women in development National Strategy for the was adopted in 2002, which goal is to decrease Integration of Women in Development social inequality and to promote women’s place and status in society. It includes four priority areas which 2005 are decision-making (with law enforcement against National Policy for illiteracy, sensitization in decision-making processes, Integrated Early Childhood and creating solidarity mechanisms), economic 2006 Development National Strategy for the participation (with creating opportunities for women, Abandonment of All Forms fighting employment discrimination, increasing of Excision access to economic resources, and capacity building), 2010 education (with decreasing inequalities between girls • National Literacy and Non-formal Education and boys, and fighting illiteracy), and health (with Strategy increasing access and use of water domestically, 2011 • National Strategy Plan improving nutrition for family and women, increasing National Gender Policy for Children in Djibouti access and use of safe motherhood program benefits). A decade later, the government adopted the 2014 National Gender Policy (2011-2021)18 seeking Vision “Djibouti 2035” to “make Djibouti an egalitarian society, without discrimination where women and men, having the 2015 Strategy for Accelebrated necessary skills and abilities, in all areas of social, Growth and Promotion of economic and political life, work equitably and in Employment perfect harmony for the development of the country”. 2016 • Economic Empowerment This national policy had two broad objectives - to and Strengthening of establish an environment in Djibouti that fosters Women's Communities 2017 • National Strategy to gender equality on the social, economic, political, • Literacy Program Accelerate the Total legal, cultural, and institutional levels; to integrate • National Action Plan on Abandonment of FGM/C Resolution 1325 gender into developmental goals of all Djiboutian • Project to Support Young activity sectors - as well as five strategic orientations. Girls and Women in The first focuses on raising gender awareness at the precarious situation 2018 household and community levels, while the second • Youth Entrepreneurship Project seeks to ensure women, men, and adolescents • Social Action Center for 2019 Women Empowerment National Community 18  It should be noted that the Ministry of Women and the Family Development Strategy is in the process of finalizing the new national gender policy for the next ten years. This should be adopted and implemented in 2023. 26 DJIBOUTI GENDER ASSESSMENT INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS have equitable access to basic social services. The illiteracy campaigns were launched between 1991 third strategic orientation focuses on promoting the and 1999 and reached more than 7,500 women. potentials and access to resources of both women and Then, the government developed in 2010 the national men equally, in addition to increasing the access of literacy and non-formal education strategy which women to factors of production and financial resources. included six main objectives such as increasing The fourth orientation seeks to reinforce equitable the basic knowledge of population in education, participation of women and men in decision-making health, hygiene, nutrition, planning and family well- bodies, whether managerial, political, or economic, being, boosting the confidence of the population in and the fifth focuses on building institutional capacity their abilities, and empowering adults. In 2017, the for the implementation of the policy. Taking this Ministry launched a national literacy program which gender policy as a reference framework on gender, targeted 3,471 women in 104 classes across Djibouti, the “Djibouti 2035” vision seeks to set policies and and comprised literacy classes in national language. strategies to promote sustainable development Djibouti has also adopted a series of policies and in Djibouti. Specifically, its goals on gender are to initiatives aiming at encouraging education of achieve gender equity and equality, and closing the children, and particularly young girls. These include gaps in school enrollment, illiteracy and integration in (i) the establishment of compulsory schooling for all development sectors as well as improving the status children aged 6-16 who are Djiboutians or residents of of women. It encompasses two quantitative goals to the republic, (ii) the organization of several awareness be reached by 2035: having 35 percent of women be campaigns to parents and mothers in efforts of economically independent (however, the government eliminating prejudices against females in education, strategy does not define the term concretely or how (iii) the distribution of food for rural household families to measure it), and 40 percent of women be members who send their daughters to school, and (iv) the of the National Assembly. The 2022-2026 Policy is granting of scholarships to young girls. currently being finalized and will be launched in early 2023. On economic empowerment, the government first established the Stratégie de Croissance Accélérée Three national strategies were specifically et de Promotion de l’Emploi19 (SCAPE) 2015-2019. established for the welfare of Djiboutian children, Overall, the strategy aims to develop inclusive and and by extension, girls. The national policy for sustainable economic growth through its policies on integrated early childhood development adopted in employment, but this strategy specifically focuses on 2005 promoted birth certificate registration of children the integration of women in professional domains, as and aimed to allow all Djiboutian children to be in well as the development of female entrepreneurship. good health, have access to hygiene, be awakened and In 2016, the Ministry and U.S. Agency for International psychologically stimulated. As one of the results, two Development (USAID) jointly created an initiative called daycare centers were created to give disadvantaged the “Economic empowerment and strengthening children opportunities to prepare for learning. In of women’s communities” seeking to strengthen 2007, the national strategic framework for the care of vulnerable women’s economic empowerment by orphans and vulnerable children was launched. And in creating income-generating activities for them. Thus, 2010, Djibouti established the national strategic plan 300 women were granted equipment which could for children (2011-2025) aiming to protect children by help start such activities and 13 work cooperatives ensuring that they have access to their rights as well as were established in 2018 across different regions in to basic social services. More specifically, it focuses on Djibouti. Beside the Youth Entrepreneurship Project four key aspects: survival, development, protection, (2018) whose goal was to provide training sessions on and participation. entrepreneurship, a Social Action Center for Women Empowerment was created in 2018 to provide In order to improve the access and level of vocational training to young girls who left school education in the country, successive waves of early. Finally, a multi-sector program was launched in 19  Strategy for Accelerated Growth and Promotion of Employment in English. 27 DJIBOUTI GENDER ASSESSMENT INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS 2017 to support young girls and women in precarious social media, national media and community-based situations and make them less vulnerable socially structures. Other interventions have included a and economically. Through this project, 220 young GBV toll free number at the police level to support women started income-generating activities and 15 reporting and case management, and a COVID-19 girls received training. specific Sexual Exploitation and Abuse Prevention (PSEA) inter-agency action plan. In 1997, FGM/C was declared a public health priority in Djibouti. Ten years later, the national Within the Ministry of Women and the Family, strategy for the abandonment of all forms of the Gender Observatory has been working for excision 2007-2011 (2006) was launched. Based several years to provide information and data on four strategic axes, it was aimed to improve on the situation of women in Djibouti in order to knowledge on FGM/C by setting up a database support decision-making in terms of strategies and develop institutional mechanisms to ensure and public policies towards gender equality. Thus, the implementation of the strategy. Moreover, the several studies have been carried out in recent years strategy called for advocacy and sensitization toward such as “The evolution of the situation of women in target groups, as well as supporting girls subjected Djibouti 2000-2018” which establishes an inventory of to FGM/C through listening, advice, legal and the situation of women between 2000 and 2015 in the judicial assistance, and psychosocial aid. In 2016, the country. The report “The current situation of women’s Ministry of Women and the Family, along with support access to credit in Djibouti” makes it possible to learn from UNICEF, developed a new national strategy about women’s access to credit in financial institutions, seeking to accelerate the abandonment of FGM/C determine the obstacles they face and identify the between 2018 and 2022. Following the first strategy institutional measures that could make it possible from 2006, this one seeks to increase engagement to overcome these obstacles. Other reports such as of influential political decision makers and leaders, “Women and Diplomacy”, “The Representation of as well as reinforce community involvement and Women in National Defense and Public Security in improve communication through media tools. The the Republic of Djibouti” and “The Representation strategy also sets up monitoring and evaluation of Women in the Civil Service” aim to draw up an systems of the abandonment of FGM/C. Finally, inventory of the situation of women in different Djibouti composed a national action plan (2018- spheres of society or public life. 2020) - based on the United Nations Security Council resolution 1325 - which calls for increasing women The Djiboutian civil society concerned with participation in decision making when it comes promoting gender equality comprises only a to settling and resolving conflicts, for prevention few non-governmental organizations (NGOs) of women and girls from gender-based violence, and organizations. First, the Union Nationale des particularly sexual violence and exploitation, and for Femmes de Djibouti20 (UNFD), established in 1977, protection and relief of women and children from aims to empowering women, protecting the health sexual violence during or after armed conflict. of mothers and children, and eradicating poverty, illiteracy, and discrimination against women in all of To address the reduction of GBV services due to its forms. In order to do so, it targets leaders to ensure the repercussion of the COVID-19 pandemic, the some laws favorable to women are being passed, and government of Djibouti and development partners engages in social protection programs, such as social implemented GBV emergency interventions that assistance and insurance labor market programs, and focused on community mobilization, advocacy, microfinance programs. While UNFD takes a more and delivery of essential services (United Nations prominent role in civil society in relation to gender Djibouti, 2020). Key measures included sensitization issues, other smaller and regional associations include on protection from GBV and child protection through the Association for the Integrated Development of 20  National Union of Djiboutian Women in English. 28 DJIBOUTI GENDER ASSESSMENT INSTITUTIONAL, LEGAL, AND REGULATORY SYSTEMS Mabla21 which focuses on the well-being, health, activities, 13 work cooperatives were established, 220 nutrition, and education of mothers and children in the young women started income-generating activities region of Mabla, “Solidarité Féminine”22 which works and 15 girls received training. While these outcomes to fight against HIV/AIDS and offers psychosocial are positive, the needs are much greater and the aid to women and their families since 1995, and the durability and sustainability of these results remains Association of Women in Tadjourah23 which focuses to be assessed. Moreover, existing data suggest that on the professional development of women, and women in self-employment are often in precarious their empowerment, as well as increasing literacy economic conditions. In terms of GBV, a national and fighting against FGM/C in the region. Similar strategy launched in 2016 seeks to accelerate the associations exist in the southern regions of the abandonment of FGM/C between 2018 and 2022 and country, such as the Ali Sabieh women’s association, a national action plan (2018-2020) was developed to the Arta women’s association for social development, fight against all forms of GBV. and the Ali Adde regional association. Compared to other countries in the region, the civil society in While the GBV is a very sensitive topic and changes Djibouti that addresses gender issues and advocates in this domain take time, they are also a function for women’s rights and empowerment remains of cultural norms and customs, and more needs to limited. To strengthen the civil society in Djibouti, the be done in this domain, especially in rural areas. Civil Society Organization Strengthening Project24 Indeed, the prevalence of FGM/C and GBV in general changed its strategy to support a single group of 12 remains very high and common in the Djibouti society, associations (four of which are involved in gender especially outside the capital. Thus, sensibilization equality and social inclusion) in order to directly campaigns should be intensified and concentrated, implement activities, including financial support and especially in rural areas. For example, greater protection, so that the entire system of civil society familiarization of the law and buy-in from community organizations becomes more coherent, vibrant, and leaders and female elders is essential, especially as sustainable. the evidence suggests that only a few cases of FGM/C have been brought to court while the prevalence rate While the national strategies show a strong will remained at 71 percent in 201925 and that the practice to improve women’s situation, more efforts must of FGM/C has been deemed illegal since 1995. Finally, be deployed to achieve tangible results towards it will be important to study in depth the evaluation26 closing gender gaps, and to meet the SDGs and of the National Gender Policy (2011-2021) which national development plan goals. For example, in aimed to “make Djibouti an egalitarian society” in the area of education, the national literacy program order to extract tangible results, lessons and actions launched in 2017 targeted 3,471 women in 104 classes for the future.27 Access to basic education and health across Djibouti. However, at the national level, these services has strongly improved over time for both men outcomes need to be more ambitious given there and women, but women appear to still be lagged are still more than half of women who are not able to in reaching higher education levels, professional read and write (the literacy rate of women aged 15 training and economic opportunities. To this end, and more is 43 percent). In the domain of economic investing in quality data that is representative of opportunities, four strategies have been developed women, and targeted around programs, projects, and since 2015 seeking to provide training and strengthen national plans is crucial to monitor, learn, and evaluate women’s economic empowerment. To that end, strategies, and tailor them based on the evidence. 300 women were granted equipment for starting 21 https://pdf.usaid.gov/pdf_docs/PA00XK1D.pdf 22 https://wfp-africa.medium.com/helping-vulnerable-women-and-children-in-djibouti-through-covid-and-by-changing-their-lives- 606b9f0600ff 23 https://www.developmentaid.org/#!/organizations/view/4038 24 https://pdf.usaid.gov/pdf_docs/PA00XK1D.pdf 25  Across all ages (0-), the rate is 71 percent. For women aged 15-49, the rate is 87.5 percent (INSD, 2020). 26  The evaluation of the 2011-2021 National Gender Policy is currently being finalized and should be available in the coming months. 27  And in particular the next national gender policy which is being developed. 29 2 DJIBOUTI GENDER ASSESSMENT CHAPTER ENDOWMENTS Photo credit: Shutterstock 30 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS Education above have ever attended school compared to 56 I percent in Djibouti-ville. However, the gender gap is n Djibouti, 59 percent of individuals aged 628 the highest in the capital with 72 percent of men having and over have ever attended school29, with a ever attended school (16 p.p difference). Notably, lower rate for women (51 percent) than for men more than half of the women aged 6 and more from (67 percent). This is a small increase compared to Djibouti-ville who have never been to school cite their 2012 where 57.5 percent of the population had ever gender as the main reason whereas in other regions, attended school, with 49 percent of women and 66 this reason is reported for less than 20 percent of percent of men. Moreover, the increase was slightly women. Compared to 2012, the proportion of women higher for women than men and the gender gap has who ever attended school has decreased in Tadjourah narrowed a bit. Rural women are much less likely to have (from 31 percent to 27 percent) and increased in ever attended school than urban women (23 percent Djibouti-ville (from 55 percent to 56 percent). compared to 55 percent), and this gap increases across generations. For instance, among 6-11 females, Access and demand for education is improving over attendance rate is 82 percent in urban and 54 percent time as younger cohorts have a higher propensity in rural areas, and among 19-24, these are 28 percent to attend school than older ones (Figure 2.1). and 5 percent respectively. However, the gender gap Among the 10-14 years old, 87 percent of girls have is higher for urban residents (16.5 p.p. difference) than ever been to school whereas among the women 60 rural ones (11 p.p. difference). Moreover, the data and over, only 8 percent had the opportunity to go to suggests that rural women have poorer access to school. The gender gap narrows across generations: school because of their gender. The most cited reason the difference between men and women is 3 p.p. for never having been to school among rural women among the 10-14 years old while it is 26 p.p. among is being a girl (30 percent) whereas it is the third-top the 25-39. Moreover, the reason for females not cited reason for urban women (15 percent).30 attending school because they are girls is lower for those aged 6-14 (12 percent) compared to those aged The proportion of women having ever attended 60+ (21 percent). Thus, the reason for not attending school varies across regions (Figures 2.1). In school has less to do with women being “girls” in the Tadjourah, only 27 percent of women aged 6 and younger generation than in the older one. 28  According to the New Education Law in Djibouti (enacted in 2000), the age of starting primary school in Djibouti is 6 years old. Cf Box 3 in page 4 of the following document for reference https://documents1.worldbank.org/curated/en/263181468748479967/pdf/ multi0page.pdf 29  According to authors’ calculations from the household survey EDAM 2017. 30  According to authors’ calculations from the household survey EDAM 2017. 31 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS Figure 2.1 Ever attended school by region and age groups Ever attended school Ever attended school Among the 6+ Among the 6+ 100 100 90 87 84 80 80 80 75 75 72 66 60 60 56 60 47 50 % % 44 38 40 40 40 40 34 37 40 27 30 25 20 20 16 8 0 0 Djibouti Ali− Dikhil Tadjourah Obock Arta 6−9 10−14 15−24 25−39 40−59 60+ ville Sabieh Male Female Male Female Source: Authors’ calculations based on EDAM 2017. Figure 2.2 Literacy rate by poverty status and age groups Literacy rate Literacy rate Among the 15+ Among the 15+ 100 100 85 80 80 74 69 68 60 60 47 % % 41 42 40 38 40 26 28 20 20 18 10 0 Non−poor Poor 0 15−24 25−39 40−59 60+ Male Female Male Female Source: Authors’ calculations based on EDAM 2017. In 2017, 43 percent of women aged 15 and over seekers population know how to read and write know how to read and write compared to 63 while it is only 30 percent among women. There are percent of men. It is an increase for both compared however high disparities across location as women’s to 2012 where literacy rates were 40 percent for literacy rate is 24 percent in the refugee village of Holl women and 60.5 percent for men. The literacy rate of Holl compared to 44 percent in Markazi. Discrepancy the Djiboutian is higher than among the population in Djiboutian women’s literacy rate is high between of refugees and asylum seekers of the same age.31 the poor and non-poor (26 percent compared to Indeed, half of 15+ men from the refugees and asylum 47 percent, respectively) (Figure 2.2). However, the 31  According to the « Rapport d’enquête profilage dans les villages de réfugiés 2019 » from the Ministry of Social Affairs and Solidarity (MASS), INSD, World Food Programme (WFP) and United Nations High Commissioner for Refugees (UNHCR). 32 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS gender gap is higher among the non-poor people completed at least primary education (25 percent than the poor (22 p.p. difference compared to 12 p.p., compared to 34 percent) and secondary education respectively). In addition, there are strong differences (13 percent compared to 23 percent). Compared to in literacy rates of women across region. Only 16 2012, the proportion of women with at least a primary percent of Tadjourah’s women aged 15 and above education has not changed but the proportion of are able to read and write while it is the case for 50 women who completed at least secondary education percent of those living in Djibouti-ville. Compared to has increased (from 8 percent to 13 percent). Almost 2012, women’s literacy rate has increased in Ali-Sabieh no women from rural areas have completed secondary (from 28 percent to 32 percent) but decreased in education and only 8 percent have completed primary Tadjourah (from 20 percent to 16 percent). Consistently education, whereas these rates are 15 percent and 28 with the trend in attendance rates, the proportion of percent respectively in urban areas (Figure 2.3). Thus, literate women decreases across age groups (Figure women living in rural areas have way less access to 2.2). While 75 percent of the 15–24-year-old women school – and higher levels of schooling – than the ones are able to read and write, this is the case for only from urban areas. However, access to higher level of 10 percent of the 60+. Moreover, the gender gap schooling is improving across generation (Figure in literacy increases across generation, with 11 p.p. 2.3). Indeed, 46 percent of women aged 15-24 have of difference among the 15-24 compared to 27 p.p. completed primary education as highest educational among the 25-39. attainment while they are 8 percent among the 60 and older. Similarly, 23 percent of 15-24 years old In Djibouti, 49 percent of the population aged 15 women have completed at least secondary education and more have no education. While 58 percent of compared to 3 percent among the 40-59 years old. women have no education at all, it is the case for 39 percent of men. Compared to 2012, the proportion Moreover, there are strong disparities in the of women without education slightly decreased (60 proportion of non-educated individuals across percent in 2012). And according to the “Report of locations (Figure 2.4). The proportion of women Profiling Survey in Refugee Villages” (2019), the refugee with no education is the highest in Tadjourah, Obock population aged 15 and over seems to be less likely and Dikhil (83, 81 and 80.5 percent, respectively) than the native to have no education for both males while the lowest is in the capital. Compared to 2012, and females (17 percent and 33 percent, respectively). the proportion of women without education has National women are much less likely than men to have decreased in all regions except Dikhil and Tadjourah. Figure 2.3 Educational attainment by area and age groups Educational attainment Educational attainment Among the 15+ Among the 15+ 100 1 100 0 25 15 3 8 27 23 31 15 12 3 7 8 15 11 14 0 3 25 80 28 87 80 21 2 92 4 3 37 78 46 84 52 76 30 3 60 60 4 4 60 60 % % 53 40 40 4 5 33 7 34 20 20 25 5 16 0 0 Male Female Male Female Male Female Male Female Male Female Male Female Urban Rural 15−24 25−39 40−59 60+ No education Less than primary No education Less than primary Completed primary Completed secondary Completed primary Completed secondary Source: Authors’ calculations based on EDAM 2017. 33 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS Figure 2.4 Proportion of non-educated by region the 4th district – the poorest district according to the Poverty and Equity Assessment – are the least likely Percentage of non−educated by region to be literate (43 percent) compared to the other Among the 15+ districts (Figure 2.6). While in all districts women are 100 less likely than men to be literate, the gender gap is 90 the highest in the 1st, 2nd and 3rd districts (22 to 24 80 70 p.p.), the less poor districts. Indeed, in these districts, 60 more than 75 percent of men are able to read and 50 write whereas it is slightly more than 50 percent for % 40 women. Regarding the educational attainment of 30 the population aged 15 and over, 20 percent of the 20 women in the 1st and 3rd districts have completed 10 0 at least secondary school while only 11 percent of Djibouti Ali Dikhil Tadjourah Obock Arta women from the 4th district did (Figure 2.6). As for ville Sabieh the literacy rate, the highest gender gap is among the Female Male population of the 2nd and 3rd districts (13 and 14 p.p. of difference, respectively). Source: Authors’ calculations based on EDAM 2017. In general, Djiboutian girls are less likely to go to However, the gender gap is the highest in the capital school than boys (Figure 2.7).33 The proportion of and the lowest in Tadjourah. Moreover, the majority of girls among students in 2019 is around 46 percent in women (87 percent) have no education in rural areas, public primary, 45 in public middle schools34 and 47 compared to about half (52 percent) in urban areas. in public secondary schools35. Compared to 2017, there is a 1 p.p. increase at the secondary level, but Women are less likely than men to experience no change at the lower level of education. There are upward mobility32 in educational outcomes with important differences between rural and urban areas, respect to their fathers’ (26 percent compared to with a higher proportion of girls at schools among 46 percent). Among the poor population, the large urban areas than rural (46 percent compared to 39 majority of women (87 percent) do not experience percent in the middle schools for example). Regional any educational mobility. There are more women, in variations are also present, especially at higher levels non-poor households, that experience both upward of education. At the primary level, girls represent at and downward mobility in education, than those least 40 percent of students in all the regions, with the in poor households most of whom experience no highest proportion in the capital (47 percent) and the mobility at all (Figure 2.5). Among both poor and non- lowest in Dikhil (42 percent). Regional differences are poor households, men are more likely to experience strongest at the secondary level. The proportion of upward mobility than women. Across the different girls among secondary students is around 48 percent age group (Figure 2.5), the younger women are much in the two richest regions of the country (Djibouti- more likely than the older ones to experience upward ville and Ali-Sabieh) and even 55 percent in Obock, educational mobility (38 percent compared to 8 while girls only represent 35 percent of the students in percent, respectively), indicating an improvement in Dikhil. In this poor region, girls have much less access education mobility across generations. to secondary education than boys. However, this is an improvement compared to 2017 where only 32 Inside the capital, educational outcomes vary percent of girls attended secondary schools in Dikhil. across districts. Djiboutians living in the capital There was also an important increase of proportion of represent around 80 percent of the whole population girls in secondary in Obock (from 46 percent in 2017 to and have quite diverse education outcomes across 55 percent in 2019) and Tadjourah (from 38 percent in the various districts. Women aged 15 and over from 2017 to 46 percent in 2019). 32  Educational mobility is characterized by downward mobility if the educational level of the child is lower than that of the parent in question, upward mobility if it is higher, and no change if it is the same. 33  According to authors’ calculations from Annuaire Statistique Éducation 2019/2020. 34  In Djibouti, middle schools correspond to lower secondary level of education and are called “collèges” in French. 35  In Djibouti, secondary schools correspond to upper secondary level of education and are called “lycées” in French. 34 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS Figure 2.5 Educational mobility by poverty status and age groups Educational mobility compared to father Educational mobility compared to father Among the 25+ Among the 25+ 100 100 53 29 19 10 59 36 37 15 23 8 87 87 80 80 79 78 72 65 60 60 59 59 % % 40 44 40 38 20 20 0 4 6 3 3 0 3 5 4 6 5 5 Male Female Male Female Male Female Male Female Male Female Non−poor Poor 25−39 40−59 60+ Downward mobility No change Upward mobility Source : Authors’ calculations based on EDAM 2017. Figure 2.6 Literacy rate and educational attainment in Djibouti-ville Literacy rate Educational attainment Among the 15+ Among the 15+ 100 100 30 20 31 17 33 20 20 11 26 18 79 77 28 80 75 80 28 27 31 26 36 66 34 62 43 38 38 60 57 60 51 53 52 4 4 6 58 5 % % 43 3 51 40 40 48 4 50 46 4 40 36 4 6 5 20 20 27 22 24 0 0 1st dist 2nd dist 3rd dist 4th dist 5th dist Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. 1st dist 2nd dist 3rd dist 4th dist 5th dist Male Female No education Less than primary Completed primary Completed secondary Source: Authors’ calculations based on EDAM 2017. Figure 2.7 Percentage of girls among students in public schools by location in 2019 100 90 80 70 60 55 50 47 46 47 45 45 45 44 46 45 47 45 43 47 48 46 48 46 43 42 39 42 43 38 40 35 30 20 10 0 Primary Middle Secondary Urban Rural All Djibouti-ville Arta Ali-Sabieh Dikhil Obock Tadjourah Source: Annuaire Statistique Éducation 2019/2020. Note: Disaggregation rural/urban is not available at secondary level as the only location of secondary schools (lycée) is in urban areas (most of the time the capital city of the region). At the country level, there are 15 secondary schools, with 6 in Djibouti-ville. In Arta, Obock, and Tadjourah, there is only one secondary school for the whole region. 35 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS BOX 2  Changes over time in education outcomes Between 2012 and 2017, the proportion of women aged 6 and above who ever attended school has slightly increased from 49 percent to 51 percent. The gender gap is slightly smaller (of 1 point) as the proportion of men who ever attended school has increased less (from 66 percent in 2012 to 67 percent in 2017) than it did for women. Looking by region, the highest increase for women (with 2 p.p.) between 2012 and 2017 is in Ali-Sabieh and Obock while the increase was 1 p.p. in Djibouti-ville and Dikhil (Figure 2.8). However, the situation has not changed in Arta and has got worse in Tadjourah, where the proportion of women who ever attended school has decreased by 4 p.p. between 2012 and 2017. That said, the gender gap has decreased the most in Tadjourah (from 14 to 11 points), as the percentage of men who ever attended school has also decreased. In fact, the gender gap is closing in all the regions except in Arta (where the gender gap has increased from 11 to 13 points) and in Ali- Sabieh (where it did not change). Figure 2.8 Attendance and literacy rate by region and age groups Ever attended school Literacy rate Among the 6+ Among the 15+ 100 100 85 81 80 72 72 80 74 66 68 63 58 60 60 55 56 60 51 47 48 50 % % 42 44 45 44 39 40 40 41 42 40 38 37 37 33 34 40 31 27 28 30 28 22 19 18 20 20 10 4 0 0 2012 2017 2012 2017 2012 2017 2012 2017 2012 2017 2012 2017 2012 2017 2012 2017 2012 2017 2012 2017 Djibouti Ali Dikhil Tadjourah Obock Arta 15−24 25−39 40−59 60+ ville Sabieh Male Female Male Female Source: Authors’ calculations based on EDAM 2012 and EDAM 2017. As for the attendance, the gender gap in the ability to read and write among the 15+ population has also closed since 2012 (Figure 2.8). Indeed, the increase in the literacy rate is twice as high for women (3 points) than men (1.5 points). However, the evolution differs by age group. The youth (15-24 years old) women are the ones for which the situation has the most improved as their literacy rate has increased by 8 points between 2012 and 2017. It is also among the 15-24 years old that the gender gap has closed the most (4 points), whereas the gender gap has increased by 4 points among the 25- 39 years old and has not changed among the oldest groups. The variation -between 2012 and 2017- of the proportion of girls currently attending school depends on the education level (Figure 2.9). The proportion of girls currently attending school has increased the most at the primary level (6-11 years old group), from 71 percent to 77 percent, while the increase is 2 points at the lower secondary level (12-15 years old). However, there is no change among the girls attending upper secondary level (16-18 years old) and even a decrease (of 3 points) at the tertiary level (19-24 years old). As the attendance of boys has also increased at the primary and lower secondary levels, the gender gap has not closed or even increased slightly (1 point of increase at the primary level), whereas it has decreased at higher levels of education. 36 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS BOX 2  continued Figure 2.9 Current attendance and educational attainment Currently attending school Educational attainment Among the 6−24 Among the 15+ 100 100 16 23 8 13 80 82 82 26 80 75 76 77 78 80 37 25 72 71 34 62 60 60 60 60 6 4 60 58 % % 40 38 40 7 4 31 28 25 40 39 20 20 0 0 2012 2017 2012 2017 2012 2017 2012 2017 Male Female Male Female 6−11 12−15 No education Less than primary 16−18 19−24 Completed primary Completed secondary Source: Authors’ calculations based on EDAM 2012 and EDAM 2017. The gender gap is also closing at low levels of educational attainment (Figure 2.9). The proportion of women aged 15 and over without education has decreased by 2 points against a 1 point decrease for men. Similarly, the gender gap has decreased from 11 to 9 points among the population whose highest level of education is primary school. However, the gender gap has increased at the highest level of education. While the proportion of women who completed at least secondary has increased from 8 to 13 percent between 2012 and 2017, the gender gap has increased from 8 to 10 points. Differences observed in enrollment to school Girls in rural areas are much less enrolled into public between girls and boys are confirmed by parity schools than in urban areas (0.63 compared to 0.84 at ratios (Table 2.1). Overall, the ratio of parity in 2019 the middle school level). Gender parity indicators in is the highest at the secondary level (0.91) and slightly rural areas decline with higher levels of education: the lower at the primary and middle levels (0.87 and parity ratio in rural areas is 0.74 at the primary level 0.84). However, there are striking differences across and 0.63 and the middle level (versus 0.88 and 0.84 public and private schools. Parity ratios are much in urban areas). Thus, girls living in rural areas have higher in private schools than in public schools at the less access to education and more difficulties to reach primary and middle levels. This suggests that families high education level than those from urban areas. who send their children to private schools are more In addition, gender differences are stronger in rural inclined to send both girls and boys to school, as areas than urban areas. compared to those who send their children to public schools. Moreover, differences in parity ratios between Looking at the parity ratio in the new registrations urban and rural areas are strong in public schools.36 suggests that these gender differences at school 36  According to the Annuaire Statistique d’Éducation 2019/2020, there are no private schools in rural areas. 37 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS Table 2.1  Gender parity ratio in public and private schools in 2019 All Public Private Urban Public Rural Public All 0.87 0.85 0.98 0.88 0.74 Primary New registrations 0.87 0.87 0.89 0.9 0.76 All 0.84 0.83 1.04 0.84 0.63 Middle New registrations 0.84 0.84 0.86 0.85 0.68 All 0.91 0.9 1.02 Secondary New registrations 0.91 0.91 0.75 Source: Annuaire Statistique Éducation 2019/2020. Note: Disaggregation rural/urban is not available at secondary level as the only location of secondary schools (lycée) is in urban areas (most of the time the capital city of the region). At the country level, there are 15 secondary schools, with 6 in Djibouti-ville. In Arta, Obock, and Tadjourah, there is only one secondary school for the whole region. are diminishing (in public schools only). While rural gap in women’s enrollment at the primary level, there is a very slight difference between new and but the urban/rural gap is higher (in favor of urban) at all registrations at the middle and secondary levels, the middle and secondary levels. the difference is higher at the primary level (Table 2.1). Thus, in 2019, new enrollments of students At higher education levels, the Ministry of Women were more favorable to girls at all levels of public and Family and the Direction de la Statistique et schools (compared to all registrations). Lara Ibarra des Études Démographiques38 (DISED) (2019)39 and Mendiratta (2019) found that gender differences observe an increase in female enrollment in in enrollment in primary and secondary education universities between 2006 and 2015. The number of declined between 2017 and 2018. Moreover, the women enrolled increased from 442 in 2006 to 5,131 in parity ratio of new registrations in public schools have 2015 and the proportion of female students increased increased at primary and secondary levels compared from 40 percent to 43 percent. However, the gender to 2017. They went from 0.83 to 0.87 at the primary difference in university courses is evident; women are level, and from 0.8 to 0.91 at the secondary level. more likely to pursue studies in technology, literature, languages, culture, communication and legal and Girls’ and boys’ net enrollment rates (NER) are economic sciences, while men are in the majority similar, highest at the primary level (97 percent), in industrial sciences, science and humanities. This and decrease with higher levels of education37 (78 confirms the idea that girls are pushed into more percent at the middle level for both girls and boys). traditionally “feminine” fields while boys are pushed The gender gap (in disfavor of girls) appears at into STEM fields. However, this does not only happen higher levels of education: boys’ and girls’ NER was at the tertiary level, as girls already constitute a minority 53 and 52 percent at the secondary level, and 18 and of students in the science stream in high school. 16 percent at the tertiary level. However, there is no statistically significative gender difference at any level In technical education, the number of girls’ of education. Moreover, there is no observable urban/ enrollment increased significantly between 2001 37  There are some discrepancies between the EDAM 2017 data on education such as the boys and girls NERs and the data from the Annuaire Statistique d’Éducation 2019/2020 such as the gender parity ratio. In addition of the year difference, this could come from the differences in the sample design. Indeed, the Annuaire Statistique d’Éducation include the whole population of children going to school in the country while the EDAM survey focuses on the national children only. 38  Directorate of Statistics and Demographic Studies in English 39 Report of the Ministry of Women and the Family in collaboration with the Directorate of Statistics and Demographic Studies published in 2019 entitled “Study on the evolution of the situation of women in Djibouti 2000 to 2018” using “Enquête Djiboutienne sur l’emploi, le secteur informel et la consommation des ménages” (EDESIC) 2015 data. 38 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS Figure 2.10 Attendance and stopping school age by area Currently attending school Average age of stopping school Among the 6−24 Among the 6−24 who dropped out of school 100 25 87 86 82 83 80 20 65 65 16 16 60 61 15 15 60 54 15 13 13 13 Age 12 13 % 43 11 11 11 40 10 9 33 30 8 8 28 7 19 20 5 7 5 0 0 Male Female Male Female Male Female Male Female Urban Rural Urban Rural 6−11 12−15 16−18 19−24 6−11 12−15 16−18 19−24 Source: Authors’ calculations based on EDAM 2017. and 2014, by 150 percent, while the number of gender gap among the urban population of children boys increased by 108 percent over the same aged 6-24 who dropped out of school.41 Among the period (Ministry of Women and Family and DISED, rural population however, girls aged 6-11 and 12-15 2019). Despite this, the proportion of girls enrolled are on average a year younger when dropping out of in technical and vocational education is still lower school. Moreover, rural girls aged 6-11 are on average than in the general population (47 percent in 2014). In 2 years younger when dropping out of school than the addition, data indicates that girls are typically found urban ones (9 years old in urban areas compared to 7 in business sectors such as accounting, secretarial years old in rural areas). Rural girls are twice more likely and administration, while boys are more likely to be to stop school because of marriage or pregnancy than in industrial and technical streams such as electrical urban ones.42 However, this is driven by the older girls: engineering, mechanics and refrigeration. 5 percent among the 19-24, 3 percent for those aged 16-18 and 0 percent for the 12-15. When looking at the proportion of children currently attending school by age group, the trend Males are more likely to enroll, attend, and remain in differs slightly. While 82 percent of boys aged 6-11 are school when compared to females due to patriarchal currently attending school in 2017, it is 77 percent of incentives. First, males are still generally considered the girls of the same age.40 This gives a slightly higher the future heads of households, and as such should gender gap at the primary level (5 p.p.) than at the be given priority in education for economic purposes middle level (4 p.p.) and the secondary level (2 p.p). in the future (Agence Française de Développement Moreover, differences are important between rural (AFD), 2014; USAID, 2004). Second, sending females and urban populations (Figures 2.10). While 43 percent to school induces higher opportunity costs for the of rural females aged 12-15 are currently attending household as women could instead raise children and school, 83 percent of the urban are at school. The engage in house and farm work. Third, socio-cultural rural/urban difference is even higher for the 16-18 age attitudes favor males over females in schools due to group with 46 p.p. of difference. There is almost no safety and security concerns (USAID, 2004). 40  According to authors’ calculations from EDAM 2017. 41  However, in its National Strategy for Social Protection 2018-2028, the MASS reports that girls experience more school dropout than boys. 42  According to authors’ calculations from EDAM 2017. 39 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS Among the refugees and asylum-seekers Although Djiboutian girls are less likely than boys population43 aged 5-14, differences of school to enroll in schools, they have on average better attendance are small at the country level (95 results at national exams (Figure 2.11). Indeed, in percent for girls compared to 96 percent for boys).44 2019, female students are more likely to pass exams However, there are disparities across location with the than their male counterparts at primary and middle lowest girls’ attendance rate in Holl-Holl (90 percent) education levels. Moreover, the gender gap is higher and the highest in Ali Addeh (97 percent) where there at the middle level, with 52 percent of success among is also no gender gap. The gender gap in school girls compared to 50 percent among boys. However, attendance of refugee children is the highest and the opposite is observed at the secondary level, with an in disfavor of girls in Holl Holl (4 p.p.) and Markazi acceptance rate of 53 percent among girls compared (3 p.p). The World Food Programme (WFP) and the to 57 percent among boys. This is quite different from United Nations High Commissioner for Refugees what was observed in 2017, as girls were performing (UNHCR) (2013) reports that primary education is better than boys at all levels of education – and even widespread in refugee camps Ali Addeh and Holl with a higher difference at the secondary level. While Holl, but secondary education is lacking. Indeed, the the success rate of girls at the secondary level has not total number of individuals attending refugee school changed between 2017 and 2019, the boys’ one has declines with successively higher levels of education.45 increased. Moreover, acceptance rates at exams have Precisely, while 2,587 individuals attended refugee strongly increased for primary education between 2017 primary schools, this number decreases to 962 at the and 2019 (from 74 percent to 87 percent among girls) middle school level and 303 at the secondary level, whereas they decreased for middle school students which is roughly an 88 percent decline in attendance. (from 85 percent to 52 percent among girls). According Furthermore, the percentage of girls among refugee to the data46, gender differences in performances, in children attending school declines with successively parity and educational results, are better in private higher levels of education: from 47 percent for schools than in public schools, revealing inequalities primary schools to 43 percent in middle schools and in the educational system. Finally, girls also tend to 31 percent in secondary schools. The same trend perform better in university exams as between 2006 occurs when considering refugee schools in Ali Sabieh and 2015, their admission rates have always been refugee camps (Ali Addeh and Holl Holl) specifically higher than those of boys (Ministry of Women and (47 percent, 42 percent, and 29 percent, respectively), Family and DISED, 2019). whereas the trend in Obock refugee camps (Markazi) is non-linear, peaking at the middle school level (48 There are also notable variations by region in percent, 57 percent, and 51 percent, respectively). acceptance rate at exams among the public Additionally, school drop-outs rates for girls are high schools (Figure 2.12). Ali-Sabieh is the region with (25 percent in Holl Holl and 16 percent in Ali Addeh) the highest success rate among girls at primary level due to household work engagements or limited access (99 percent) while only 56 percent of girls pass their to sanitary towels (WFP and UNHCR, 2013). In efforts exams in Obock. Besides, Obock is one of the few of controlling drop-outs, families are encouraged regions in which the gender gap is in disfavor of to enroll their girls by receiving vegetable oil if they girls at both primary and middle levels, and with a attend 4 out of 5 school days in a week. Further, special very high difference for the primary level (19 p.p. of summer courses are organized for girls who fail their difference). Girls’ acceptance rate at middle school year so that they do not have to repeat it (WFP and is the lowest in Djibouti-ville (48 percent) and the UNHCR, 2013). highest in Tadjourah (76 percent). 43  As of early 2019, Djibouti has been home for roughly 29,000 refugees, 44 percent of which are from Somalia, 36 percent from Ethiopia, and 17 percent from Yemen. The main refugee camps have been Holl Holl, Ali Addeh and Markazi, where refugees have been living in close proximity to nationals and have had amicable terms with them (Hamida et al., 2020). As a result, the two groups shared similar social and demographic characteristics. 44  According to the « Rapport d’enquête profilage dans les villages de réfugiés 2019 » from MASS, INSD, WFP and UNHCR. 45  According to the Annuaire Statistique Éducation 2019/2020. 46  Annuaire Statistique d’Éducation 2019/2020. 40 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS Figure 2.11 Acceptance rate at exams in public and private schools in 2019 (percent) 100 94 94 87 86 86 85 83 80 80 57 60 52 51 53 50 49 40 20 0 Girls Boys Girls Boys Girls Boys Primary (OTI) Middle (BEF) Secondary (Bac) All Public Private Source: Annuaire Statistique Éducation 2019/2020. Note: OTI, Objectifs Terminaux d’Intégration, is an exam for assessing knowledge at the end of primary school (it is not a selective criterion for entering into middle school). BEF, Brevet d’Études Fondamentales, is an exam taken at the end of middle school to access general secondary education. Bac is the exam taken at the end of secondary education. Dublic/private disaggregation is not available at the secondary level for girls and boys. Figure 2.12 Acceptance rate at exams in public schools by region (percent) 100 99 98 86 87 87 88 90 85 86 81 80 80 76 75 76 70 72 72 70 68 62 60 56 57 57 58 51 52 54 48 49 50 45 40 30 20 10 0 Girls Boys Girls Boys Primary (OTI) Middle (BEF) All Djibouti-ville Arta Ali-Sabieh Dikhil Tadjourah Obock Source: Annuaire Statistique Éducation 2019/2020. Note: OTI, Objectifs Terminaux d’Intégration, is an exam for assessing knowledge at the end of primary school (it is not a selective criterion for entering into middle school). BEF, Brevet d’Études Fondamentales, is an exam taken at the end of middle school to access general secondary education. Regional disaggregation is not available at the secondary education level. While school attendance appears to be relatively percent). Girls are more likely to work in Ali Addeh high among the refugees and asylum seekers (11 percent) and Holl Holl (9 percent), while in Markazi population, around 9 percent of children aged and Djibouti-ville the proportion is much lower, at 2 5-14 years old are working, with a slightly higher percent. This is much higher than among the national proportion among girls (9 percent) than boys (8 population where 0.2 percent of the 5-14 years 41 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS old children are reported to be working.47 Similar percent of the 20-24 got married before 18 and 6 to refugees, Djiboutian girls are more likely than percent before 15.49 Amongst female refugees living boys to work (0.3 percent compared to 0.1 percent, in Djibouti, the prevalence rates for child marriage respectively) with a higher proportion among the rural has been relatively low and girls with any kind of girls (0.55 percent) than the urban ones (0.2 percent). education have a 44 percent lower likelihood of being However, when looking at child work as a reason for married compared to their uneducated counterparts not going to school, boys are more likely than girls (Hamida et al., 2020). to have stopped school in order to be employed (1.6 percent compared to 0.8 percent among the 6-24 Benchmarking the situation of women and girls’ years old who dropped out of school). education in Djibouti to comparable countries, Djibouti fares better in some areas and lags behind Another reason for girls to stop studying is in others. In terms of girls’ primary school enrollment associated with early marriage and family (Figure 2.13), Djibouti stands behind Cabo Verde and obligations, where in 2017, 4 percent of Djiboutian São Tomé and Príncipe at 62 percent compared to women were married before 14 years old and 14 92 percent and 93 percent for the NER (72 percent percent before 18.48 Rural women are much more compared to 100 percent in Cabo Verde and 105 likely to get married before 18 years old than urban percent in São Tomé and Príncipe for the gross ones as 27 percent of rural women get married before enrollment rate (GER). The gap between Djibouti adulthood compared to 11 percent of urban ones. and comparable countries is even higher at the Moreover, getting married before 18 is much less secondary level (Figure 2.14): the GER is 48 percent in frequent among men (10 percent in rural areas and 3 Djibouti, 97 percent in Cabo Verde and 96 percent in percent in urban areas). The trend of early marriage São Tomé and Príncipe. Djibouti’s female enrollment among girls is still persistent among the youth as 26 rate in primary school is much lower than the MENA Figure 2.13 Girls at the primary education level in comparable countries (percent) 100 93 95 92 92 88 90 86 80 70 62 60 55 48 49 49 48 49 48 49 50 47 45 40 38 30 20 8 10 9 10 6 7 4 0 DJI CPV STP ERI MENA LOW MID UPP MID WORLD Girls' primary school enrollment (net) Proportion of girls in primary school Girls out of school (primary school age) Source: WDI 2017. Numbers are not available for Ethiopia, Somalia, and Yemen. Note: DJI = Djibouti, CPV = Cabo Verde, STP = São Tomé and Príncipe, ERI = Eritrea, MENA = Middle East and North Africa region, LOW MID = Lower middle income countries group, UPP MID = Upper middle income countries group. 47  Education numbers for refugee and asylum seeker population comes from the « Rapport d’enquête profilage dans les villages de réfugiés 2019 » from MASS, INSD, WFP and UNHCR, and data for the national population is from authors’ calculations based on EDAM 2017. 48  According to the EDAM 2017 data. 49  According to the WDI data, 1.3 percent of Djiboutian women aged 20-24 were first married by age 15 in 2017, a figure that is much lower than in Ethiopia (16 percent in 2011), Eritrea (13 percent in 2010) or Somalia (17 percent). This proportion increases at 5 percent when looking at marriage by age 18, but is still way behind neighboring countries (41 percent in Ethiopia and Eritrea). 42 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS Figure 2.14 Girls at the secondary education level in comparable countries (percent) 97 96 100 91 90 78 80 75 66 70 60 52 53 48 45 47 48 48 48 50 45 47 40 30 20 10 0 DJI CPV STP ERI MENA LOW MID UPP MID WORLD Girls' secondary school enrollment (gross) Proportion of girls in secondary school Source: WDI 2017. Numbers are not available for Ethiopia, Somalia and Yemen. Note: DJI = Djibouti, CPV = Cabo Verde, STP = São Tomé and Príncipe, ERI = Eritrea, MENA = Middle East and North Africa region, LOW MID = Lower middle income countries group, UPP MID = Upper middle income countries group. Girls’ secondary school gross enrollment rate is not available in Djibouti. average (92 percent for the NER), but it is higher than Cabo Verde (76 years) and São Tomé and Príncipe its neighboring country Eritrea (49 percent). The trend (73 years). Although Djibouti’s life-expectancy has is similar at the secondary level. The proportion of risen during the last 20 years, it remains more than 10 girls among pupils in primary education in Djibouti years lower than the average MENA region figures (76 is 46.5 percent, not too far from the MENA average, years), as well as the lower middle-income countries Cabo Verde, and São Tomé and Príncipe (48, 48 and (71 years) and the upper middle-income countries (79 49 percent, respectively) and higher than Eritrea (45 years). Further, the increase over time has been slower percent). However, at the secondary level, Djibouti (45 than in comparable economies, such as Cabo Verde percent) is behind its neighbor Eritrea (46.5 percent), and Belize (World Bank, 2018). Djibouti had one of the MENA average (47 percent), Cabo Verde (52 the highest ratios in the world for maternal mortality percent) and São Tomé and Príncipe (53 percent). and the highest in the region and across economically While Djibouti has a lower rate of out-of-school girls comparable nations: 730 out of every 100,000 live in age of primary school than its neighbor Eritrea (38 births (Martinelli and Olle-Goig, 2012). However, the percent versus 49 percent), it stands far behind the risk of maternal death in the country has steadily MENA average at 7 percent, Cabo Verde at 8 percent decreased, as it was 1 out of 140 in 2017 compared to and São Tomé and Príncipe at 6 percent. 1 out of 48 in 2000. This is much less than in Ethiopia (1:55) or Eritrea (1:46) but lags behind the average for the MENA region (1:570). Additionally, Kireyev Health50 (2017) estimated that 1 in 19 women are at maternal risk of dying in Djibouti. This is partially due to the The life expectancy at birth of women has increased fact that only about half of women receive any form from 58.6 years in 2000 to 69.4 years in 2020, which of postnatal care (World Bank, 2018). Furthermore, is 4 years higher than that of men. This is similar to approximately 9 percent of children die before the the life expectancy of women in neighboring countries age of 5, and among the extremely poor females, 9 such as Ethiopia and Eritrea (both at 68.5 years in 2020) percent die at childbirth (Martinelli and Olle-Goig, but lower than that of comparable countries such as 2012). 50  Health statistics and data are limited in Djibouti, especially as there are no recent DHS or MICS. Indeed, the latest MICS is from 2006. 43 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS Figure 2.15 Need and use of health services in the last 30 days per region Needed health services Used health services in the last 30 days. Among those in the last 30 days who needed health services in the last 30 days 60 100 50 80 40 60 60 60 55 53 52 30 48 49 % % 41 38 40 20 29 12 12 22 23 20 10 6 7 7 6 5 6 5 5 5 3 0 0 Djibouti Ali Dikhil Tadjourah Obock Arta Djibouti Ali Dikhil Tadjourah Obock Arta ville Sabieh ville Sabieh Male Female Male Female Source: Authors’ calculations based on EDAM 2017. At the national level, there is no observable and men respectively reported in June/July 2021 that difference in the access of health care between intend to get vaccinated. Therefore, women seem to men and women (Figures 2.15). In most regions, be more reluctant to take the COVID-19 vaccine as women and men are equally likely to report needing 52 percent do not intend to do so (compared to 36 health services in the past 30 days. Compared to men, percent of men). women are more likely to access health services in Djibouti-ville and Obock but less likely in all the other The prevalence of HIV among the 15-49 women has regions. In particular, in Tadjourah, women are much drastically decreased since the 2000s as it went from less likely than men to use health services when they 4.8 new infections per 1000 uninfected people52 in need it (29 percent compared to 41 percent). While 2000 to 0.15 per 1000 in 2020.53 The gender gap the non-poor females are more likely to need health has also closed from 1 point to 0.01 point during the services during the last 30 days than the poor ones (6 last 20 years. According to WFP (2019), the infection percent compared to 4 percent), they are also more rate is 1.5 percent for women compared to 1.1 percent likely to use them when needed (57 percent compared for men. Thus, women are still overrepresented among to 47 percent). However, the gender gap in the use of the persons aged 15 and over living with HIV as they health care is in favor of women and higher among the represent 56.5 percent of this population. Among the poor (8 p.p. of difference) than the non-poor (3 p.p. of youth as well, the prevalence of HIV has decreased difference). Moreover, women living in urban regions from 4.5 per 1000 uninfected women of 15-24 years are more likely than men to receive health services old in 2000 to 0.14 in 2020. Additionally, the gender when needed whereas the opposite is observed for gap among the youth decreased from 1.87 points to rural population. On average, women appear less 0.04 point these last 20 years. However, the results likely to be vaccinated against COVID-19 than men of the EDSF/PAPFAM data reported in the report (9 percent compared to 24 percent) but there is no by Ministry of Women and Family and DISED (2019) gender difference in the intention of getting the highlight a drop in the number of non-single women vaccine.51 Indeed, 31 percent and 32 percent of women declaring that they have a good knowledge of the 51  From the Knowledge, Attitudes and Practices (KAP) survey on Covid-19 and its vaccines in MENA region (June/July 2021) by WHO and UNICEF. 52  The prevalence of HIV is defined by the WDI as follow: number of new HIV infections among uninfected populations ages 15-49 expressed per 1,000 uninfected population in the year before the period. 53  Data from the WDI and gender statistics data from the World Bank. 44 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS disease of AIDS (79 percent in 2002 against 67 percent of 25 in Arta. The average age for pregnancy is also in 2012). This drop of more than 11 p.p. was about the higher among the non-poor than among the poor, so same in urban and rural areas. Beyond recognizing that poor women are younger when they are pregnant sexual intercourse as a form of HIV transmission, compared to their non-poor counterparts. This is the majority of these women were unaware of other confirmed by the fact that poor women have a higher modes of transmission, such as non-use of condoms probability to be pregnant at the time of the interview (75 percent), injections (62 percent), transmission from than the non-poor ones (3 percent compared to 2 mother-child (59 percent) and blood transfusions (59 percent). The fertility rate is estimated at around 3.1 percent). The percentage of misinformed people was children per women around 2018 (World Bank, 2018). particularly high in rural areas. Moreover, some women The adolescent (15-19) fertility rate has decreased from still believed that religion could cure AIDS. 33 births per 1000 women in 2000 to 18 in 2019. This is the lowest adolescent fertility rate of the neighboring According to l’Enquête Nationale de Prévalence countries (63 in Ethiopia and 49 in Eritrea) and even des Handicaps54 (ENPH) carried out in 2019 by the compared to the average of the MENA region (39 per National Agency for Persons with Disabilities and 1000 women of 15-19 years old). Among the pregnant the INSD, the national disability prevalence rate in or breastfeeding women, the non-poor are more likely general is 8.5 percent in Djibouti, with a slightly to have a pregnancy booklet, meaning that they are higher prevalence in women than men (9 percent medically followed. Regional variations are important versus 8 percent). The regions where the disability (Figures 2.16). While more than 9 women out of 10 prevalence rate is the highest are Ali Sabieh (13 living in Arta, Ali-Sabieh and Djibouti-ville have a percent) and Obock (11 percent), while the rate is the pregnancy booklet, it is the case of only 7 women out lowest in Dikhil (3.5 percent). The ENPH report (2019) of every 10 living in Dikhil. paints the socio-demographic portrait of people with disabilities in Djibouti. It turns out that women with Although rates of prenatal care and assisted disabilities have a higher probability of never having deliveries have increased, there are significant attended school (74 percent) than men with disabilities gaps between women.55 Between 2002 and 2012, (59 percent), as well as being illiterate (73 percent the percentage of ever-married pregnant women for women with disabilities compared to 57 percent (aged 15-49) who received skilled antenatal care at for men with disabilities). In addition, women with a least once increased significantly, from 77 percent to disability are more likely to have never been to school 88 percent. In rural areas, this percentage was much and to be illiterate than those without a disability. The lower than in urban areas both in 2002 (35 percent vast majority of disabled women aged 15 and over against 92 percent) and in 2012 (66 percent against are inactive (85 percent compared to 73 percent for 95 percent). It seems that the higher the educational men) and when they are active, half are unemployed (7 level of the pregnant woman, the more she requests percent). Thus, only 8 percent of active women living prenatal care and makes visits. In addition, living in with a disability work compared to 18 percent of men urban areas is correlated with higher use of prenatal in the same situation. Finally, women with a disability care than rural areas, and pregnant women there are are more likely to be elderly and widowed than those more likely to reach the total number of recommended without a disability (22 percent versus 9 percent). visits. The age at first pregnancy is similar for women Although contraceptive use has increased in living in rural and urban areas (24 years old) with Djibouti, rates are lower than in other MENA notable regional variation. For example, women countries and there are disparities at the national living in Dikhil, which is the poorest region, have their level56. Contraceptive use among married women first child around the age of 22 compared to the age more than doubled between 2002 and 2012, from 9 54  National Disability Prevalence Survey in English. 55  Ministry of Women and Family and DISED (2019). 56  Ministry of Women and Family and DISED (2019). 45 DJIBOUTI GENDER ASSESSMENT Figure 2.16 Possession of pregnancy and vaccination booklet by region Possession of pregnancy booklet Possession of vaccination booklet Among 12+ pregnant or breastfeeding women Among 0−5 years old children 100 93 93 100 91 85 82 80 73 80 75 72 67 63 62 60 60 57 43 42 42 % % 39 40 39 40 40 20 20 0 0 Djibouti Ali Dikhil Tadjourah Obock Arta Djibouti Ali Dikhil Tadjourah Obock Arta ville Sabieh ville Sabieh Male Female Source: Authors’ calculations based on EDAM 2017. percent to 19 percent. Although this is a significant However, girls in urban areas are much more likely to increase, it is still below the average for Sub-Saharan have a vaccination booklet than the ones living in rural Africa, which is 26 percent, as well as that for areas (70 percent compared to 46 percent). In terms North Africa and the Middle East (61 percent). The of mortality, in 2020, the number of infant deaths prevalence of modern methods of contraception among girls is 43 per 1000 live births compared to tripled during this period (6 percent to 18 percent) 51 among boys. For children under 5, mortality is and the use of traditional methods decreased at estimated at 51 deaths per 1000 live births among girls the same rate. Although contraceptive rates have compared to 61 per 1000 among boys. In addition, increased in both rural (0.4 percent to 10 percent) the mortality rate of girls has decreased over the and urban (12.5 percent to 21 percent) areas, the rate years. For instance, the mortality rate of infant girls remains significantly lower in rural areas. In addition, has decreased by 31 points between 2000 and 2020, the percentage of unmet needs for family planning and 43 points among children under 5. However, remains high at nearly 50 percent of all needs. according to World Bank data from WDI, female Married women who belong to the poorest class of infant mortality rate of Djiboutian girls remains one the population are more likely to have unmet needs of the highest of the comparable countries (41.5 per than other women. In addition, married women living 1000 live births in Yemen, 30 in Ethiopia, 25 in Eritrea) in rural areas are 1.6 times more likely to have unmet and is much higher than the average of the MENA needs for contraception to space births than those region (16 infant deaths per 1000 live births). Among living in cities. the girls under 5, Djibouti’s mortality rate is lower than the Somalian and Yemenite ones, but higher than in According to the - limited - existing data on health Eritrea, Ethiopia and the MENA average (20 per 1000 in the EDAM 2017 survey, girls do not seem to live births). suffer from gender difference in the follow up of vaccination during childhood and have a lower Girls aged 6 months to 5 years are less likely to mortality rate than boys. Indeed, 65 percent of be subject to stunted growth57 (20 percent of girls children aged 0 to 5 have a vaccination booklet and compared to 22 percent of boys) and there is no no difference is observed between girls and boys. gender difference in the proportion of severe forms 57  Stunted growth is defined as the height/age ratio (WHO, 2006). 46 DJIBOUTI GENDER ASSESSMENT (6 percent of children) according to the results of boys). Similarly, 8 percent of girls aged 6 months to the SMART survey carried out by the INSD58 for the 5 years suffer from acute malnutrition60 compared to Ministry of Health in 2019. Among those living in 12 percent for boys. Further, girls are twice less likely Djibouti-ville, girls from Balbala are much more subject than boys to suffer from severe acute malnutrition to stunted growth than girls from the other districts (21 (1.7 percent compared to 3.4 percent). According to percent compared to 13 percent) while in regions, the the WHO standards (2006), Djibouti’s severe acute proportion of girls with stunted growth is the highest malnutrition is considered a serious situation for girls (38 percent) in Obock while the lowest is in Ali Sabieh (between 1 and 1.9 percent) and an urgent one for boys (22 percent). Thus, World Health Organization (WHO) (more than 2 percent). Among the refugee children norms of ranking of stunted growth considers the aged 6 months to 5 years, the trend is similar to that of situation in Obock as serious (stunted growth between the national population. Girls are less likely than boys 30 and 39.9 percent) whereas the situation in the to suffer from stunted growth (25 percent compared richest districts of Djibouti-ville is considered normal to 30 percent), ponderous insufficiency (16 percent (proportion of stunted growth inferior to 20 percent). compared to 23 percent) and acute malnutrition (9 Girls from 6 months to 5 years old are also less likely to percent compared to 10 percent). But compared to be underweight59 (16 percent of girls compared to 18 the national population, refugee girls are more likely percent of boys). This is also the case for severe forms to suffer from stunted growth (20 percent compared of underweight (3 percent for girls and 4.5 percent for to 25 percent). 61 BOX 3  Access to health and education services There are no strong differences between households headed by women and men in distance to services and in likelihoods of using various modes of transportation to reach them (Figures 2.17). Rural households with a female at its head are less likely to be located at less than 1 km to a health center61 than the ones with a male head (34 percent compared to 39 percent, respectively) (Figure 2.18). Moreover, urban households with a female head are slightly more likely (16 percent) to be located at more than 5 km to a health center than the ones with a male head (12 percent). In general, rural households with a woman at its head are more likely than the ones headed by a man to use a car to go to health facilities (such as health center, pharmacy, maternity center, or hospital) and schools (primary and secondary). For example, 22 percent of rural female headed households use a car to go to the health center while the figure is 15 percent for rural male headed households (Figure 2.18). In urban areas, households with a female head are more likely (25 percent) than those with a male head (22 percent) to use public transportation to a health center. This appears to be consistent with the fact that in the law, Djiboutian women can choose where to live, apply for a passport, travel outside their home, and outside the country in the same way as men (WDI, 2020). 58  « Institut national de la statistique de Djibouti » in French. 59  Ponderous insufficiency or underweight is defined as the weight/age ratio (WHO, 2006). 60  Acute malnutrition is defined as the weight/height ratio (WHO, 2006). 61  The same trend is observed for all the amenities, therefore the analysis focuses here on health center as an example. 47 DJIBOUTI GENDER ASSESSMENT ENDOWMENTS BOX 3  continued Figure 2.17 Distance and transport to main services Distance to services Transport to services Among households Among households 100 100 5 6 7 7 8 9 10 11 4 4 7 8 21 23 35 34 32 34 40 38 10 10 31 31 11 13 16 16 18 19 10 11 18 17 21 22 5 5 80 44 40 38 36 80 1 1 44 46 51 50 9 9 83 84 45 41 43 40 1 2 20 17 21 19 60 60 69 66 15 16 19 19 % % 49 48 1 1 2 2 2 1 1 2 40 40 48 48 46 49 40 41 46 47 32 30 40 41 29 32 27 27 20 24 25 20 0 0 Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Heal. cent. Pharma. Mater. Hosp. Prim. sch. Sec. sch. Heal. cent. Pharma. Mater. Hosp. Prim. sch. Sec. sch. 0−1 Km 1−5 Km By foot By bicycle/moped 5−10 Km 10+ Km By car By public transportation Source: Authors’ calculations based on EDAM 2017. Figure 2.18 Distance and transport to health center by area Distance to a health center Transport to a health center Among households Among households 100 100 2 4 21 17 22 25 10 5 10 12 22 15 80 47 42 22 80 17 8 0 7 1 2 2 75 60 23 60 69 71 27 65 % % 40 42 40 40 39 34 20 20 0 0 Male Female Male Female Male Female Male Female Urban Rural Urban Rural 0−1 Km 1−5 Km By foot By bicycle/moped 5−10 Km 10+ Km By car By public transportation Source: Authors’ calculations based on EDAM 2017. 48 3 DJIBOUTI GENDER ASSESSMENT CHAPTER ECONOMIC OPPORTUNITIES Photo credit: Shutterstock 49 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES Overview of the labor market same extent, maintaining the gender gap over time. G While women are more likely to be inactive in 2017 ender differences in the labor market in than in 2012, they are less likely to be unemployed. In Djibouti are striking (Figures 3.1). More than particular, the proportion of discouraged unemployed half of women aged 15 and above are out of women had decreased from 21 percent to 13 percent, the labor force and not in education in 2017, suggesting that some women went from being in compared to a quarter of men in the same age group discouraged unemployment to being out of the labor who are in the same situation. While the proportion force and not in education. of unemployed discouraged women is similar to the men’s (13 and 12 percent, respectively), women are This report follows the national definitions62 of less likely to be unemployed and currently searching labor force participation, unemployment, and for a job. Moreover, women aged 15+ are less likely to employment, which differ from internationally be in a formal employment than in an informal one (4 agreed definitions such as those of the ILO. The percent compared to 7 percent), while among men, national definition excludes the search requirement 21 percent work in the formal sector and 15 percent from the definition of the labor force participation in the informal sector. Indeed, women are five times and is therefore defined as those available for work. less likely to work in the formal sector than men. The This effectively includes “discouraged unemployed” proportion of those who are out of the labor force but individuals in the labor force. Indeed, based on studying is rather similar among women and men (13 national definitions shown in Table 3.1, among women and 15 percent, respectively). in the age of 15 and above, 33 percent participate in the labor force, 63 percent are unemployed, and 37 Compared to 2012, the proportion of individuals percent are in employment, and among men of the out of the labor force and not in education has same age group, 61 percent participate in the labor increased substantially. This increase is higher force, 38 percent are unemployed, and 62 percent for women than for men (16 p.p. more for women are in employment (Table 3.2). The ILO defines the compared to 12 p.p. for men), leading to more than half labor force as the employed and unemployed (55 percent) of women not being active nor studying individuals, the latter of which includes only those who in 2017. The proportion of inactive and studying actively search for jobs and excludes those who are individuals has decreased in both groups to the “discouraged unemployed”. Based on international 62  See for example the National Development Plan (2020-2024) and associated statistics. This is consistent with World Bank report “Challenges to Inclusive Growth. A poverty and equity assessment of Djibouti (2019)”. 50 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES definitions, among women in the age of 15 and employment, and among men of the same age group, above, 18 percent participate in the labor force, 37 47 percent participate in the labor force, 22 percent percent are unemployed, and 63 percent are in are unemployed, and 88 percent are in employment. Figure 3.1 Summary of status of the active age population by gender and year Activity status among the 15+ in 2012 Activity status among the 15+ in 2017 Male Female Male Female 4% 15% 6% 7% 26% 9% 27% 21% 7% 39% 9% 19% 13% 15% 55% 16% 21% 15% 13% 13% 11% 11% 17% 12% Working (formal or public sector) Unemployed ILO Inactive and student Working (informal sector) Unemployed discouraged Inactive and not in education Source: Authors’ calculations based on EDAM 2017 and EDAM 2012. Note: The working category includes employment as well as self-employment. For those in self-employment, informality is defined as not complying with at least one of these conditions: having (i) accounting records, (ii) a trading license, (iii) a subscription to the chamber of commerce. For those in employment, informality is defined as not complying with at least one of these conditions: (i) having a contract, (ii) being subscribed to the national social security. Unemployed ILO represents individuals who are unemployed in the labor force; they are available to work and actively searching for a job. On the other hand, unemployed discouraged refers to individuals who are available to work, but not searching for a job. They are not considered by the ILO as being part of the labor force. See Tables 3.1.1 and 3.1.2 for more details. Table 3.1  National and ILO definition of labor market participation National definition ILO definition Active Inactive Active Inactive Employed YES NO YES NO Unemployed non-discouraged: available to work YES NO YES NO and searching for a job Unemployed discouraged: available to work and YES NO NO YES not searching for a job Neither employed nor unemployed NO YES NO YES Table 3.2  Comparison between national and ILO definitions on labor market outcomes (percent) National definition ILO definition All Male Female All Male Female Labor force participation among the 15+ 45 61 33 32 47 18 Unemployment among the active 15+ 47 38 63 27 22 37 Employment among the active 15+ 53 62 37 73 88 63 Source: Authors’ calculations based on EDAM 2017. 51 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES Characteristics of labor force percent). This could be explained by several reasons. participation First, those regions are among the most deprived areas and have the least access to electricity supply. Second, school enrollment, educational attainment, Djibouti suffers from a low labor force participation, and literacy rates are the lowest in these areas, with more than half of the working age (15 years especially for women. For Obock, this low school old and above) population out of the labor force. attendance could stem from the fact that schools in Thus, Djibouti experiences what is known as the MENA these areas are less accessible due to distance. Third, paradox: although there are some improvements the age at first marriage is lower in those areas and in education gender gaps, labor market outcomes the pregnancy rates are the highest of the country, do not reflect a parallel trend (International Fund for both of which could impact labor force participation. Agricultural Development (IFAD) and ILO – in MENA). Among women, it is only 32 percent of those in the The labor force participation rate does not follow working age who are engaged in the labor force, a linear trend with the years of education for both compared to 59 percent of men. Compared to 201263, females and males (Figure 3.3). Specifically, the the labor force participation has strongly decreased trend is generally increasing between 0 and 3 years for both women and men (45 percent and 65 percent, of education, decreasing from 3 to 11 years and then respectively). And compared to 199664, women’s labor increasing again from 11 years of education. While 45 force participation has not changed (33 percent) and percent of women with secondary education or more the men’s one has slightly decreased from 62 percent. participate in the labor force, it is the case for only 28 As the decrease has been higher for women than percent of the uneducated. Further, the gender gap men, the gender gap has increased from 20 p.p. to 27 narrows at higher education levels. It is worth noting p.p. between 2012 and 2017. Urban women are much that the majority of women (60 percent) have no more likely than rural ones to participate in the labor education, 37 percent have 5+ years of education, 21 force (33 percent compared to 24 percent) but the percent have 10+ years of education, and 0.6 percent gender gap is the same among the rural population have 15 years of education. The participation in the and the urban one (Figure 3.2). Further, the difference labor force also varies by age groups (Figure 3.3). The is quite lower between poor and non-poor women 25-39 years old women have the highest participation as 28 percent of poor women are active in the labor rate with 39 percent, compared to the 15-24 years market compared to 32 percent among the non- old with 26 percent of participation in the labor poor. However, the gender gap is higher among the force. The more children there are in the households, non-poor people than the poor (29 p.p. of difference the less likely a woman is to work. While 38 percent compared to 20 p.p. respectively). of the women without any child (0-18 years old) in the household participate in the labor force, it is 26 Women living in Obock are the least likely to percent for those with at least 4 children. Moreover, engage in the labor force (18 percent), while women with a child aged 0-6 years old (30 percent) are those from the capital have the highest probability less likely to participate in the labor force than those to do so (34 percent). It is particularly striking that with a child aged 0-15 years old (33 percent) or 0-18 in 2012, women’s labor force participation rate was years old (36 percent). 41 and 42 percent in Tadjourah and Obock, while it has been divided in half between 2012 and 2017. The analysis shows that more educated women Thus, approximately four out of every five women have a higher likelihood to participate in the labor in Tadjourah and Obock are out of the labor force, market65; and this relationship appears to be stronger and the labor force participation is also the lowest for women than for men (Table 3.3). This shows that among men in these two regions (between 35 and 43 women’s labor force participation depends more 63  EDAM 2012. 64  EDAM 1996. 65  Labor market participation is defined as being 15 years old and above– i.e. in the working age – and either working (employed) or available for work (unemployed). Following the national definition of unemployment, the discouraged unemployed individuals are included and considered active. This definition differs from the ILO’s one. 52 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES strongly on their levels of education compared to that negative for women. The education level of women’s of men’s. Married women seem to be less likely to mothers does not have any significant correlation with participate in the labor market than non-married ones. their probability to be active whereas men with a more While the relation between being married and being educated mother are more likely to participate in the active in the labor market is positive for men, it is labor force. Figure 3.2 Labor force participation by region and area Labor force participation Labor force participation Among the 15+ Among the 15+ 100 100 80 80 60 61 63 60 60 57 60 51 % % 43 40 34 35 40 33 31 25 27 24 20 18 20 20 0 0 Djibouti Ali Dikhil Tadjourah Obock Arta Urban Rural ville Sabieh Male Female Male Female Source: Authors’ calculations based on EDAM 2017. Figure 3.3 Labor force participation by years of education and age Labor force participation Labor force participation Among the 15+ Among the 15+ 100 100 90 90 80 80 70 70 60 60 50 50 % % 40 40 30 30 20 20 10 10 0 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 15 20 25 30 35 40 45 50 55 60 65 70 Years of education Age Male Female Male Female Source: Authors’ calculations based on EDAM 2017. 53 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES Table 3.3  Determinants of labor force participation among the working-age (15+) population (1) (2) (3) All Men Women -6.297*** Female (1.353) 6.524*** 7.797*** 5.534*** Age (0.241) (0.347) (0.337) -0.084*** -0.098*** -0.072*** Age squared (0.003) (0.005) (0.005) -4.876*** -6.410*** -3.232* Extreme poor (1.252) (1.898) (1.661) -0.167 -4.027* 2.368 Rural (1.552) (2.274) (2.109) Reference group: Djibouti-ville -5.851*** -4.180* -7.088*** Ali-Sabieh (1.634) (2.336) (2.275) -3.665** -2.097 -4.923* Dikhil (1.818) (2.548) (2.556) -13.952*** -18.700*** -10.863*** Tadjourah (2.023) (3.153) (2.622) -16.399*** -18.201*** -14.769*** Obock (1.946) (2.962) (2.561) 0.511 2.237 -1.030 Arta (1.833) (2.593) (2.548) Reference group: mother with no education 4.332** 5.503** 3.533 Mother primary education (1.729) (2.224) (2.575) -4.108 -2.621 -4.646 Mother secondary education (2.624) (3.500) (3.943) 0.743*** 0.614*** 0.742*** Years of education (0.098) (0.125) (0.156) 21.330*** 15.312*** -16.232*** Married (1.633) (2.072) (1.671) -40.146*** Female*Married (1.787) -0.001 0.050 -0.025 Household size (0.180) (0.245) (0.262) 0.603 0.958 0.265 Number of 0-6 children (0.460) (0.634) (0.652) -62.734*** -82.697*** -52.989*** Constant (4.257) (6.072) (5.749) Observations 10,182 4,727 5,455 R2 0.238 0.269 0.070 Source: Calculations based on EDAM 2017. Note: Weighted OLS regressions with robust standard errors in parentheses. * p<0.1, ** p<0.05, *** p<0.01. 54 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES Figure 3.4 Labor force participation and unemployment in comparable countries 100 90 80 75 76 70 63 60 54 57 53 50 46 46 % 44 44 43 40 39 40 36 37 31 30 23 23 25 21 22 2119 20 13 13 12 10 7 3 6 8 6 6 6 0 DJI CPV STP ERI ETH SOM YEM MENA LOW MID UPP MID WORLD Labor force participation of women aged 15-64 Percentage of women in the labor force Unemployment rate of women aged 15-64 in the labor force Source: WDI 2017. Note: DJI = Djibouti, CPV = Cabo Verde, STP = São Tomé and Príncipe, ERI = Eritrea, ETH = Ethiopia, SOM = Somalia, YEM = Yemen, MENA = Middle East and North Africa region, LOW MID = Lower middle income countries group, UPP MID = Upper middle income countries group. Comparing Djibouti to comparable countries, the WDI data suggests that in 2017 women’s labor Characteristics of unemployment force participation in Djibouti (54 percent66) is Among the labor force participants, half are at a similar level to the one in Cabo Verde (57 unemployed.67 Precisely, the unemployment rate for percent) while it lags far behind neighboring those aged 15 and above was 47 percent in 2017. This countries such as Eritrea and Ethiopia (75 and 76 is an increase from the unemployment rate of the 15+ percent) (Figure 3.4). However, Djiboutian women’s active population at 44 percent in 1996 (World Bank participation in the labor force is higher than in (1997) using EDAM 1996) but a decrease compared to neighboring countries Somalia and Yemen (23 and 2012 (EDAM 2012) (48.5 percent). Thus, unemployment 6 percent), as well as the average rate for the MENA in Djibouti is characterized as being structural, which region (22 percent) and the comparable country São is confirmed by the fact that more than 2/3 of the Tomé and Príncipe (44 percent). The percentage of unemployed individuals have been searching for a job women in the labor force is almost twice as high for over a year. The gender gap in unemployment is in Djibouti as it is in the MENA region (40 percent high68, as the female unemployment rate is 63 percent versus 21 percent), and is higher than São Tomé and compared to 38 percent for men. Combined with the Príncipe (36 percent), but lower than Eritrea and low labor force participation rate, this means that a Ethiopia (both at 46 percent) and Cabo Verde (44 very low percentage of women aged 15 and above percent). were working in Djibouti in 2017 (precisely, only 66  The numbers used to compare Djibouti with comparable and neighboring countries differ from the ones coming from the household survey EDAM4 2017, even if they are from the same year (2017). Indeed, in order to be able to make consistent comparison between countries, benchmarking numbers come from WDI database that uses ILO estimates and ensures that definitions are the same across countries. 67  Following the national definition of the labor market outcomes, the unemployment rate is defined as the proportion of the active population that is available to work. Thus, this definition includes the discouraged unemployed who are not searching for a job. This definition is different from the ILO’s one. 68  Gender wage gap is not analyzed in this report due to limited availability and poor quality of wage data in the EDAM4 survey and the EDESIC 2015 survey. Specifically, in the EDAM4 (2017) an analysis of the distribution of the hourly wages data showed a large number of outliers. This could be a result of individuals misreporting the unit of time worked (hour, day, week, month) which results in inaccurate calculations of their hourly wages. 55 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES Figure 3.5 Unemployment rate by region and poverty status Unemployment rate Unemployment rate Among the 15+ active population Among the 15+ active population 100 100 80 80 75 73 80 63 58 57 60 58 60 52 60 51 % % 42 40 39 40 34 34 29 28 20 20 0 0 Djibouti Ali Dikhil Tadjourah Obock Arta Non−poor Poor ville Sabieh Male Female Male Female Source: Authors’ calculations based on EDAM 2017. 11 percent were employed). However, the gender can be partly explained by the low levels of education gap69 has slightly decreased from 30 p.p. in 2012 to and the limited opportunities in this region. 25.5 p.p in 2017. This is partly due to the increase of men’s unemployment rate from 66 percent in 2012 to Women’s unemployment is the highest for those who 63 percent in 2017 (women’s unemployment rate has completed primary but not secondary (79 percent), been stable over the years). However, the situation while it is the lowest for the most educated women has worsened compared to 1996 where the gender (less than 50 percent if completed secondary). gap was only 11 p.p. of difference and 51 percent of As for the labor force participation, the relationship women were unemployed. between unemployment and level of education is not linear (Figure 3.6). Strikingly, women with no education Poor women are much more likely to be unemployed are much less likely to be unemployed than the ones than non-poor ones (80 percent compared to 60 who completed primary (61 percent compared to 79 percent). Furthermore, the unemployment rate percent, respectively). From a labor demand point of non-poor women is higher than that of poor of view, it appears that there is stronger demand for men (albeit slightly), standing at 60 percent and 58 uneducated women and highly educated women. percent, respectively (Figure 3.5). The gender gap Notably, a female with a secondary education or in unemployment is higher in rural areas than urban above has a higher probability to be unemployed ones (32 p.p. compared to 25 p.p.) and the urban- than a male with no education at all. Further, females rural gap is higher for women than for men (12 p.p. with no education have an unemployment rate that compared to 6 p.p.). This could be partially explained is nearly double than their male counterparts. The by the fact that individuals living in rural areas have gender gap across the different education categories lower levels of literacy and educational attainment is also palpable. than those in urban areas. Further, it could also be attributed to fewer employment opportunities in the The unemployment rate of women generally rural areas. There are also some regional variations declines with age, starting with 89 percent for the in the unemployment rate, as the lowest women’s 15-24 year-olds to 33 percent for the 60+ (Figure 3.6). unemployment rate is in Ali-Sabieh with 52 percent Moreover, 63 percent of prime-aged women (25-39 and the highest is in Tadjourah with 75 percent. This year-olds) are unemployed. Within the working age, 69  The employment rate of men minus the employment rate of women, expressed in p.p. difference. 56 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES Figure 3.6 Unemployment rate by education level and age groups Unemployment rate Unemployment rate Among the 15+ active population Among the 15+ active population 100 100 89 84 79 80 74 80 61 63 60 60 51 48 46 % % 43 40 40 37 32 30 33 24 20 20 20 0 0 No education Less primary Primary Secondary 15−24 25−39 40−59 60+ Male Female Male Female Source: Authors’ calculations based on EDAM 2017. the negative correlation between unemployment cited by 9 percent of men. Poor women are much and age could be attributed to the fact that the new more likely to be in NEET than non-poor ones, at 64 entrants have a limited set of skills, but do not take percent compared to 50 percent respectively (Figure up job offers with employment conditions that do not 3.7). Notably, education seems to be strongly linked to match their expectations. Furthermore, new jobs in NEET status (Figure 3.7). While more than 80 percent the formal sector are rarely created, and retirement of women who have no education or less than primary from formal sector jobs does not happen often. are in NEET, they are around 35 percent among the Finally, the high unemployment rate among the ones who completed at least primary education. youngest age cohort (15-24) could be attributed to delays in school-to-work transitions. However, given Once unemployed, women are more likely than that only 26 percent of women and 32 percent of men men to be discouraged, meaning that most of aged 15-24 participate in the labor force, it does seem them are no longer actively searching for work. that a significant portion of the youth, and especially Indeed, on average 66 percent of women available for females, is completely disconnected from education a job are discouraged unemployed, compared to 52 and employment. percent of men. Poor unemployed women are more prone to being discouraged than the non-poor, as 76 Indeed, 50 percent of the 15-24 years old are percent of poor unemployed women are discouraged classified as Not in Employment, Education or compared to 64 percent among the non-poor (Figure Training (NEET)70, with 53 percent of women and 46 3.8). The gap between genders for the poor and non- percent of men. This is a strong increase compared poor is similar, with the former having a 2 p.p. larger to 2012 where 45 percent and 32 percent of female gap. Furthermore, older women are shown to be more and male youth were in NEET. The majority of women likely to be discouraged (Figure 3.8). While 61 percent being NEET are inactive and the main reason is family of the youngest unemployed women (15-24 years old) obligations (21 percent), whereas this reason is only are discouraged, the rate is 94 percent among the 70  The share of NEET is defined as the percentage of individuals aged 15-24 (the total youth population) who are not in employment, education, or training (ILO and OECD). Those in education include those studying part-time or full-time, but exclude those in non- formal education and those engaged in an educational activity of very short duration. Employment includes all those who have been paid for at least one hour in the week preceding them filling the survey. It also includes all those who have been absent from work on a temporary basis. Therefore, NEET youth are those who are either unemployed or inactive and not in education or training (OECD). Cf ILO’s definition: https://data.oecd.org/youthinac/youth-not-in-employment-education-or-training-neet.htm 57 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES Figure 3.7 NEET by poverty status and educational attainment Not in employment, education, or training Not in employment, education, or training Among the 15−24 Among the 15−24 100 100 88 91 86 83 80 80 64 61 60 60 50 % % 43 40 40 37 36 35 32 20 20 0 0 Non−poor Poor No education Less primary Primary Secondary Male Female Male Female Source: Authors’ calculations based on EDAM 2017. Figure 3.8 Type of unemployment by age groups and poverty Type of unemployement Type of unemployement Among the 15+ unemployed Among the 15+ unemployed 100 100 52 61 49 66 60 75 60 94 50 64 60 76 80 80 60 60 % % 48 51 50 40 40 39 40 40 40 34 36 20 25 20 24 0 6 0 Male Female Male Female Male Female Male Female Male Female Male Female 15−24 25−39 40−59 60+ Non−poor Poor Not discouraged Discouraged Not discouraged Discouraged Source: Authors’ calculations based on EDAM 2017. Note: The non-discouraged unemployed are the individuals who are available to work and have searched for a job in the last 30 days while the discouraged unemployed are the ones who are available to work but have not searched for a job during the last 30 days before the survey. oldest group (60 and over). In addition, the gender men are more likely to search for a job for more than gap increases across generations (9 p.p. among the a year, the opposite is observed among women (64 15-24 compared to 34 p.p. among the 60+). percent for the poor unemployed women compared to 71 percent for the non-poor). Among all the regions, Around 67 percent of the discouraged and non- it is in Tadjourah that women are the most likely to be discouraged unemployed have been searching for unemployed for more than a year (81 percent) while a job for over a year, and women are more likely to in Dikhil only 40 percent of women searched for over be searching for over a year than men (69 percent a year. Generally, the highest educated unemployed compared to 66 percent). While the poor unemployed individuals have been searching for a job for shorter 58 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES periods of time compared to their lowest educated Figure 3.9 Length of unemployment by education level counterparts (Figure 3.9). Indeed, 59 percent of unemployed women with at least secondary education Searching for a job for over a year searched for a job for more than a year compared to Among the 15+ unemployed 72 percent among the ones with no education. 100 80 The level of education is also correlated with the 70 72 69 67 69 reasons for not searching for a job among the 60 60 58 59 inactive persons aged 15 and over (Figure 3.10). % While 60 percent of inactive women without education 40 do not search for a job because of family obligations, the rate is twice lower among the ones who completed 20 secondary education (29 percent). 0 No education Less primary Primary Secondary Around 29 percent of refugees and asylum seekers71 aged 15 and over72 are working, with the Male Female rate for women being 21 percent and for men being Source: Authors’ calculations based on EDAM 2017. 38 percent. The limited access to job opportunities in the camps results in refugees migrating to the capital to work. Women for example tend to work as Figure 3.10 Reason for not searching for a job housemaids in the capital to prevent increased risk by education level of exposure to sexual exploitation and prostitution. Reason for not wanting to find a job or start an activity Firewood collection points are usually far from Among the 15+ inactive persons refugee camps, with a distance of roughly 12 to 13 km 100 30 18 24 21 13 10 16 12 on average, thereby increasing the threats for girls of 7 0 10 5 being exposed to gender-based violence (WFP and 80 21 12 35 29 37 15 11 UNHCR, 2013). This limited access to work increases 35 60 57 67 food insecurity, especially when the head of household 60 62 58 % is a female, and the lack of income is exacerbated 40 48 by the limited access to credit. For instance, while 29 34 250 women in Ali Addeh received micro financing, 20 this service was unavailable in Holl Holl (WFP and 10 10 0 1 1 UNHCR, 2013). Although the government has been Male Female Male Female Male Female Male Female keen on ensuring food is adequately distributed, No education Less primary Primary Secondary especially for children younger than five years old as Studies Family obligations well as pregnant/breastfeeding women, food security Not working age, illness, handicap Other remains an issue. Considerable effort has been put Source: Authors’ calculations based on EDAM 2017. to guarantee that both female and male refugees receive equal treatment (UNHCR, 2000). First, a loans outlets for products of refugees, and to promote scheme for refugee women has been implemented safe access to reproductive health services. Further, by UNHCR. Second, a committee for women was through the National Refugee Law in 2017, Djibouti formed aiming to expand women’s participation in has mandated increased integration of refugees in food and camp management amongst other affairs. health, education, and the labor market. For instance, Third, research has been conducted to identify more than 324 refugees have obtained residency and the resettlement of women at risk, to spot market working permits as of 2018 (Hamida et al., 2020). 71  Numbers are from the « Rapport d’enquête profilage dans les villages de réfugiés 2019 » from MASS, INSD, WFP and UNHCR. 72  In this report of 2019, the active population is defined differently from the analysis in the current report. Indeed, all the 15 years old and above individuals are considered as the active population. 59 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES BOX 4  Deep Dive: Djibouti-ville and economic opportunities The labor force participation of women varies across the five districts of the capital (Figure 3.11). Those from Balbala, the 4th and 5th districts, are the least likely to be active (33 and 32 percent, respectively) compared to 47 percent of women from the 1st district who participate in the labor force. Moreover, the 5th district is also the one where the gender gap is the highest, with 31 p.p. of difference between men and women. Once in the labor market, women from the 4th district are much more likely than others to be unemployed (Figure 3.11). Indeed, 72 percent of the 4th district active women are unemployed while the rate is 50 percent in the 3rd district. In addition, the gender gap is the highest in the 2nd district with 28 p.p. of difference compared to 20 p.p. in the 3rd district. While women from the 4th district are the least likely to employed, once employed, they are also the least likely to work in the formal private sector as well as the public sector, compared to women from other districts (Figure 3.12). Indeed, the large majority (83 percent) of employed women from the 4th district are working in the informal sector, while half of women from the 2nd district are. Moreover, the gender gaps in sectors of employment are also the highest in the 4th district, with a 40 p.p. of difference in the informal sector. Both women and men from the 5th district are the most likely to work in the public sector (43 percent and 55 percent, respectively). Figure 3.11 Labor force participation and unemployment Labor force participation Unemployment rate Among the 15+ Among the 15+ active population 100 100 80 80 72 66 65 63 59 61 61 60 56 60 57 50 % % 44 44 40 38 40 36 36 39 31 33 31 30 20 20 0 1st dist 2nd dist 3rd dist 4th dist 5th dist 0 1st dist 2nd dist 3rd dist 4th dist 5th dist Male Female Male Female Source: Authors’ calculations based on EDAM 2017. 60 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES BOX 4  continued Figure 3.12 Sector and type of employment Sector of employment Type of employment Among the 15+ employed Among the 15+ employed 100 100 43 35 49 41 44 28 47 14 55 43 4 9 2 6 3 4 2 7 1 5 44 48 35 61 44 50 20 4 38 50 46 80 80 83 10 3 60 8 62 60 0 8 5 63 19 57 19 10 4 1 3 35 6 14 51 % % 52 14 29 8 8 4 40 43 9 40 17 36 38 37 15 18 33 36 36 34 8 30 13 20 20 24 24 25 22 15 11 11 0 7 0 Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. Male Fem. 1st dist 2nd dist 3rd dist 4th dist 5th dist 1st dist 2nd dist 3rd dist 4th dist 5th dist Informal Formal Public Employer Freelancer Daily laborer Waged worker Other Source: Authors’ calculations based on EDAM 2017. In comparison to comparable countries, the women is higher in Djibouti. The employment status unemployment rate of women in 2017 – as reported of Djiboutian women is quite different from women by the WDI database – in Djibouti (11.5 percent73) in neighboring countries as they are less likely to be is much higher than in its neighboring countries self-employed and more likely to work as waged and Eritrea and Ethiopia (6.7 and 2.8 percent), whereas salaried workers than in its neighboring countries. it is lower than the MENA region average (18.5 Working women in Djibouti are also engaged in percent), Cabo Verde (13 percent), and São Tomé different branches than in countries around: they are and Príncipe (21 percent). Once working, women in less likely to work in agriculture and more likely to Djibouti are less likely to be in a vulnerable employment work in services than in the neighboring countries. (defined as contributing family workers and own- account workers) than women from comparable and neighboring countries. While 43 percent of employed women are in a vulnerable position in Djibouti, the Characteristics of employment rate is up to 58 percent in São Tomé and Príncipe, and Djiboutian women are more likely than men to work more than 80 percent in the neighboring countries in the informal sector74 (63 percent compared to Eritrea, Ethiopia, and Somalia. However, compared to 41 percent), in precarious employment conditions the MENA average (27 percent) and Cabo Verde (31 and without recourse to social security benefits. percent), the proportion of vulnerable employment of The remaining working women are employed in the 73  The numbers used to compare Djibouti with comparable and neighboring countries differ from the ones coming from the household survey EDAM4 2017, even if they are from the same year (2017). Indeed, in order to be able to make consistent comparison between countries, benchmarking numbers come from WDI database that uses ILO statistics and ensures that definitions are the same across countries. 74  The sectors of employment are divided into 3 categories: private formal sector, private informal sector, and public sector. This report follows the definition of the Poverty and Equity Assessment of Djibouti by the World Bank 2019 report for defining these sectors. In the private sector (as opposed to the public sector), employers, contractors, and freelance or daily workers are classified as working in the formal sector if they (i) have accounting records, (ii) a trading license, and (iii) a subscription to the chamber of commerce. Salaried individuals, family helpers and apprentices are considered employed if they (i) have a contract, and (ii) are subscribed to the national security. The rest of the individuals working in the private sector are considered to be part of the informal sector. 61 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES public sector (31 percent) and in the formal private in senior civil service occupations77. Women represent sector (6 percent). On the other hand, 48 percent of less than twice as many men in all positions, the largest men are working in the public sector and 11 percent in difference being within the position of Secretary the private formal. Compared to 2012, the proportion General for which the proportion is 1/10. Moreover, of women working in the informal sector increased the higher the level of responsibility, the less women (from 58 percent to 63 percent), while the proportion are employed. working in the formal private sector decreased by more than 50 percent (from 13 percent to 6 percent). Women in rural areas are more likely to be working In general, informal workers face higher risks and in the informal sector than urban ones (80 percent receive less benefits than their counterparts in the compared to 62 percent) (Figure 3.13). They are also formal sector (OECD, 2019). For instance, informal less likely to be employed in the public sector (16 workers face higher occupational risks (due to the percent compared to 36 percent), while there is no absence of occupational safety and health registers) significant difference for the proportion of women and receive less social protection services, making working in the formal sector. The gender gap of them more vulnerable. Thus, female informal workers working in the informal sector is higher for the urban do not enjoy maternity benefits. Moreover, due to the workers than the rural ones, whereas the gender gap lack of contracts, individuals in the informal sector of working in the public sector is higher in the rural face higher degrees of economic uncertainty as well areas. Almost all the poor women are working in as lack of stability. As females are more likely than men the informal sector (94 percent), with only 6 percent to work in the informal sector in Djibouti, this makes in the public sector and none in the formal private them more prone to work risks and less entitled to sector. This is a strong difference with poor men, as benefits and protection by construction. Nevertheless, 37 percent are working in the public sector and 59 Djibouti has passed a law in 2019 on the social and percent in the informal sector. Moreover, the gender solidarity economy (SSE)75 seeking to eradicate gap is higher among the poor than the non-poor (35 poverty and expand social protection in Djibouti p.p. difference compared to 21 p.p. difference for through creating jobs and income opportunities in the informal sector). The preponderance of women rural areas and formalizing the informal economy in working in the informal sector can be attributed, in Djibouti, amongst other measures. part, to the difficulty of women in accessing financial services. Indeed, men who are 15 years of age or older Even within the public sector, it is clear that are twice as likely as women78 to own an account in a the ministries do not all have the same level of financial institution or with a mobile money-service feminization76. In 2017, the Ministry of Women and provider (16.6 percent compared to 8.8 percent, Family had the highest proportion of female employees respectively). This is much lower than in Somalia and (54 percent). Other ministries with a high proportion Ethiopia, where the percentages of women aged of women are the Ministry of Communication (46 15 and older owning a bank account are 34 and 29 percent), the State Secretariat for Social Affairs (44 percent respectively, but higher than in Yemen, where percent), the Ministry of Labor (41 percent), the the rate is only 2 percent. Ministry of Health (41 percent) and the Prime Minister (41 percent). Conversely, the Ministries of Agriculture The highest proportion of women working in the and Transport have only 13 percent and 14 percent informal sector is in Ali-Sabieh and Dikhil at 84 female staff respectively. Between 2000 and 2017, percent, whereas these rates in the other regions some departments saw a considerable increase in range between 60 percent and 64 percent. As female staff, such as the Ministry of Communication (28 expected, one of the highest proportions of public percent), the Ministry of Muslim Affairs (25.5 percent) workers is in the capital (33 percent among women) as and the Ministry of Labor (17 percent). However, the well as Tadjourah and Obock (34 percent). The more gaps between women and men remained significant educated a woman is, the less likely she is to work in 75 https://www.ilo.org/global/topics/cooperatives/sse/WCMS_732087/lang--en/index.htm 76  Ministry of Women and Family and DISED (2019). 77  Study on the representation of women in the public service, Ministry of Women and Family and UNFPA (2019). 78  According to the 2020 Human Development Report by UNDP. 62 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES the informal sector (Figure 3.13). While 90 percent the other hand, women without education or who did of the non-educated women are working in the not complete primary have almost no chance of being informal sector, the rate is 48 percent for the women employed in the formal sector. The less educated the who reached primary school and 16 percent among individuals are, the higher the gender gap is in sectors those who are the most educated. Moreover, there of employment. In terms of age groups, the 25-39 is no gender gap in the proportion of people with women are the most likely to work in the public sector a secondary education working in the public sector (41 percent) and the least likely to be employed in (both women and men at 72 percent). Women who the informal sector (52 percent). On the contrary, the completed at least secondary education are also the oldest ones are the most likely to be working in the most likely to work in the formal sector (12 percent). On informal sector (87 percent). Figure 3.13 Sector of employment by area and educational attainment Sector of employment Sector of employment Among the 15+ employed Among the 15+ employed 100 100 49 32 36 16 34 7 33 27 48 45 72 72 3 80 4 80 90 80 3 6 4 6 11 70 60 60 62 60 56 59 15 7 12 % % 48 40 40 39 38 14 12 20 20 14 16 0 0 Male Female Male Female Male Female Male Female Male Female Male Female Urban Rural No education Less primary Primary Secondary Informal Formal Public Informal Formal Public Source: Authors’ calculations based on EDAM 2017. Figure 3.14 Type of employment by poverty status and educational attainment Type of employment Type of employment Among the 15+ employed Among the 15+ employed 100 100 2 6 3 6 4 11 3 42 2 4 64 67 52 37 40 11 41 54 15 51 45 80 5 80 61 5 60 60 63 3 12 10 6 1 40 % % 6 28 42 6 33 40 40 26 14 16 16 1 5 4 18 30 20 24 20 26 18 28 18 17 17 13 11 10 6 0 0 Male Female Male Female Male Female Male Female Male Female Male Female Non−poor Poor No education Less primary Primary Secondary Employer Freelancer Daily laborer Waged worker Other Source: Authors’ calculations based on EDAM 2017. Note: The freelance workers correspond to the category of “travailleurs indépendants” in French. Freelancers are part of the self-em- ployed category, which also includes employers and daily laborers. 63 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES In addition to mainly working in the informal the 15-24 and the 25-39) than the oldest (8 percent sector, women are more than twice as likely as and 4 percent for the 40-59 and 60+). men to be freelancers (42 percent compared to 17 percent), and less likely to be employers and waged Among working women, 58 percent are considered workers. Compared to 2012, the proportion of women self-employed79, compared to 46 percent of men. who are freelancers has increased (from 31 percent to This status of employment is associated with greater 42 percent in 2017), whereas the proportion of female vulnerability and more precarious working conditions. waged workers has decreased from 57 percent to 35 Indeed, the self-employed women are more likely percent. The type of employment varies according to be poor than the waged workers (13 percent to the poverty status of the employed (Figure 3.14). compared to 4 percent). Moreover, 69 percent of the Around 60 percent of poor women are freelancers, self-employed women have no education compared while the rate is 40 percent among the non-poor. to 35 percent of the waged women. Further, compared Moreover, poor women are much less likely to to self-employed men, self-employed women are be waged workers than the non-poor (11 percent much more likely to have no education (69 percent compared to 37 percent). Further, poor women are compared to 48 percent). Thus, self-employed women more likely to work as daily laborers and employers are much less likely to have a secondary education than the non-poor ones. In addition, the gender gap level (15 percent) compared to the waged women (44 in the type of employment is higher among the poor percent). Among the rural population, 68 percent of people than the non-poor, for almost all the types of women are self-employed compared to 48 percent employment (for example, there is 29 p.p of difference of men, whereas the difference between women and among the poor waged workers compared to 15 p.p. men is 11 p.p. among the urban population. Moreover, among the non-poor). the difference is even higher by poverty status as 82 percent of poor women are self-employed compared The more educated a woman is, the more likely she to 57 percent of poor men. The older women are, the is to be a waged worker and the less likely to be more likely they are to work as self-employed. Indeed, freelance worker (Figure 3.14). Indeed, 57 percent 64 and 69 percent of 40-59 and 60+ years old are self- of uneducated women are freelancers compared to 3 employed compared to 47 and 53 percent of the 15-24 percent of women with at least a secondary education and 25-39 years old. level, while 79 percent of the most educated women are waged workers compared to 25 percent of On average, waged men are more likely to have a the ones without education. In addition, the daily formal contract than women (80 percent compared laborers are only among the least educated women. to 69 percent), but among those who had contracts, Therefore, this group usually works in very precarious women are slightly more likely to have an open- situations, and is often unaware whether they will work ended contract than men (89 percent compared the next day. There are notable differences in types to 87 percent). Poor women on average are much of employment held by women among generations. more likely to be in precarious employment as only Almost half of the 15-24 years old (46 percent) are 28 percent of them have a formal contract compared waged workers whereas only 15 percent of the 60 to 71 percent of non-poor ones. Among the younger years and more are. There is a generational shift in cohorts (15-24 and 25-39), women are slightly more the primary type of employment from waged worker likely to work with a formal contract, while the 40-59 to freelancers between the oldest (40-59 and 60+) aged women are much less likely to have a contract and young generations (15-24 and 25-39). Indeed, 59 than their male counterparts (64 percent compared to percent and 54 percent of the 60+ and 40-59 groups 81 percent) (Figure 3.15). Consistently, non-educated are freelance workers while 22 percent of the 15-24 are. women are less likely to have a formal contract than The cohorts before age of 40 are also more likely to be the very educated ones (37 percent compared to 97 employers (22 percent and 19 percent, respectively for percent) (Figure 3.15). Moreover, almost all the waged 79  According to the ILO, the self-employed category of employment includes the employers, the freelancers and the daily laborers. 64 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES Figure 3.15 Contracts of waged workers by age groups and educational attainment Has contract Type of contract Among the 15+ waged workers Among the 15+ waged workers 100 100 81 83 81 36 64 19 18 14 14 5 3 80 75 77 76 4 3 15 12 91 95 64 80 17 58 75 60 71 60 12 64 52 % % 40 40 28 9 20 20 27 24 0 0 15−24 25−39 40−59 60+ Male Female Male Female Male Female Male Female No education Less primary Primary Secondary Male Female Open−ended Fixed−term Verbal/No contract Source: Authors’ calculations based on EDAM 2017. Figure 3.16 Social security coverage of waged workers by educational attainment and poverty status Registered to social security Registered to social security Among the 15+ employed Among the 15+ employed 100 100 80 74 72 80 60 57 60 54 49 42 % % 40 34 40 37 37 32 20 20 11 10 0 0 No education Less primary Primary Secondary Non−poor Poor Male Female Male Female Source: Authors’ calculations based on EDAM 2017. women with a secondary education level (95 percent) be registered to social security than employed men have an open-ended contract, whereas the rate is (33 percent compared to 51 percent). Moreover, only 29 percent among the ones without education. The 19 percent of rural employed women have a social majority of waged women with some education but coverage compared to 34.5 percent of urban ones. The who did not complete primary have a formal contract; difference is even more striking when disaggregating nevertheless, they are more likely to have a fixed- by poverty status (Figure 3.16). While 36 percent of the term contract than an open-ended one (58 percent non-poor employed women are registered to social compared to 24 percent, respectively). security, less than 10 percent of the poor ones are. Moreover, the gender gap is higher among the poor Although less than half of the population has a than the non-poor (26 p.p. of difference compared to social coverage, employed women are less likely to 17 p.p.). Registration to social security seems to also 65 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES Figure 3.17 Firm size and branches of employed by education level Firm size Branches of activity Among the 15+ employed Among the 15+ employed 100 100 51 10 59 27 67 43 79 78 25 13 31 9 13 10 9 9 9 27 45 72 72 80 7 48 12 80 80 1 34 69 29 33 60 60 64 15 % % 18 40 7 42 12 40 45 41 40 13 17 14 36 29 20 10 14 14 20 18 17 19 19 10 5 4 4 3 0 Male Female Male Female Male Female Male Female Male Female Male Female Male Female Male Female No education Less primary Primary Secondary No education Less primary Primary Secondary 1 person 2−3 persons 4−10 persons 11+ persons Services Public administration Other Source: Authors’ calculations based on EDAM 2017. be correlated with the education level of the women percent) than the non-poor (42 percent). Moreover, (Figure 3.16). Indeed, 72 percent of the working the more a woman is educated, the more likely she is women who completed secondary school have a to work in a large firm with more than 10 persons (78 social coverage, and the proportion is decreasing percent among the secondary level women compared with education level: 49 percent among the ones who to 10 percent among the non-educated) (Figure 3.17). completed primary, 31 percent for those who did not Thus, the less educated women are, the more likely complete primary school, and 12 percent among the they are to work in small firms who offer vulnerable women without education. This might be due to the positions that are more subject to shocks. Employed type of employment educated individuals have access women are mostly working in the branch of services to: they are more likely to work in the formal private or (58 percent) while men work in majority in the public public sectors, and to have a contract. Nevertheless, administration (48 percent). As for the size of firms, the it might be that those who are less educated are also branch of activity varies with the level of education less aware of the importance of having a contract and (Figure 3.17). Thus, 80 percent of the non-educated being registered to social security. Similarly, the gender women work in the services branch while they are gap in social security registration decreases with level only 20 percent among the women who completed of education. Among the most educated, the gender secondary education. For the refugees and asylum gap in only 3 p.p., whereas it is 22 p.p among the non- seekers population, branches of activity differ across educated. This suggests that for equivalent levels of location and gender.80 Women are mostly working education, women are more likely to face precarious in domestic services in Ali Addeh and Holl Holl (25 situations (informal sector, without a formal contract and 23 percent respectively) while they are working and without social coverage) than men. in majority in trade and other commercial sectors in Markazi (34 percent) and Djibouti-ville (59 percent). Women are more likely than men to be working in On the other hand, men work primarily in construction small firms with one employee only (44 percent). and public works in Ali Addeh and Holl Holl (20 and 18 Conversely, men are mainly working in very large firms percent respectively), fishing in Markazi (43 percent), of 11 persons and more (63 percent). Poor women are and trade and other commercial sectors in Djibouti (53 more likely to be employed in self-owned firms (68 percent). 80  According to the « Rapport d’enquête profilage dans les villages de réfugiés 2019 » from MASS, INSD, WFP and UNHCR. 66 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES BOX 5  Changes over time in economic opportunities (EDAM 2012/2017) Between 2012 and 2017, the labor force participation of both men and women aged of 15 years old and more has decreased, from 65 to 59 percent and from 45 to 32 percent, respectively. However, this decrease has not affected each age group the same way (Figure 3.18). The lowest decrease is registered among the youngest (15-24 years old) with a 7 p.p. difference between 2012 and 2017 for the women and only 1 p.p. for men. However, while there was no gender gap in labor force participation among this age group in 2012, the gender gap has widened to a 6 p.p. difference in disfavor of women in 2017. The 25-39 years old women are the most affected by the decrease of labor force participation with a 17 p.p. difference between 2012 and 2017, whereas the difference is only 10 points for men. Moreover, the gender gap in labor force participation – which was already high – considerably increased for the two middle age groups (8 points more for the 40-59 years old and 7 points more for the 25-39 years old). Not only did the labor force participation of women decrease for every age, but the gender gap also widened for every age group. Women aged 15 and more who participate in the labor force are less likely to be unemployed in 2017 than in 2012. Indeed, their unemployment rate has decreased from 66 to 63 percent whereas the men’s one has increased from 36 to 38 percent between 2012 and 2017. Therefore, at the whole active population level, the gender gap has closed from 30 to 25 p.p. difference between men and women. However, there are differences across age groups (Figure 3.18). Whereas among all the age groups, active women are less likely to be unemployed in 2017 than in 2012, the opposite is true among the youth of 15-24 years old as their unemployment rate has increased from 82 to 89 percent. Thus, in 2017 young women are less likely to participate in the labor market and yet, less likely to find a job compared to 2012. Besides, the proportion of youth women in NEET has increased by 8 points between 2012 and 2017. On the other hand, women of 40-59 years old, who are probably the most experienced, have seen their situation regarding unemployment improve the most. Indeed, their unemployment rate has decreased by 11 points – the highest decrease among age groups – between 2012 and 2017, and at the same time the gender gap has closed by 11 points – the highest closing among age groups. Figure 3.18 Labor force participation and unemployment rate by age group Labor force participation Unemployment rate Among the 15+ Among the 15+ active population 100 100 85 89 84 82 84 80 74 78 80 76 67 63 60 56 60 52 % % 47 44 39 43 40 38 40 36 37 3333 32 32 34 32 33 26 22 24 18 20 20 15 20 0 2012 2017 2012 2017 2012 2017 2012 2017 0 2012 2017 2012 2017 2012 2017 2012 2017 15−24 25−39 40−59 60+ 15−24 25−39 40−59 60+ Male Female Male Female Source: Authors’ calculations based on EDAM 2012 and EDAM 2017. 67 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES BOX 5  continued Between 2012 and 2017, the proportion of women working in the informal sector has increased by 5 p.p. whereas the proportion of men has increased by 3 points (Figure 3.19). Thus, the gender gap has increased among the informal workers. On the other side, the proportion of women working in the private formal sector has decreased by half, from 13 percent to 6 percent, keeping the gender gap in this sector identical between 2012 and 2017 (5 p.p. difference). Among the workers of the public sector however, the gender gap has slightly increased from 16 to 17 p.p. difference. Indeed, even if both women and men are more likely to work in the public sector in 2017 compared to 2012, the women’s proportion has increased less than the men’s one. In the meantime, the proportion of waged workers decreased among both women and men between 2012 and 2017 (Figure 3.19), but the decrease is higher among women (19 points) than men (15 points). The consequence is a higher gender gap among the waged workers in 2017 (16 points) than in 2012 (12 points). On the other hand, the proportion of freelancers has increased from 31 to 42 percent among women between 2012 and 2017, and from 9 to 17 percent among men. In that type of employment, the gender gap in favor of women has increased from 22 p.p. in 2012 to 25 p.p. in 2017. Therefore, the trend of women to be over-represented in the informal sector and in freelance types of employment while less likely to work in public and formal sectors and as waged workers and employers, was strong and intensified over time. Figure 3.19 Sector and type of employment Sector of employment Type of employment Among the 15+ employed Among the 15+ employed 100 100 45 48 29 31 1 2 2 6 68 51 54 35 80 80 13 6 60 63 60 18 58 3 11 % % 42 40 7 41 40 7 38 17 31 17 20 20 23 9 14 0 5 5 0 2012 2017 2012 2017 2012 2017 2012 2017 Male Female Male Female Informal Formal Public Employer Freelancer Daily laborer Waged worker Other Source: Authors’ calculations based on EDAM 2012 and EDAM 2017. 68 DJIBOUTI GENDER ASSESSMENT ECONOMIC OPPORTUNITIES Female entrepreneurship and access effect on employment, especially for women, and an improvement in their socio-economic situation and to the world of finance lifestyle. About 71 percent of loan recipients were women and 18 percent were widowed or divorced. In Djibouti, women are much less likely than men Of those who received the microloan, 27 percent to have a bank account (14 percent versus 38 reported using the money to improve their family’s percent of men).81 Indeed, Ministry of Women and quality of life, 10 percent said the loan gave them Family and UNFPA (2019)82 point out that women more independence, and 26 percent saved money face many challenges when trying to access bank or used it to pay off a debt. Nevertheless, figures credit. According to risk managers of local banks, from EDESIC83 (2015) suggest that only 45 percent only 10 percent of loans are granted to women. of women are aware of the microfinance services Although banks do not discriminate against women available in Djibouti, and only 7 percent have and the criteria and conditions for granting loans are attempted to access credit. Moreover, of those who the same for both sexes, women are more reluctant applied, only 8 percent passed. The figures suggest to approach banking institutions. Also, it is more that women are better informed and apply for more difficult for independent women and women at the credit than men, even though the latter have a higher head of SMEs/SMIs to acquire the funds they need approval rate. because they have to find alternative means of financing their business, such as family aid or self- Among women-owned micro, small, and medium financing. Financial institutions lack the necessary enterprises (MSMEs), 58 percent believe support systems and there are no national financial accessing finance is the biggest challenge84, education programs specifically tailored to women, whereas 57 percent and 51 percent believe finding further compounding the low financial inclusion of customers and generating sales are the biggest women. Commercial and Islamic banks have taken challenges respectively. The challenges for males steps to create specialized windows for women are different: with access to finance only being the and to train their branch managers or sales people. third biggest challenge (48 percent), behind finding However, more needs to be done to change the customers (51 percent) and competition (49 percent). current mentalities and practices of bankers and Women-owned businesses have significantly lower create mechanisms for accompaniment adapted to propensity of owning bank accounts than their male the needs of women. counterparts (54 percent compared to 71 percent). This is partially because women-owned enterprises According to the Ministry of Women and Family are usually micro-sized (80 percent of women and UNFPA report (2019) on the financial inclusion businesses are sole-proprietorships), informal (42 of women in Djibouti, the number of microloans percent of women businesses are informal), and granted to women by Caisses Populaires and young (42 percent of women businesses have Caisses d’Épargne Crédit is increasing over time. been operating for less than 6 years). In terms of These institutions are more lenient in the guarantees management positions, the Enterprise Survey (2013) they require from their customers, unlike local in Djibouti shows that 14 percent of firms have banks. With the emergence of the Balbala Islamic female top managers, compared to an average of Microfinance Unit, women can access credit even 4 percent in MENA region and 21 percent in Lower if they adhere to Islamic principles. This initiative is Middle Income Countries. expected to expand nationwide in the near future. In 2015, ADDS and DISED surveyed 400 people Women-owned businesses also have higher rates of who had received microloans and found a positive using the tontine system and borrowing from family 81  UNDP (2020). National Strategy for Financial Inclusion in Djibouti, 2021-2026. 82  Report entitled “Study on the current situation of women’s access to credit in Djibouti”. 83  “Enquête Djiboutienne sur l’emploi, le secteur informel et la consommation des ménages” in French. 84  Djibouti - MSME Finance Needs Assessment and Policy Recommendations 2021 (English). Washington, D.C. : World Bank Group. 69 DJIBOUTI GENDER ASSESSMENT members85 (53 percent compared to 33 percent), which headed households in rural areas are considered are informal methods of financing, and lower rates of poor, which is significantly higher than the male using loans (20 percent compared to 25 percent) and headed households, standing at 17 percent (Kireyev, long-term loans (8 percent compared to 15 percent). 2017). While 51 percent of female-headed households The main constraint behind the lack of usage of (long- suffer from food insecurity, the rate for male-headed term) loans is the lack of collateral (63 percent of households is lower, standing at 45.5 percent (WFP, women businesses) and the lack of understanding of 2019). loan application procedures (53 percent compared to 33 percent). In order to improve bank and loan usage The MPI allows to capture a broader understanding for women MSMEs, more targeted support is required of deprivations households face. This index for them to acquire assets and to access nonfinancial combines monetary poverty and indicators on services (for instance, development services and education and access to basic services and networking). In terms of general challenges, women infrastructure. At the country level, 28 percent of the in business cite family responsibilities as the biggest population is estimated to live in a multidimensionally challenge (53 percent), followed by the perception poor household, with no difference between women of women as being incompetent in entrepreneurial and men. The MPI of urban women is higher than skills (48 percent), the stereotype of entrepreneurship those of urban men (19 percent compared to 18 as being a masculine job (33 percent), and the lack of percent). Consistently, women living in Djibouti-ville female role models in the business field (20 percent) are slightly more likely than men to be considered highlighting the role of social norms as barriers to as multidimensionally poor (19 percent compared to entrepreneurship. 18 percent). In all the other regions, women are less likely to experience multidimensional poverty than men. The MPI is also higher among women than men Poverty aged 15-24 (27 percent and 18 percent, respectively) and 25-39 (27 percent and 23 percent, respectively). In 2017, 21 percent of the population in Djibouti The reverse is observed for the older age-groups. was below the extreme poverty line86. Women are on average slightly less likely to be living in When looking at the different indicators of the extremely poor households than men (21 percent MPI, differences between women and men compared to 22 percent), but rural women are much arise. Women are more likely than men to live in a more likely to be extremely poor than urban ones (63 household deprived of improved sanitation (46 percent compared to 14 percent). The gender gap percent compared to 45 percent). Further, women are (in favor of women) is the highest in Tadjourah (7 p.p. more likely (32 percent) to live in a household with no of difference) and the lowest in Dikhil (0.6 p.p.) and individuals aged 13 and over with primary education Djibouti-ville (0.8 p.p.). Moreover, the probability to be than men (29 percent). However, women and men extremely poor decreases with the level of education. have on average the same probability to be deprived Around 25 percent of the women without education of access to electricity and improved drinking water are in extreme poverty, whereas the rate is 14 percent (34 percent and 7 percent respectively), as well as for women with a primary level of education and 4 living in a monetarily poor household. There is also percent among those with a secondary education. no gender difference in the likelihood of living in a household where at least one school-aged child is not According to Kireyev (2017), poverty appears enrolled in school (18 percent). to affect female and male-headed households proportionately across the nation as well as in urban areas. Nevertheless, 29 percent of female- 85  Ibid. 86  The extreme poverty line is estimated using the EDAM4-IS survey and the reference consumption basket at DF 111,783 per year (DF 306 per day) – see Challenges to Inclusive Growth: A Poverty and Equity Assessment of Djibouti (2019) for technical details. 70 4 DJIBOUTI GENDER ASSESSMENT CHAPTER VOICE AND AGENCY Photo credit: Shutterstock 71 DJIBOUTI GENDER ASSESSMENT VOICE AND AGENCY Gender based violence prevalence said they had experienced physical violence from their husband/partner87, and 5 percent in the last and drivers twelve months (INSD and Ministry of Women and A the Family 2020). According to the same survey, two lthough there is no national-level data on out of five non-single women (40 percent) answered all forms of GBV in Djibouti, some forms of that this violence was justified. There are also spatial GBV persist, such as FGM/C, forced/early disparities in the prevalence of IPV. Table 4.1 shows marriages, sexual violence and IPV. Vulnerable that physical, sexual and emotional violence are populations, such as refugees and undocumented more common in rural areas than in urban areas. migrants, are particularly affected by GBV during travel However, urban women are more likely to report and upon arrival in refugee camps and urban areas. having a controlling partner (34 percent) than rural There are also large spatial disparities between rural women (19 percent). In refugee camps, women have and urban areas in terms of the prevalence of such been particularly affected by domestic violence due violence. Underlying factors and amplifying factors to marital stress, which stems from lack of livelihoods of GBV, such as social norms, traditions and customs and limited economic opportunities, among other partly explain these disparities. factors (Green et al. 2017). Intimate partner violence Economic violence is also particularly widespread and frequently reported to service providers in IPV, which refers to physical, emotional, and Djibouti. Economic violence refers to a form of sexual violence and controlling behaviors by domestic violence that involves making or attempting an intimate partner, is not widely reported and to make a person financially dependent by maintaining documented in Djibouti. Indeed, according to the control over financial resources, withholding access OECD (2019), domestic violence and marital rape, to money, and/or forbidding attendance at school or often underreported, are the most common forms employment (UN Women, n.d). In Djibouti, women are of gender-based violence in Djibouti; the latest law often affected by economic violence from an intimate of March 2020 (Law no. 66/AN/719/8e) does not partner. For example, cases of economic violence and mention them explicitly. The latest National Violence lack of resources represented more than half of all GBV Against Women Survey 2020 found that 7 percent of cases (54 percent) reported to the hotline (“cellule never-single women and 6 percent of single women d’écoute”) of the UNFD between 2015 and 2020. 87  Despite the lack of data, under-reporting is very common and likely on this topic. Therefore, these statistics could be biased downwards. 72 DJIBOUTI GENDER ASSESSMENT VOICE AND AGENCY Table 4.1  IPV prevalence among ever-married women aged 15-64 IPV type National (%) Urban (%) Rural (%) Lifetime physical violence 8 8 9 Physical violence in the last 12 months 5 4 8 Lifetime sexual violence 5 5 7 Sexual violence in the last 12 months 4 3 6 Lifetime emotional violence 9 9 9 Emotional violence in the last 12 months 5 5 5 Controlling partner 31 34 19 Source: INSD, 2020. Note: Questions on IPV are asked about violence commited by a husband/partner. Table 4.2  Reasons for not seeking help Reasons for not seeking help Prevalence (%) Shame 37 Fear of being divorced 20 Fear of aggravating violence 13 Fear of family reaction 10 Perception of violence as “normal” 8 Fear of not being understood 8 Fear of husband keeping children away from her 2 Fear of not obtaining help despite seeking help 2 Source: INSD, 2020. IPV remains a taboo topic and a private matter (24 percent), and NGOs (24 percent). The family is thus solved among families due to socio-cultural norms the primary support system and pathway used by IPV (UNDP, 2018). Incidents of IPV, domestic and family victims to seek help. Formal institutions and services violence are usually addressed within the family or are not the most common pathways for women to get customary system rather than the formal legal system. support. Among IPV survivors who did not use any If the problem is resolved, the woman continues living of these pathways – who represent 79 percent of IPV with her husband. Only when families do not find a cases, they report not seeking help for the following solution that satisfies both sides, the case might main reasons: shame (37 percent), fear of being proceed to the formal justice system. divorced (20 percent), fear of aggravating violence (13 percent), and fear of family reaction (Table 4.2). These Women who experienced IPV in Djibouti often do findings suggest that social and gender norms play a not seek help. Only 21 percent of IPV survivors among role in help-seeking behavior and GBV reporting. This the age group 15-64 years old reported seeking help situation is exacerbated by the fact that a significant and external support, mostly from the survivor’s family portion of women believe that it is acceptable for a (71 percent), the husband’s family (55 percent), men husband to physically punish his wife for a multitude and women in the justice sector (36 percent), neighbors of reasons.88 For example, in 2012, non-single women 88  Ministry of Women and Family and DISED (2019). 73 DJIBOUTI GENDER ASSESSMENT VOICE AND AGENCY aged 15-49 believed that physical violence is accepted refugees face major security risks and protection if the wife goes out without telling her husband in 49 issues, including physical and sexual abuse by percent of cases, if the wife neglects her children (52 traffickers, hunger, and psychosocial stress (Green et percent) or argues with her husband (53 percent), for al. 2017). Undocumented migrants from Ethiopia and refusing to have sex (52 percent), for burning the food Somalia who are in transit in Djibouti towards Yemen (37 percent) or not paying attention to the husband and Gulf countries may be subject to forced labor and (53 percent). sex trafficking in Djibouti City, the Ethiopia-Djibouti Trucking Corridor, and Obock, which is the main point Non-partner violence: rape, sexual abuse and of departure and arrival to/from Yemen (U.S. State Department 2020). exploitation Non-partner violence refers to gender-based physical, Forced/early marriage sexual, and/or psychological violence by someone other than a partner (other family member, stranger, Forced/early marriage is common in Djibouti, colleague, friend, etc.). Non-partner violence can especially in rural areas. In 2019, the prevalence of include rape, sex trafficking, and other forms of sexual forced/early marriage under 18 years was more than abuse and exploitation. twice as high in rural areas (27 percent) as in urban areas (10 percent), with a national average of 13 There is a lack of nationally representative data on percent (Table 4.3) (INSD, 2020). non-partner violence in Djibouti. The 2020 national survey on violence against women found that the Family considerations and community-related prevalence of lifetime physical violence against single traditions are the main drivers of forced/early women was higher in urban areas (6 percent) than in marriage in Djibouti. The primary reason for child rural areas (2 percent). This finding can be due to a marriage is the tradition of Absouma, which refers lack of reporting and documentation to assess the to parents organizing the marriage of their daughter magnitude of the issue (INSD, 2020). Rape cases are with a male relative with the aim of strengthening ties particularly underreported and only represented 0.4 between villages, families, and communities (INSD, percent of cases reported to the UNFD between 2015 2020). 46 percent of rural and 31 percent of urban and 2020. early marriages reportedly derive from this tradition (Table 4.4). In some ethnic communities in Djibouti, Vulnerable populations, especially female migrants refusal to marry can also bring shame to the family and refugees, are at risk of non-partner violence, and may result in punishment (e.g. social stigma, including sexual violence, abuse, and exploitation exclusion). Other causes of forced/early marriage while on the move and upon arrival in Djibouti. include pressure from parents, especially in urban Djibouti’s peace and stable situation in a volatile areas (24 percent), economic/financial reasons, and region prone to conflict and insecurity has made a lack of agency (35 percent in urban areas and 34 the country a favored destination for migrants and percent in rural areas). Moral, psychological, and refugees from neighboring Ethiopia, Somalia, Eritrea physical threats from parents represent 7 percent of and Yemen, who represent 20 percent of the country’s forced marriages in urban areas and 12 percent in population. With a population of one million, Djibouti rural areas. hosts more than 30,000 refugees and about 150,000 people on the move in Djibouti-ville alone (United Early/forced marriages are usually not reported Nations Djibouti, 2020). The country has also been a to service providers. For example, the UNFD only transit hub used by smugglers and traffickers to the handled 7 cases of early marriage and 11 cases of Arabian Peninsula and the Gulf states. During transit forced marriage between 2015 and 2020 – which to/through Djibouti, migrants, asylum seekers, and represent 0.3 percent of GBV cases.89 89  National Union of Djiboutian Women – GBV cases of the CEIO, 2020. 74 DJIBOUTI GENDER ASSESSMENT VOICE AND AGENCY Table 4.3  Prevalence of early marriage Early Marriage, by age National (%) Urban (%) Rural (%) 10-14 years old 3 2 9 15-17 years old 10 8 18 Total 13 10 27 Source: INSD, 2020. Table 4.4  Causes of early/forced marriages in urban and rural areas Causes Urban (%) Rural (%) Family considerations 24 6 Threats from parents (moral, physical, psychological) 7 12 Community traditions (including Absouma and arrangements among 31 46 families) Unwanted pregnancy 3 2 Other causes (economic/financial reasons, lack of agency, etc.) 35 34 Source: INSD, 2020. Female genital mutilation/cutting (15-19) had an overall prevalence rate of 90 percent compared to 94 percent among the oldest age cohort While the government of Djibouti has criminalized (45-49). The difference in FGM/C prevalence between FGM/C, it remains highly prevalent in Djibouti. the youngest and equivalent oldest age cohort is even 2017 UNICEF and DHS Survey data estimated that more striking in 2019: the youngest cohort (12-17) 93 percent of women aged 15 to 49 years old have had an overall prevalence rate of 72 percent whereas undergone FGM/C in Djibouti. A more recent 2020 the equivalent oldest cohort (40-49) had an overall national survey by the National Institute of Statistics prevalence rate of 95 percent. This finding suggests of Djibouti and the Ministry of Women and the Family that younger women are becoming less exposed to revealed that the prevalence of FGM/C among women FGM/C at a faster rate than their older counterparts. aged 15-24 has dropped by 98 percent in 2002 to 82.5 The fact that FGM/C prevalence rates for women aged percent in 2019 (Table 4.5) (INSD, 2020). The single 12-17 and 18-23 in 2019 are lower than women aged largest drop in FGM/C prevalence occurred between 15-19 in 2006 further supports this claim. 2012 and 2019 when it dropped from 96.3 percent to 82.5 percent. There are however major spatial disparities in FGM/C prevalence. While the practice of FGM/C has FGM/C prevalence varies by age as FGM/C significantly decreased in urban areas from 93 percent is more prevalent among older than younger in 2006 to 69 percent in 2019, FGM/C prevalence women, indicating that newer generations are remains higher in rural areas and only decreased from less exposed to this practice compared to their 96 percent to 79 percent in the same period (Table older counterparts. In 2006, the youngest age cohort 4.6). Moreover, the most mutilating forms90 of FGM/C 90  The WHO has classified FGM/C into 4 types: Type I: Excision of the prepuce with partial/total excision of the clitoris (known as sunna). Type II: Excision of the prepuce and clitoris with partial/total excision of the labia minora. Type III: Partial/total excision of the external genitalia and stitching the vaginal opening with its consequent narrowing (known as infibulation or pharaonic form). Type IV: Includes several types of genital manipulation: pricking, piercing, incising and stretching the clitoris or labia; burning the clitoris; scraping the tissue surrounding the vagina; introduction of corrosive substances or of herbs into the vagina (Martinelli and Olle-Goig, 2012). 75 DJIBOUTI GENDER ASSESSMENT VOICE AND AGENCY Table 4.5  FGM/C prevalence of women 15-24 at the national level (2002-2019) Year 2002 2006 2012 2019 FGM/C prevalence (%) 97.8 93 96.3 82.5 Source: INSD, 2020. Table 4.6  FGM/C prevalence by area 2006 2019 Urban 93 69 Rural 96 79 Source: EDIM , 2007; INSD, 2020. 91 are more widespread in rural than urban areas. For FGM/C is strongly rooted in socio-cultural norms, instance, type III is more likely to be practiced in rural religion, and tradition. While the 2020 national areas (28 percent) compared to 14 percent in the survey found that a majority of the population thinks capital city (Petit, 2018). 91 that the practice of FGM/C should end, 22 percent of women and 20 percent of men believe that the Spatial disparities in FGM/C prevalence also exist practice should remain. Three levels of resistance at the regional level. Rates of FGM/C are particularly have limited behavior change and attitudes towards high in Arta (76 percent), Dikhil (82 percent), and FGM/C at societal, family/community, and individual Tadjourah (88 percent) (Figure 4.1). FGM/C is also levels (Sougueh Guedi and Cavatorta, 2018). At the particularly prevalent in the refugee camps of Holl- societal level, anti-FGM/C efforts are perceived as an Holl and Ali Addeh (Green et al. 2017). attack against traditions. At the family level, pressures from relatives that link FGM/C and girls’ suitability for marriage explain the perpetuation of FGM/C. At Figure 4.1 FGM/C prevalence by region (2012-2019) the individual level, false perceptions about women’s health, sexuality, and body are also other factors that have inhibited the abandonment of the practice (Sougueh Guedi and Cavatorta 2018). Moreover, parents with higher levels of education, lower Obock illiteracy rate, or richer are more likely to forgo FGM 2012: 95% 2019: 74% (OFPRA, 2017). While 56 percent of women from rich Tadjourah households have an unfavorable opinion concerning 2012: 85% FGM, it is 35 percent among poor households. 2019: 88% Djibouti City National response on GBV prevention and 2012: 79% 2019: 69% management Dikhil Arta 2012: 76% 2012: 69% A National Protocol to address GBV was established 2019: 82% 2019: 76% Ali Sabieh in 2018 with several line ministries (Women and 2012: 70% Family, and Justice and Health), the National Police 2019: 61% and Gendarmerie and the NGO, i.e., UNFD. This joint protocol aims to ensure that GBV survivors receive Source: Adapted from INSD, 2020. urgent care and coordinated support from service 91  « Enquête Djiboutienne à Indicateurs Multiples » in French. 76 DJIBOUTI GENDER ASSESSMENT VOICE AND AGENCY providers. The protocol aims at the provision of a holistic package, including free health care, psycho- Intra-household decision making social support, protection, and legal aid. It also includes While women actively participate in decisions principles of confidentiality to ensure that service around food purchases and buying equipment and providers keep data about survivors confidential, safe furniture, investment and decisions concerning and secure, and focuses its approach on human rights choosing a person for marriage are often made and survivor-centered response. by men. When it comes to decisions on food purchases, both men and women share the decision- Currently, there are two referral service providers making process in 37 percent of households. Women that address GBV in Djibouti. The UNFD operates exclusively make decision on food expenditure in a support center and hotline (Cellule d’Ecoute, 36 percent of households and men exclusively in 27 d’Information et d’Orientation - CEIO) that assists percent (Figure 4.2). Households from rural areas women and girls who are survivors of violence and are more likely than urban ones to take decisions orient them towards the appropriate medical and legal about food expenditure by both women and men care and services. Their services are currently limited (50 percent compared to 34 percent) (Figure 4.2). In to Djibouti City and the refugee camps of Ali Addeh, fact, in urban areas, women are more likely to decide Holl-Holl, and Markazi. The Ministry of Women and the alone regarding food expenditure. When it comes to Family manages a hotline and has social workers for buying equipment or furniture, women are the only children who are survivors of violence. Similarly, their decision makers in 37 percent of the households while services are limited and accessible only in Djibouti men decide alone in 28 percent of the cases (Figure City. All survivors who are 17-year-old and younger 4.3). Regarding education and health decisions, are referred to Ministry of Women and the Family case both men and women participate in the decision for workers. Both institutions are at the front line of the the majority of households (35 percent). However, National Protocol on GBV but remained understaffed. men are the main decision makers when choosing a person for marriage for their children (39 percent) and With regards to FGM/C, the government of the decision is made by both men and women in 34 Djibouti has led several sensitization campaigns percent of the households (Figure 4.3). Decisions on at both national and local levels on FGM/C investment and on accepting or inviting household prevention. In 2017, the Ministry of Women and the visitors are made in majority either by men only or Family launched the 2018-2022 National Strategy for by both women and men (34 percent for investment the Total Abandonment of FGM/C, which includes and 36 percent for visitors). Thus, decisions on food the declaration of a fatwa on abandonment of purchases, education, and health are in the majority FGM/C and sensitization campaigns through schools of the households made jointly by women and men, and NGOs using radio, television, and door-to-door whereas equipment decisions are made mainly by campaigns (UNDP, 2018). Advocacy and awareness women only and choosing a partner in marriage campaigns in the media in local languages, are mainly done by men only. This could be partly including radio and television, and a traveling show explained by the fact that, according to Ministry of “Plus jamais ça à ma fille” reached inland areas Women and Family and DISED (2019), about a quarter to sensitize about the harmful effects of FGM on of households in the country are headed by women, the health of mothers and children (Fédération with the percentage being slightly higher in rural Internationale des Ligues des Droits de l’Homme, areas (26 percent) than in urban areas (23 percent). In 2011). These campaigns also included discussions addition, women are often heads of households due with local communities and the creation of a network to widowhood or divorce. of religious leaders to promote the abandonment of FGM that hold awareness campaigns at mosques and community associations. 77 DJIBOUTI GENDER ASSESSMENT VOICE AND AGENCY Figure 4.2 Decision makers on food expenditure Decision on food expenditure Decision on food expenditure All households Urban Rural 37% 36% 34% 25% 39% 50% 25% 27% 28% Decision makers Decision makers Women only Men only Both Women only Men only Both Source: Authors’ calculations based on EDAM 2017. Figure 4.3 Decision makers on equipment expenditure and marriage Decision on equipment/furniture expenditure Decision on a person for a marriage All households All households 27% 35% 37% 34% 27% 39% Decision makers Decision makers Women only Men only Both Women only Men only Both Source: Authors’ calculations based on EDAM 2017. Participation in political life92 and to the 5 percent of women in ministerial position in 2005, no women in 2000 in the national parliament, access to decision-making spheres and only 11 percent female parliamentarians in 2003 after the first enaction of the law on female quotas In 2020, 26 percent of seats in Djibouti’s national stating that 10 percent of each party seats must be parliament were held by women (Figure 4.4) and held by either sex (a second law93 enacted in 2018 held there were 13 percent of women in ministerial level the proportion of women candidates at 25 percent). positions (WDI). This is an important increase compared Thus, women entered parliament for the first time in 92  There is no gendered data available on the political participations to elections. 93  Law no. 219/AN/18/7ème L. 78 DJIBOUTI GENDER ASSESSMENT VOICE AND AGENCY 2003, there were 15 women elected in 2018 and there (28 percent), and the number is not reported for all the are currently 17 in the National Assembly after the other neighboring and comparable countries. reappointment of the cabinet.94 In addition, women parliamentarians created in January 2020 a caucus Several reports from the Gender Observatory96 dedicated to women and gender. But although the focus on the representation of women in decision- proportion of women employed in various government making spheres such as the civil service, the functions has increased, their representation does defense and public security forces, the armed not reflect their true weight in the population. For forces, the gendarmerie and national police, as example, in 2018, the Government was made up of well as diplomacy. The Gender Observatory notes a 24 members, including 3 women, and the National slight improvement in the proportion of women in the Assembly remains predominantly male (17 women out public sector, even though the civil service remains of 65 deputies). The Supreme Court includes 3 women predominantly composed of men. The effort of the members, while the Constitutional Council has 2 Djiboutian government is underlined since the number women out of 6 members. And at the regional level, of women in all sectors is increasing, however women there were 48 women out of 168 elected. Compared are more concentrated in ministries with a social to neighboring countries, Djibouti stands behind vocation (education and health) and the gender gaps Ethiopia which has 39 percent of seats held by women are widening even more in the senior civil service. Thus, in the national parliament and 40 percent of women the higher the level of responsibility, the less women in ministerial positions. However, the proportion of are present. A sharp increase in the number of women women in national parliament in Djibouti is higher than is also noted within the army, the gendarmerie and the MENA average (18 percent) and higher than most of the national police, while remaining a minority. For the neighboring countries. On the other hand, looking example, the rate of female soldiers rose from 7 percent at the percentage of women at ministerial positions, in 2002 to 11 percent in 2017. But this increase was not Djibouti falls quite behind its neighboring countries accompanied by a rebalancing of the various hierarchical (18 percent in Eritrea and Somalia for example). At bodies. With regard to the diplomacy sector, women the local level, the share of seats held by women in remain very under-represented and little change has Djiboutian local governments was estimated around been observed. Thus, women represent 8 percent 29 percent95 by UNDP in the Human Development of executive staff working in diplomatic missions, Report, 2020. This is slightly higher than Cabo Verde permanent representations and consular posts. Figure 4.4 Proportion of seats held by women in national parliament 50 40 30 20 10 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Source: WDI. 94  According to IPU Parline https://data.ipu.org/node/49/data-on-women?chamber_id=13377. 95  Over the period 2017-2019. 96  Synthesis of studies carried out by the Gender Observatory on the situation of Djiboutian women (2021). 79 5 DJIBOUTI GENDER ASSESSMENT CHAPTER GENDERED COVID-19 IMPACTS Photo credit: Shutterstock 80 DJIBOUTI GENDER ASSESSMENT GENDERED COVID-19 IMPACTS T o understand the impact of COVID-19 and the associated government measures on Djiboutian households’ welfare, a phone survey was designed and carried out in Djibouti between July 2020 and April 2021 across four different with female breadwinners (72 percent) were slightly more likely to face such stress than those with male breadwinners (67 percent). Similarly, the food insecurity has been impacted differently according to the gender of the breadwinner. In March/April 2021 waves.97 Containment measures had a different (wave 4), households with a female breadwinner were impact on the ability to work according to the gender less likely to have an adequate food consumption (82 of the breadwinner, with female breadwinners less percent compared to 89 percent) and were more likely likely than males to have worked the week prior to the to have a poor food consumption (6 percent compared survey (wave 2, Sept/Oct 2020). Around 76 percent of to 2 percent). However, a few months before, children female breadwinners worked that week, compared to from households with a female breadwinner were 78 percent for male breadwinners. But work stoppage more likely to eat three meals a day than those from was less prevalent among females compared to males households with a male breadwinner (Dec 2020/ (12 compared to 14 percent, respectively). This stems Jan2021, wave 3). It might suggest that households from the fact that female breadwinners were less likely with a female breadwinner prioritize children’s food to work even before COVID or that women are more security even if they do not have an adequate food likely than men to work in small businesses that are less consumption. The main strategies to cope with the likely to have shut down during COVID-19 lockdowns. activity decrease since March 2020 varied according to Moreover, women’s and men’s labor income were not the gender of the breadwinner. In Dec 2020/Jan2021 affected the same way. Among those who worked less, (wave 3), households with female breadwinners were female breadwinners (41 percent) were less likely to more likely to purchase on credit (31 percent) or to receive no payment than males (48 percent), whereas adopt no coping mechanisms (25 percent), than a relatively smaller share of female breadwinners households with a male breadwinner (11 percent and received their full paycheck than male ones (wave 1, 16 percent, respectively). July 2020). The lockdown measures and high demand for The COVID-19 pandemic has induced a situation in healthcare for confirmed cases of COVID-19 may Djibouti wherein many households were struggling have restricted the access of households to the to make ends meet. In July 2020 (wave 1), households health systems. In July 2020 (wave 1), a relatively 97  The sample size of national respondents was 1,486 in wave 1, 1,460 in wave 2, 1,383 in wave 3, and 1,561 in wave 4. More information about the sampling and the details of the survey is available here: https://www.worldbank.org/en/country/djibouti/brief/ monitoring-the-impact-of-covid-19-on-households-in-djibouti-through-high-frequency-phone-surveys 81 DJIBOUTI GENDER ASSESSMENT GENDERED COVID-19 IMPACTS larger share of households with a female breadwinner The COVID-19 pandemic and social distancing (40 percent) could not access healthcare compared measures have had major impacts on already limited to households where the breadwinner is a male (34 GBV prevention and response services in Djibouti. percent). Therefore, the COVID-19 pandemic has The pandemic led to a reduction in case management exacerbated existing vulnerabilities in Djibouti. There (including clinical management of rape) and limited has been an increase in cases of domestic violence child protection capacity due to the reduction of in- during the COVID-19 confinement and quarantine person social support services, especially for refugees requirements, according to the National Gendarmerie’s and migrants in Ali Sabieh and Djibouti-ville (United special GBV unit (U.S State Department, 2021). Nations Djibouti, 2020). For example, according to Similarly, UNHCR’s 2020 COVID-19 survey in refugee UNHCR 2020, prevention activities, medical assistance, camps suggested that 42 percent of forcibly displaced security support, and psychosocial support have largely people reported that domestic violence increased been reduced or suspended. Other services such as during the pandemic due to the following reasons: risk reduction activities have been reoriented towards idleness, loss of livelihoods, increases in stress and the COVID-19 health response. Overstretched health anger, and abusive consumption of khat (UNHCR, services have also affected women’s access to sexual 2020). and reproductive healthcare, potentially leading to increases in unplanned pregnancies. 82 6 DJIBOUTI GENDER ASSESSMENT CHAPTER CONCLUSION AND POLICY RECOMMENDATIONS Photo credit: Shutterstock 83 DJIBOUTI GENDER ASSESSMENT CONCLUSIONS AND POLICY RECOMMENDATIONS D jibouti has made some progress in advancing its gender agenda in recent years, but still has a long way to go to promote women’s ability to reach their full potential. Based on the identified constraints and opportunities for women in Djibouti, safer environment at school. As for maternal mortality, international evidence points to the importance of access to contraception and promoting the use of trained midwifery and medical services during delivery in reducing maternal mortality. Reducing early the report’s policy recommendations fall within three marriage and adolescent fertility contributes both to areas: closing gaps, creating opportunities, and better educational and health outcomes. Poverty, bolstering agency. lack of information, distance to medical facilities, and inadequate services also contribute to increased maternal mortality. Closing gaps Reducing rural-urban gaps in education, as well A key gap to address is that of education through as in economic opportunities, in Djibouti, for investments and improvement in access to both boys and girls must receive priority policy schooling of girls and reducing maternal mortality. attention. The data shows that women in Tadjourah, Despite a narrowing gender gap in school attendance, for instance, are twice less likely to have ever attended girls in Djibouti are still less likely to attend schools school than those in Djibouti city, and the least likely than boys. Indeed, while 82 percent of boys aged 6-11 to participate in the labor force compared to women were attending primary school in 2017, it is 77 percent across the country. These women face poor access of the girls of the same age, with a smaller proportion to electricity, compounded by multiple deprivations in middle and secondary schools. Investing in and and high poverty rates. They are also among the improving access to education is crucial not only to youngest to marry and have the highest fertility rates. improve labor market outcomes but also to foster the It is therefore important to identify and relieve some intrinsic value of education, improve intergenerational of the constraints, social and economic, that prevent mobility, and achieve inclusive development in girls from achieving their full potential. Closing Djibouti. Investments in literacy programs and lifelong the urban-rural gap by reducing poverty, equitably education opportunities, such as vocational or on- distributing resources, and investing in services and the-job training, could also bring significant benefits, in infrastructure improves the economic and human particularly to older women who have missed out on capital outcomes for women and men alike. Indeed, formal education as children. Evidence suggests that international evidence suggests that fiscal policies investing in good sanitation facilities also contributes aimed at improving infrastructure and transport have to reducing household chores for girls and provide a a positive effect on economic outcomes for women. 84 DJIBOUTI GENDER ASSESSMENT CONCLUSIONS AND POLICY RECOMMENDATIONS Closing data gaps enables evidence-based policy digital transformation by 2035. Regulatory systems decisions and identifies drivers and determinants must align with the efforts to improve access to of gender inequalities. For instance, including credit. For instance, the government could consider gender disaggregated data on asset ownership, developing a national financial inclusion strategy, access to finance and credit, norms and intra- aligning the central bank policies to provide financial household decision making information, allow for packages with fewer barriers, and increasing women’s better diagnostics of labor market policies to enable financial literacy and inclusion. In addition, improving women’s access to economic opportunities. The the regulatory environment for microfinance and government of Djibouti could consider conducting non-conventional financing, with a focus on women’s a Women’s Entrepreneurship Development (WED) access, would be beneficial. Assessment to determine the key barriers to women’s entrepreneurship, and developing policies Creating opportunities by promoting a healthy and accordingly. Further, Djibouti lacks data on social robust private sector, with equal opportunities for norms and attitudes such as the World Values Survey, all, and providing incentives for formal businesses Arab barometer or Afrobarometer among others, that who provide social security and decent work for could be instrumental in designing, implementing, employees. Djibouti’s economy is structured in large and evaluating policies. Data on marginalized part around the public and informal sectors. While communities, particularly migrants, undocumented formalizing the informal sector may not be feasible, persons, and nomadic groups is lacking and requires and indeed create distortionary spillovers, creating particular attention. Further, the government may incentives for new businesses, particularly in the consider conducting market research to understand government’s priority areas of digitalization and the needs of the various women market segments trade, would align with the national development and develop, pilot, and roll out financial packages plan, boost growth and economic diversification, tailored to their needs that have procedural and promote decent work for men and women. flexibility and rely on collaterals and cash flow based Diagnosing firm-level decisions, quantifying the underwriting methodology, without the necessity formality premium, and understanding the supply and of male sponsorship. Therefore, investing in better demand dimensions of formality are crucial. Indeed, quantitative (and qualitative) data would be crucial to after an economic census is conducted, an investment identify opportunities and constraints that can help in firm-level data collection as well as up-to-date labor define targeted actions to enhance decision making force surveys would be instrumental in the design and improve the status of women in households, and implementation of policies around private sector markets, formal and informal institutions. development. Moreover, better opportunities for women could Creating opportunities emerged from an improved access to land, internet and ICT. Indeed, these are valuable assets allowing Improving access of women to productive inputs and women to empower themselves and take ownership credit is necessary to promote entrepreneurship. of their development and economic inclusion. For Given the limited access of women compared to men example, internet and ICT are crucial to improve to finance and credit, and productive inputs (e.g. land), access to information, find and access educational and Djibouti may consider improving financial inclusion economic opportunities, as well as marketing products to allow women more independence in starting and networking. Digital literacy programs could be businesses. Some of the areas that are identified from deployed in order to empower women to achieve a portfolio review of World Bank operations include autonomy and prevent exclusion. More generally, improving the ecosystem of women’s entrepreneurship both men and women would benefit from access to in areas such fisheries and food processing. Other areas technology to improve the flow of information and include digital development, subject to investment enable a free and fair market place for goods, services, in human capital, which would align with Djibouti’s and economic opportunities. Other legal reform to development vision for improved connectivity and consider is inreducing gaps in inheritance law which 85 DJIBOUTI GENDER ASSESSMENT CONCLUSIONS AND POLICY RECOMMENDATIONS inhibit women from achieving their full potential and against violence. For example, children’s judges keeps women at an unequal footing with respect to work is constrained by a lack of human and financial men. Inheritance, while a right, is also instrumental in resources (UN Human Rights Committee, 2013). allowing intergenerational mobility, and adequacy of The justice system is also inefficient, partly due to a resources to create opportunities. lack of experienced legal staff and a lack of digital archives about judicial decisions (White and Case LLP and Child Rights International Network, 2015). Bolstering agency The lengthy and slow legal processes also deter survivors from accessing justice.98 Prisoners in pre-trial Empowering women’s voice, agency, and detention often wait a number of years for their trials documentation are priority policy areas. Djibouti has to begin (U.S Department of State, 2020). Legal staff made substantial progress in the past two decades in and other public administration personnel also often improving political representation of women. Building lack knowledge and awareness about how to handle on this progress, broadening and empowering civil GBV cases although judges, court clerks, and police society organizations who advocate for women’s rights officers have been trained to enforce article 333 of and work to relieve their challenges is an effective the Penal Code that criminalizes FGM/C, for example way to advance gender and ensures that government (Fédération Internationale des Ligues des Droits de policies are in tune with women’s concerns and l’Homme, 2011). needs. Women in marginalized communities, such as migrants and undocumented groups, face particular Investment and capacity-building in key actors in challenges, many of which are related to their legal the implementation of the National Protocol. Major status and documentation. Particularly policy attention barriers continue to hinder the institutionalization of a must be placed on regularization of the situation of multisectoral approach in terms of procedures, service such groups, many of whom have been in Djibouti delivery, and access to justice. The key actors involved decades, is essential. in the implementation of the protocol often lack basic resources and training, and as a result, the protocol is Empowering government actors involved in the also not always implemented. For example, the police National Protocol to address GBV, mainly those and gendarmerie reportedly do not take cases of GBV involved as first responders to GBV survivors, seriously and do not communicate to survivors their providing them with necessary resources and rights.99 Overall, the police rarely intervenes in IPV building there are necessary to enforce laws and incidents and there is no witness or victim protection implement policies against GBV. While Djibouti has system to protect them from perpetrators. Confusion enacted a number of laws against various forms of among various service providers about their roles and GBV, improving the capacity of key actors involved responsibilities also remain. There is also to date no in the GBV response is crucial. Line ministries and centralized system or database in place to collect services that are at the forefront of the GBV response, and share data among service providers. Investing mainly the Ministry of Health and the Ministry of in the existing key service providers to GBV survivors Justice, often lack sufficient resources and capacity to (health, police, CEIO and social workers) both in ensure coordination, monitoring, and implementation terms of human resources and capacity-building (i.e. of national strategies, especially from the national to case management, first aid psycho-social support) is the local level (OECD, 2019). fundamental. Strengthening the capacity of the justice system Bridging the gaps between the legal system and to deal with GBV, including the enforcement of customary laws, which hinder law enforcement, the criminalization of FGM/C by law. The justice by raising awareness and sensitizing local system faces similar resource and capacity constraints communities. Survivors often avoid the formal as other government actors in safeguarding them justice system and prefer to establish settlements 98  Interview with UNICEF, June 2021. 99  Interview with UNFPA, June 2021. 86 DJIBOUTI GENDER ASSESSMENT CONCLUSIONS AND POLICY RECOMMENDATIONS between families (UNDP, 2018). Discrepancies Finally, adopting the implementing decree for Law include lower compensations for female victims, a No. 66/AN/719/8e of March 2020 and equipping lack of private hearings, and women’s exclusion in GBV survivors with knowledge of their rights representing their own cases.100 Incidents of GBV are and legal procedures to access justice, in part often addressed at the family and community levels due to weak education levels and literacy rates through traditional conflict resolution mechanisms, (Government of Djibouti, 2018). Cultural norms, lack of including compensation to the victim or the victim’s resources and discriminatory attitudes and behaviors family. For example, the Makaban (Afar judge) often prevent women from access to justice, health applies the Maro (traditional set of rules) to settle services or other type of support. For example, the criminal cases and conflict within the family, between Ministry of Women and the Family noted that some neighbors, within a clan, or between clans. Cases can women do not file complaints against their spouse or range from homicides to civil cases, except disputes other people close to them because of social stigma, related to marriage and divorce, which are dealt family norms, religion, and a lack of resources. Some by the Sharia judge (Al-Ma’adoun al-Shari) (Alemu, forms of GBV, including psychological, emotional, and 2013). In the Issa customary system, the Begti (Issa economic violence, are not addressed through the judge) works with tribal leaders to apply the rules of justice system because women do not report these Xeer -- the Somali customary law -- which also does types of violence. Continuous awareness-raising about not guarantee gender equality and equal protection women’s rights and the existence of service providers, (Lifos, 2018). coupled with activities aiming at behavior change, involving men and boys, is fundamental. 100  See Annex 3. 87 DJIBOUTI GENDER ASSESSMENT BIBLIOGRAPHY BIBLIOGRAPHY AFD (2014). Profil genre Djibouti. https://plate- article/34/4/4534/6129701?casa_token=7DKNxn- forme-elsa.org/wp-content/uploads/2014/02/Pro- L9zkkAAAAA:muas38yFcTYUVWuP5d-YA3AvQ_Yb- fil-Genre-Djibouti.pdf 65B5iF56WO7ZZlq7zHZ1kUwB_fHT5PGCmAD5T9M- J4pc4bOekyg Alemu, M. (2013). The Mada’a and Mablo of the Afar: Customary System of Conflict Transformation. Dahir, M. (2015). Researching the Legal System of the Republic of Djibouti. https://www.nyulawglobal.org/ Allendorf, Keera. (2007). “Do Women’s Land globalex/Djibouti.html Rights Promote Empowerment and Child Health in Ne- pal?” World Development 35 (11): 1975–88. Djibouti Ministry of Women and the Family. (2019). https://www.sciencedirect.com/science/article/pii/ Rapport d’Examen National Approfondi Etabli à l’oc- S0305750X07001416 casion du 25ème anniversaire de la quatrième Con- férence mondiale de Beijing sur les Femmes Et les 5 Andrabi, Tahir, Jishnu Das, and Asim Ijaz Khwaja. ans du Programme de développement durable des (2002). “The Rise of Private Schooling in Pakistan: Nations Unies à l’horizon 2030.  Catering to the Urban Elite or Educating the Rural Poor?” Background Paper for the Pakistan Poverty As- Doss, Cheryl R. (2006). “The Effects of Intrahouse- sessment, World Bank, Washington, DC. Processed. hold Property Ownership on Expenditure Pat- http://economics-files.pomona.edu/Andrabi/Re- terns in Ghana.” Journal of African Economies 15 search/Pakschool%20March29.pdf (1): 149–80. https://academic.oup.com/jae/arti- cle/15/1/149/753288?login=true Annuaire Statistique Éducation Djibouti 2019/2020. (2020). http://mail.education.gov.dj/index. Duflo, Esther. (2003). “Grandmothers and Grand- php?option=com_k2&view=item&task=down- daughters: Old-Age Pensions and Intrahousehold load&id=42_4c405e1b517c7d660d3208d0f4e87172&I Allocation in South Africa.” World Bank Economic temid=1700&lang=en Review 17 (1): 1–25. https://academic.oup.com/wber/ article/17/1/1/1676291?login=true Anos Casero, P., & Seshan, G. K. (2006). Public-private sector wage differentials and returns to education in Dumas, Christelle, and Sylvie Lambert. (2010). “Educa- Djibouti.  World Bank Policy Research Working Paper, tional Achievement and Socio-Economic Background: (3923). https://openknowledge.worldbank.org/bit- Causality and Mechanisms in Senegal.” Journal of stream/handle/10986/8660/wps3923.pdf?sequence=1 African Economies 20 (1): 1–26. https://www.research- gate.net/publication/5162647_Educational_achieve- Attanasio, Orazio, and Valérie Lechene. (2002). “Tests ment_and_socioeconomic_background_causality_ of Income Pooling in Household Decisions.” Review and_mechanisms_in_Senegal of Economic Dynamics 5 (4): 720–48. https://www.sci- encedirect.com/sdfe/reader/pii/S1094202502901912/ EDSF/PAPFAM. (2012). Deuxième Enquête Djibouti- pdf enne sur la Santé de la Famille. http://ghdx.healthda- ta.org/record/djibouti-family-health-survey-2012 Ben Hamida, A., Hunersen, K., Houssein Barkat, H., Doualeh Aden, Y., Hersi Doualeh, R., Elmi, N., Met- Fédération internationale des ligues des droits de zler, J., & Robinson, W. C. (2021). Prevalence of l’homme. (2011). Cahier d’exigences pour le respect child marriage among Yemeni and Somali refugee des droits des femmes à Djibouti https://tbinternet. and host communities in Djibouti, 2019. Journal of ohchr.org/Treaties/CEDAW/Shared%20Documents/ Refugee Studies. https://academic.oup.com/jrs/ DJI/INT_CEDAW_NGO_DJI_49_8567_E.pdf 88 DJIBOUTI GENDER ASSESSMENT BIBLIOGRAPHY Government of Djibouti. (2018). The National Strat- Institut national de la statistique de Djibouti, Ministère egy for Social Protection 2018-2022. https://sociales. de la femme et de la famille, UNFPA, UNICEF and UE gouv.dj/uploads/Categorie/9b486a2fb64f519b7cb- (2020). Rapport de l’enquête nationale sur les mu- f155e82bf1c67.pdf tilations génitales féminines et les violences faites aux femmes. https://famille.gouv.dj/uploads/publica- Green, S., Snyder, J., Flowers, K. (2017). Refugees, tions/606621408520670ffecb10cdfde29479.pdf Migrants, and the Food Insecure in Djibouti. Center for Strategic and International Studies (CSIS). https:// Jeyaseelan, L., Shuba Kumar, Nithya Neelakantan, www.jstor.org/stable/pdf/resrep23161.8.pdf?ab_seg- Abraham Peedicayil, Rajamohanam Pillai, and Nata ments=0%2Fbasic_search_gsv2%2Ftest&refreqid=- Duvvury. (2007). “Physical Spousal Violence against fastly-default%3A43ed42554e9dd9d17484bdd70d25a Women in India: Some Risk Factors.” Journal of b2d Biosocial Science 39 (5): 657–70. https://www.cam- bridge.org/core/services/aop-cambridge-core/con- Lifos – Centre for Country Information and Country tent/view/807F053280CD8DC8D26D5E7B870C9999/ Analysis in the Migration Area (2018). Somalia: the po- S0021932007001836a.pdf/physical-spousal-violence- sition of women in the clan system. Swedish Migration against-women-in-india-some-risk-factors.pdf Agency (SMA). https://lifos.migrationsverket.se/doku- ment?documentAttachmentId=45863 Katz, Elizabeth, and Juan Sebastian Chamorro. (2003). “Gender, Land Rights, and the Household Economy Haddad, Lawrence, John Hoddinott, and Harold Al- in Rural Nicaragua and Honduras.” Paper presented derman. (1997). Intrahousehold Resource Allocation at the Annual Conference of the Latin American and in Developing Countries: Models, Methods, and Pol- Caribbean Economics Association, Puebla, Mexico, icy. Baltimore: Johns Hopkins University. https://www. October 9. ifpri.org/publication/intrahousehold-resource-alloca- tion-developing-countries-0 Kireyev, M. A. P. (2017). Growth Inclusiveness in Dji- bouti. International Monetary Fund. https://www.eli- Hamida, Amen B., Kara Hunersen, Hemeda H. Barkat, brary.imf.org/view/journals/001/2017/093/001.2017. Yacin D. Aden, Robleh H. Doualeh, Nika Elmi, Janna issue-093-en.xml Metzler and W. Courtland Robinson (2020). Prevalence of child marriage among Yemeni and Somali refugee Kishor, Sunita, and Kiersten Johnson. (2004). Profiling and host communities in Djibouti, 2019. Domestic Violence: A Multi-Country Study. Calver- ton, MD: ORC Macro. https://dhsprogram.com/pubs/ Hindin, Michelle J., Sunita Kishor, and Donna L. Ansara. pdf/od31/od31.pdf (2008). “Intimate Partner Violence among Couples in 10 DHS Countries: Predictors and Health Outcomes.” Koenig, Michael A., Rob Stephenson, Saifuddin DHS Analytical Studies 18, U.S. Agency for Interna- Ahmed, Shireen J. Jejeebhoy, and Jacquelyn Camp- tional Development, Washington, DC. https://dhspro- bell. (2006). “Individual and Contextual Determinants gram.com/pubs/pdf/AS18/AS18.pdf of Domestic Violence in Northern India.” Ameri- can Journal of Public Health 96 (1): 132–38. https:// Hoddinott, John, and Lawrence Haddad. (1995). “Does ajph.aphapublications.org/doi/pdfplus/10.2105/ Female Income Share Influence Household Expendi- AJPH.2004.050872 tures? Evidence from Côte D’Ivoire.” Oxford Bulletin of Economics and Statistics 57 (1): 77–96. https://onlineli- Lundberg, Shelly J., Robert A. Pollak, and Terence J. brary.wiley.com/doi/abs/10.1111/j.1468-0084.1995. Wales. (1997). “Do Husbands and Wives Pool Their tb00028.x Resources? Evidence from the United Kingdom Child Benefit.” Journal of Human Resources 32 (3): 463–80. IFAD and ILO. (n.d.). Promoting women’s empow- http://apps.olin.wustl.edu/faculty/pollak/R_Lund- erment in the Middle East and North Africa. https:// berg_Pollak_Wales_1997_JHR_Do_husbands.pdf www.ilo.org/wcmsp5/groups/public/---ed_emp/docu- ments/publication/wcms_563865.pdf Martin, Sandra L., Kathryn E. Moracco, Julian Gar- ro, Amy Ong Tsui, Lawrence L. Kupper, Jennifer L. 89 DJIBOUTI GENDER ASSESSMENT BIBLIOGRAPHY Chase, and Jacquelyn C. Campbell. (2002). “Do- Rubalcava, Luis, Graciela Teruel, and Duncan Thom- mestic Violence across Generations: Findings from as. (2009). “Investments, Time Preferences, and Public Northern India.” International Journal of Epidemiol- Transfers Paid to Women.” Economic Development ogy 31 (3): 560–72. https://academic.oup.com/ije/arti- and Cultural Change 57 (3): 507–38. https://www.jour- cle/31/3/560/629773 nals.uchicago.edu/doi/pdfplus/10.1086/596617 Martinelli, M., & Ollé-Goig, J. E. (2012). Female genital Schady, Norbert, and José Rosero. (2008). “Are Cash mutilation in Djibouti. African health sciences, 12(4), Transfers Made to Women Spent Like Other Sourc- 412-415. https://www.ajol.info/index.php/ahs/article/ es of Income?” Economics Letters 101 (3): 246–48. view/85069 https://www.sciencedirect.com/science/article/pii/ S0165176508002498 Ministry of Women and Family and DISED (2019). Étude sur l’évolution de la situation de la femme à Dji- Sougueh Guedi, I. and Cavatorta, G. (2018). Les obsta- bouti 2000 à 2018 cles socioculturels à l’élimination de toutes les forms de MGF à Djibouti. FGM/C: From Medicine to Criti- Ministry of Women and Family and UNFPA (2019). cal Anthropology, https://hal.archives-ouvertes.fr/hal- Étude sur la représentation de la femme au sein de la 02274978 fonction publique. Thomas, Duncan, John Strauss, and Maria-Helena Ministry of Women and Family and UNFPA (2019). Henriques. (1990). “Child Survival, Height for Age, and Étude sur la situation actuelle de l’accès des femmes Household Characteristics in Brazil.” Journal of Devel- au crédit à Djibouti. opment Economics 33 (2): 197–234. https://www.sci- encedirect.com/science/article/pii/0304387890900224 OECD. (2019). Social Institutions and Gender Index: Djibouti. https://www.genderindex.org/wp-content/ Thomas, Duncan. (1990). “Intra-Household Resource uploads/files/datasheets/2019/DJ.pdf Allocation: An Inferential Approach.” Journal of Hu- man Resources 25 (4): 635–64. https://www.jstor.org/ OFPRA. (2017). Les mutilations génitales féminines stable/145670?seq=1 (MGF). https://www.ofpra.gouv.fr/sites/default/files/ atoms/files/25_dji_mgf.pdf Thomson Reuters Foundation (2018). Djibouti: The Law and FGM/C. https://www.28toomany.org/static/ Petit, V. (2013). Acteurs et rhétoriques autour d’une media/uploads/Law percent20Reports/djibouti_law_ mutation sociale : Pratiques de l’infibulation et de report_v1_(july_2018).pdf l’excision à Djibouti. Colloque AISLF: Sociologie des mutations, mutations des société. https://hal.ar- United Nations. (2019). Gender equality, justice in law chives-ouvertes.fr/hal-00840324/document and practice: Essential for sustainable development.  Petit, V. (Ed.). (2018). Population Studies and Devel- US Agency for International Development. (2004). Sit- opment from Theory to Fieldwork (Vol. 7). Springer. uation Analysis of Basic Education in Djibouti. https:// https://link.springer.com/content/pdf/10.1007/978-3- genbase.iiep.unesco.org/workspace/applis/epidoc/ 319-61774-9.pdf fichiers/EPIDOC/24989_E025965.pdf Quisumbing, Agnes R., and John A. Maluccio. (2000). United Nations Djibouti. (2020). UN and Partners “Intrahousehold Allocation and Gender Relations: COVID-19 Response Plan: Djibouti. https://reliefweb. New Empirical Evidence from Four Developing Coun- int/sites/reliefweb.int/files/resources/DJI_COVID-19_ tries.” Discussion Paper 84, Food Consump- tion and Emergency_Appeal_22072020_Draft%2025%20 Nutrition Division, International Food Policy Researcy Aout%20FR.pdf Institute, Washington, DC. https://ageconsearch.umn. edu/record/16460/ UNDP (2018). Djibouti: Gender, Justice & The Law. https://www.undp.org/content/dam/rbas/doc/Gen- der percent20Justice/English/Full percent20reports/ 90 DJIBOUTI GENDER ASSESSMENT BIBLIOGRAPHY Djibouti percent20Country percent20Assessment per- U.S State Department. (2020). Trafficking in Per- cent20- percent20English-min.pdf sons Report: Djibouti. https://www.state.gov/re- ports/2020-trafficking-in-persons-report/djibouti/  UNDP. (2020). Human Development Report 2020–the Next Frontier: Human Development and the Anthro- U.S State Department. (2021). 2020 Country Re- pocene – Djibouti. ports on Human Rights Practices: Djibouti. https:// www.state.gov/reports/2020-country-reports-on-hu- UNDP (2020). National Strategy for Financial Inclusion man-rights-practices/djibouti/ in Djibouti, 2021-2026. WFP and UNHCR. (2013). Joint Assessment Mission: UNDP. (n.d). Climate Change Adaptation : Djibouti. Ali Addeh and Holl Holl Camps, Djibouti. https://www. https://www.adaptation-undp.org/explore/eastern-af- unhcr.org/uk/protection/health/54d3762b21/djibouti- rica/djibouti joint-assessment-mission-ali-addeh-holl-holl-camps- djibouti-18.html UNFPA. (2017). Djibouti: Scorecard on Gender-based violence. https://arabstates.unfpa.org/sites/default/ WFP. (2019). Draft Djibouti country strategic plan files/pub-pdf/UNFPA percent20- percent20Djibouti (2020–2024). percent20WEB.pdf White & Case LLP and Child Rights International Net- UNHCR Djibouti. (2020). Rapport de l’évaluation des work (CRIN). (2015). Access to justice for children: Dji- besoins pour la réponse aux violences basées sur le bouti. https://archive.crin.org/en/library/publications/ genre (VBG) durant la crise sanitaire de la COVID-19. djibouti-access-justice-children.html UNHCR Global Report (2000). East and Horn of Africa, World Bank. (1997). Djibouti-Crossroads of the Horn regional overview. https://www.unhcr.org/en-us/pub- of Africa: poverty assessment. https://documents1. lications/fundraising/3e2ebc1e23/unhcr-global-ap- worldbank.org/curated/en/557421468748480282/pdf/ peal-2000-east-horn-africa-regional-overview.html multi-page.pdf UN Human Rights Committee. (2013). Concluding World Bank. (2018). Economic Transformation in Dji- observations on the initial report of Djibouti: Human bouti: Systematic Country Diagnostic. https://elibrary. Rights Committee. https://www.refworld.org/publish- worldbank.org/doi/abs/10.1596/31301 er,HRC,,DJI,53354d0a4,0.html World Bank. (2019). Challenges to Inclusive Growth: UNICEF and UNFPA. (2015). Rapport sur l’Évaluation A Poverty and Equity Assessment of Djibou- de la mise en œuvre de la Stratégie Nationale pour ti. https://openknowledge.worldbank.org/han- l’Abandon Totale à Toutes Formes d’Excision. https:// dle/10986/33032?show=full www.unfpa.org/sites/default/files/pub-pdf/UNFPA_ FGM_FR_WEB_FINAL_FRENCH.pdf World Bank and AFD (2021). Djibouti - MSME Fi- nance Needs Assessment and Policy Recommen- UN Women. (n.d). Frequently Asked Questions: dations (English). Washington, D.C. : World Bank Types of violence against women and girls. https:// Group. http://documents.worldbank.org/curat- www.unwomen.org/en/what-we-do/ending-vio- ed/en/289941616568791338/Djibouti-MSME-Fi- lence-against-women/faqs/types-of-violence nance-Needs-Assessment-and-Policy-Recommenda- tions USAID (2004). Situation analysis of basic education in Djibouti. U.S Department of Labor. (2019). Findings from the Child Labor and Forced Labor Reports. https://www. dol.gov/agencies/ilab/resources/reports/child-labor/ djibouti 91 DJIBOUTI GENDER ASSESSMENT ANNEXES 92 DJIBOUTI GENDER ASSESSMENT ANNEX Annex 1. International and regional human rights treaties LEGAL INSTRUMENTS Signed Ratified Acceded Reservations INTERNATIONAL International Covenant on Civil & Political Rights (1966) 2002 (ICCPR) International Covenant on Economic, Social & Cultural 2002 Rights (1966) (ICESCR) Convention on the Elimination of All forms of 1998 Discrimination Against Women (1979) (CEDAW) Convention Against Torture & Other Cruel, Inhuman or 2002 Degrading Treatment or Punishment (1984) (CTOCIDTP) Convention on the Rights of the Child (1989) (CRC) 1990 1990 With reservations (later withdrawn) Hague Convention on the Civil Aspects of International - - - - Child Abduction (1980) REGIONAL African Charter on Human & Peoples’ Rights (1981) 1991 1991 (ACHPR) (Banjul Charter) African Charter on the Rights and Welfare of the Child 1992 (1990) (ACRWC) African Charter on Human and Peoples’ Rights on the 2003 2005 Rights of the Women in Africa (2003) (ACHPRRWA) (Maputo Protocol) 93 DJIBOUTI GENDER ASSESSMENT ANNEX Annex 2. Non-exhaustive gender relevant national laws • Nationality Code, Law No. 79 of 2004 (Law No. 79/AN/04/5th L). • https://www.ecoi.net/file_upload/1504_1218035297_code-de-la-nationalite-djiboutienne.pdf. • Constitution of Djibouti of 1992, with amendments to 2010. • https://www.constituteproject.org/constitution/Djibouti_2010.pdf. • Penal Code, Law No. 59 of 1995 (Law No. 59/AN/94). • http://www.gendarmerie.dj/Le%20Code%20P%C3%A9nal2.pdf. • Family Code, Law No. 152 of 2002 (Law No. 152/AN/02/4th L). • https://unstats.un.org/unsd/vitalstatkb/KnowledgebaseArticle50735.aspx. • Labour Code of 2006 (Law No. 133/AN/05/5th L). • https://www.ilo.org/wcmsp5/groups/public/---ed_protect/---protrav/---ilo_aids/documents/ legaldocument/wcms_126983.pdf • Law No. 55 of 2009 (Law No. 55/AN/09/6th L) on Violence Against Women Specifically on Female Genital Mutilation. • https://evaw-global-database.unwomen.org/es/countries/africa/djibouti/2009/loi-n-55-an-09-6eme-l- relative-a-la-violence-contre-les-femmes-notamment-les-mutilations • Law No. 133 of 2016 (Law No. 133/AN/16/7th L) on Combating Trafficking in Persons and the Smuggling of Migrants. • http://www.presidence.dj/texte.php?ID=133&ID2=2016-03-24&ID3=Loi&ID4=6&ID5=2016-03-31&ID6=n. • Law on the Status of Refugees in Republic of Djibouti of 2017, (Law No. 159/AN/16/7th L). • https://www.ilo.org/dyn/natlex/natlex4.detail?p_lang=en&p_isn=103613&p_count=1&p_classification=17 • Law N° 66/AN/719/8th L of 2020 on the protection, prevention and support for women and children victims of violence. https://www.presidence.dj/texte.php?ID=66&ID2=2020-02-13&ID3=Loi&ID4=3&ID5=2020-02-13&ID6=n. 94 DJIBOUTI GENDER ASSESSMENT ANNEX Annex 3. Legal gaps in criminal and personal status laws CRIMINAL LAWS Domestic Violence While the Penal Code of 1995 criminalizes acts of violence, Djibouti does not have a separate law that specifically criminalizes domestic violence. Marital Rape While the Penal Code has specific provisions on rape and sexual assault, marital rape is not prohibited and criminalized. Abortion for Rape Survivors Abortion that is not for therapeutic purposes is a criminal offence under Art. 447-448 of the Penal Code, including for survivors of rape. Sex work and anti-prostitution Prostitution is prohibited by Art. 394-398 of the Penal Code. Women who sell sex may be deterred from reporting rape or violence because they risk prosecution for selling sex. PERSONAL STATUS LAWS Minimum Age of Marriage While Art. 13 and 14 of the Family Code, Law No.152 (2002) provide that the minimum age of marriage is 18, a minor can marry with the consent of a guardian or with a judge’s approval if consent is not granted. Male Guardianship over Art. 7 of the Family Code provides that the marriage is formed with the Women consent of both spouses and the guardian of the woman. Marriage and Divorce Women do not enjoy equal rights in marriage and divorce. The wife owes a duty of obedience to her husband. The husband must support the wife financially. The husband may file divorce without providing grounds. The wife may file for divorce without grounds by renouncing her financial rights. Polygamy Polygamy is permitted by Art. 22 of the Family Code. Custody of Children The mother has custody rights, and either parent may grant custody after a divorce. A mother with custody can lose custody if she remarries, unless a court decides it is in the best interest of the child, or if the new husband is within the prohibited degree of relationship to the child. Guardianship of children Fathers have the right to control matters related to education and cultural upbringing. Inheritance Under the Family Code, Sharia rules of inheritance apply. Women have a right to inheritance but in many cases receive less than men. Daughters receive half the share that sons receive. Source: Adapted from UNDP, Djibouti: Gender Justice and the Law, 2018. 95