DOMINICAN REPUBLIC GENDER ASSESSMENT NOVEMBER 2023 Poverty and Equity Global Practice WORLD BANK © [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 This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org Table of Contents List of Figures ............................................................................................................................................... 3 List of Tables ................................................................................................................................................ 4 ACKNOWLEDGEMENTS .......................................................................................................................... 6 GLOSARY.................................................................................................................................................... 7 INTRODUCTION ........................................................................................................................................ 8 1.UNCOVERING THE GAPS ................................................................................................................... 13 1.1 ACCUMULATION OF ENDOWMENTS ........................................................................................... 13 1.2 ACCESS TO ECONOMIC OPPORTUNITIES ................................................................................... 20 1.3 WOMEN’s AGENCY .......................................................................................................................... 30 1.4 CONCLUSION ..................................................................................................................................... 38 2. PRIORITIZATION OF GAPS................................................................................................................ 39 3. DEEP ANALYTICAL DIVES ............................................................................................................... 42 3.1 VIOLENCE IN SCHOOLS .................................................................................................................. 42 3.1.1: PREVALENCE AND IMPACTS OF SCHOOL VIOLENCE ......................................................... 43 3.2 TEENAGE PREGNANCIES ................................................................................................................ 56 3.3 FEMALE LABOR FORCE PARTICIPATION AND WAGE GAPS ................................................. 74 REFERENCES ........................................................................................................................................... 94 ANNEX 1. The legal and institutional framework for gender equality ...................................................... 99 ANNEX 2: A characterization of poverty by gender, 2019 ...................................................................... 102 ANNEX 3 – Prioritization of gender gaps ................................................................................................ 103 A3.1 Methodology .................................................................................................................................... 103 ANNEX 4. Risk factors associated with adolescent pregnancy ............................................................... 107 ANNEX 5. Initiatives aimed at developing agency, aspirations, and alternative life projects in the Dominican Republic ................................................................................................................................. 108 ANNEX 6. National Institute for Comprehensive Early Childhood Care (INAIPI) ................................ 109 ANNEX 7. Ñopo wage gap decompositions ............................................................................................ 110 2 List of Figures Figure 1: gender Inequality Index, 2021 ......................................................................................... 9 Figure 2: The WDR analytical framework ................................................................................... 10 Figure 3: Mortality rate (per 1,000 live births) among infants and under-5 children 2019 .......... 13 Figure 4: School enrollment, gross, gender parity index (GPI) 2019 .......................................... 14 Figure 5: Completion rates 2019................................................................................................... 14 Figure 6: Harmonized tests scores in LAC Countries, 2020 ........................................................ 15 Figure 7. Expected years of education and learning-adjusted years of school – 2020 ................. 16 Figure 8. Share of youth (15-24 years) not in education, employment, or training (%), 2020.... 17 Figure 9: Youth not attending school, 2019-2020 ........................................................................ 17 Figure 10. Self-reported reasons for leaving school (18-23 years old)......................................... 17 Figure 11. Adolescent fertility rate ............................................................................................... 18 Figure 12. Teenage pregnancies by income quintile, 2013. ......................................................... 18 Figure 13: Maternal mortality rate (per 100,000 women), modeled estimate .............................. 19 Figure 14: Female vis-à-vis male labor force participation. ......................................................... 21 Figure 15: Unemployment rate by gender and level of education (2019) .................................... 22 Figure 16. Employment distribution (percentage) across economic sectors and occupations in 2019............................................................................................................................................... 23 Figure 17. Hourly wage gaps and share of women employment by economic sector, 2019 ........ 24 Figure 18. Average weekly hours worked by economic sector, 2019 .......................................... 24 Figure 19. Average weekly hours worked by age, 2019............................................................... 24 Figure 20: Female/male ownership of an account at a financial institution ................................. 26 Figure 21. Household composition among monetary poor households, 2019.............................. 28 Figure 22. Incidence of multidimensional poverty, 2000-21by household configuration and gender of head, % households, 2019. ........................................................................................... 28 Figure 23. Contribution of selected factors to changes in poverty rates in the Dominican Republic, by national/urban/rural domains ................................................................................... 29 Figure 24: Evolution of poverty by gender, 2001-2020 ............................................................... 30 Figure 25. Decomposition of poverty changes in 2019-2020, by household composition........... 30 Figure 26. Child marriage in LAC: % of women (aged 20-24 years) married or in union, 2019. 31 Figure 27: Women should only work if their partner does not earn enough ................................ 33 Figure 28: Ever experienced physical violence, % of women by partner/husband, 2007-2013 ... 34 Figure 29: National representation of women (%) ....................................................................... 37 Figure 30: Local representation of women (%) ............................................................................ 38 Figure 31. Share of students self-reported as victims of bullying ................................................ 44 Figure 32. Share of students affected by violence, comparison by gender (%)............................ 44 Figure 33. Share of students affected by violence, by region and income group (%) .................. 44 Figure 34. Share of students affected by violence, comparison by gender (%).. Error! Bookmark not defined. Figure 35. Index of learning outcomes by quintiles of violent acts .............................................. 45 Figure 36. Index of learning outcomes by quintiles of social interaction environment ............... 46 3 Figure 37. Selected enablers of school violence and policy recommendations ............................ 49 Figure 38. Marriage status of female population aged 15-19, by poverty status and pregnancy experience ..................................................................................................................................... 57 Figure 39. Pregnancy status of the female population aged 15-19, by poverty status .................. 59 Figure 40. Adolescent girls aged 15-19 by school attendance status at the time of pregnancy, by multidimensional poverty status ................................................................................................... 60 Figure 41. Female population aged 15-19 that stopped attending school according to educational level reached, by poverty level ..................................................................................................... 61 Figure 42. Knowledge of selected health care services among female adolescents (15-19), by pregnancy experience and multidimensional poverty status ........................................................ 63 Figure 43. Drivers of teenage pregnancy and selected policy recommendations ......................... 64 Figure 44. Reasons for not returning to school after pregnancy among 15–19-year-olds, by poverty status ................................................................................................................................ 70 Figure 45. gender wage gap in the Dominican Republic .............................................................. 76 Figure 46. Decomposition of the corrected gender wage gap ...................................................... 76 Figure 47. Wage gap decomposition in Dominican Republic (Total employment for population aged 20-55) – 2019 ....................................................................................................................... 78 Figure 48. Evolution of employment and participation rates, by gender (2000 - 2021)............... 79 Figure 49. Reasons for not working or searching for a job among economically inactive population in 2021 ........................................................................................................................ 80 Figure 50. Average hours worked, and care activities by socio-economic group and gender (2016) ............................................................................................................................................ 80 Figure 51. Hourly gender wage gap (2002-2021) ......................................................................... 81 Figure 52. Share of (informal) women employed by economic sector (2019) ............................. 81 Figure 53. Decomposition of changes in the employed population Q2 2019-2020 for the longitudinal sample, by gender, social characteristics, and location. ........................................... 82 Figure 54. Change in the probability of job loss between 2019 – 2020 (Q2) ............................... 83 Figure 55. Drivers of jobs losses and wage gaps and associated key policy recommendations ... 85 Figure 66. Internal efficiency indicators of adult and youth education program vis-a-vis regular high school program, 2018-2019 .................................................................................................. 88 List of Tables Table 1: A lifecycle approach to assess gender inequality ........................................................... 12 Table 2: Characteristics of workers pre-COVID (2019) by gender. ............................................. 20 Table 3. Adolescent girls by marriage and schooling status, ages 15-19 (%) .............................. 32 Table 4. Key selected sex gaps for the Dominican Republic with respect to four groups of countries ........................................................................................................................................ 39 Table 5. Violence in schools: percentage of teachers according to perception of school climate (public sector) ............................................................................................................................... 43 Table 6. Violence in schools: percentage of parents according to frequency of reported violent acts (public sector) ........................................................................................................................ 43 Table 7: Potential impact of violence in school on student learning, PISA ................................. 47 Table 8: Marginal potential impacts of violence in school on students’ experience in school, PISA .............................................................................................................................................. 47 Table 9: Marginal potential impacts of violence in school on various health outcomes .............. 48 4 Table 10: Simulated benefits of ending violence in schools......................................................... 48 Table 12: Estimates of the returns to educational attainment, 17 years of data............................ 61 Table 14. Percentage of workers with a labor income less than or equal to the minimum wage, by firm size, Q1 2019......................................................................................................................... 90 List of Boxes Box 1: What’s behind high maternal mortality in the Dominican Republic?............................... 19 Box 2: Are wage gaps in the Dominican Republic similar across the population? ...................... 25 Box 3: Is there a way out to tackle GBV in the Dominican Republic? ........................................ 35 Box 4. Factors contributing to violence against children, including in schools ........................... 50 Box 5. Implementation of the Peace Culture Protocol ................................................................. 53 Box 6. Lifetime losses in earnings for women who had become pregnant as adolescents and drop out of school may be significant. .................................................................................................. 61 Box 7. Theories of Change for Interventions Targeting Adolescent Girls ................................... 65 Box 8 Guatemala’s Abriendo Oportunidades program ................................................................ 67 Box 9. The Effects of Supérate school transfers on high school completion ............................... 68 Box 10. What happened to women participation and employment during the pandemic crisis? . 82 Box 11. Communities of Care Program: Towards the establishment of a National System of Care in the Dominican Republic ........................................................................................................... 86 5 ACKNOWLEDGEMENTS This study was prepared by a team led by Alejandro de la Fuente and comprised of Christian Gomez, Carmen de Paz, Magdalena Lizardo, Gloria Ortega, Mariel Siravegna and Quentin Wodon. Overall guidance for this project was received by Ximena del Carpio and Alexandria Valerio. Peer reviewers for the report were Gabriel Demombynes, Jonna Lundvall and Karla Dominguez. The team greatly benefited from conversations with and comments from Miriam Muller, Pedro Rodriguez, Eliana Rubiano Matulevich, Monica Yañez Pagans, Juan Baron, Jose Antonio Pellerano, Madiery Vásquez, Jose Mola, Jimena Llopis Abella, and Ana Maria Munoz Boudet. Preliminary findings and recommendations from this study were presented through a series of consultations with government counterparts from the following institutions: Ministry of Women; Ministry of Economy, Planning and Development; Ministry of Education; National Statistical Office; Cabinet of Social Policy; Ministry of Presidency; National Anti-poverty Program Supérate; and CONANI. The team thanks all of them for their comments and suggestions. Laurence Blair edited this report. 6 GLOSARY Social Subsidies Administrator in Dominican ADESS MICs Multiple Indicator Cluster Surveys by UNICEF. Republic BRAC Non-governmental development organization MMR Measles, Mumps, and Rubella CAFI Child and Family Care Centers MPI Multidimensional Poverty Index CAIPI Comprehensive Early Childhood Care Centers MSMEs Micro, Small and Medium Enterprises CCT Conditional Cash Transfer MSP Ministry of Public Health Nationally determined contributions are at the CEPREV Centro de Prevención de la Violencia in Nicaragua NDC heart of the Paris Agreement and the achievement of its long-term goals. CICH Behavioral Intervention Center for Men NEET Not in Employment, Education or Training CIPAF Research Center for Women's Action OBPP Olweus Bullying Prevention Program European Budgetary support program for the CN Caregivers Network PAPSE II education sector of the Dominican Republic CONADIS National Council for Disability PER World Bank Public Expenditure Review CONANI National Council for Childhood and Adolescence PISA Programme for International Student Assessment National Plan for Equality and Sex Equity 2019- CPN Centers of First Level of Health Care PLANEG 2030 Committee for the Assured Availability of National Plan for the Reduction of Pregnancies in DAIA PREA-RD Contraceptive Supplies Adolescents Basic Education Program for Adults and Distance DHS Demographic and Health Surveys PREPARA High School Maternal, Child and Adolescent Health Division of National Program for the Comprehensive Health DIMIA PRONAISA the MSP Care of Adolescents Continuous Labor Force Survey in Dominican Strategic and Special Projects Program of the ECNFT PROPEEP Republic. Started from 2017. Presidency Conditional Cash Transfer Program in Dominican ENDESA DHS survey for Dominican Republic PROSOLI Republic Labor Force Survey in Dominican Republic. From ENFT PUCMM Pontificia Universidad Católica Madre y Maestra 2000-2016. Ugandan program aiming to change community ENHOGAR Life Standard Survey in Dominican Republic SASA attitudes, norms and behaviors around gender, violence against women and children. FGM Female genital mutilation SDSS Dominican Social Security System GANA Cabinet for Children and Adolescents SENASA National Health Insurance GBV gender Based Violence SNS National Health Services GGI gender Gap Index SP Social Protection HCI Human Capital Index SRH Sexual and Reproductive Health United Nations Special Representative of the ICT Information and Communication Technology SRSG Secretary-General on Violence against Children ICV Life Quality Index STD Sexually transmitted diseases Instituto Dominicano de Evaluación e Investigación IDEICE STEM Science, Technology, Engineering, and Math de la Calidad Educativa Monetary incentive for school attendance in ILAE STI Sexually transmitted infections elementary school The government's main poverty reduction ILO International Labor Organization SUPERATE strategy in the Dominican Republic. Third Regional Comparative and Explanatory National Institute for Comprehensive Early INAIPI TERCE Study is a regional study on the quality of Childhood Care education in Latin America. INFOTEP National Institute of Professional Technical Training TSS Social Security Treasury IPV Intimate Partner Violence UMC Upper middle-income countries JEE Extended School Day Program UNAP Primary Care Units KOICA Korea International Cooperation Agency VAW Violence Against Women LARC Long-acting reversible contraceptives VIAS Violence in and around schools 7 INTRODUCTION Achieving inclusive growth and maximizing poverty reduction in the Dominican Republic requires closing existing gender gaps: from early childhood to working age, and further still into old age. Using a lifecycle approach, this gender Assessment attempts to uncover, better understand, and deliver some policy recommendations for the main challenges in this area, with a focus on the three main dimensions of endowments, economic opportunity, and agency. The rationale The Dominican Republic experienced robust economic growth since the turn of the century up to 2019, becoming a regional success story. The country’s economy grew at an average rate of 5.7 percent per year from 2004-2019 and had the fourth-highest GDP growth rate among countries in Latin America and the Caribbean (LAC) between 2008 and 2016 (Winkler, H and Montenegro M., 2021). However, the pace of poverty reduction has not been proportional to economic growth. Poverty fell from 50 percent in 2004 to 32 percent in 2016. This decline in poverty continued from 2017 to 2019 under a new (methodologically adjusted) series, from 29 to 25 percent. While representing a significant improvement, this rate of poverty reduction is comparable only to the average experience of the Dominican Republic’s slower-growing LAC peers. Considering that the Dominican Republic’s per capita GDP grew almost twice as fast as the regional average, a much faster reduction in poverty could have been expected. This suggests that the nature of GDP growth in the Dominican Republic could have been more inclusive. In particular, gains have not been commensurate between women and men. The poverty rate has been consistently higher among women than for men between 2003 and 2019 and little progress has been accomplished closing this poverty gap between men and women. In 2019 the poverty rate for women was 22 percent (vs 19 for men). Across different types of household configuration (i.e., unipersonal, adult couple with children, other combinations of adults with and without children, etc.) women-headed households systematically report higher poverty rates (see Annex 2). The health and economic crises triggered by the COVID-19 pandemic increased poverty, especially among women. The brunt of job losses stemming from the pandemic and containment measures in the Dominican Republic fell on women, resulting in higher female poverty rates than among males for all age groups. During the first semester of 2020, poverty increased by 2.6 percentage points among women (compared to June 2019), but only by 2.2 points among men. And in the same year, for every 100 poor men there were 118 poor women, up from 116 in 2019. Women in cities were particularly affected by this increase in poverty (MEPYD, 2020). The historical and recent imbalances in poverty rates between women and men in the Dominican Republic are likely rooted in multiple gender disadvantages. Unequal investments in health and education between girls and boys start from an early age, and are then amplified through limitations in access to productive and income generating activities in the transition into adulthood, and in the constraints on the decision-making capacity of Dominican women and girls that prevents them from fully participating in all public spheres of life. Indeed, international rankings indicate that the Dominican Republic still has a long way to go in closing existing gender gaps. For example, the country ranks 106 out of 170 countries worldwide in the 2021 UNDP gender Inequality Index, which covers indicators from all three dimensions of gender equality 8 (reproductive health, empowerment, and the labor market).1 This is well below the LAC average, and at the level of other poorly performing regional peers such as Venezuela and Guatemala (see Figure 1). The Dominican Republic also ranks low in the gender Gap Index compiled by the World Economic Forum that focuses on economic aspects: it came 89th out of 156 in 2021, close to Venezuela and Brazil.2 Figure 1: gender Inequality Index, 2021 0.7 0.6 Gender Inequality Index 0.5 0.4 0.3 0.2 0.1 0 HTI CRI MEX ECU LCA PAN NIC BRA VEN URY ARG PER GTM JAM STP CHL CUB TTO SLV BOL COL DOM HND PRY Source: UNDP. Note: 1 imparity, 0 parity Aside from being a moral imperative, reducing gender inequality makes sense from an economic and social perspective.3 It has been widely demonstrated that boosting gender equality can enhance productivity and improve development outcomes for the next generation. By contrast, allocating resources based on gender has been shown to lead to inefficient outcomes and diminished productivity. As an example, in the presence of well-functioning labor markets, educated women would enter the labor force and make an efficient use of their talents and skills not only to their benefit, but also to that of their families, communities, and society as a whole. Higher female labor force participation, for instance, is associated with lower infant mortality rates and higher life expectancy.4 Children born to very young mothers tend to have higher risks of under-five malnutrition and mortality. Part of the reason is that some young mothers may simply not be ready yet to give birth. When mothers are themselves poorly nourished, this may put their children at higher risk of intrauterine growth restriction. gender disparities tend to be starkest among the poorest in society, contributing to the intergenerational transmission of disadvantage, and ultimately undermining efforts to attain shared prosperity. Gender inequality can have sizeable costs for the wealth of the country. Instead of measuring losses from inequality as annual flows (the GDP approach), estimations for this study based in human capital (the wealth approach) show that the loss of human capital wealth due to gender inequality represented $185.4 billion in 2018, equivalent to 2.2% of GDP for that year.5 As women account for 41 percent of human capital wealth versus 59 percent for men, women’s human capital could increase from US$313.1 billion to US$454.6 billion if women earned as much as men. According to other recent estimations, the potential 1 The UN gender inequality index focuses on maternal mortality ratios, adolescent birth rate, share of women in parliaments, population with at least some secondary education, and labor force participation rates. 2 The Global gender Gap Index scores countries from 0 (imparity) to 1 (parity) across four thematic dimensions —Economic Participation and Opportunity, Educational Attainment, Health and Survival, and Political Empowerment. However, it particularly focuses on economic aspects such as wage gaps and occupational segregation. 3 Achieving gender equality and empowering all women and girls is the fifth sustainable development goal. 4 Mateo Díaz and RodriÌ?guez-Chamussy (2016) in World Bank (2018); World Bank (2012). 5 Human Capital Wealth is defined as the present value of the future earnings of today’s labor force for individuals aged 15 and above. 9 productivity gains from incorporating women who completed some level of education have not entered the labor market could amount to as much as 4 percent of GDP in the Dominican Republic. In particular, the country may be foregoing as much as 40 percent of its human development potential due to gender inequality.6 Closing existing gender gaps will be imperative to boosting shared prosperity and maximizing poverty reduction over the coming years. Gender inequality can drag on the Dominican Republic’s aspiration to become a high-income country in 2030, especially considering its current stage in the demographic transition. As an early dividend country, the difference between a potentially detrimental youth bulge and a beneficial, dynamic young workforce depends on whether the Dominican Republic can close existing gender gaps from early childhood up to working and old age.7 The country could benefit from a demographic dividend if it could further reduce the fertility rate and create employment for young workers, especially for women. This could reduce population growth and thereby increase human capital wealth per capita and well-being. It is therefore imperative to uncover and better understand existing gender gaps in the country as a first step to tackling them through existing or new policies. Doing so will in turn boost shared prosperity and maximize poverty reduction. Framework This assessment applies the analytical framework devised by the World Development Report (WDR) 2012 on gender Equality and Development, and the World Bank Group (WBG) gender Strategy 2016-2023. gender equality is both a pre-requisite for and a result of inclusive growth. According to this framework, achieving inclusive growth requires closing the gaps between men and women in three main areas: • Endowments, which primarily include investments in education and health, but can also encompass physical and financial capital in the form of land and access to credit, respectively. These endowments can determine the capability of women to fulfill their potential as individuals and in society. • Access to economic opportunity, which refers to the possibility to take part in paid and productive employment, entrepreneurship and business management, and to access productive assets and obtain equal returns for those activities through wages and profits. • The agency of women and girls, which is their capacity to make decisions for themselves, and act on those decisions. When this is weak, it translates into unequal outcomes in the other two areas. Informal institutions (e.g., social norms and aspirations), markets (e.g., market failures), formal institutions (e.g., laws and services), household decisions (e.g., in the distribution of key resources), and policies all combine to mediate the opportunities of women vis-à-vis men in terms of endowments, economic opportunity, and agency (see Figure 2). Figure 2: The WDR analytical framework 6 UNDP (2019) 7 World Bank (2016). Early dividend counties are characterized by falling fertility rates and an increasing labor force. In order to fully reap the benefits of the first demographic dividend generated by these favorable circumstances, early dividend countries are recommended to invest in maternal health, education that does not leave girls behind, and women ’s empowerment and access to family planning. In the future, and as the population ages, women’s access to economic opportunities will be similarly important. 10 Source: Adapted from World Bank (2012) When combined with a life-cycle lens, the analytical framework expands the identification of gender gaps, allowing for better diagnostics and improving the efficiency of policy design (see table 1). This assessment aims to identify priority gaps and barriers along three crucial periods of life: • Gender gaps start to build up during early years and childhood, mainly in the form of differences in human capital investments in key endowments such as health and education. These will lead to potential discrepancies in cognitive and soft-skills development that will shape long-life opportunities. Gender gaps during this period are reflected in differences in access to and attainment in education from pre- primary to the lower-secondary level, in early and child health outcomes, in the incipient presence of patriarchal social norms among school aged children, and in early manifestations of gender-based violence, including forced child marriage. • Over adolescence and youth, early and childhood discrepancies tend to consolidate, translating into wider or harder to overcome gaps in health and education outcomes and agency. This is a time when gaps in agency emerge, through social norms that relegate women to the household and care activities or prevent young women from using contraception, translating into high rates of teenage pregnancy. Indeed, in the transition to the upper secondary or higher levels, where most dropout occurs, women can be (often permanently) excluded from the educational system and employment (NEETs). In the rare cases where girls pursue education beyond the secondary level, gender segregation by field of study (also partly based on internalized aspirations) will often condition their self-selection into lower- productivity and lower-earning occupations in the future. • During adulthood, and more specifically over the peak reproductive and work years, gender gaps become particularly palpable in economic opportunities, mostly in connection with differences in access to the labor market and productive assets such as finance or land, constraints on entrepreneurship and on time use, and high rates of maternal mortality. Over the entire period of adulthood, including old age, the challenges faced by women are also manifested in the form of limited agency, for instance evidenced by their low political representation, their exposure to different forms of gender-based violence, the high prevalence of patriarchal social norms constraining women to certain roles, and their higher vulnerability to poverty. 11 Table 1: A lifecycle approach to assess gender inequality Early years and childhood Adolescence and youth Adulthood and old age Endowments -Gaps in early care -Teenage pregnancy -MMR and maternal health -Gaps in mortality and health -Early drop-out from secondary -Incidence of chronic outcomes education condition and mortality -Gaps in enrollment and -gender segregation by field of -Fertility rates, life attainment in primary and lower study expectancy and ageing secondary education Economic opportunity - (Potentially permanent) -Gaps in access to labor exclusion from both the markets and wages educational system and the -Poor quality of employment labor market (NEET Constraints on use of time phenomenon) Limited access to assets -Scarcity of time to engage in -Gaps in entrepreneurship work activities -Lack of SP (especially in old age) Agency -GBV against girls -Consolidation and -Limited political -Early unions and marriage internalization of traditional representation -Building constrained social norms and aspirations -GBV against pregnant aspirations due to patriarchal that relegate women to HH and women and in old age social norms in the family, care activities -Prevalence of patriarchal schools, and media -GBV against teenagers social norms that prevent (cyberviolence, harassment, adult women from becoming sexual violence) active members of society - Legal and formal institutional framework, policies, infrastructures, and service delivery. There is a good correspondence between the indicators on gender gaps that result from the proposed analytical framework and the Plan Nacional de Igualdad y Equidad de Genero 2019-2030 (PLANEG III), which is part of the National Development Strategy of the DR and has detailed knowledge of the specific gender challenges faced by the country and is organized in six pillars: Education for Equality; Integral Health for Women; Economic Autonomy; Citizenship, Democracy and Political and Social participation; Sex-Based Violence Against Women; and Digital Technologies for Women’s Autonomy (see Box A3.1). Broadly speaking, the indicators explored in section 1 under endowments correspond to the pillars of education and health within PLANEG III; those under economic opportunities fall under the pillar of economic autonomy; and those under agency relate to those contained under political and social participation; gender-based violence and women’s autonomy. This gender Assessment will be organized in three sections. Section 1 will describe and quantify the prevailing gaps between men and women throughout the lifecycle (noting that for some age groups the information may be thin or absent), in the fields of endowments, economic opportunities and agency, while attempting to identify some of the major factors that explain them. Section 2 will propose some gender gaps that should be prioritized. The prioritization will be informed through benchmarking and contrasting key selected indicators on gender gaps in the country against those in other groups of countries. This data will guide the team’s judgement and subsequent engagement, through virtual consultations with key stakeholders and government counterparts to validate the priorities. Section 3 will examine in-depth some of the identified priorities to better understand their drivers followed by concrete policy recommendations. This approach will be supplemented with consultations and reviewing key policy documents to assess the virtues and shortcomings of existing programs and policies to tackle gender disparities. 12 1. UNCOVERING THE GAPS This section summarizes the key findings of a review on work completed inside and outside the World Bank. The review was supplemented by some analytical work and helped the team develop a list of gender gaps organized around three main domains of endowments, economic opportunities, and agency throughout the life cycle of people following the proposed analytical framework and in connection to DR’s defining characteristics. 1.1 ACCUMULATION OF ENDOWMENTS This sub-section reviews the existing gender gaps in the Dominican Republic with regards to health outcomes, teenage pregnancy, educational enrolment, and attainment. Differences in the accumulation of endowments or human capital, especially in the form of health and educational outcomes since early childhood and up to adolescence and youth, will shape the skills and competencies of girls vis-à-vis boys. Poor outcomes in these areas will translate into constrained life-long opportunities for economic, social, and political participation among women. The analysis highlights the strikingly high incidence of adolescent pregnancies in the country, which imposes a large (both direct and opportunity) cost on the Dominican society. Partly in connection with this phenomenon, maternal mortality is also concerning and comparatively high, as is the rate of young women who do not work or study (NEETs). 1. Some gender gaps to the advantage of girls exist in early-years health and in education during childhood In the Dominican Republic, boys show poorer early health outcomes than girls, while gaps in access to pre-school are negligible. Child mortality rates in the Dominican Republic are well above the LAC and UMC averages (see Figure 3), and close to those registered in countries with much lower incomes. This may be associated with inadequate pre-natal care and persistently high rates of adolescent pregnancies. Infants at birth and under-5 boys register higher mortality rates than girls, as well as slightly higher rates of malnutrition (height for age), wasting, severe wasting, and overweight (under-5s only).8 With regards to access to early education, nevertheless, gender parity has been a constant for the past decade. In 2019, the gross enrolment ratio in early childhood education was about 56.4 percent for boys and close to 57.9 percent for girls.9 Figure 3: Mortality rate (per 1,000 live births) among infants and under-5 children 2019 35 30.6 Per 1000 live births 30 25.6 25.2 25 21.2 17.9 20 15.3 14.5 14.4 15 12.3 12.3 12.2 10.3 10 5 0 Female Male Female Male Infant Under-5 Dominican Republic Latin America & Caribbean Upper middle income countries Source: Health, Nutrition and Population statistics, World Bank 8 Based on the latest available data - WB 2013 9 Education Database, World Bank 13 Beyond the early years, gender gaps (favoring girls) exist in educational enrolment. The gender Parity index (GPI), calculated as the quotient of the number of females by the number of males enrolled, show almost identical enrollment for girls and boys in primary education. However, a gender gap to the advantage of girls emerges at the secondary and tertiary levels: disparity for the female-male enrollment rate is around 1.09 at the secondary level and 1.4 for tertiary education (Figure 4). The share of women enrolled and, consequently, the GPI in tertiary education for Dominican Republic (1.44 women per men), is particularly high as compared to the LAC and UMC averages (1.16 and 1.23, respectively, in 2019). A gender gap to the detriment of boys is also reflected in attainment and completion rates. Despite similar completion rates for boys and girls (93.9 and 92.7 respectively) at primary level, the gap grows at secondary level: 87 percent of girls vis-à-vis 80 percent of boys completed lower secondary education (Figure 5). Progression rates into secondary education show a similar trend: 94.8 percent among girls compared to 91.6 percent among boys in 2017. Figure 4: School enrollment, gross, gender Figure 5: Completion rates 2019 parity index (GPI) 2019 1.6 1.60 1.4 1.40 % of relevant age group 100 Gender Parity Index 1.2 1.20 1.0 1.00 80 0.8 0.80 0.6 0.60 60 0.4 0.40 0.2 0.20 40 Female Male Female Male 0.0 0.00 Primary Lower secondary Primary Secondary Tertiary Dominican Republic Dominican Republic Upper middle income countries Upper middle income countries Latin America and the Caribbean Parity Latin America and the Caribbean Source: WDI and ECNFT Source: WDI and ECNFT Note: UMC – Upper Middle-Income Countries. Tertiary estimates for DR are based on 2018 figures. Girls outperform boys at school on learning outcomes, but only at lower educational levels. The learning gap between boys and girls in the Dominican Republic varies across the educational level, except for reading, where girls outperform boys in all grades. In general, the sex learning gap between girls and boys in math is smaller than in reading, as found in the PISA results for 2018 (OECD, 2019b). In lower educational levels, the girls outperform boys in both reading and math, but in higher levels, boys outperform girls in all subjects tested except reading. This result is similar to the findings by Treviño et al. (2015) when they analyzed the results of countries participants in the Third Regional Comparative and Explanatory Study (TERCE). Specifically, the sex learning gaps in the Dominican Republic across educational levels are similar in both international (comparing students in primary, TERCE 2013, with 15-year-old students mostly in secondary, PISA 2018) and national assessments. Girls and boys alike fare poorly in enrolment and completion rates when compared to other countries. Female secondary education completion rates for those aged 25 years or over appear especially low when comparing the average rates over the last decade: 39 percent for DR, compared with 44 percent for LAC and 52 percent in UMC countries. The numbers are even more acute in the case of boys: 32 percent in DR compared to 40 percent in LAC and 47 in UMC. 14 The Dominican Republic lags other countries in educational achievement and learning as well. According to Barón and Mola (2019), the performance of Dominican students is lower than expected given the country’s level of GDP per capita. When comparing harmonized test scores, for instance, the DR has the lowest level among comparator countries in LAC (figure 6). The performance is poor for girls and boys alike, but more so for boys: Dominican girls scored on average 4 percent higher than boys, the largest such gap for LAC countries with available data. According to the Dominican Republic’s national standardized test, just 12 percent of children in third grade achieved satisfactory levels in Spanish, and 27 percent in mathematics, in 2017. And the Dominican Republic has the lowest scores for reading, mathematics, and science among 15-year-olds when compared to other LAC countries that participate in the Program for International Student Assessment (PISA). Scores in these areas moreover displayed some of the strongest drops among PISA-participating countries since 2015: for example, average reading test scores in the DR declined in 2018 (World Bank, 2021). Figure 6: Harmonized tests scores in LAC Countries, 2020 500 Harmonized test scores 450 400 350 300 250 200 150 100 CHL SLV MEX CRI ECU COL GTM ARG HND PAN DOM Female Male Source: Human Capital Index - WB The low academic performance of Dominican students is mainly explained by low preschool attendance, high repetition rates, and the low efficiency of the public education sector measured as the ratio between learning outcomes and expenditures per pupil (World Bank, PER). Both Treviño et al. (2015) and Castillo and Mola (2016) use TERCE information to analyze the factors that explain students' poor performance in the Dominican Republic relative to the rest of LAC. Better-off children have better learning outcomes and most of the variation on student performance is explained by differences across schools. This indicates that students with similar economic and social backgrounds are clustered in the same schools. In addition, preschool attendance, positively correlated with learning outcomes and school attendance, is low among Dominican children.10 Many students do not pass the academic year because they accumulate learning lags over the years. The quantity and quality of education could be a crucial bottleneck in accessing jobs. When comparing the expected years of education of the future labour force, the DR has one of the lowest levels among comparator countries (figure 7), about 10 percent lower than that of Chile. When adjusting for the quality of education, the gap becomes even larger. The learning-adjusted expected years of education for the Dominican Republic represent about 72 percent of the corresponding value for Chile for girls and 67 percent for boys. The poor quality of education cuts across the entire population: According to PISA 2018, socio- 10 León (2018) finds similar results for the Dominican Republic using 15-year-old students tested in PISA 2015 information. The preschool attendance rates of 3rd and 6th graders in the Dominican Republic are 17.6% and 19.3%, respectively, compared with 32.6% in the region. Additionally, in the country, 26.9% and 35.4% of students in third and sixth grade did not attend school at least two times a month versus 28.6% and 23.8% in the region, respectively. 15 economic status explains just 9 percent of the variance in reading performance (OECD average: 12%) and the average difference between advantaged and disadvantaged students in reading is 65 points, a large gap albeit somewhat better than the average of 89 points in OECD countries. Figure 7. Expected years of education and learning-adjusted years of school – 2020 Female Male 14 14 12 12 10 10 8 Years 8 Years 6 6 4 4 2 2 0 0 COL DOM MEX PER PAN COL CHL CRI ECU URY MEX HND DOM ARG NIC PER PAN SLV CRI ECU ARG URY CHL HND NIC SLV Expected Years of School Learning-Adjusted Years of School Expected Years of School Learning-Adjusted Years of School Source: Human Capital Index – WB. Note: Expected Years of School is calculated as the sum of age-specific enrollment rates between ages 4 and 17. Learning-adjusted years of school are calculated by multiplying the estimates of expected years of school by the ratio of most recent harmonized test scores to 625. In sum, Dominican girls stay longer in school and learn more than boys, but many of them do not complete their basic education and perform poorly when compared to girls in other countries in the region. As girls enter adolescence, gender roles and social norms may affect their ability to stay in school and to transition into work, particularly among the poorest. This is analyzed in the next section. 2. Large gender gaps detrimental to young women arise before transitioning into the labor market, some driven by exceptionally high rates of teenage pregnancy NEET status is a growing and concerning phenomenon in LAC and globally.11 The transition from school to work is crucial in life, with young people between 15 and 24 that are neither in school, nor employed, nor in training being of special concern. Young people out of school and work represent around a fifth of that age group globally and continue to be a large and particularly vulnerable group in LAC, especially in rural areas .12 Faced with the scarcity of quality education and poor employment prospects, these young people can be easily trapped in a circle of low aspirations, poverty, and violence. Emerging evidence from developing countries indicates that this condition tends to become chronic, translating into long-term and often permanent labor market exclusion.13 Young women are more likely than men to become NEETs, largely in connection with social norms and traditional roles often leading to early relationships and marriage, teenage pregnancy, and to a skewed distribution of household and care responsibilities.14 Indeed, women are up to four times as likely as young men to be out of school and work in some LAC countries.15 The Dominican Republic is no exception: Despite their educational advantages as per several indicators, Dominican women are clearly over-represented among NEETs. According to recent 11 People aged 15-24 who are neither in employment, education, nor training. 12 Alfani et al. (2020); de Hoyos et al. (2016) 13 Alfani et al. (2020); Machado and Muller (2018) 14 De Hoyos et al. (2016) 15 Székely and Karver (2016); de Hoyos et al. (2016) 16 estimations, by 2020 there were around 600,000 thousand NEETs in the Dominican Republic and roughly 65 percent of them were women. The share of young women not in education, employment, or training, has been systematically higher than that of young men since 2004. In 2020 around 39 percent of young women were NEETS compared to 25 percent of young men (see Figure 9). These percentages are above the LAC average, although the gender gap is very similar. Figure 8. Share of youth (15-24 years) not in Figure 9: Youth not attending school, 2019-2020 education, employment, or training (%), 2020 50 60 48.3 48.0 40 50 35.4 37.1 Percentage 40 Percentage 30 27.0 29.4 30 20 19.121.6 20 10 10 0 0 CRI ECU PER DOM COL PRY ARG(Urban) URY CHL MEX BRA BOL Female Male Female Male 12 to 23 years 18 to 23 years 2019 2020 All Female Male Source: Lac Equity Lab and WDI, World Bank. But what drives these high rates of NEET women in Dominican Republic? Teenage pregnancy is strongly linked with not working nor studying among Dominican adolescents. A recent World Bank study found that 44.5 percent of the adolescents who reported having been pregnant were neither employed nor attending school or university, compared to just 12.5 of those who reported never having been pregnant. That same study found that while 83.2 percent of adolescent girls who were never pregnant were in school or college, only 45.1 percent of those who reported having been pregnant did.16 In another recent UNDP study, around 14 percent of pregnant girls in the Dominican Republic reported being expelled from schools.17 Family reasons, which very likely include becoming pregnant and household chores, are the main reason for school dropout for girls and young women according to the most recent Labor Force Surveys in the country (ENCFT). By contrast, lack of interest in school and joining work are the primary reasons for men to leave their studies (figure 11). The different reasons for leaving school reported by Dominican girls and boys reflect differentiated gender roles in society. Figure 10. Self-reported reasons for leaving school (18-23 years old) Male Female 16 World Bank (2020) 17 UNDP (2018) 17 Lack of interest Familiar reasons Work Lack of interest Expensive Expensive Other Work Familiar reasons Completed studies Physical or mental disablity Other Other Without documentation Completed studies Other Bad results Physical or mental disablity Without documentation Bad results Age Never enrolled Never enrolled Age 0 20 40 0 10 20 30 40 50 2020 2019 2020 2019 Source: Author’s calculation based on ECNFT. Teenage pregnancy in the Dominican Republic is exceptionally high by global standards.18 Although it has decreased since 1996, the adolescent fertility rate in the Dominican Republic, at 93 births per 1,000 women ages 15-19 in 2018, was much higher than the LAC average (62.1) and almost three times the average for Upper-middle-Income countries (32.8) (Figure 11). Figure 11. Adolescent fertility rate Figure 12. Adolescent fertility rate by wealth quintile, 2019. 160 121 140 Per 1000 women ages 15- 19 101 120 81 100 61 41 80 21 60 1 40 2010 1996 1998 2000 2002 2004 2006 2008 2012 2014 2016 2018 20 Dominican Republic Latin America & Caribbean 0 1 2 3 4 5 Upper middle income countries Source: WDI Source: ENHOGAR-MICS, 2019 3. In adulthood, gender gaps in health are evident, especially in sexual and reproductive health. Non-communicable diseases (NCD)19 explain 72 percent of total deaths in the country and risk factors vary significantly by gender. Both cardiovascular disease and diabetes are among the world’s leading causes of death and continue to be on the rise as well as in Dominican Republic (Castro, B. et. al., 2018). Hypertension is the second leading cause of death in the Dominican Republic, while diabetes is the sixth leading cause of death. According to the WHO (2019) Dominican adult men aged 30 – 70 years old have higher risks of premature mortality than women (22 percent vs. 16 of women in the same age group). For men, the top risk factors are the harmful consumption of alcohol (12 percent) and tobacco (18 percent) and high blood pressure (22 percent). All these are more preponderant than in women (2, 8 and 18 percent 18UNDP (2018) 19 NCDs are non-infectious and non-transmissible chronic diseases that are divided into four categories – cardiovascular disease (CVD), diabetes (DM), cancers, and chronic respiratory diseases 18 respectively). By contrast, physical inactivity affects 42 percent of women and one in three women aged 18 or more years old are obese. These rates are significantly lower among men (33 and 20 percent, respectively). While mortality between 16 and 60 years old tends to be higher among Dominican men, maternal mortality is a concerning and growing phenomenon in the country. The male mortality rate (probability of dying between 16 and 60 years old) in 2018 was 197.3 per 1,000 adults, compared to 114.6 among women. Mortality due to chronic conditions (between ages 30 to 70) is also higher among men (21.7 compared to 16.5 for women), as it is for mortality due to suicide (16.8 compared to 3.1) and air pollution (50 compared to 36).20 However, Dominican women between 15 and 49 still die at a very high rates relative to other countries in the region (and especially other UMI countries) due to pregnancy and delivery complications: 95 per 100,000 women died giving birth in 2017. Indeed, there has been a substantial increase in maternal mortality over the past 20 years, making the gap with other countries grow larger (see Figure 14). Figure 13: Maternal mortality rate (per 100,000 women), Lack of access to maternal services does modeled estimate not seem to explain the high level of maternal deaths. The percentage of births attended by skilled staff was 99.6 percent in 100 2015 (above the LAC and UMC averages of 94.7 and 97.7 percent respectively in 80 2016). 98 percent of women received Per 10000 women prenatal care in 2014, at the level of UMC countries on average in 2016 and above the 60 LAC average. By contrast, the low quality of services, limited female empowerment, 40 and the high prevalence of unsafe abortions Dominican Republic Latin America & Caribbean have been signaled as the main reasons Upper middle income behind the striking level of maternal 20 mortality in the Dominican Republic (see 2001 2003 2005 2007 2009 2011 2013 2015 2017 box 1 below).21 The Criminal Code penalizes abortion under any circumstances Source: WDI (art. 317), including for any person who assists it in any way. Box 1: What’s behind high maternal mortality in the Dominican Republic? A recent qualitative study of maternal mortality in the Dominican Republic found that the lack of knowledge of both health staff and service users about the right to respectful and quality attention could amount to “obstetric violenceâ€?. It also confirmed the insufficiency of support for pregnant women during delivery, and of adequate equipment and materials. The low empowerment of women, their low socioeconomic status, and their exposure to unsafe abortions were also key factors.22 The illegal nature of abortion under any circumstances in the country and the prosecution of those who help may lead families and women to only seek medical help when it is too late. Low educational levels and the high incidence of poverty are other important factors.23 Finally, it is worth noting the high prevalence of c-sections in the country; c- 20 Health, Nutrition and Population, World Bank 21 Cruz (2013); Instituto Tecnológico de Santo Domingo (2016) 22 Báez et al. (2017) 23 Cruz (2013); Instituto Tecnológico de Santo Domingo (2016) 19 sections expose women to a greater risk of mortality. In 2018 over 58 percent of all deliveries in the Dominican Republic were c-sections, one of the highest rates in the world.24 1.2 ACCESS TO ECONOMIC OPPORTUNITIES The accumulated endowments by women and men over their childhood and youth, be it in the form of health or education, will largely define their prospects and opportunities to generate income and be productive during adulthood. The analysis presented in this sub-section reveals that, despite comparatively higher levels of endowments among Dominican women, they are faced with especially high barriers to becoming economically autonomous. Unjustified gender gaps in wages, occupational segmentation by gender into less productive and more poorly paid activities, shorter working lives, and higher unemployment and unpaid roles, all combine to contribute to the higher incidence of poverty among women of all age groups in the Dominican Republic as compared to men. 1. Women’s opportunities to earn a living through employment are constrained vis-à-vis those of men, especially during their peak reproductive years Employed women are more educated and hold more formal and salaried jobs than men. According to the ECNFT 2019, the proportion of high-skilled women workers among employed women is double the proportion of males (36 vs 18 percent). Employed women register a lower informality rate (36 percent) than men (51 percent). Their jobs are more suitable to be performed from home than male jobs. And the share of female jobs located in urban areas is higher than the share for men (Table 2). Table 2: Characteristics of workers pre-COVID (2019) by gender. Female Male All (1) (2) (3) (1) (2) (3) (1) (2) (3) Hours Labor Hours Labor Hours Labor Employment worked income Employment worked income Employment worked income (average) (average) (average) (average) (average) (average) All 100.0 36.7 18109 100.0 42.0 22185 100.0 39.9 20533 Age group 15-24 13.1 38.9 11769 16.5 41.3 13033 15.1 40.5 12590 25-34 25.1 39.1 16568 24.7 43.4 21336 24.8 41.7 19383 35-44 25.3 37.2 19312 22.1 43.3 25220 23.4 40.7 22625 45-64 31.2 34.7 21548 29.9 41.4 25428 30.5 38.7 23813 65 and over 5.3 30.2 15521 6.8 36.7 24304 6.2 34.5 21254 Education Low skilled 63.7 36.6 11303 81.6 42.4 18495 74.3 40.4 15997 High skilled 36.3 37.0 30079 18.4 40.1 38628 25.7 38.4 33719 Informality Formal 63.5 37.6 23116 49.0 43.8 27121 54.9 41.0 25241 Informal 36.5 35.3 9411 51.0 40.3 17465 45.1 38.6 14826 Area Rural 14.6 36.6 12599 20.1 41.2 17045 17.9 39.7 15574 Urban 85.4 36.8 19052 79.9 42.2 23481 82.1 39.9 21614 Category 24 INTEC (2016) - BC 2015 20 Self 29.7 32.3 12277 50.0 39.8 21502 41.8 37.7 18844 employed Salaried 70.3 38.7 20584 50.0 44.2 22873 58.2 41.5 21751 Work from home suitability No 81.6 36.7 13406 89.6 42.3 19111 86.4 40.2 16926 Yes 18.4 37.0 39036 10.4 39.1 48802 13.6 38.0 43458 Presence of children No 43.1 35.9 20150 52.6 40.9 21508 48.7 39.2 21020 Yes 56.9 37.3 16565 47.4 43.1 22930 51.3 40.6 20071 Essential sector No 83.8 36.4 17782 78.7 40.7 21998 80.8 38.9 20224 Yes 16.2 38.7 19971 21.3 46.7 23157 19.2 44.0 22068 Source: Author’s calculations based on ENCFT-2019 Note: High-skill education is defined as population with at least one degree of tertiary education. Essential sectors are defined following Decreto 134-20 and Decreto 153-20, which announced the list of economic and essential service sectors allowed to operate throughout the lockdown. These included: health services, private security work, press and media, electricity distribution, transportation, and distribution of goods, supplies and fuel, energy, water, telecommunications, and solid waste collection services. Work from home suitability is defined as those workers in the top 25 percent of the WFH index distribution (see Cueva et al. 2021 for the definition and computation of the WFH index). Unfortunately, Dominican women participate in the labor market far less than men. Twenty years ago, the participation of men in labor markets was twice as high as that of women. Women have been gradually catching up. But even in 2019, only around 51 percent of women were working or actively looking for a job compared to around 78 percent of men (see Figure 14). The male participation rate has kept very close to the average rate observed in other Upper-middle-Income countries and LAC peers. By contrast, the female participation rate has remained much lower, only recently starting to catch up with the LAC average levels. The pandemic crisis impacted hardly the participation of women in DR, by 2021 after a partial recovery for both men and women the participation gap has remained invariant. Figure 14: Female vis-à-vis male labor force participation. 90 80 DR Male 70 60 50 Percent 40 DR Female 30 20 10 0 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 2021 DR Female DR Male LAC Female LAC Male UMC Female UMC Male 21 Source: World Bank WDI (ILO estimate, % of population +15 years old) Women’s inactivity in labor markets is partly explained by their disproportionate shouldering of household and family care responsibilities. In 2016, around one third of women were devoted to household and family care activities, while 39 percent were not on the job market because of their studies. By contrast, two thirds of inactive men were studying, while just less than 2 percent of inactive men were homemakers or carers. The inactivity rate is especially high among women in their principal reproductive years (25-49 years old). This high rate of female economic inactivity is due to the social role assigned to them: women bear the brunt of the responsibility for family care, which limits their ability to join the labor market.25 26 The latest available data shows that Dominican women invest over three times more hours than men in care activities, and slightly above the observed regional average. The 2016 time-use survey shows that women allocate more than 31 hours to unpaid work per week, compared to 9.6 hours among men. This difference is larger for rural women, and in poorer regions. Adding up the number of paid and unpaid working hours per week, women work 4 hours more than men. 27 Around one in three Dominican women are economically dependent. Despite the growing number of women who earn an income, 31 percent of women still depend on someone for their subsistence. This is in line with the LAC average (30.8 percent), but over twice the share observed for men (13.5 percent). Rural women (37.3 percent) and those aged 15-24 (50.5 percent) were even more likely to be economically dependent than men. Economic dependence weakens the bargaining position of women at home and heightens their vulnerability.28 Figure 15: Unemployment rate by gender and level At an economically active age, Dominican of education (2019) women are more likely to experience unemployment. The unemployment rate for women, at 8 percent, was more than twice than for men (at 4 percent) in 2019. For young people 16 14 14 (ages 15-24), the rates are much higher among 12 both groups, but the female rate is still around Percentage 12 10 10 9 7 7 8 twice that of males. There are also differences in 8 5 unemployment rates by education level and 4 3 4 gender. The unemployment rate for women with 0 basic education was 2.5 times higher in 2019 than Female Male Female Male the rate for comparable men (10 vs 4 percent), and DR LAC around twice that observed among men at the intermediate and advanced levels (Figure 15). Basic Intermediate Women with tertiary education are more likely Advanced than men to have a job, but gender segregation Source: WDI by sector and occupation persists. While tertiary-educated men are 6.6 percentage points more likely to have a job, the corresponding figure for women is 34.4 percentage points. On the other hand, women with low-to-medium levels of education do not have many opportunities in the labor market. 29 Most working women (90 percent) are engaged in service-related activities such as commerce and hotels, bars and restaurants. This is followed by manufacturing (8.1 percent) and agricultural activities (2 percent). Men have more diversified jobs: 69 percent of occupied men were in the service sector, 19.9 percent were found 25 INTEC (2016) - BC 2015 26 INTEC (2016), Observatorio de Políticas Sociales y Desarrollo (2017b) 27 ONE (2018) in PLANEG III; World Bank (2020) 28 ENHOGAR has a time use module, but it has not yet been analyzed, as far as the authors of this note are informed. 29 World Bank (2020) 22 in agriculture, and 10.9 percent in industry.30 However, in recent years there has been an increase in the number of women working as managers, administrators, professionals, and office employees.31 Women are concentrated in sectors and occupations traditionally associated with lower productivity and lower pay. As figure 16 shows, more than half of the women employed in the country are occupied as service and unskilled workers in areas such as retail commerce and domestic work. Women account for 37 percent of the jobs in the ICT sector and less than 3 percent of positions that require technical and technological skills. In addition, only 2.9 percent of company directors are women.32 The growing and significant share of women engaged in domestic work is especially concerning because domestic workers are poorly protected by law.33 Figure 16. Employment distribution (percentage) across economic sectors and occupations in 2019 Source: Author’s calculations based on ECNFT-2019 Note: Darker green cells represent higher shares of employment in the specific sector and occupation. Women also earn less than men. The average labor income for women is lower than for males across a broad range of sociodemographic characteristics, including age, area, educational attainment, and informality status. The biggest remuneration differences are presented among informal, low skilled, and self-employed work, where average male labor income can be up to 75 percent more than the female average (table 2). 30 Observatorio de Políticas Sociales y Desarrollo (2017b), INTEC (2016) 31 Observatorio de Políticas Sociales y Desarrollo (2017b) 32 PLANEG III (CIPAF and ENFT) 33 INTEC (2016) 23 Even when employed in the same sector, women make less money. Employed women make on average 85 percent of men’s total labor income earnings (table 2). This pattern is more acute in sectors where women have higher shares of participation (figure 17). The wage gap is even larger in the informal sector, where women make only 60 percent of men’s earnings on average.34 The raw gender pay gap (i.e., the simple difference between hourly earnings across genders) has remained largely stable since 2000, hovering between 6 and 16 percent (Winkler and Montenegro, 2021). Wage gender gaps in the Dominican Republic are large even by regional standards, as noted in chapter 2 of this report. Figure 17. Hourly wage gaps and share of women employment by economic sector, 2019 1.8 Construction 1.6 Hourly wage gap ratio (female/male) 1.4 Public administration Dominican pesos Transport and 1.2 communications Finance and banks Health 1 Electricity and water 0.8 Agriculture Education Industry 0.6 Services Commerce Hotels, bars, restaurants 0.4 0 10 20 30 40 50 60 70 80 Share of women in sector Source: Author’s calculations based on ENCFT-2019 Note: The size of the circles represents the size of the sector in terms of total employment. Note 2. The agriculture sector includes fishing and mining. Dominican women work fewer hours than their men counterparts. In addition to the lower hourly payments received, women work fewer hours across all economic sectors.35 Men work 5 more hours per week than women on average (table 2). This gap increases to up to 7 hours in sectors with high female participation like commerce and hotels, bars and restaurants (figure 18). Across the life cycle, the weekly worked hours gap widens with age, starting with about 2 hours for those between 15-24 years old and reaching more than 6 hours from age 35 onwards (figure 19). Figure 18. Average weekly hours worked by Figure 19. Average weekly hours worked by economic sector, 2019 age, 2019 34 MEPyD (2017) in PLANEG III 35 With exception of those women employed in construction, where their participation is only 3% of total employment in that sector. 24 45 Hotels, bars, restaurants Transport and… Commerce 40 Industry Weekly hours Services 35 Public Administration Finance and banks Health 30 Construction Agriculture 25 Electricity and water Education 0 10 20 30 40 50 20 15 to 24 25 to 34 35 to 44 45 to 54 55 to 64 65 and over Average hours worked Male Average hours worked Female Male Female Source: Author’s calculations based on the ENCFT. Box 2: Are wage gaps in the Dominican Republic similar across the population? Most studies on gender wage gap use the difference between the median (or mean) wage of men and the median (or mean) of women to measure wage inequality between sexes. However, the size of the gap can vary widely at different parts of the wage distribution. For example, the gender wage gap could be larger at the top quantiles of the wage distribution, which may suggest the existence of a glass ceiling that prevents women from securing the best-paid jobs. Similarly, large wage gaps at the bottom of the distribution may imply the existence of a “sticky floorâ€? that restricts women’s career advancement. As such, analysis should be extended beyond the mean or median gender pay gap and analyze the entire distribution of wages instead. In section 3.3, we use data from the Encuesta Nacional Continua de Fuerza de Trabajo to gain more insights into wage differences in the Dominican Republic across the wage distribution. We do this analysis controlling for selection into employment to avoid potential biases derived from sample selection. To that end, we use the methodology proposed by Arellano and Bonhomme (2017) which models selection into employment in a context of quantile regression using a copula-based approach. Furthermore, the literature on gender gaps has also focused on decomposing the differentials in wage between men and women. This technique aims to identify what part of the wage gap stems from observable labor characteristics that can vary across gender, and what part is left unexplained. The latter, referred to in the literature as the “structural effectâ€?, is the portion of the gap explained by differing pay by gender (Albrecht, Bjorklund and Vroman 2003). The former is known as “composition effectâ€? and it is the portion of the gap attributable to gender differences in individual characteristics. We decompose the selection- corrected gender wage gap for the Dominican Republic following Machado and Mata (2015). In general, developing rigorous indicators for the gender wage gap across the distribution is crucial for promoting gender equality and empowering women, especially in Latin America. The COVID-19 crisis has imposed new challenges on women and will likely amplify pre-existing vulnerabilities on their labor market situation. Evidence suggests that, in times of crisis, Dominican women tend to be expelled from the labor market at higher rates than men.36 At the same time, women in the Dominican Republic have jobs more amenable to working from home than men, given that their jobs tend to be less physically intensive.37 A dedicated analysis of the labor market impacts of COVID-19 on women in 13 LAC countries using high frequency survey data found that female workers were 44 percent 36 UN Women (2017) 37 Word Bank (2020) 25 more likely to lose their jobs than male workers at the onset of the crisis. As the crisis evolved, temporarily unemployed workers started to go back to work, but the difference in employment losses between men and women persists. In the Dominican Republic, employment rates dropped by 4.3 percentage points for both women and men in 2020 relative to 2019. In 2022, employment had not fully recovered when compared to 2019 levels, and the recovery has been faster for men. Female employment remains 1.2 percent below pre- crisis employment levels while male employment is 0.9 percent lower.38 The deep dive on labor participation during the Covid-19 crisis will explore in detail the drivers of this lower participation for women. Formal institutions, especially with regards with the law, play a key role as facilitators and barriers to gender equality in the job market. Evidence from the Women, Business and the Law database across the world shows that, for instance, where the law ensures greater equality of economic opportunity between women and men, female labor force participation is higher. A correlation also exists between legal reforms and a reduction in the wage gap. Other studies indicate that when both women and men are legally granted equality of opportunity, additional positive development outcomes ensue, including higher investments in health and education among women and their children, lower MMR, and lower rates of STDs.39 Only two differences persist in Dominican legislation between women and men in this area, as per the WBL 2020. First, the government does not provide 100% maternity leave and shared parental leave is not in place. In addition, periods of absence from work due to childcare are not accounted for in pension benefits (see Annex 1). 2. Women’s constrained access to productive assets and entrepreneurship further limit their capacity to generate their own income in adulthood Access to financial products including credit is high Figure 20: Female/male ownership of an among Dominican women compared to the regional account at a financial institution average, albeit lower than for men. The percentage of Dominican women with an account at a financial 80 institution increased between 2011 and 2017 and has a 60 relatively good standing in relation to women within other groups of countries; however, it remains lower 40 than among men (figure 20): In 2017, 54 percent of 20 women had an account at a financial institution compared to 58 percent among men. This gap was larger 0 on average in LAC, where only 52 percent of women 2011 2014 2017 compared to almost 59 percent of men had an account. Male Female The average share of women with access to accounts, at 69 percent, was much larger among UMC comparator countries, but the gender gap was even larger: as much Source: WDI as 77 percent of men in UMC countries on average had an account. Around 42 percent of commercial credits in the Dominican Republic correspond to women, despite them being assessed as more responsive debtors. However, there seems to be growing interest from commercial banks in this group. As an example, almost half of the mortgage and commercial credit of the largest bank in the country is dedicated to women. 40 Yet women are underrepresented among entrepreneurs and business ownership. Around 32.2 percent of companies had female participation in ownership in 2016, still below the 45 percent observed (in 2019) 38 See https://www.bancentral.gov.do/a/d/2541-encuesta-continua-encft 39 World Bank (2020) 40 Fernández (2018) in PLANEG III 26 for LAC or 37 percent for UMC countries.41 Although only 13 percent of small and medium-sized enterprises are fully owned by women,42 the share of micro-businesses owned by women appears to be higher: around 62 percent of individual owners of micro, small, and medium businesses were women in 2013. This is especially the case among newer (under 10 years old) companies. However, female-owned businesses show less proclivity to grow than those owned by men.43 Indeed, according to a recent study, around 51.3 percent of micro-businesses in the Dominican Republic are owned by women, and roughly half of them have been formalized compared to close to two thirds of those owned by men.44 Women are less present than men in business management. Only 21.2 percent of firms had a female top manager in 2016, compared to 22 and 20 percent on average in LAC and UMC countries, respectively.45 Only 5.4 percent of firms managed by women in the manufacturing and services sectors in the Dominican Republic engage in export sales, significantly below the 14.7 percent average for firms managed by women across LAC and the 13.4 percent of male-managed Dominican firms that export. 46 Around half of the middle and senior management of companies in the Dominican Republic is female.47 COVID-19 seems to have affected female-led businesses differently. A recent World Bank study on the impact of COVID-19 using data generated by the Business Pulse surveys found larger closures of businesses led by women; with this gap larger for medium and small size businesses and in the accommodation and restaurant sector. It also found that the businesses led by women respond more strongly (i.e., reducing the number of employees) to negative expectations.48 Another study on the impact of COVID on micro and small businesses led by women in the Dominican Republic concludes that female-owned businesses have been particularly affected. Most of these operate in commerce and services, specifically beauty parlors and hairdressers, clothing and shoes shops, and restaurants and cafes. More than half of these women (51.4 percent) reported not being able to work from home at all. Around 14 percentage points more businesses owned by women report substantial declines in revenue during the pandemic compared to those owned by men.49 3. The lower capacity to generate incomes among Dominican women has led to the femininization of poverty in the country, especially during the peak reproductive years and in old age Economic and poverty gains since the turn of the century have not been commensurate between women and men in the Dominican Republic. The country grew 5.7 percent per year between 2004 and 2019, and the poverty rate dropped steadily from 50 percent in 2004 to 32 percent in 2016. The decline in poverty continued from 2017 to 2019 under a new (methodologically adjusted) series, falling from 29 to 25 percent. However, women tend to register higher poverty rates across all age groups. An examination of household composition shows that for the entire population and for all age groups female poverty rates are higher than males (see Annex 3). Marital status is also correlated with poverty rates for women, with divorced women appearing to be more vulnerable.50 According to WBL 2020, the legal framework in the Dominican Republic does not recognize the same rights for women as men when it comes to divorce. 41 WDI 2021 42 ONE (2015) in World Bank (2018) 43 See PLANEG III 44 UNDP (2020) 45 WDI 2021 46 World Bank (2018) 47 WDI (2021) 48 Apedo-Amah et al. (2020) 49 UNDP (2020) 50 World Bank (2018) 27 Households with children and female heads are overrepresented among the monetary poor. As section 3.3 will show, this is likely attributable to lower wage and job rates for women than men. Multidimensional poverty shows a lower incidence in women and in female-headed households. 62 percent of households in the country contain children, but 88 percent of poor households have children. Single- adult female households with children account for 13 percent of poor households: more than double their representation among total households (figure 21). Lower participation rates, fewer working hours, high dependency ratios, and wage gaps all contribute to the feminization of monetary poverty in the country. Figure 21. Household composition among monetary poor households, 2019. Notes: 1) The percentages in the cells refer to the share of the category among poor households, the numbers in parentheses refer to the share of the category among all households. 2) Smaller boxes refer to households with senior members only and unipersonal households. By contrast, the incidence of multidimensional poverty, as measured by the Quality-of-Life Index (ICV), is lower among women and in female-headed households. In 2019, according to the ICV, 19 percent of men were moderately poor and 4 percent extremely poor. These percentages were 17 percent and 2.7 percent, respectively, for women. The same trends are observed for female-headed households’ when compared with male-headed households, indistinctly of the household configuration (see Figure 22). However, it is worth noting that the reduction in multidimensional poverty between 2000 and 2021 was faster among males and male-headed households. Figure 22. Incidence of multidimensional poverty, 2000-21by household configuration and gender of head, % households, 2019. Multidimensional Multidimensional extreme poor (% poor (% households) households) Share of Male Female Male Female female head head head head heads Adult couple with children 16.4 23.7 18.4 4.6 1.1 Other adult combinations without children 38.6 15.7 12.3 3.2 0.7 Adult couple with children and other adults 35.3 16.3 22.0 2.0 1.4 Multiple adults, only female with children 18.8 23.2 22.5 2.1 1.5 Adult couple, no children 30.3 23.1 12.7 4.0 1.1 One adult, female with children 91.4 25.0 22.3 3.4 3.7 28 Other adult combinations with children 84.3 22.1 20.7 2.2 1.5 Another household configuration 45.8 25.3 19.6 3.4 2.6 Unipersonal 32.9 44.5 16.9 16.2 2.7 Senior only, no children 44.4 28.3 30.4 6.6 5.1 Total 37.1 25.0 19.0 5.6 2.2 Source: Authors, based on information provided by SIUBEN calculations using data from ENFT and ENCFT. Note: Multidimensional Poverty is proxied by the Quality-of-Life Index (ICV) and is defined as those individuals who are in households categorized as ICV1 and ICV2. Increasing labor participation and labor incomes among women explain almost one third of the poverty reduction in the country in recent years. Labor income and employment are overall the highest contributors to poverty reduction on recent years (figure 23). Despite the lower remuneration and inclusion rates of women in labor markets, the closing gap in female participation rates and incomes contributed 30 percent of the reduction in poverty rates between 2008 and 2016, and 28 percent between 2017 and 2019. Figure 23. Contribution of selected factors to changes in poverty rates in the Dominican Republic, by national/urban/rural domains 2008 - 2016 2017 - 2019 2019 - 2020 8.0 2.0 1.3 1.4 Poverty change (percent points) Poverty change (percent points) 0.8 1.1 Poverty change (percent points) -2.0 -4.1 -3.1 0.0 4.0 1.0 -5.4 -0.8 -0.7 1.4 1.2 1.0 -2.2 -1.8 0.8 -2.1 0.6 -6.0 -3.5 -2.1 1.8 2.1 -1.9 1.1 -2.3 0.0 -1.9 -1.2 -5.3 -0.6 -0.5 -4.0 -0.5 -0.6 -3.5 -10.0 -2.9 -4.1 -1.1 -0.9 -1.0 -0.5 -5.6 -2.9 -1.0 -4.0 -2.5 -14.0 -0.7 -1.5 -8.0 -8.0 -18.0 National Urban Rural National Urban Rural National Urban Rural Source: Author’s estimations based on the LFS Encuesta Nacional de Fuerza de Trabajo (ENFT, 2001-2016) and Encuesta Continua Nacional de Fuerza de Trabajo (ECNFT, 2017-2020) The sanitary and economic crises triggered by the COVID-19 pandemic led to a deep contraction in labor market participation, particularly among women, putting previous gains at risk. More than 300,000 people stopped working or looking for employment between 2019 and 2020. Women represented 42 percent of the labor force in 2019 but accounted for 54 percent of those who stopped working or looking for a job in 2020. In other words, the female participation rate fell by around 8 percentage points while male participation did so by 5 percentage points. Job losses led to income drops and increased poverty with pronounced effects on women, particularly for single mothers and those with children. Women bore the brunt of job losses stemming from the pandemic. As a result, of the 270,000 people that joined the ranks of the poor during 2020, 160,000 were women. In relative terms, poverty increased by 2 percentage points among men, but by 2.6 percentage points among women (Figure 24). By household composition, almost 4 out 5 households that felt into poverty in 2020 contained children (figure 25), and households with only one female adult with children 29 (predominantly single mothers) represented almost 20 percent of the households that fell into poverty in 2020. Figure 24: Evolution of poverty by gender, 2001-2020 60 50 Poverty rate 40 30 24.6 20 22.0 10 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Male Female Source: Authors calculation based on ECNFT. Figure 25. Decomposition of poverty changes in 2019-2020, by household composition Source: Authors calculation based on ECNFT. Note: The sum of all areas represents the total poverty increase from 2019 to 2020. To summarize, monetary poverty appears to be highly feminized in the Dominican Republic. Although female labor force participation has improved over time, it is still much lower than that of men. Lower participation rates in labor markets, fewer working hours, and wage gaps lie at the heart of the femininization of poverty. Women still concentrate in lower productivity sectors and occupations that are traditionally regarded as “femaleâ€?, such as services and trade. 1.3 WOMEN’s AGENCY Women’s opportunities to accumulate endowments, to generate their own income and, ultimately, to fulfill their potential as individuals and members of society, depend to a large extent on their capacity to make decisions and act on them – put another way, their agency. When agency is weak, it translates into unequal outcomes in access to health, education, and economic opportunities. At the same time, as women’s and girls´ human capital and capacity to generate their own income increase, so will their autonomy and empowerment in the household and community. Important gaps in this area remain in the Dominican 30 Republic throughout the lifecycle, as signaled by the high incidence of child marriage, other forms of gender-based violence, low female political participation, and the persistence of patriarchal social norms from a young age. 1. Dominican girls are highly exposed to a particularly concerning form of gender-based disenfranchisement during childhood: Early and forced marriage Child marriage has large impacts on a wide range of development outcomes for girls who are married early, their children and families, and society at large. Child brides are at greater risk of experiencing a Figure 26. Child marriage in LAC: % of women (aged 20- range of poor health outcomes, having 24 years) married or in union, 2019. children at younger ages when they are not yet ready to do so, dropping out of Before age 15 Before age 18 school, earning less over their lifetimes, and living in poverty compared to their peers who marry at later ages. Child brides are more likely to experience intimate partner violence, have restricted physical mobility, and limited decision- making ability. Most fundamentally, child brides may be disempowered in ECUADOR MEXICO TRINIDAD AND PANAMA COSTA RICA PERU SAO TOME AND EL SALVADOR DOMINICAN BOLIVIA JAMAICA BRAZIL HAITI NICARAGUA COLOMBIA GUATEMALA SAINT LUCIA HONDURAS PARAGUAY URUGUAY ways that deprive them of their basic REPUBLIC PRINCIPE TOBAGO rights to health, education and safety.51 Child marriage therefore has very important social costs. It has been estimated that through its effects on Source: UNICEF, Global database (2020) education, child marriage could reduce women´s earnings in adulthood by 17 percent in the Dominican Republic. Eliminating this practice would on the contrary generate gains up to 1.15 percent of the national income, and a reduction in poverty rates – from 30.5 to 27.7 percent.52 The Dominican Republic has the highest prevalence of child marriage in LAC. It is especially common in rural areas and among poorer and less educated girls. According to the National Household Survey (ENHOGAR-MICS 2019), 32 percent of women ages 20-24 at the time of the survey were married or in a union before the age of 18 (9 percent before their 15th birthday). This is the highest rate of child marriage in LAC (see figure 26).53 Incidentally, these girls are more likely to drop out of school than their peers who marry later, and tend to complete fewer years of education (UNICEF 2019). Male and Wodon (2016) indicate that child marriage in the Dominican Republic is associated with lower wealth, lower education levels, and lower labor force participation. Child marriage has also been associated with intrafamiliar violence, and violence against girls.54 Importantly, early marriage or sexual union can lead to earlier pregnancy and school dropout, reinforcing the need to tackle this issue. Table 3 shows that in the latest DHS survey available, one in ten girls is both in school and married (in a formal or informal union). This is a much larger proportion than in many other countries, suggesting that married girls have better opportunities to remain in school than elsewhere. However, the share of girls who are married and out of school is almost as large. 51 Wodon et al. (2017) 52 World Bank (2017) 53 UNICEF 2019 54 INTEC (2016) 31 Table 3. Adolescent girls by marriage and schooling status, ages 15-19 (%) Shares (%) All girls ages 15-19, DHS Data In school, not married, ages 15-16 32.9 In school, not married, ages 17-19 37.3 Out of school, not married, ages 15-16 2.1 Out of school, not married ages 17-19 8.7 Married, in school, any age 10.4 Married, not in school, any age 8.5 Total 100.0 Cultural and gender norms, and low perceived returns to girls’ schooling, feeds into the joint decision- making process about marriage and schooling. Addressing child marriage clearly emerges as one of the priorities from the analysis presented in the next section. 2. Discriminatory informal institutions (social norms) take root from an early age, already constraining to the agency of women during childhood and youth Patriarchal informal institutions can play a key role as facilitators or constraints to gender equality across a wide range of outcomes. Social norms can influence expectations, values, and behaviors, and, as such, can shape the agency of women. Evidence from different contexts indicates the important role those social norms can play, for instance on women’s mobility and bargaining power within the household or with regards to family formation, an area where they tend to be particularly binding.55 Social norms, together with formal institutions, help to mold the internalized aspirations and motivations of girls, young women and female adults, their beliefs about what they can and cannot do, or even their perceived preferences. Evidence from a qualitative study on the NEET phenomenon in Brazil, for instance, confirms that young women’s capacity to form aspirations in relation to work or education is shaped by their social context and the perceived possibilities of what is normal, acceptable, and socially desirable.56 Negative gender stereotypes are already deeply entrenched among younger generations. In a study about violence in schools, 8 out of 10 participants agreed with the statement that women often “drive men crazyâ€?, which could be seen to justify violence against them (74 percent of female and 80 percent of male students). 57 Around 76 percent of male students also reported believing that women often go to the police with false claims of violence just to cause trouble for men. 58 Research has found a strong association between egalitarian social norms and female labor force participation.59 Across LAC (including the Dominican Republic), there is a clear positive association between the share of women in the labor market across countries and the share of Latinobarómetro respondents that disagreed with the statement that women should only work when their male partner does not make enough money (figure 27).60 About 43 percent of Dominican participants in the 2015 55 World Bank (2012) 56 Machado and Muller (2018) 57 INTEC (2016) 58 UNFPA (2015). The study included 1,200 surveys in each area, reaching a total of 7,768 surveys carried out between April and May 2015. The average age in the sample was 15.08 years old, and 58.1 percent of respondents were female. 59 Fortin (2005); Goldin (2006); and Fernández and Fogli (2006) 32 Latinobarómetro agreed with the statement that women should only work if the man does not make enough money. This is one of the largest shares among all the LAC countries included in that survey (see Figure 27). Moreover, the change in the share of participants agreeing with this statement since 2008 is among the smallest in the region, having fallen by just 0.3 percent. This suggests that traditional gender norms are remarkably entrenched in the Dominican Republic.61 Figure 27: Women should only work if their partner does not The transmission of traditional earn enough gender norms occurs not only in the family but also at school. A recent BRA study by Ziffer et. al. (2018) found that CHL staff working in Dominican schools PER URY can hold very traditional views on VEN gender roles and associated ECU ARG characteristics. For example, many PRY think that boys are “naturallyâ€? more COL predisposed to subjects like BOL CRI mathematics than girls. These beliefs NIC are not confirmed by performance DOM MEX results.62. Both fourth and eighth SLV grade-students participating in the PAN HND study also hold traditional conceptions GTM of the roles of women vis-à-vis men. 63 0 20 40 60 80 100 % These patriarchal views consolidate Strongly Agree Agree Disagree Strongly Disagree over adulthood, operating as Don't Know Source: Latinobarómetro 2015 in Muller and Sousa (2018) barriers to women´s political and economic participation. Around 50 percent of men and 37 percent of women reported believing that men are better political leaders than women in 2014 (the remaining shares of the male and female population do not agree with that statement). Around 32 percent of respondents agreed with the statement that politics was for men, 27 percent disagreed with the statement that women should participate in politics more and 35 percent disagreed with the statement that women have the same capacity as men to hold office.64 3. In adulthood, violence against women remains a widespread and extreme constraint to agency Gender based violence is one of the most extreme manifestations of the lack of agency of women, with relevant adverse impacts in all areas of life for them and their families. Gender based violence prevents women’s and girls’ full participation in society, constrains their access to education and economic participation, and undermines efforts to achieve gender equality. Besides the often very serious physical harm resulting from gender-based violence from a partner, the emotional and psychological effects for victims are also multiple and profound. Children of women survivors also suffer lasting consequences, for instance in the form of poorer educational attainment and (mental) health,65 and they are more likely to become victims or perpetrators of violence themselves. Exposure to physical violence is higher among lower-income groups, which may also have negative implications for their access to services and hence 61 Muller and Sousa (2018) 62 Based on a study developed by CEIE (Consorcio de Evaluación e Investigación Educativa) integrated by the Instituto Tecnológico de Santo Domingo (INTEC), la Pontificia Universidad Católica Madre y Maestra (PUCMM) and Albany University. 63 Ziffer et al. (2018) 64 UNDP (2016) 65 Observatorio de Políticas Sociales y Desarrollo (2017) 33 increase their overall vulnerability. GBV has not only devastating consequences for women and their families, but also entails important social costs: according to some estimations, amounting to as much as 3.7 percent of GDP for some countries. According to the latest survey data available (ENDESA 2013), as many as 35 percent of women ages 15- 49 who are married or in union have experienced some form of violence by their partner in their life; emotional violence is the most common (31 percent) followed by physical violence (19 percent). The share of those experiencing physical violence goes up to goes up to 25.1 percent when only considering younger women (15-19 years old). More importantly, the share of women who have ever experienced physical violence by a partner increased between 2007 and 2013 for almost all age groups, and especially so among younger (15-19 years) and middle-aged (40-49 years) women (see Figure 28). Diverse factors are associated with violence. The incidence of violence against women tends to be higher among women with primary and secondary education (versus those with tertiary education); it also decreases with income. Widows or divorced women and women with five or more children are more likely to have experienced violence. Violence is also more common in urban areas.66 Reported cases of gender-based violence have actually been increasing in the Dominican Republic. The number of complaints has grown from around 8,000 in 2015 to over 22,000 in 2021 (MEPYD, 2021). This trend is also reflected in earlier years through the change observed in the ENDESA survey results between 2007 and 2013. At the same time, it is hard to interpret these data: they could indicate the increasing incidence of this phenomenon, but also heightened social awareness and/or increased trust in institutions by female survivors. Figure 28: Ever experienced physical violence, % of women by partner/husband, 2007-2013 % of women who have experienced some kind % of women who have experienced some type of physical violence by age of physical violence by educational level 30.0 25.0 25.0 20.0 20.0 15.0 15.0 10.0 10.0 5.0 5.0 0.0 0.0 without Primary Secondary University / 15_19 20_29 30_39 40_49 Total education Technical 2007 2013 2007 2013 66 https://tbinternet.ohchr.org/Treaties/CCPR/Shared%20Documents/DOM/INT_CCPR_ARL_DOM_28381_S.pdf 34 % of women who have experienced some kind of physical violence by wealth quintile 30.0 20.0 10.0 0.0 Bottom Second Middle Fourth Top Total 2007 2013 Source: ENDESA 2013 Box 3: Is there a way out to tackle GBV in the Dominican Republic? The share of female survivors of GBV who report looking for help has stayed at around 40 percent over the last 12 years up to 2013 according to the DHS. Most of the women who searched for assistance did so among family (50 percent) and friends (10 percent). The share of women victims that looked for help from institutions grew from 4 to 11 percent between 2007 and 2013. It must be noted that there is a need to improve data collection and analysis on this phenomenon in the Dominican Republic, as in other countries, to better understand its incidence, determining factors, and implications. Several different services for GBV survivors exist in the Dominican Republic. Services for women include the Línea Vida and the Casas de Acogida. The Casas de Acogida are shelters that provide temporary protection to women survivors and their children under 14 years old. Between 2008 and 2016 these houses hosted around 2070 women and 1647 children, girls and adolescents. In addition, in 2008 the Centro de IntervencioÌ?n Conductual para Hombres (CICH) was created to provide therapeutical and psychoeducational services to male perpetrators at the judge’s request. The Unidades de AtencioÌ?n Integral a ViÌ?ctimas de Violencia de GeÌ?nero, Intrafamiliar y Delitos genderuales provide counselling, medical assistance, and practical support for victims through specialized teams. There are 17 of these units working across the country.67 68 However, there seems to be a general lack of institutional capacity and coordination to deal with this problem. Different reports indicate that judicial institutions were not prepared to deal with the large and growing number of cases in recent years. Indeed, the Attorney General’s Office reported that the caseload far exceeded prosecutorial capacity, such that only a small fraction of complaints went to court. Civil society organizations also denounce the lack of capacity of the judicial system to adequately prosecute perpetrators, the lack of sufficient shelters, and, overall, the need for further capacity and resources for the adequate enforcement of laws in this area.69 In addition, the number of protection orders issued relative to that of complaints is low. In 2015, 67,224 of gender and intra-family complains were registered, but only 15,093 protection orders were issued. 70 The high level of gender-based violence in the Dominican Republic is seen in sociological research as the byproduct of the cultural conceptions of and roles assigned to men and women. Sociological 67 Observatorio de Políticas Sociales y Desarrollo (2017) 68 No information was available on the effectiveness or outreach capacity of these institutions. 69 https://www.state.gov/documents/organization/265794.pdf 70 INTEC (2016); https://tbinternet.ohchr.org/Treaties/CCPR/Shared%20Documents/DOM/INT_CCPR_ARL_DOM_28381_S.pdf 35 studies contend that the popular codes and rituals of manhood in the Dominican Republic are based on concepts that imply domination over femininity.71 The use of violence is ultimately part of the indoctrination of both men and women since childhood, reflected for instance in the wider freedom afforded to boys and young men compared to women, the encouragement of girls to be submissive, delicate or passive, or the assumption that males are better suited for sciences and mathematics, while women are better at caring roles. 72 Any efforts to change these adverse beliefs and social norms must begin early on in life. Addressing gender-based violence requires an integral approach across sectors, with a focus on prevention, from the early years and up to old age. Prevention efforts should start in the early years, during childhood and in the crucial time of adolescence, and can take various forms with a focus on education (including edutainment), skills building and empowerment, and provision of safe spaces for girls. Recent reviews of interventions indicate the unique potential of comprehensive school-based programs that combine different methods and tools, as well as individual and group awareness-raising activities, that are targeted not only at students but also parents and staff). Similarly, youth drop-in centers that include general health information, such as the Steppingstone initiative in South Africa, appear to be promising. Working with men and boys, communities, and leaders, are components of successful approaches. Another noteworthy program is the Bell Bajao! (Ring the Bell) campaign in India. The campaign was launched in 2008 to call on men and boys to take a stand against domestic violence by performing a simple bystander intervention – ringing the doorbell when they witnessed or heard of domestic violence taking place. The campaign’s integrated cultural, organizational, and media strategy sought to make the issue part of mainstream conversations; increase knowledge about and change community attitudes towards domestic violence and HIV-positive women; and alter individual behavior. The effort resulted in a 49 percent increase in the number of people aware of the Protection of Women from Domestic Violence Act in India, and a 15 percent increase in access to services for survivors. Engaging with communities to shift norms also matters. The SASA! Program (“Now!â€? in Kiswahili) developed in Uganda is a good example of how norms can be challenged. The program employs multiple strategies to build a critical mass of engaged community members, leaders, and institutions, including local activism, media and advocacy, communication materials, and training. In comparison to control communities, SASA! communities reported a reduction in violence against women of 52 percent. The share of women and men who believe it is acceptable for women to refuse sex increased by 28 percent, and the share of men and women who believe that physical violence against a partner is unacceptable increased by 50 percent.73 Essentially, SASA! works with key stakeholders at the community level to deconstruct power in intimate partnerships. Effective response and protection systems that avoid further harm and victimization are also required to help survivors of violence and prevent the escalation or recurrence of violence. A focus on behavioral elements is required with regards to response efforts, both during childhood and for adult women. Parenting and health interventions with women, for instance, have shown positive effects with regards to the occurrence of IPV and in changing related attitudes among women and their partners. Behavioral interventions with older couples show similarly positive impacts. Health screening plus (including counselling and referral) systems in primary health care settings are also promising. Access to justice and adequate protection mechanisms, such as specially trained police departments or staff, are also key in the efforts to end the pandemic of GBV. 71 Observatorio de Políticas Sociales y Desarrollo (2017) 72 Observatorio de Políticas Sociales y Desarrollo (2017) 73 Abramsky et al. (2014). 36 4. Despite recent improvements, the relatively low political representation of Dominican women remains an indication of and contributing factor in their limited agency The political participation of women is both a reflection of and a factor in agency. Low levels of political representation tend to suggest that women face constraints vis-à-vis men, often related to the prevalent patriarchal social or even formal norms that confine them to roles of mothers and caregivers. At the same time, raising the political representation of women can boost the prominence of issues that chiefly affect them (including child mortality, maternity leave, childcare, and violence against women), as shown by the experience of high-income countries. 74 Women’s representation in parliament and in ministerial positions is relatively low in the Dominican Republic. The share of women represented in parliament has grown from 11.7 percent in 1998 to 28 percent in 2020. Although this share is slightly above the UMC average of 26 percent, it is also below the LAC average of 33 percent (see Figure 29, left-hand panel). The percentage of women appointed as ministers has also barely increased over the last three electoral cycles: 2008-2012, 2012-2016, and 2016-2020 (see Figure 29, right-hand panel). Indeed, the Dominican Republic is one of the LAC countries with the fewest women ministers: out of 22 ministries, only 3 were headed by women in 2020. The share of women in ministerial positions, at 17 percent in 2019, was well below the LAC and UMC averages.75 On the other hand, local institutions appear to be more representative. The share of women mayors has risen from around 2 to almost 13 percent over the last two decades (see Figure 30, left-hand graph), reaching the regional average. The representation of women in municipal councilor positions has also increased over time, and has been above the LAC average for decades, at around 34 percent since 2009 (see Figure 30, right-hand graph). Figure 29: National representation of women (%) Proportion of seats held by women in national parliaments Ministers by sex (%) (%) 40 100 30 80 Percent 60 20 40 10 20 0 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 2021 0 I II III Dominican Republic Male Female Latin America & Caribbean Upper middle income countires Source: Source: WDI; ECLAC ** I:2008-2012 II:2012-2016 III:2016-2020 Despite legal provisions towards a more balanced representation of women in politics, Dominican women still face important obstacles. Candidate lists in local and national elections must be 40 percent women by law, and primary elections for the main positions of síndico and vice-síndico (mayor and vice 74 World Bank (2012) 75 WDI 37 mayor) must feature women candidates.76 Quotas generally strengthen the position of women in political parties, increase the number of capable, educated women with the necessary professional experience in the field of politics and management of states, and have a symbolic function in terms of women’s political participation.77 However, it has been highlighted that the political context in the Dominican Republic does not seem to promote parity, as increasingly greater resources are required to compete, while internal party mechanisms attempt to block the advancement of women to executive and elective positions.78 A patriarchal culture also still limits women’s participation, as they need to spend much more time on care and household related activities than men, for instance. 79 Figure 30: Local representation of women (%) % of female mayors % of local councilors 15.0 40.0% 10.0 30.0% 20.0% 5.0 10.0% 0.0 0.0% 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 Dominican Republic LAC Dominican Republic LAC Source: ECLAC Although the Dominican authorities have put into place an advanced institutional framework for the promotion of gender equality since the Beijing Conference for Women in 1995, constraints remain. In 1999 the Secretary of State for Women was established; with the 2010 Constitution it became the Ministry for Women, in charge of establishing the norms and coordinating the implementation of policies, plans, and programs aimed at promoting gender equality.80 The ministry oversees Provincial and Municipal Offices for Women.81 However, the budget allocated to the Ministry is low, at around 0.08 percent of the national budget in 2018, down from 0.11 percent in 2013.82 Little advancement has been observed in gender mainstreaming across all policies, related to the lack of adequate human and financial resources.83 Some important legal reforms also appear to be pending in this area. 84 1.4 CONCLUSION Using a lifecycle approach in combination with a focus on the three main dimensions of endowments, economic opportunity, and agency, this section has identified some of the main existing gender gaps in the Dominican Republic. 76 INTEC 2016; https://www.poderjudicial.gob.do/documentos/PDF/leyes/LEY_ley_num._33_18_1_7.pdf 77 PoskoÄ?ilová 2015 78 UNDP 2016. The resistance of party structures to comply with parity in candidate lists and include women as candidates for the vice-síndico and síndico (mayor and vice-mayor) positions led the Electoral Committee to establish requisites on how to present the candidacies in order to be approved in the last elections (2010 and 2016). 79 INTEC 2016 80 Observatorio de Políticas Sociales y Desarrollo (2017); Cruz (2013); PLANEG III. 81 PLANEG III 82 http://hoy.com.do/ministerio-de-la-mujer-con-79-mm-para-prevenir-violencia/ 83 Cruz (2013) 84 INTEC (2016) 38 The average educational enrollment and attainment of girls is higher than that of boys. But discrimination and social norms prevent girls from reaping those benefits (better endowments than boys) when entering adolescence and shape the terms of female labor force participation. Women are less likely than men to join the labor force and to work for pay. When they do, they are more likely to work in salaried and formal jobs, but they tend to work less hours and be attached to occupations that have lower pay. These disadvantages translate into substantial gender gaps in earnings, which in turn decrease women’s bargaining power and voice. In addition, many girls are married or have children before the age of 18, before they may be physically and emotionally ready to become wives and mothers. Women and girls also face higher risks of gender- based violence in their homes, at work, and in public spaces. Their voice and agency are often constrained compared to that of males, whether this is within the household, at work, or in national institutions. This also affects their children. For example, children of young and poorly educated mothers often face higher risks of dying by age five, being malnourished, and doing poorly in school. Fundamentally, gender inequality disempowers women and girls in ways that deprive them of their basic human rights. This lack of inclusive growth also operates as a constraint to poverty reduction and shared prosperity. 2. PRIORITIZATION OF GAPS This section briefly reports the results of an exercise that was applied to evaluate the standing of the Dominican Republic against other countries in relation to a series of gender gaps to identify some of the main challenges on gender faced in the Dominican Republic. This exercise follows and builds on the stocktaking of evidence presented in the first section of this report. Annex 3 shows the methodology and results for the full set of selected indicators. The Dominican Republic would appear to be among the worst performers globally, and in relation to LAC, Upper-Middle Income countries and structural peers, when it comes to child marriage and teenage pregnancies; school performance (measured through harmonized test scores); labor force participation and wage gaps in the private sector; high time and cost constraints to open a business; and low savings for retirement and access to digital technologies for saving purposes. With one of the highest shares of teenage pregnancy in LAC and the world, it comes as no surprise that Dominican Republic remains in the 100th percentile (the closer the indicator is to 100, the higher the severity of the constraint being faced) for child marriage and teenage pregnancies with respect to the LAC region and the structural peers. The country is also in the 83rd percentile in terms of lack of savings for old age when compared to upper middle-income countries. Furthermore, harmonized test scores are among the lowest globally (table 4). Table 4. Key selected sex gaps for the Dominican Republic with respect to four groups of countries Upper Structural LAC World Middle Peers region Income Starting a business: Cost - Women (% of income per capita) - Score 87 63 84 90 Harmonized Test Scores, Female 99 100 100 97 Harmonized Test Scores, Male 98 99 100 98 Completed post-secondary, population 25+, female (%) (cumulative) 84 37 83 79 39 Completed upper secondary, population 25+, female (%) (cumulative) 77 58 77 73 Women who were married by age 18 (% of women ages 20-24) 100 100 100 100 Adolescent fertility rate (births per 1,000 women ages 15-19) 87 100 81 100 Saved for old age, female (% age 15+) 82 58 83 79 Female to male wage ratio in the private sector (using mean) 64 65 63 63 Note: Structural peers are Sri Lanka, Tunisia, Ecuador, Peru, Guatemala, Honduras, Romania and Fiji. The main challenges were then corroborated and benefited from insights shared at stakeholder consultations in close consultation with the national authorities. The consultations and discussions provided a rich set of ideas and insights that have been incorporated into this Diagnostic, but also helped to define the three analytical deep dives that would be undertaken in this report. The resulting deep dives are as follows: (i) school violence, particularly in connection with academic underperformance, (ii) teenage pregnancies; and (iii) wage gaps and labor force participation. (i) Violence in schools - It was brought to the team’s attention during government consultations that school violence is a serious but often overlooked topic in the DR. Counterparts expressed that the report could serve as a platform to raise awareness on this issue. Widespread violence in schools can act as a deterrent for both girls and boys to attend and complete education, negatively affecting their performance and their overall wellbeing, with important implications over the lifetime. While violence levels are higher among boys, which could explain to some extent their poorer performance vis-à-vis girls, dropping out of school and academic underperformance can have higher societal repercussions when it happens for girls, as it for instance can indirectly increase their likelihood to marry and have children early in life. In other words, school violence may lead not just to low educational outcomes for boys and girls, but also other negative externalities especially for girls (like teenage pregnancies if they dropout and the cascading negative effects of this, GBV, NEETs). For these reasons, the team felt that addressing school violence to improve school outcomes among boys, but particularly among girls, is crucial and merits a deeper understanding.85 (ii) Teenage pregnancy - The overall need to reduce child marriage, early unions and teenage pregnancies in the Dominican Republic is conclusive from the analyses and consultations. The prioritization exercise conducted confirms that teenage pregnancy emerges as one of the priority areas when comparing the average fertility rate over the last 10 years in DR with the average in LAC, UMC, and structural peers. The costs that these phenomena have for the girls themselves but also for societies at large – in terms of development outcomes – have been systematically proven. It is well understood what keeps girls from getting married at young age and what drives teenage pregnancies in the Dominican Republic and elsewhere. Having a poor education, low aspirations, and limited economic opportunities appear to increase the risk of teenage pregnancy at the individual level, while the main societal factors are patriarchal social norms and a (related) culture of child marriage. However, it is much less well understood how much different interventions contribute to these goals. (iii) Wage gaps and labor force participation - The overall need to reduce labor market participation and wage gaps between men and women is undoubtedly a key entry point into reducing poverty at this stage of Dominican Republic’s development, especially when considering the relevant role that labor income has played in doing so during the past years. Importantly, the deliberate focus on the average wage gap at the mean obscures the potential heterogeneity across the wage distribution and the need for differentiated policy solutions. Having information across quantiles can show what segments are the most affected and prioritize interventions. This deep analytical dive will therefore attempt to identify potential recommendations to improve the policy and legal framework to expand the opportunities for women to participate in the labor market and to reduce unjustified disparities in wages. 85 https://www.togetherforgirls.org/wp-content/uploads/2021/07/Ending-Violence-in-Schools-An-Investment-Case.pdf 40 These final prioritized constraints do not imply that other challenges or constraints should be forgotten. All the issues discussed in the first section are important to ensuring inclusive growth in Dominican Republic. The purpose of the prioritization is to identify actionable constraints that are likely to have the greatest bearing on eliminating absolute poverty and improving prosperity over the next few years. Thus, they are necessary but unlikely to be sufficient for across-the-board progress. The next section of this Assessment seeks to understand in more detail the drivers behind some of these key gaps and policy solutions to overcome the gaps. 41 3. DEEP ANALYTICAL DIVES This section seeks to gain a more granular understanding of three areas of focus when addressing gender as identified in the previous chapters: (i) school violence, particularly in connection with underperformance, (ii) teenage pregnancies; and (iii) wage gaps and labor force participation. 3.1 VIOLENCE IN SCHOOLS86 Ending violence in school is essential to reaping the benefits from education and ensure children’s well-being. Receiving a quality education is the right of every child, as enshrined in the Convention on the Rights of the Child (1989). Education plays a unique role in promoting respect for human rights and contributes to building safe and inclusive societies that do not condone violence, but rather provide children with the skills they will need as adults to solve conflicts peacefully. Moreover, education plays a fundamental role in countries’ ability to achieve the targets set forth under the Sustainable Development Goals.87 It is the foundation of countries’ future economic development , in driving human capital wealth (the value of the future earnings of the labor force), which itself accounts for two thirds of the changing wealth of nations.88 Unfortunately, violence in and around schools (VIAS) remains widespread in developing and developed countries alike, causing children not only to fail to learn but even to drop out of school.89 Failing to prevent VIAS will affect not only children today, but also their future families, their communities, and society as a whole. Sexual violence is defined as any sexual act, intimidation, attempt to obtain a sexual act, unwanted sexual comments, or advances against another individual using coercion. In addition to the lasting physical and psychological damage that sexual violence frequently exacts on victims, it also can have subsequent and lasting negative effects on victims’ health in terms of sexually transmitted infections (STI) and other health issues. Emotional or psychological violence may include verbal and emotional abuse, such as isolating, rejecting, ignoring, insulting, making up lies, humiliating and threatening. It can be inflicted in schools by students, but also by teachers through forms of punishment that denigrate, scapegoat, threaten, scare, or ridicule students, again frequently with long-term consequences for the children’s health and well-being. While emotional violence may be less visible than physical or sexual violence, it can also have significant and lasting impacts on students. Physical violence is any form of physical aggression with intent to hurt another person. It can manifest itself from student to student, student to teacher, or teacher to student. One manifestation of teacher to student violence is corporal punishment, which remains used in many schools as a form of discipline or punishment. Physical violence is the most common form of violence in schools and is highly visible to all students. In addition to those three categories of violence, one can also highlight particular types of violence. These include bullying, corporal punishment, and intimate partner violence. Bullying is defined as unwanted, repeated aggressive behavior among students, and is among the most common forms of violence in school settings. Bullying can produce serious and lasting physical, mental, and emotional problems among victims. Cyberbullying (over digital devices) has become a front-and-center issue that schools, educators, parents, and communities must struggle with. Corporal punishment is also a common form of violence against children worldwide, particularly in schools as a means of discipline and punishment. It includes any punishment in which physical force is used and intended to cause some degree of pain or 86 This note is adapted from Wodon et al. (2021). 87 Wodon et al. (2018). 88 Lange et al. (2018). 89 UNICEF (2018). See also UNESCO (2019). On violence against children more generally, see Hillis et al. (2016). See also UNICEF (2014), Office of the SRSG on Violence against Children (2016), Hillis et al. (2016), UNICEF (2017, 2019), Know Violence in Childhood (2017), and UNESCO (2019). 42 discomfort, however light, as well as cruel and degrading non-physical forms of punishment. Intimate partner violence or IPV refers to any behavior within an intimate relationship that causes physical, psychological, or sexual harm to those in the relationship. Dating violence between students is a form of IPV and violence in school, while sexual harassment on the way to or from school by an individual who is not an intimate partner is a form of violence, but not of IPV. 3.1.1: PREVALENCE AND IMPACTS OF SCHOOL VIOLENCE 1. There is a high prevalence of school violence from childhood through to adolescence in the Dominican Republic, especially among boys. Violence remains commonplace in schools. No data exists on students’ perception about school violence in early grades. However, the national diagnostic assessment for grade 3 (children aged 8-10) asks teachers and parents for their perception of violence in schools. Two thirds of third-grade teachers say verbal bullying among students are a problem (Table 5). Other problems identified by an important share of teachers are: (i) disorder and disruption of classes; malicious rumors and gossip; physical fights; and insults and threats. Threats by teachers against students aged 8-10 and gun violence are seen as less common in Dominican schools. However, 7 out of 10 parents cited physical fights between students as a problem and more than half perceived that threats or insults between students were an issue (See Table 6). Far fewer were concerned about theft within the school. Given the prevalence of these reports by both teachers and parents and how young and vulnerable children are in this grade, any intervention to deal with violence in school should prioritize primary education. Table 5. Violence in schools: percentage of teachers according to perception of school climate (public sector) It is not a It is a minor It is a moderate problem problem problem Disorder and disruption of classes by students 49.1 33.3 14.77 Malicious rumors and gossip between students 41.84 41.67 13.96 Verbal fights between students 30.17 50.38 15.56 Physical fights between students 40.2 41.37 13.24 Gun violence inside the school 88.59 4.89 1.73 Robberies inside the school 72.75 17.45 4.56 Acts of vandalism against the school 85.7 6.88 2.39 Violence in the surroundings of the school 66.99 18.14 7.84 Insults and threats between students 48.89 36.44 10.22 Threats from student to teacher 80 11.51 4.06 Threats from teacher to student 90.49 4.07 1.68 Source: Department of Evaluation of Education Quality. Ministry of Education of the Dominican Republic Table 6. Violence in schools: percentage of parents according to frequency of reported violent acts (public sector) A few Many times, Never times or always Theft within the school 57.68 31.28 4.13 Fights between students (yelling, hair pulling, hitting) 21.17 57.45 14.79 Threats or insults between students 33.63 45.12 13.23 Damage to the school caused by students (scratching of walls, breaking of tables, glass, chairs, computers, etc.) 41.59 39.47 11.66 Source: Department of Evaluation of Education Quality. Ministry of Education of the Dominican Republic 43 The Dominican Republic is also the country with the highest level of bullying among 12-year-old students in Latin America. Almost half students (49 percent) reported being victims of at least one form of bullying, the highest level among the 15 Latin American countries that participated in the Third Regional Comparative and Explanatory Study (TERCE) (see Figure 31).90 The country ranked first in four out of six types of bullying documented by TERCE (feeling afraid, being threatened, being hit and feeling left out). The Dominican Republic is one of the few countries in the sample where the prevalence of bullying in private schools (52 percent) is slightly higher than in public schools (49 percent). Figure 31. Share of students self-reported as victims of bullying 60% 49% 48% 47% 50% 44% 44% 44% 43% 43% 43% 43% 40% 37% 35% 33% 33% 31% 30% 20% 10% 0% Dominican Mexico Colombia Paraguay Guatemala Costa Rica Argentina Peru Uruguay Nicaragua Ecuador Panama Brazil Honduras Chile Republic Source: Chávez et al. (2021). School violence in the Dominican Republic also affects teenagers, especially boys. Figures 32 and 33 provide estimates of violence in schools among teenagers (15-year olds) in the Dominican Republic based on data from the 2018 round of PISA (Programme for International Student Assessment).91 The first figure provides estimates for different types of violence by gender, and the second figure compares overall levels of school violence in the Dominican Republic with levels observed in other countries. In the PISA assessment, children are asked if, during the past 12 months, (1) they were left out on purpose by other students; (2) other students made fun of them; (3) they were threatened by other students; (4) other students took or destroyed things that belonged to them; (5) they were hit or pushed around by other students; or (6) other students spread nasty rumors about them. (1) and (2) are not strictly speaking measures of violence and thus are not reported here, but they are included as controls in the regression analysis presented subsequently.92 Judging by the data compiled by the PISA assessment, one in every two boys in the Dominican Republic is affected by various forms of violence, and at a far greater rate than in comparator countries. Figure 32. Share of students affected by Figure 33. Share of students affected by violence, violence, by region (%) by gender (%) 90 Chávez et al. (2021). 91 The PISA assessment used in this analysis includes 15-year-old students in a sample of approximately 70 countries, most of which are more developed than the Dominican Republic. 92 Number (6) can be classified as emotional or psychological violence. 44 60 60 49.5 49.9 49.9 49.7 49.7 50 46.6 43.1 44.0 50 41.3 43.1 41.3 40 36.3 36.0 33.1 36.0 Percent 40 28.0 32.4 32.2 31.4 30 Pereentage 25.2 24.3 23.5 30 22.0 20 20.7 20.5 20.5 19.0 20 10 10.7 10 0 Threatened by Things Hit by others Nasty rumors 0 others destroyed Female Threatened by Things Hit by others Nasty rumors others destroyed Dominican Republic Latin America and the Caribbean Upper middle income countries World Source: PISA Boys suffer violence more often, but the share of girls affected is not negligible (figure 34). Moreover, school violence is detrimental to both girls and boys through different outcomes (for girls: GBV, teenage pregnancies if they dropout and the cascading societal negative effects of this, NEETs, and for boys: joining labor force at younger ages without completing their education). 2. School violence is associated with worse student performance School violence affects child learning among both girls and boys. An index on school violence was constructed using data on parents’ perceptions about the frequency of violent acts at school from the national diagnostic assessment for grade 3 (listed on Table 6). Figure 35 shows that the higher the frequency of violent acts in schools, the lower students’ learning outcomes in reading and math. The association is similar both for boys and girls, with steady gender gaps across quintiles. Figure 34. Index of learning outcomes by quintiles of violent acts 320 310 Average score 300 290 280 270 260 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Reading Math Boys Girls Source: own estimates using third grade national diagnostic assessment. Notes: An index was computed using principal-component factor and based on information reported by parents about violent acts in schools. Cronbach's alpha is 0.7811. Gaps between boys and girls are all significant at 1%. By contrast, a positive environment for social interaction at school is associated with better learning outcomes. The national diagnostic assessment for grade 3 also asked parents how much they agreed with 45 certain statements about the existence of positive social interactions at their child’s school, such as a respectful dynamic between students. As shown in Figure 36, the better the social environment at schools, the higher the learning outcomes both for both in reading and math. The breakdown by gender suggests that an improved social atmosphere benefits boys and girls alike. Figure 35. Index of learning outcomes by quintiles of social interaction environment 320 315 310 305 Average score 300 295 290 285 280 275 270 265 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Reading Math Boys Girls Source: own estimates using third grade national diagnostic assessment. Notes: Q5 stands for the top quintile of the social interaction index. An index was computed using principal-component factor and based on information reported by parents about the interactions between and within teachers and students, and the interaction climate in the school. Cronbach's alpha is 0.8900. Gaps between boys and girls are all significant at 1%. Dominican teenagers also learn less due to school violence. Data from the PISA assessments can be used to assess the potential impact of violence in school on students’ test scores. The assessment is implemented among 15-year-old children to assess their ability to use their reading, mathematics and science knowledge and skills to meet real-life challenges. The global average score for all students and countries participating in PISA is 500 points.93 The performance of the Dominican Republic on PISA is on the low side in comparison to OECD countries. As noted in the Dominican Republic’s OECD country profile, for reading literacy, its 15-year-olds score 342 points versus 487 points for OECD countries on average.94 For mathematics, the average score in the country is at 325 points versus 489 points in OECD countries. Finally, in sciences, the average score is 336 points, also compared to 489 points in OECD countries. Regression analysis is used to assess the potential marginal impact of VIAS on learning outcomes. While the analysis is based on associations (correlations) and may therefore not indicate causal effects, it is nevertheless instructive. The regression analysis is done for the pooled sample of all countries, as well as for the Dominican Republic’s data only. Results tend to be more stable with the global sample due to larger sample size, but the results for the Dominican Republic show that in many cases, the conclusions remain the same (key results are provided in Table 7). 93 Test scores are scaled so that the international average for all countries is 500 points, and the standard deviation is 100 points when weighting all countries equally. This implies that two in three pupils are in a range of 400 points to 600 points, and most students are in the 250 to 750 range. An average score of 500 does not mean that a student is doing well, as many students may still not achieve basic reading, math, or science proficiency. 94https://gpseducation.oecd.org/CountryProfile?primaryCountry=DOM&treshold=10&topic=PI#:~:text=In%20reading%20literac y%2C%20the%20main,30%20points%20higher%20for%20girls). 46 Diverse forms of school violence are negatively associated with student performance in DR, with threats the most harmful. When a student is threatened by other students, this is associated with a reduction in performance on the mathematics test of -23.0 points within the global sample (Table 7). Effects for the reading and science tests are similar. When considering only the sample for the Dominican Republic, effects are statistically significant only for the performance in science, with a smaller negative effect (-11.1 points). Other types of violence are also negatively associated with student performance. The magnitude of the effects is smaller at the global level, but larger for some effects in the Dominican Republic sample only. Curiously, in the case of nasty rumors, these are associated with better student performance in the Dominican Republic. Still, overall, there are clear signs that violence in schools affects learning negatively. Whether at the global level or for the Dominican Republic, the negative correlations of the proxies for violence in school are typically as large – and in some cases larger – than the potential impact of a wide range of other variables on student performance. This suggests that violence in school matters. For example, a higher quintile of well-being of the household to which the student belong is associated with better PISA scores – by 15 to 35 points. But this is roughly equivalent to the negative effects of some forms of violence in schools. For the top quintiles this is slightly larger than some of the effects of violence in schools, but not by much. The effects of violence in school are also of a similar magnitude, and in some cases larger, than the effects of parental education. These results are tentative because they depend on the specification used for the regressions and they need not imply causality (only associations are measured) but seem intuitive: violence at school is likely to be highly disruptive to the learning environment, regardless of prior student advantages. Table 7: Potential impact of violence in school on student learning, PISA Math Reading Sciences Global Dom. Rep. Global Dom. Rep. Global Dom. Rep. Indicators of violence I was threatened by other students -22.980 NS -28.394 -18.160 -24.728 -11.063 Others destroyed my things -1.398 NS -6.225 -7.619 -2.978 NS I was hit by other students -9.959 -26.246 -14.815 -33.320 -10.798 -25.433 Others spread nasty rumors about me -3.664 11.847 -3.295 16.796 -3.604 6.334 Source: Authors’ estimations using PISA data. Note: NS: Not statistically significant. 3. School violence worsens student well-being School violence makes the entire student experience at school more difficult, not just learning. Regression analysis is also used to assess the potential marginal impact of VIAS on a range of student outcomes, including their experience of school. In the PISA data, which is gathered only in secondary schools with 15-year-olds, information is available on whether students (1) have no close friends, (2) feel lonely, (3) feel that other students do not accept them as they are and (4) find it easy to make friends. PISA also surveys students’ perceptions as to whether they feel that their teachers (5) make them confident, (6) listen to them, and (7) understand them. Table 8 provides the results for the potential impacts of violence on the variables of interest. The interpretation of the coefficients is in terms of percentage points at the margin. For example, controlling for other factors, being threatened by others is associated at the margin with an increase in the probability of feeling lonely of 3.1 percentage points. The regression analysis is done for the pooled sample of countries as opposed to the Dominican Republic because, due to larger sample sizes, this tends to generate more reliable results. Most coefficients in Table 8 are statistically significant, suggesting negative potential impacts of VIAS. By and large, effects are systematic. Table 8: Marginal potential impacts of violence in school on students’ experience in school, PISA Threatened Things Hit Nasty by others destroyed by others rumors 47 I have no close friends NS NS NS 0.004 I feel lonely 0.031 0.007 0.006 0.041 Other students do not accept me as I am 0.03 NS NS 0.048 It is not easy for to me to make friends 0.013 NS NS 0.012 My teachers do not make me confident 0.012 0.018 0.01 0.032 My teachers do not listen to me 0.012 0.018 NS 0.033 My teachers do not understand me 0.013 0.019 NS 0.037 Source: Authors’ estimations using PISA data. Note: NS means statistically not significant at 0.1 level. More importantly, violence in school also affects the health and well-being of students. Table 9 provides the key results from the regression analysis. For example, having things stolen or destroyed is associated with an increase of 2.2 percentage points in the likelihood of finding it hard to sleep. Some associations in Table 9 are not statistically significant and a few have the opposite effect of what would be expected. Still, most effects are statistically significant and suggest potential negative effects of violence in school on student wellbeing. Estimates from a global study point at a rate of 5% of drop-outs due to school violence – though estimates likely vary by country a lot. Table 9: Marginal potential impacts of violence in school on various health outcomes Threatened Things Hit Nasty by others destroyed by others rumors Perceived health Difficulty sleeping NS 0.022 NS 0.089 Poor self-assessment of health 0.004 NS NS 0.004 Headaches -0.028 NS -0.018 0.079 Stomachache NS 0.034 NS 0.072 Back pain NS 0.035 0.021 0.062 Psychological well-being Feeling depressed NS 0.015 NS 0.124 Feeling irritable -0.046 NS NS 0.095 Feeling nervous -0.041 0.017 NS 0.073 Feeling dizzy 0.026 0.042 0.018 0.096 Feeling anxious NS 0.021 NS 0.111 Source: Authors’ estimations using PISA data. Note: NS means statistically not significant at 0.1 level. Ending violence in school could have large beneficial effects for children. Simulations suggest that up to a quarter of the problems or issues encountered by secondary-school students could be due to violence. Table 10 provides baseline values (column a). Simulated values are estimated assuming violence is eliminated. The differences between baseline and simulated values then indicates how much of the difficulties faced by students could be due to violence, and these differences are reported in the Table in absolute terms (column b), and as shares of base line values (column c). Those shares are often high, suggesting that major potential benefits to student’s experience, perceived health, and wellbeing could result from ending violence in schools. Table 10: Simulated benefits of ending violence in schools Predicted share Potential impact Share associated under baseline of violence with violence conditions in school in school (a) (b) (c) Boy Girl All Boy Girl All Boy Girl All Experience in school I have no close friends 2.6 3.6 3.1 0.8 1.3 1.1 21.7 25.5 23.6 I feel lonely 25.5 27.0 26.2 7.7 9.7 8.7 21.0 25.3 23.1 Other students do not accept me as I am 23.8 23.9 23.9 6.1 7.4 6.7 18.5 22.3 20.4 It is not easy for to me to make friends 27.5 23.5 25.5 4.9 5.6 5.2 14.1 18.7 16.4 48 My teachers do not make me confident 24.4 25.7 25.0 2.8 3.5 3.2 9.7 11.5 10.6 I feel my teachers do not listen to me 27.4 28.7 28.1 2.9 3.5 3.2 8.9 10.4 9.7 I feel my teachers do not understand me 26.4 27.0 26.7 3.1 3.7 3.4 9.9 11.5 10.7 Perceived health Difficulty sleeping 54.7 46.4 50.5 7.4 8.9 8.1 11.3 15.3 13.3 Poor self-assessment of health 0.7 0.7 0.7 0.3 0.4 0.4 25.7 32.3 29.0 Headaches 74.5 54.5 64.5 3.9 5.1 4.5 4.8 8.2 6.5 Stomachache 70.1 48.3 59.2 5.0 7.2 6.1 6.4 12.3 9.4 Back pain 56.6 43.4 50.0 6.1 7.6 6.9 9.3 14.3 11.8 Psychological well-being Feeling depressed 69.1 53.8 61.5 8.5 11.6 10.1 10.6 17.0 13.8 Feeling irritable 70.9 52.0 61.5 5.9 7.7 6.8 7.5 12.3 9.9 Feeling nervous 78.9 61.7 70.3 4.7 7.0 5.9 5.6 10.0 7.8 Feeling dizzy 51.9 41.4 46.6 8.3 10.1 9.2 13.0 18.7 15.8 Feeling anxious 60.7 48.7 54.7 8.2 10.2 9.2 11.3 16.6 14.0 Source: Authors’ estimations using PISA data. 3.1.2: FACTORS LEADING TO SCHOOL VIOLENCE AND PROMISING INTERVENTIONS This section documents and discusses policy options to reduce school violence in the Dominican Republic. Each policy recommendation aims to tackle key emerging findings associated with the increase in school violence documented below. As a preview, the key recommendations for this deep dive on school violence include: (a) the introduction of conflict-management modules in teacher-training programs, to provide teachers with the necessary skills and tools for positive discipline and classroom management; (b) the universalization of the Rules for Harmonious School Coexistence and the School Peace Culture Protocol, which as described below are the two strategies around which efforts to help children improve their social and emotional skills coalesce; (c) setting up a violence observatory within the monitoring and evaluation system of the Ministry of Education (MINERD), integrated with the National Plan for gender Equity 2020-2030 (PLANEG III), in order to follow up on how indicators develop in this area; and (d) the production and dissemination of specialized studies on the determinants of school violence to improve understanding of what drives this phenomenon in the Dominican Republic (figure 37). Figure 36. Selected enablers of school violence and policy recommendations 49 Overall drivers of school violence Many children face violence in their daily lives, at home, at school or in their community (see Box 4). This section focuses on interventions that have been shown to be effective in countering physical, verbal, and psychological violence in schools. Box 4. Factors contributing to violence against children, including in schools The Social Ecological Model framework helps in disentangling the complex factors leading to violence and thereby identifying some of its root causes. Risk factors for violence are recognized at the levels of individual, relationship, schools, community, and society. By recognizing shared responsibility for violence prevention, the model helps to focus on ways to change mindsets and behaviors. At the individual level, childhood disruptive behavior including in schools (opposition, aggression and hyperactivity with corresponding poor self-esteem, poor emotion management, impulsiveness, lack of empathy), and cognitive deficits in part reflecting the school environment (low IQ, inattentiveness, poor school performance) are associated with violence and delinquency in adulthood. For young children, poor nutrition (stunting), exposure to toxic substances such as lead or mercury, prenatal and perinatal complications, and head injuries can all cause brain defects, neurological dysfunctions, and learning disabilities, all of which may in turn lead to violent behavior. For mothers, substance abuse (i.e., alcohol and drugs) during pregnancy may hamper child neural/brain development and predispose children to aggression and antisocial behaviors. Other factors that increase the chances of (experiencing) violent behavior in children involve their gender, genderual orientation, and low socio-economic status. Age also matters, as younger students are more likely to be victims of violent behavior at the hands of older students. Older students are meanwhile at a greater risk of initiation to violent crime. Too much unstructured free time and association with delinquent peers are additional individual risk factors. At the family level, a wide range of factors may lead to victimization or perpetration of violence. In particular, child maltreatment has been shown to make children far more likely to be a perpetrator or victim later in life. It is also associated with a higher risk of delinquency, even more so than unemployment, educational attainment, gun ownership, or exposure to lead through paint or gasoline. In addition, too soft or too harsh discipline at home, corporal punishment, parental deviance, domestic violence, witnessing violence at home, parental rejection, and poor supervision may all lead to higher risks of violence later in life. Issues such as food insecurity may also be risk factors for violence. At the school level, risk factors include lack of resources, lack of organizational capacity, and a school climate characterized by anti-social behaviors. Lack of student bonding with their schools, use of corporal punishment, and more generally a lack of positive role models and caring adults also tend to result in higher risk of violence. Dropping out, student absenteeism, and the presence of alcohol and drugs in the school are also negative factors. Poor school governance and classroom environment, including dilapidated infrastructure and lack of safety making it easier for outside perpetrators to enter schools, are also detrimental. Gang activity within schools and student access to various types of weapons are especially detrimental. At the community level, the presence of gangs is a major risk factor, as is the prevalence of crime and violence in the community. Availability of firearms, high levels of unemployment or underemployment, high population density, poverty, and norms conducive to violence and gender inequality also tend to be associated with higher risks of violence in schools. At the societal level, wars or dictatorships, post-conflict situations, norms condoning the use of violence to solve conflict, patriarchal systems, and political climates marked by impunity and corruption all tend to be associated with higher risks of violence. Violent portrayals in the media can also affect behaviors for both children and adults alike, as can weak police and legal systems and cultural norms condoning violent behaviors, for example by husbands towards their wives and other dependents. Source: Wodon et al. (2021). 50 Ending violence in schools requires multifaceted interventions. As noted above, risk factors for violence include factors at the individual, family, school, community, and societal level. The accumulation of risk factors often explains why an individual behaves more violently or is more prone to be victimized than others. Instead of looking for a single best intervention that is most effective in preventing violence, it often makes sense to combine interventions that can both mitigate the most salient risk factors and enhance relevant protective factors in a given context and for a specific group. Adopting a lifecycle approach is also useful because risk factors leading to violence in school evolve over time in a child’s life. Early childhood interventions are essential in preventing violence in school and often have high returns. Cost-benefit analyses for center-based interventions (typically preschools, although many programs also include home visiting, parenting advice, health and nutrition services, and referrals for social services) show that such programs tend to have high returns on investment, with benefit-to-cost ratios ranging from 2.04 to 16.14.95 Some programs implemented recently do not have as high benefit-to-cost ratios, but this may be due in part to the fact that some of the long-term benefits from these programs in adulthood could not yet be measured. The renewed emphasis that the Dominican Republic is giving to health and nutrition services through conditional cash transfers from the program ALIMENTATE and to childcare services are good examples in this respect (see Box 11). In primary and secondary schools, programs to help children improve their social and emotional skills show promise. A recent synthesis of cost-benefit analyses for these types of programs suggests benefit-to-cost ratios ranging from 3.46 to 13.91.96 A key area of focus should be to reduce bullying and other types of negative interaction among students. Reviews suggest that intensive and long-lasting programs are needed to change behaviors, with parental sessions contributing to success.97 For the Olweus Bullying Prevention Program (OBPP), a benefit- cost ratio of 6.94 has been found when start-up costs are not included.98 For the KiVa Antibullying Program, analyses suggest benefit-cost ratios well above one, depending on the country.99 Main Efforts to Date to Promote Conflict Management and Prevention In the Dominican Republic, efforts to help children improve their social and emotional skills coalesce around two strategies: the Rules for Harmonious School Coexistence and the School Peace Culture Protocol. The Rules for Harmonious Coexistence is an umbrella strategy supported by the national legal and regulatory framework, which includes the Constitution of the Republic, the 2030 Agenda for compliance with the SDGs, the General Education Law 66/97, and Law 1-12: National Development Strategy 2030, Law 68-13 that establishes the Code for the Protection of the Rights of Children and Adolescents, and the National Agreement for Educational Reform (2014-2030), with decrees 228-13 and 84-15 that support it. The Rules for Harmonious School Coexistence, approved in 2013, were articulated by the National Council for Children and Adolescents (CONANI) with support from UNICEF. The Rules advocate a paradigm shift away from a punitive model to restore discipline and solve conflicts in the school. Restorative practices are encouraged so that boys and girls can internalize new ways of acting such as the positive redefinition of conflict, school mediation, positive discipline and emotional regulation. 95 Dalziel et al. (2015). 96 Belfield et al. (2015). 97 Farrington and Ttofi (2009). 98 Highmark Foundation (2018). 99 See Huitsing et al. (2019) and McDaid (2017). 51 The Rules comprise a series of recommendations to strengthen the role of parents, teachers, and students in schools in the use of mediation to prevent violence, the promotion of a culture of peace, and the peaceful resolution of conflicts, among other strategies (see Box 5). The head of the school together with a management team (made up of a Director, Deputy Director, Counselor, pedagogical coordinator, and teachers) are responsible for their implementation, which consists of: (i) disseminating standards and regulations for all members of the educational community, and (ii) monitoring and ensuring their compliance and verifying that each teacher socializes them with the classroom at the beginning of each school year. Mediation is part of the Rules of Harmonious School Coexistence and has been introduced in the Dominican educational sector as a basic tool to foster the implementation of non-violent conflict resolution models. At its core, mediation rests on two elements: a) the empowerment of students to become decision-makers and reach lasting agreements, and b) the recognition and better understanding of the needs, interests and values of the other, which will contribute to a culture of tolerance, active listening and mutual respect. Parties in conflict or their relationships can be altered during the mediation process, changing their way of communicating and relating to each other and to the rest of society. The mediator in this dynamic is the teacher who becomes a companion of the process and the instrument for channeling agreements. The foundations for the application of the mediation model have already been laid through the competency-based approach contained in the Dominican school curriculum.100 In particular, the Problem-Solving Competence and the Ethical and Citizen Competencies (MINERD, 2014) seek to train teachers on how to tackle situations of violence encountered in school on a daily basis while demonstrating empathy, solidarity, and respect in their relationships with students. The Rules for Harmonious Coexistence order the creation of school mediation teams for the peaceful resolution of conflicts. A Management Council comprised by Regional, District and Public and Private Educational Center Directors are responsible for creating the Mediation Team. The mediation team must be comprised of orientation and psychology staff, pedagogical coordinator, a member of the teaching staff from the school and two students representing the course councils. The goal of the mediation team is to come up with timely and appropriate measures for coexistence that respect the rights of students, known as the School Coexistence Standard. The measures determined by the mediation team must be approved by the Management Team prior to their application. Between 2016 and 2017, a pilot plan was executed for the application of the School Coexistence Standards in 32 primary and secondary schools in Boca Chica, a municipality in the Province of Santo Domingo (MINERD, 2018). The evaluation, according to respondents, shows a positive impact on the dynamics of peer mediation and student dynamic. The school communities involved recognized the value of mediation. Therefore, this action could be scaled up to the rest of the country, with a monitoring and evaluation strategy in place to measure results and impact. The School Peace Culture Protocol also aims to curb school violence and helps to implement the Rules of Harmonious School Coexistence. The Protocol is developed through the National Strategy for Peace 100 Within the framework of the institutional commitment to improve the quality of education in the Dominican Republic, the rules of harmonious coexistence are a transversal axis. The curriculum constitutes one of the fundamental qualitative components (MINERD, 2021). In the curricular revision developed in the country, and through consensus, a series of Fundamental Competences were identified and defined, two of them are identified having a direct relationship with the achievement of Harmonious coexistence in the school based on the Peace Culture Strategy: Ethics and Citizenship, and the Resolution of Conflicts. The curricular review and update constitutes a first step in the progressive adoption of the competency approach, in compliance with what is established in the regulatory framework (MINERD, 2016) and would allow peace culture strategy to be incorporated with all its components into the curricular content. 52 Culture, a MINERD initiative that has been formulated in coordination with UNICEF and seeks to promote a culture of peace with the national educational community. This strategy is designed considering the Competencies Approach of the Dominican Curriculum (MINERD, 2014); and is aligned with the National Agreement for Educational Reform. The School Peace Culture Protocol organizes the strategies, techniques, procedures, and instruments to promote a culture of peace in the school community. In doing so, the Protocol offers a procedural guide to implement the Rules for Harmonious School Coexistence in the educational system; as well as the mechanisms for managing discipline. Knowledge Gaps and Low Implementation of Violence Prevention Programs The information available does not allow us to ascertain the number of schools that implement the School Peace Culture Protocol. Likewise, in the MINERD Strategic Plan, only one proxy indicator was identified to measure the implementation of the Rules for Harmonious School Coexistence: that is, the number of high school students impacted by the school harmonious coexistence strategy and there is no budget assigned to measuring this element. At the operational level, few activities explicitly suggest a detailed implementation of the Rules and the Protocol. Box 5. Implementation of the Peace Culture Protocol Implementation Phases Actions to develop 1. Socialization and The actors involved in the process are guided so that they know their role in the general induction development of the strategy, the methodology, the resources, and the tools that will be used. 2. Training Strengthening the skills of regional directorates, educational districts, and schools to promote a peace culture in the school environment. 3. Diagnosis A diagnosis of the situation of school coexistence and violence against children and adolescents in the school and community context will be carried out in educational centers. These results are key to defining the interventions, techniques and communication campaigns that will respond to or improve said situation. 4. Improvement plan and Defines the actions that are to be integrated into the School Improvement Plan. Applies application the set of training tools that have been made available to the educational community during the training phase. 5. Monitoring, follow- Collect the necessary evidence to reflect on the progress and results of the up, and evaluation interventions, techniques, campaigns, etc. designed to improve school coexistence and eliminate violence against children and teenagers. Source: Own elaboration based on the School Peace Culture Protocol (MINERD, 2018) Similarly, the Rules of Harmonious School Coexistence have only been implemented very gradually and imperfectly. Even though the mandates are clear, there is no central authority at MINERD that verifies if the rules are being implemented and disseminated in every school. Ideally, these mandates should be subject to periodic discussion and reflection as part of the school curriculum on skills for life and human and religious formation, directly supervised by the Regional Directorate of Education and by the Educational District of MINERD. Social norms that normalize violence 53 Supporting teachers to enhance their skills in positive discipline and classroom management is also effective. Multiple studies in the Dominican Republic reveal that teachers assume that verbal and physical violence in school are normal; they have "internalized a model of violent relationship" (Pacheco-Salazar, 2019) as part of their daily lives and as a disciplinary mechanism with students. Teachers may simply be reproducing what they have learned from those who taught or trained them, or may be reflecting "the contexts of violence from which the teachers themselves come" (López, 2020). Students deal with school violence by defending themselves violently, partly because teachers themselves ignore how to handle these behaviors. Violence manifests itself in verbal violence (backbiting or gossip, insults, use of nicknames and threats), cyberbullying, sexual violence, and physical violence. In relation to the latter, the use of bladed or pointed weapons (razors, knives, daggers, needles, hairpins) seems to be widespread among students to attack, threaten and for use in "self-defense" (Pacheco-Salazar, 2019). This is explained by three main factors: the realization that school authorities do not offer effective protection or support against violence; the weak institutional mechanisms in the school to address violence; and the lack of emotional education, since students do not have the tools to manage their feelings or conflicts in a peaceful and harmonious way. Providing teachers with skills to improve their relationship with students and manage behaviors reduces disruptive and aggressive behaviors in the classroom and enhances prosocial behaviors later in life. By contrast, punitive interactions tend to feed a vicious circle of violence, delinquency, and further exclusion. The Global Initiative to End All Corporal Punishment of Children and the WHO handbook on school violence prevention provide useful resources on positive discipline for teachers and schools.101 When teachers and the entire school community understand that respect and trust are key pillars for child’s healthy development and that corporal punishment is not only counterproductive but negatively impacts a child’s learning ability, the school climate can be transformed. The Dominican Republic has recently trained teacher and other MINERD staff on an anti-violence methodology developed in Nicaragua called CEPREV. This program has aligned well with existing regulations and guides in the Dominican Republic and could be scaled up to train more stakeholders in the education sector with the aim of reducing school violence. Since the Ministry structures already exists (e.g., teacher training programs, school principal programs, inspectors, school counselors, and a student counselling directorate within the Ministry), the marginal cost of scaling up this program would be low. Families and communities need to be part of school programs. Engaging with parents of adolescents that display behavioral problems can yield results even in a relatively short period of time. But parenting programs should follow evidence-based practices, including focusing on positive discipline, positive communication, and opportunities for increased bonding among family members. As with teachers, providing alternative tools and skills to caregivers in dealing with their children can help break the intergenerational cycle of violence. Effectively engaging with parents requires choosing wisely among alternative programs. The most challenging part is to keep parents engaged long enough to produce sustained behavioral change. While some parenting interventions have step-by-step protocols, resources and tools with sessions in the classroom, and involve sharing of practical, how-to information with parents and caregivers, others may simply consist of newsletters or unstructured engagement with uncertain results. Engaging with the community is also important to ensure safe passage to schools by identifying hot spots where children may feel vulnerable and placing adult monitors on those spots. To the team’s knowledge, no interventions of this nature have been implemented in the Dominican Republic thus far. Whole-school approaches can also help reduce violence in school at a limited cost. Engaging with the entire school community is often beneficial. A whole-school approach uses multiple strategies to develop 101 World Health Organization (2019). 54 a common vision and shared values and rules for the school, and works through the curriculum, teacher training, parental engagement, and student learning towards a safe and inclusive school climate and respectful school values. One such example is the Good School Toolkit in Uganda. The program was developed by the NGO Raising Voices in consultation with teachers and children in 600 schools. It works through six sequential steps to provide behavioral change techniques to teachers, school staff and students. It focuses on goal setting, positive discipline, empathy, reflection, and the practice of new behavioral skills. Teachers and school staff are trained and supported throughout implementation. Evaluations suggest that after 18 months of implementation, the program reduced the risk of physical violence by teachers and school staff against students by 42 percent; halved the number of teachers who reported using physical violence against students; and improved students’ connectedness and sense of safety, wellbeing and belonging with their school. The program also increased teachers’ satisfaction in their role at school .102 Importantly, if the Good School Toolkit program were implemented at scale (e.g., on a national level), unit costs for implementation would be low.103 A final important recommendation is that a system of indicators on school violence needs to be put in place. New diagnostic student evaluations by MINERD collect data on the school environment, bullying, and other forms of violence. However, it is necessary to design additional indicators, and create an observatory of school violence and oversight for the implementation of related regulations.104 This could potentially be tried as part of the System of Indicators for the Follow-up of the Commitments of the National Pact for Educational Reform (ISPE for its Spanish acronym), which establishes the indicators, procedures and institutional responsibilities associated with compliance with the Educational Pact.105 However, currently within the proposed indicators to monitor ISPE the team could only identify 4 direct and proxy indicators that allow for monitoring the implementation of the rules for harmonious school coexistence.106 Some of these indicators have already been incorporated into the strategic planning instruments currently being implemented by MINERD. To summarize, beyond efforts in individual schools, strategies to end violence in school should be led by MINERD with other ministries or agencies. To sustainably shift norms, parent associations, teacher unions, religious groups, and political actors all need to participate and be heard. Several guides already exist in that respect. MINERD has developed a series of materials to tackle school violence, although their implementation requires consistent funding allocations, monitoring, and close coordination to ensure that progress is being made. These policy documents follow international standards and have been supported by international development partners including UNICEF and the Organization of Inter-American States (OIS), they involve all stakeholders (parents, teachers, the community), and they are recognized under the 102 See Devries et al. (2015) and Naker (2018). 103 Greco (2018). 104 See: IDEICE (2014). Estudio de Prevalencia, Tipología y Causas de la Violencia en los Centros Educativos de Básica y Media de la República Dominicana. Instituto Dominicano de Evaluación e Investigación de la Calidad Educativa, IDEICE. Santo Domingo, D.N., República Dominicana; Parada, H., A. L. Ibarra, R. Burgos y M.E. Asuad (2017). Incidencia de violencia en adolescentes de 13 a 15 Años en las escuelas públicas del sistema de Educación de la República Dominicana. Santo Domingo, D.N., República Dominicana. 105 The National Pact for the Reform of the Educational Sector [2014] is conceived as a comprehensive strategy that seeks to achieve the transformations required in the Dominican educational system (Decree 84-14). The Pact is based on a set of commitments aimed at democratizing access to the educational system from the pre-university to the higher level, including technical-professional education, guaranteeing the quality and relevance of learning at all levels of knowledge, dignifying the teaching career, establishing an evaluation system, guaranteeing the modernization of the educational system and adapting the financing and mobilization of resources for education. 106 Percentage of educational centers that promote gender equity (indicator #4242 C 1 80); percentage of educational centers with behavior professionals that offer psychological support services for children and adolescents in conditions of social and emotional vulnerability (indicator #4260 A 1 82); and percentage of educational centers that apply the norms of harmonious coexistence of the educational system (indicator #70151 C 1 133). 55 policies and laws that define the functioning of the education sector. These documents and policies include training materials for conflict resolution, guides for mediators and students on a culture of peace, norms and regulations for harmonious coexistence in public and private schools, and protocols for the promotion of a culture of peace in school, and they incorporate a strong gender focus.107 The MINERD also produces diagnostics on these issues, such as the new diagnostic student assessment that collects data on school environment, bullying, and other forms of violence.108 Studies of similar programs in Central America have shown positive results as documented by observational studies.109 The available analyses for some of these promising interventions suggest high benefits to costs ratios. While these ratios are sensitive to assumptions used in the analyses, results suggest that reducing violence in and around schools is a smart economic investment. While most of the available analyses are from developed countries, there is little obvious reason why similar programs should not yield significant benefits in developing countries as well. 3.2 TEENAGE PREGNANCIES As noted in chapter 2, the incidence of teenage pregnancies in the Dominican Republic is high by regional and global standards. Such phenomenon has pernicious consequences for women themselves and for their newborns throughout life, making the need for action urgent. The first part of this section identifies the key factors associated with teenage pregnancies in the Dominican Republic that could later serve as entry points for policy recommendations. The second section provides evidence on “what worksâ€? to delay marriage and reduce teenage pregnancies. This review is based on both international evidence and an analysis of existing programs in the Dominican Republic. Relative to past publications (World Bank 2021b and World Bank 2022), this gender Assessment adds value to the treatment of teen pregnancy in at least two ways: (i) it shows how teen pregnancies are clearly interlocked with opportunities and/or limitations faced by girls and women throughout their life cycle (for instance, it puts into perspective how onerous early unions and early pregnancies could became for later stages in life by reducing the women's endowment of human capital and agency); and (ii) it maps and analyzes in much greater detail the institutional framework and policies (for instance, limitations in terms 107 See: MINERD (2013, Julio). Normas del Sistema Educativo Dominicano (En cumplimiento de los artículos 48-49, Ley 136-03. Aprobadas por el Consejo Nacional de Educación). 2da. Edición. Santo Domingo, D.N., República Dominicana. Santo Domingo, D.N., República Dominicana; MINERD (2016). Programa niños y niñas constructores/as de paz. Ministerio de Educación de la República Dominicana. Santo Domingo, D.N., República Dominicana; MINERD (2016). La mediación como herramienta de resolución de conflictos en el sistema educativo dominicano manual de entrenamiento para facilitadores. Ministerio de Educación de la República Dominicana. Santo Domingo, D.N., República Dominicana; MINERD (2017, July) Escuela por una cultura de paz: Guía estudiantes mediadores. Ministerio de Educación de la República Dominicana. Santo Domingo, D.N., República Dominicana; MINERD (2016) Guía práctica para el desarrollo de competencias en mediadores. Ministerio de Educación de la República Dominicana. Santo Domingo, D.N., República Dominicana; Civolani, K. (2016). Diseño de Propuesta de Mesas de Paz y Mediación Escolar. Organización De Estados Iberoamericanos y Ministerio de Educación de la República Dominicana. Santo Domingo, D.N., República Dominicana; MINERD (2016, Abril). Construyendo La Paz En La Escuela: Cuidado Y Buen Trato. Talleres para trabajar procesamiento positivo del conflicto en Educación Primaria. Modulo 1. Santo Domingo, D.N., República Dominicana; MINERD (2016, Abril). Construyendo La Paz En La Escuela: Cuidado Y Buen Trato. Guía Procesamiento Positivo del Conflicto para trabajar con estudiantes. Módulo 2. Santo Domingo, D.N., República Dominicana; and MINERD (2016, Junio). Protocolo Para La Promoción De La Cultura De Paz y Buen Trato En Los Centros Educativos. Santo Domingo, D.N., República Dominicana. 108 See: IDEICE (2014). Estudio de Prevalencia, Tipología y Causas de la Violencia en los Centros Educativos de Básica y Media de la República Dominicana. Instituto Dominicano de Evaluación e Investigación de la Calidad Educativa, IDEICE. Santo Domingo, D.N., República Dominicana; Parada, H., A. L. Ibarra, R. Burgos y M.E. Asuad (2017). Incidencia de violencia en adolescentes de 13 a 15 Años en las escuelas públicas del sistema de Educación de la República Dominicana. Santo Domingo, D.N., República Dominicana. 109 Gonzalez, K. (2013), Impacto del Modelo de Prevención de la Violencia del CEPREV y Formación de Multiplicadores. CEPREV. Doi: http://www.ceprev.org/uploads/1/4/0/7/14076803/investigacionmetodologiaceprev.pdf 56 of scope, sustainability in the provision of services and integrality) that have been implemented in the DR to meet the needs of girls and adolescents. 3.2.1: CORRELATES AND DRIVERS OF TEENAGE PREGNANCIES 1. The main societal factors that appear to increase the risk of teenage pregnancy are patriarchal social norms and child marriage. There is wide social tolerance of child marriage and early unions in the Dominican Republic.110 This is indicated by the large share of child marriages in the country, the highest in the LAC region. According to the National household survey (ENHOGAR-MICS, 2019), in 2018, one third of girls in the Dominican Republic ages 20-24 at the time of the survey marry before the age of 18. Involvement in an early marriage or sexual union is strongly linked with the probability of adolescent pregnancy. Seven in every ten adolescents (71.5 percent) married or in an early sexual relationship also reported being pregnant at least once. By contrast, pregnancies are much less prevalent among those who are single. Early marriage or union is more common among those living in extreme poverty (figure 38). Early marriage or sexual union could also lead to school dropout. Girls under an early marriage or union are more likely to drop out of school than their peers who marry later, and they tend to complete fewer years of education (UNICEF 2019).111 Traditional cultural and gender norms, and perceived lesser returns to girls’ schooling, feed into the joint decision-making process about marriage and schooling. One out three pregnant adolescents reported agreeing to become pregnant at that time and not having postponed childbearing. Figure 37. Marriage status of female population aged 15-19, by poverty status and pregnancy experience 120 97.2 100 80.9 80 71.5 70.3 36.8 Percent 8.9 6.7 29.7 60 31.8 29.3 40 28.5 29.7 43.6 19.1 36.4 6.1 5.2 20 25.5 29.7 13.8 14.4 2.8 8.4 14.0 12.4 6.9 8.9 6.9 0 Ever pregnant Never pregnant Ever pregnant Never pregnant Ever pregnant Never pregnant Ever pregnant Never pregnant Married or united Separated or widow Single, never married or in a Total relationship Extremely poor Moderately poor Not poor but vulnerable Not poor Source: Authors based on ENHOGAR 2018. 110 Báez et al. (2017) 111 ENHOGAR 2018 57 Other gender norms contribute to poor outcomes for adolescent girls (and boys) . Traditional gender norms promote an unequal and passive role for women in relation to their sexuality. Combined with a lack of access to information and comprehensive health services, including on SRH, this weakens women’s autonomy and exposes them to a greater risk of affective and sexual coercion, teenage pregnancy, abusive relationships, and unprotected gender and violence that lead to HIV and maternal mortality.112 These gender norms are often acquired within the family, at school and community spaces, and through media and virtual spaces and are learned and reinforced throughout life (Plan, 2017; Plan y UNICEF, 2014; Save the Children, 2019). Various studies in the Dominican Republic show that the perception of "being a woman" among girls and adolescents (International Plan, 2017) is associated with being a mother or wife. Likewise, these norms frame as acceptable relationships between girls and adults to escape dysfunctional families and violence, often due to family imposition, with the hope that things will be better (Plan and UNICEF, 2014). They do not promote emotional, affective, and economic autonomy as something to which girls can aspire. As for men, these norms also promote a vision of authoritarian masculinity where fathers and brothers (and other family members) can exert violence that often expels girls from their homes (Vargas, 2021). In particular, male roles characterized by an ambivalent paternity between physical absence and non-affective presence, and an authoritarian exercise of male presence in the home (Vargas, 2019), are common in Dominican households. This authoritarian masculinity is part of the violence that usually expels girls from their homes, only to later reach them again within the framework of affective relationships (unions). The emancipation associated with early unions, marriage, and adolescent motherhood, grant a level of relative autonomy to girls and adolescents that places them in a situation of even greater dependency. In some scenarios, early unions are a response to the need to escape highly violent, restrictive, and authoritarian family environments and a previous pregnancy situation (Save the Children, 2019). Women who have ever been forced to have sexual intercourse are 1.86 times more likely to get pregnant before the age of 18 compared to women who have not suffered this type of violence. Likewise, the age of first sexual abuse is strongly related to adolescent pregnancy. The probability is 3.26 times higher among women who were also abused before the age of 18 compared to women abused after that age (Marques-García, 2019). 2. Schooling can discourage teenage pregnancy by increasing the economic stakes of dropping out and by providing sexual and reproductive information. Adolescent pregnancy rates are especially high in regions bordering Haiti and tend to decline as income levels rise. Households in poverty have the highest pregnancy rates across all maternal age groups and over a lifetime. In 2018, the share of pregnant women aged 15 to 19 was more than five times higher in households living in extreme poverty (44.3 percent) than in non-poor households (7.7 percent) The shares 112 In the Dominican Republic, families are encouraged to marry their daughters at an early age as an accepted cultural practice. Partnering or marrying an older man is perceived as a social duty among girls and young women. There is also strong social pressure to marry off daughters as soon as possible to prevent her from becoming sexually active before marriage: a woman who does so is seen as dishonoring her family and her community. Parents make the decision to marry off their daughters at an early age because they have stereotyped ideas about sexuality and the role of women in society. In these circumstances, early marriage and union is seen to protect girls from the risk of sexual violence, to prevent unmarried older girls from being criticized as “impureâ€?, to restore family honor in cases of violence, or hide real or perceived non-heterosexual orientation (Disla, 2020; Vargas, 2021). Existing sex norms also promote pregnancies and early unions in girls, boys and adolescents, since from an early age the differences in roles emphasize greater freedoms in public spaces for men and subject girls and adolescents to a greater burden of domestic labor (Tineo, 2014). Perceptions about child marriage (MI), early unions and teenage pregnancy are often transmitted from mothers to daughters and between sisters to help them fulfill their social duty of being "good wives and mothers". 112 It is also a product of adult-centric patriarchy. The adult world (families, school and community spaces, etc.) tend to invalidate the experiences, knowledge, and concerns of children and adolescents. Sex violence is internalized from an early age in family dynamics, with recurrent patterns of aggressor fathers/stepfathers towards mothers, whose children then replicate with their own partner and within the constituted home (Plan and UNICEF, 2014, Plan 2020, Disla 2020). Constant exposure to media reinforces stereotypes (racial, sex, class, age) and fetishize adolescent girls as usable objects by the male adult world (Vargas, 2021). 58 of adolescents aged 15-19 who reported never being pregnant at the time of the survey was substantially higher among non-poor households’ (figure 39). The rural adolescent pregnancy rate is 23.7 percent, well above the urban rate of 18 percent. The highest adolescent pregnancy rates are registered in the regions bordering Haiti: Enriquillo (27.7 percent), Cibao Noroeste (25.6 percent), and El Valle (23.1 percent). Figure 38. Pregnancy status of the female population aged 15-19, by poverty status 120 100 80 Percentage 60 44.3 35.7 40 18.8 19.1 20 7.7 0 Extremely Moderately Not poor but Not poor nor Total poor poor vulnerable vulnerable Poverty status Ever gave birth Loss or miscarriage Pregnant for the first time Never pregnant Ever pregnant Source: Authors based on ENHOGAR 2018. Note: The DR’s Unique Beneficiary System (Sistema Único de Beneficiarios, SIUBEN) assigns households to one of four socioeconomic groups based on a multidimensional Quality of Life Index (Indice de Calidad de Vida, ICV). Households in extreme poverty are classified as ICV-1; moderately poor households are classified as ICV-2; non-poor but vulnerable households are classified as ICV-3; and economically secure households are classified as ICV-4. Living in an urban area, being income poor, and having a low level of education appear to increase the risk of teenage pregnancy at the individual level. Using information from ENDESA we estimate a probabilistic model on the determinants of late adolescent pregnancies in DR. Results show that despite the higher rate of teenage pregnancy in rural areas of the country, when individual characteristics, education and labor status are considered as controls in the model, urban adolescents appear to face a higher probability of becoming pregnant (Annex 4)113. Women in municipalities with higher rate of adolescent population and more positive perceptions of adolescent motherhood are more likely to become an adolescent mother themselves. The family structure also matters: if the father is present within the adolescent girl’s household it statistically decreases the probability of late adolescent pregnancy. Having a low education appears to increase the risk of becoming pregnant. Based on the ENHOGAR 2018 survey, UNFPA (2021) shows that the conditional probability of not being a mother is higher in adolescents who attend school (0.846) than in those who do not attend (0.631). Adolescents who attend school also show lower conditional probabilities of being intentional mothers (0.04) and unintentional mothers (0.11) than adolescents who do not attend school (whose probabilities are 0.09 and 0.279, respectively). The risk of becoming pregnant is 6 times higher among women with only a primary education compared to women with higher education. 113 To assess the significance of key risk factors in the likelihood of late adolescent pregnancy, a series of weighted probabilistic linear models were estimated taking into consideration variables associated with five groups of factors: place of residence, household characteristics, personal life, and educational and labor status. 59 Conversely, teenage pregnancies (and child marriage) are recognized as one of the main factors leading girls to drop out of school prematurely, especially among the poor. ENHOGAR 2018 reports that among the poorest girls who became pregnant as teenagers, more were out of school when becoming pregnant (38 percent) or dropped out of school during the pregnancy (27 percent) than remained in school while pregnant (30 percent). School attendance is at further risk again after childbirth. Among adolescents who were attending school when they become pregnant, 21.6 percent did not continue attending school during pregnancy (figure 40). While for some adolescents this was a temporary exit (8.4 percent), for most it was permanent (21.7 percent). Similarly, of those who remained in school during pregnancy, 12.6 percent did not return to school after childbirth.114 The situation is worse among poor adolescents. Of those who attended school at the time of pregnancy, 42 percent stopped attending during pregnancy and did not return after childbirth, and 13.5 percent gave up continuing to attend after childbirth. Nonetheless, 54.3 percent of adolescents who were attending school when they became pregnant ultimately stuck with their schooling: 45.9 percent stayed in school while 8.4 percent returned to school after childbirth, suggesting that girls are interested in continuing their education if the conditions exist to do so. Figure 39. Adolescent girls aged 15-19 by school attendance status at the time of pregnancy, by multidimensional poverty status 100 5.0 9.1 10.9 18.7 10.7 80 38.0 24.7 24.1 24.3 16.7 60 20.1 18.6 23.5 21.6 40 27.0 20 41.0 46.7 46.0 43.5 30.0 0 Extremely poor Moderately poor Not poor but Not poor nor Total vulnerable vulnerable No information Not at school when became pregnant At school when became pregnant and out of school during pregnancy At school when became pregnant and during pregnancy Source: Own elaboration based on ENHOGAR 2018. The public policy challenge is therefore to create the conditions for adolescents who become pregnant to remain in or return to school. Many teenagers drop out of school during pregnancy because of embarrassment. They suddenly face psychological pressure for which they were unprepared. Among those who stopped attending school during pregnancy, 5 percent indicated that they felt ashamed, a percentage that rises to 12 percent in adolescents within extremely poor households. Being in school may reduce teenage pregnancies by increasing the economic stakes of dropping out. Education leads to a greater accumulation of human capital that increases the opportunity cost of an eventual teenage pregnancy. The educational opportunity cost for Dominican women who became mothers at 10-19 years is substantial. UNFPA (2021) estimates losses in earnings due to lower educational attainment caused by pregnancies at USD 51 million annually (RD $ 2,569 million) for women who became mothers as adolescent. In addition, female heads of households who became mothers as adolescents also seem to lose USD 8 million in earnings annually due to a higher risk of unemployment.115 These types of estimations typically assume that labor markets would be able to absorb a larger supply of better educated women. 114 Econometric analysis using instrumental variables conducted for a global report suggests that child marriage and teenage pregnancies may also have a negative effect on schooling at the secondary level. Wodon et al. (2017). 115 Preliminary estimates from Socioeconomic Consequences of Adolescent Pregnancy in the DR, UNFPA 2021. 60 Box 6. Lifetime losses in earnings for women who had become pregnant as adolescents and drop out of school may be significant. The personal benefits of work go well beyond earnings, but earnings are crucial for standards of living and are a handy way of measuring the potential cost of not educating girls. There is a large literature (based on earnings regressions) on the potential impact of educational attainment on earnings that applies to boys and girls alike. Summary estimates of such regressions for the Dominican Republic using 17 years of data are provided in Table 12. Educational attainment is measured in levels, which are better at capturing non-linearities than years of schooling. Average gains associated with each level of educational attainment are provided, treating all years equally. In addition, consideration is given to whether workers are literate (as subjectively declared). For women with primary education (partial or completed) who say they are not literate, the expected gain in earnings versus no education is only 12 percent on average and is often not statistically significant. This suggests that without effective learning, primary education does not lead to gains in earnings. When women have primary education and declare themselves to be literate, the average gain is larger, at 28 percent. As expected, the gains are higher at the secondary and tertiary levels, at 59 percent and 127 percent respectively. In general, differences in the returns between men and women are small. Table 11: Estimates of the returns to educational attainment, 17 years of data All Men Women Primary, not literate 12.2% 10.4% 18.9% Primary, literate 28.4% 31.3% 25.0% Secondary 59.1% 61.8% 59.3% Tertiary 126.8% 128.1% 133.7% Source: Authors’ estimates based on ENFT and ENCFT. For both men and women, the gains from educational attainment start to be substantial only for secondary and tertiary education. Why are the gains from primary education comparatively small? On the demand side, employers may require workers to have skills that a primary education does not provide. On the supply side, even workers with a formal primary education may not have meaningfully acquired skills that they should have like basic literacy and numeracy. As noted in the World Development Report (World Bank, 2018), many education systems are witnessing a learning crisis wherein enrollment and attendance in school do not automatically ensure that sufficient learning is taking place. This is also clear from the differential in returns to primary education based on whether workers are literate or not. Schools (and family) also facilitate access to information on reproductive biology. This in turn can lead to more informed decisions about sexual and reproductive health and reduce pregnancies by improving the agency and decision-making of the adolescent. Family and school are the main sources of information on sexual and reproductive health for teenagers. According to ENHOGAR 2018, girls and teenagers learn how to take care of their sexual and reproductive health mostly from parents (64.4 percent) followed by their teachers at school (20.7 percent). Teachers are more commonly a source of information on sexual and reproductive health for adolescents who live in poverty. Importantly, adolescents aged 15-19 who have never been pregnant report greater exposure to reproductive physiology courses taught at school (81.3%) than adolescents who have been pregnant (67.8%). The level of exposure to these courses is lower among the poor. Courses on reproductive physiology are offered precisely at the level of schooling that is generally less attended by poor teenagers. As figure 41 below shows, more than eighty percent of girls aged 15-19 from the poorest backgrounds stopped attending school before second grade of high school, which is when courses on reproductive physiology are taught in the DR. Figure 40. Female population aged 15-19 that stopped attending school according to educational level reached, by poverty level 61 120 100 - 1.6 4.0 3.2 2.9 14.7 80 Percentage 38.8 60 45.6 68.4 62.7 89.7 40 33.9 26.0 22.7 20 22.0 20.4 13.1 4.11.5 0.66.3 8.8 0 3.6 0.2 2.4 Extremely poor Moderately poor Not poor but Not poor Total vulnerable Primary Education 1st cycle Primary Education 2nd cycle Higher Education 1st cycle Higher Education 2nd cycle Terciary Education Source: Authors based on ENHOGAR 2018. 3. Insufficient access to sexual and reproductive knowledge and services, or lack of economic opportunities, also play a role in teenage pregnancies. For most Dominican teenagers pregnancy is an unplanned event. In 2018, survey data suggest that only around one third (36 percent) of pregnant teenagers aged 15-19 wanted to become pregnant, while 62 percent would have preferred to postpone pregnancy. Among those who gave birth and were in school when becoming pregnant, more than two thirds (69.2 percent) would have rather had a child later. Among adolescents who gave birth and were not in school at the time of pregnancy, the desirability of the pregnancy was higher (35.6 percent), but most would still have wanted to postpone the birth (63 percent). Life skills and SRH knowledge can help girls avoid unwanted pregnancies. These data underscore the need for better information and access to sexual and reproductive health services.116 Family planning and access to contraception have been identified as key preventive factors of teenage pregnancy. According to ENHOGAR 2018, only 46 out of 100 adolescents used a contraception method in their first experience of sexual intercourse. The ENDESA survey for 2013 also shows important gaps in access to contraception, especially among younger women: 27 and 21 percent of 15-19 and 20-24 year-olds respectively said they struggled to access it.117 The lack of adequate use of contraception, in connection with insufficient knowledge or maturity, was the main reason of pregnancy reported by pregnant girls in a recent qualitative study conducted in the Dominican Republic.118 Yet teenagers seek little counseling on preventative sexual health measures, except among those living in extreme poverty. In 2018, only one in five adolescents requested a health service at a health center for adolescents. Most searched for advice once pregnant or in an emergency. Prevention is not a regular practice among adolescents. However, there is an acute demand for contraceptive counseling and methods among teenagers who have never been pregnant and are in dire poverty. In other words, among adolescents aged 15-19 who have never been pregnant and are aware of the existence of comprehensive adolescent health care units in hospitals and Centers of First Level of Health Care (CPN for their acronym in Spanish), the percentage in extreme poverty who attend those units seeking counseling on contraceptive methods (9.5 percent) and obtaining contraceptives (22.4 percent) is significantly higher than in adolescents living in 116 World Bank (2020) 117 INTEC (2016) 118 UNDP (2018) 62 moderate poverty or the non-poor (figure 42). The main health establishments where adolescents demand reproductive care services are public hospitals (66 percent), followed by private centers or clinics (25.9 percent).119 Figure 41. Knowledge of selected health care services among female adolescents (15-19), by pregnancy experience and multidimensional poverty status 100 6.9 14.2 10.6 20.4 17.5 23.3 22.4 24.5 25.0 31.2 37.1 36.4 31.9 80 18.9 42.8 37.9 33.5 17.2 5.0 19.5 18.8 10.3 9.6 9.3 7.4 9.5 6.8 5.5 6.8 6.5 60 Percent 5.0 5.7 8.1 7.8 7.4 9.9 8.3 15.4 22.4 40 61.2 68.2 67.4 61.9 55.1 55.0 55.5 53.1 58.7 52.7 58.8 49.7 47.3 44.1 20 36.9 0 Total Total Total Not poor but vulnerable Not poor but vulnerable Not poor but vulnerable Extreme poor Extreme poor Extreme poor Moderate poor Moderate poor Moderate poor Not poor nor vulnerable Not poor nor vulnerable Not poor nor vulnerable Ever pregnant Never pregnant Total Gynecological consultation Search for contraceptive methods Pregnancy consultation Contraceptive counseling Other Source: Authors based on ENHOGAR 2018. Unfortunately, the demand among poor girls to access comprehensive health care services is unmet, despite the important role played by publicly provided services. Most girls aged 5-19 do not see the need to attend a specialized health clinic. However, many living in extreme poverty are unaware of the existence of specialized health care clinics for adolescents. The ENHOGAR 2018 shows that 56 percent of adolescents aged 15-19 were not aware of the existence of comprehensive adolescent health care units in the public health system. This percentage exceeds 60 percent among adolescents living in extreme poverty. The inexistence or distance of comprehensive health care centers and spaces in or from communities also seems to prevent poor girls from seeking support, particularly when already pregnant. In the Dominican Republic, living in an urban area and having a poor education appear to increase the risk of teenage pregnancy at the individual level, while the main societal factors are patriarchal social norms and child marriage. Better opportunities to access information and services on sexual and reproductive health, and more equal labor market opportunities for women, have also been identified as key preventive factors of teenage pregnancy. 3.2.2: POLICY RECOMMENDATIONS TO ADDRESS TEENAGE PREGNANCIES This section documents and discusses selected policy options to reduce teenage pregnancies in the Dominican Republic. Each policy recommendation aims to tackle key factors associated with the increase in teenage pregnancies documented in the previous section. As a preview, the specific recommendations on teenage pregnancy based on the key findings emerging from this deep dive include: (a) ensure the enforcement of Law 1-21 on child marriage; (b) keep and expand the Aprende and Avanza programs aimed at promoting the completion of primary and secondary schooling within the Supérate strategy, to increase the opportunity costs of dropping out of schools; (c) 119 Establishments of the National Health Service (SNS) 63 evaluate ongoing life skills programs such as the Girls Club to consolidate and scale up successful interventions; (d) scale up the PRONAISA program to all level-one units; (e) introduce SRH education in schools at all levels; (f) expand low-cost digital modalities to deliver SRH services; (g) incorporate SRH services intro the Family Health Insurance program; and (h) mandate an annual medical visit by children and adolescents ages 6-15 years for preventive purposes as a condition to access SP cash transfer programs such as Aprende and Avanza (figure 44). Figure 42. Drivers of teenage pregnancy and selected policy recommendations The Dominican Republic has spent more than two decades trying to consolidate (with varying degrees of progress) policies to reduce teenage pregnancy, end child marriage, and educate girls. The country has constitutional, legal and regulatory mandates aimed at protecting and guaranteeing the rights of adolescents. Inter-sectoral and inter-institutional plans and policies have been formulated since 2010 to address the economic, social, and cultural determinants associated with adolescent pregnancy and early unions. These plans and policies include the Strategic Plan for the Prevention of Pregnancies in Adolescents: Towards a National Policy (Plan-EA 2011), the National Plan for the Reduction of Pregnancies in Adolescents PREA-RD 2019-2023, the National Plan for the Elimination of Child Marriage and the Reduction of Early Unions, the National Plan for gender Equity 2020-2030 (PLANEG III), the National Strategic Plan for a Life Free of Violence for Girls and Women (2020), the "Coordination in the prevention of gender Violenceâ€? agreement, and the National Policy of prevention and attention to early unions and adolescent pregnancies (2021). Interventions have been implemented, among others, for SRH education, the prevention of child abuse, and care for adolescent mothers. Most recently, the Cabinet for Children and Adolescents (GANA) was created as a coordinating body in 2020. Overall, while laws and broad-brush policies can help to reduce teen pregnancies and end child marriage, specific interventions are required. This section summarizes international evidence as well as experiences in the Dominican Republic related to three main types of interventions: (1) programs that provide girls with life skills and reproductive health knowledge; (2) programs that expand girls’ economic opportunities; and (3) programs designed to ensure that girls remain in school or that enable them to return to school. Each type of program is based on a different theory of change (Box 6). The summary of findings is based on a review by Botea et al. (2017). To qualify for the review, interventions had to (i) target girls aged 10–19, either exclusively or as part of a larger group; (ii) equip girls with life skills and SRH knowledge, economic opportunities, or educational opportunities; (iii) demonstrate results in terms of improving the health of young women, especially SRH, or delaying marriage or childbearing; and (iv) have 64 been tested in a developing country (see Kalamar et al., 2016, for another review of the international evidence). Box 7. Theories of Change for Interventions Targeting Adolescent Girls Life skills and SRH knowledge: By increasing their knowledge, life skills coaching can raise girls’ awareness of the risks associated with becoming pregnant at an early age and increase their desire and ability to avoid early pregnancies through family planning. Through such channels, life skills may lead to better health outcomes for girls and their children. By increasing girls’ confidence and self -esteem, life skills may also help expand their aspirations, which may increase their motivation to delay marriage and childbearing. Finally, life skills can increase the communication and decision-making skills of young women and increase their abilities to negotiate their marriage and childbearing preferences with partners. However, while life skills and SRH knowledge may empower girls, they may not be sufficient to delay marriage and childbearing if social norms curtailing the agency of girls are not also addressed. Economic opportunities: Programs to increase young women's earnings potential may increase their ability to plan and improve their marriage and childbearing decisions in three ways: (1) Improvement in a woman’s ability to make an economic contribution expands her accepted role beyond that of gender and reproduction. This can increase a girl’s desire to delay marriage or childbearing. The transformation of girls from economic liabilities into assets in the eyes of their communities and families can also alleviate the external pressures on them to marry or have children early. (2) The loss in earnings associated with childrearing is an opportunity cost that may increase women’s desire to limit or space births and to exercise reproductive control. (3) Increased earnings may suppleme nt a young woman’s bargaining power within the household and allow her to effectively exercise reproductive control by negotiating delays in genderual debut or marriage, and to better negotiate the terms of gender, such as the use of contraceptives. Creating income-generating opportunities for women can therefore, in addition to the direct economic benefit, also contribute to female empowerment by widening a woman’s personal choice and control over SRH outcomes. Incentives for girls to participate in schooling or delay marriage: In many communities, the economic, cultural, and social environment does not offer adolescent girls viable alternatives to marriage. Once girls drop out of school, possibly because of the school’s poor quality or high cost, parents may find it difficult to identify any option other than marrying off their daughters. In such communities, providing better-quality and more affordable primary and secondary education may be one of the best ways to delay marriage and teenage childbearing. Programs to keep girls in school may also lead to cultural tipping points in communities that make it easier for more and more girls to stay in school and thus delay marriage and teenage pregnancies. Some interventions have also aimed at delaying marriage by providing financial incentives conditional on not marrying early, often resulting in girls staying in school for longer. Source: Botea et al. (2017). Life Skills and Access to Sexual and Reproductive Health Knowledge (SHR) Life skills and access to SRH knowledge interventions and services can be highly beneficial for girls. In many countries, adolescent girls are less likely than older women to access SRH services, including modern contraception and skilled assistance during pregnancy and childbirth. International evidence shows that SRH knowledge would generate multiple benefits, enabling girls to stay healthy, make independent decisions about their health, avoid unintended pregnancies, finish their education, engage in productive work, and choose to have fewer and healthier babies when they are ready. Another intervention trialed is the “safe space clubâ€? for girls. These clubs convene under the guidance of a trusted adult mentor at a specific time and place and have proven effective when implemented well. By combining opportunities to socialize and have fun with access to mentors, the clubs are attractive to girls and offer a platform for other services. Clubs can be held in a variety of settings, including schools or community centers. Girls can discuss a range of issues under the guidance of the mentors, including those related to SRH. The clubs facilitate the delivery of life skills, including “softâ€? or socio-emotional skills such as critical thinking and problem solving, 65 negotiation, and communication (for example, within the household). One of the objectives is often to boost self-awareness and self-esteem so that girls can explore and fulfill their own aspirations. Often, safe space clubs are also used to facilitate the delivery of “hardâ€? skills as basic literacy, numeracy, and business skills. Several pilot projects promoting life skills and access to SRH knowledge have been implemented in the Dominican Republic. These projects are aimed at adolescent girls, but some incorporate boys (see Annex 5). A first recommendation consists of evaluating the intervention methodologies and impact of these pilot programs to decide what gets scaled up or consolidated. Consolidation is vital to avoid program overlaps for the same adolescent population, and given the limited scale of most projects. Evaluations can help determine if it is worth continuing with ongoing interventions. In doing so, more participatory, and interactive methodologies can help gauge adolescents’ own perceptions and involve parents and their knowledge in the programs. Life skill programs should be absorbed by central public bodies to guarantee their continuity. Scalability is conditional on the availability of financial and human resources and the establishment of inter- institutional agreements between public institutions to carry out joined-up work. Both goals can be achieved if the pilot projects become hosted by a state institution which can make them part of its respective institutional strategic plans.120 Successful pilot projects should work closely with municipalities. The experience of the Girls Club shows the merits of collaboration with city councils and local governments. In the Dominican Republic, most vice mayors’ offices in municipalities are occupied by women, and generally show a high commitment to supporting programs aimed at girls and adolescents. In addition, municipalities should allocate 4 percent of their total budget to financing projects that contribute to gender equality and support adolescents. Programs and projects that encourage the development of new masculinities and healthy life attitudes, as well as alternative life plans for boys, should also be created. As well as MINERD, this may also involve entities like the Ministry of Sports and Recreation and the Ministry of Culture. While international experience suggests that life skills and SRH knowledge programs are beneficial, they may not be sufficient by themselves to delay marriage and childbearing. Outcomes in impact evaluations have included increases in girls undergoing HIV testing or counseling; greater use of modern contraception or other methods of family planning; a reduction in parents’ desire for FGM for daughters in countries where the practice is prevalent; a reduced risk of intimate partner violence when a program also reaches out to men; higher self-esteem; and gains in specific skills taught in safe-space sessions, such as basic literacy, numeracy, and financial skills. However, without additional supportive interventions to enable girls to participate in schooling or employment or otherwise improve their livelihood options, it is not clear that safe spaces alone can delay marriage and childbearing (perhaps because that is not always the primary goal of such programs). Therefore, it is important to consider developing and/or consolidating programs where safe spaces have been combined with measures to improve livelihood opportunities or offer incentives to remain in school, which are usually more effective in delaying marriage and childbearing. Increase Livelihood Opportunities and Economic Incentives for Schooling 120 They can be assigned to the rectory of MINERD, MSP, Ministry of Woman, Ministry of Youth or others. 66 Apart from focusing on life skills and SRH knowledge, a second category of programs emphasizes both empowering girls and providing livelihood opportunities. These programs are particularly appropriate for girls who are not in school and would otherwise have no income-generating skills. Two groups of interventions are distinguished in the review by Botea et al. (2017): (1) livelihood interventions and (2) interventions to improve financial literacy and access to financial services. These programs are often successful in terms of increasing earnings, employment, or savings. Several programs have also resulted in increased use of modern contraceptives and improved SRH knowledge and some delayed marriage and reduced pregnancies, usually more so than for life skills/SRH knowledge programs alone. For example, the BRAC Uganda Empowerment and Livelihoods for Adolescent Girls proved to have the following impacts: (1) The likelihood of girls engaging in income-generating activities went up by 32 percent; (2) Among sexually active girls, self-reported routine condom use went up by 50 percent; (3) Fertility rates went down by 26 percent.; and (4) Reports of unwanted gender plunged by 76 percent. There were also reductions in teenage pregnancies and child marriage, and a shift in community gender dynamics (Bandiera et al. 2014; Buehren et al. 2016). The lesson from those programs is that adding a livelihood dimension to life skills and SRH knowledge programs may help delay marriage and childbearing. The focus on economic opportunities may also help to ensure the regular participation of girls in the programs. The Dominican Republic has experience with successful programs aimed at increasing the employability of poor young adolescents. The Youth and Employment Program, implemented between 2003-2010, was aimed at the population aged 16 to 29 who were out of school, with incomplete secondary education and from the poorest 40% of households. At least 45% of the participants had to be female. The program included classroom-based vocational and life-skills training as well as an internship with a private- sector employer. The program showed a positive impact on labor participation and on the probability of working (the subsequent employment rate was 8.4 percent higher than in the control group) especially among women.121 In the medium-term the impacts dissipated, suggesting the need for a continuous monitoring process for the target population. Both in the short term and in the long term, women showed a higher level of satisfaction with their employment situation and the incidence of pregnancy was reduced, mainly in women between 16 and 19 years of age. More recently, an economic inclusion component for adolescents and young people has been implemented as part of the Progresando Unidos Program and contained affirmative actions for women (i.e., at least 50 percent of beneficiaries had to be women), but there were no positive impacts on employment or better quality jobs for women in particular. The new SUPERATE program, the successor of the DR’s principal anti-poverty strategy Progresando con Solidaridad (PROSOLI), also has a marked emphasis on capacity building for economic inclusion among women and young people. Box 8 Guatemala’s Abriendo Oportunidades program The issue of early marriages has become a policy priority for the Dominican government. The Dominican Republic has committed to eliminate child, early and forced marriage by 2030 in line with the Sustainable Development Goals. In January 2021, the Government completely banned child marriage, was previously legal for girls at 15 and boys at 16 years old with parental consent (and even earlier with judicial consent).122 This represents a major step towards the end of this harmful practice. However, the experience from other countries suggests that, while a first necessary condition, legal change is often not enough to prevent early marriage on the ground, especially in areas or among groups where it is still socially accepted. Working to shift social and gender norms and expanding access to education and other opportunities will also likely be required.123 Some good practices exist in the region, such as Guatemala’s Abriendo Oportunidades program. The program aims to help rural girls to stay in school and develop skills, and offers coaching on their sexual and reproductive rights. 121 See Vezza et al. (2014) 122 https://www.girlsnotbrides.org/child-marriage/dominican-republic/ 123 Wodon et al. (2017) 67 Almost all participating girls did not marry for the duration of the program and most of them expressed a desire to postpone childbirth after the program. It also helped to bring about positive change in beliefs and views regarding early marriage in the communities. The success of Abriendo Oportunidades has led to its replication in other LAC countries including Mexico and Belize.124 A third category of programs aims to ensure that girls remain in school, enable them to return if they have dropped out, or directly delay marriage. There have been numerous interventions in more than 29 low-income countries to keep girls in school and delay marriage (Kalamar et al. 2016; Ozler et al. 2021; Bastagli et al. 2016). In some cases, evaluations demonstrate that programs that provide incentives for girls to remain in school often succeed in delaying marriage or childbearing. While most of these programs are designed to keep girls in school, some are also designed to enable girls who dropped out to return to school. Also, conditional cash transfers can be effective to incentivize girls’ schooling, promote health, and support families during shocks. These incentives are often conditional on children’s attendance at school or participation in preventive medicine programs. A significant body of research shows that such transfers have been effective in promoting participation in schooling by children in developing countries. While not all programs succeed everywhere, the evidence is quite convincing that in comparison to the other two types of programs reviewed above (Life skills and SHR; livelihood opportunities and long-term economic incentives), those focusing directly on schooling for girls (or in some cases using immediate financial incentives), may be more successful in delaying marriage and childbearing. In the Dominican Republic, Supérate, a program that serves 1.3 million households, provides such opportunities. Making cash transfers to poor households conditional on them sending their children to school began in 2004 with the launch of the Solidarity Program, which granted the Incentive for School Attendance (ILAE) transfers to encourage school attendance at the primary education level. In 2012, a new conditional cash transfer called Student Voucher Studying to Progress (BEEP) was established to encourage attendance at secondary level. ILAE and BEEP were renamed in 2021 as Aprende and Avanza, respectively, under a renewed comprehensive strategy to fight poverty and promote inclusion called Supérate. Maintaining and expanding the Aprende and Avanza programs within Supérate is recommended (See Box 9) given the positive impact of these CCT programs on high school attainment and their reduction of teenage pregnancy and early unions. Similar programs in other contexts have been evaluated and show positive impact on teen pregnancy (Schultz 2004; Baird et al. 2011; Baird et al. 2013; Bergstrom, K., & Ozler, B. (2021). Box 9. The Effects of Supérate school transfers on high school completion In 2013, Supérate incorporated Avanza (then called BEEP), a second school transfer financed by the Ministry of Education and targeted at high school students in beneficiary households. Avanza implied an increase in school transfers of up to 3.3 times the amount received relative to Aprende (then called ILAE), the basic school transfer (which started in 2005 and was conceived for all eligible households with school-aged children and then targeted exclusively at primary school students after the creation of BEEP). In a recent work (Hernandez et al, 2022) the authors take advantage of the partial coverage of Avanza (due to restrictions in the Ministry of Education school enrollment verification process and Supérate budget constraints) to study the effects on high school completion of receiving additional school transfers at this level. The database gathered for this study combines three administrative datasets: the Ministry of Education’s National Exams database, the historic payroll data of Supérate provided by the Social Subsidies Administrator (ADESS), and two waves of SIUBEN’s census data. SIUBEN’s first wave w as collected mainly between 2004 and 2008 and contains information on both beneficiary and non-beneficiary households before the former were included in the program. 124https://www.popcouncil.org/research/abriendo-oportunidades-opening-opportunities;https://www.girlsnotbrides.org/wp- content/uploads/2016/02/Case-Study-Empower-girls-Population-Council-Guatemala.pdf 68 The estimation sample includes beneficiary students who took the mandatory tests to graduate from primary education between 2010 and 2013, as they could have been exposed to the additional transfers implied by Avanza for at least one year of their high school education.125 Furthermore, the study compares the incidence of high school completion between students whose households receive Avanza transfers versus students whose households receive Aprende transfers only, but who otherwise share similar program participation probabilities at baseline.126 Results suggest that receiving Avanza is, on average, associated with an 11.7 to 13.2 percentage-point higher probability of completing high school relative to not receiving these additional transfers. Considering average graduation rates among the matched control groups, the estimated effects represent an increase in the high school graduation rate of 23 to 25.3 percent compared to the counterfactual (i.e., with no Avanza transfers). Furthermore, the effects are slightly larger in rural areas among male students compared to their peers in urban areas, although the opposite is true for female students. In addition, the transfers appear to play an especially important role during the last high school year of targeted students. Effect of 10 additional US$ in high school transfers on delaying age for having first child (in months) Relative to incomplete secondary 0.35 education 0.4 Relative to completed primary 0.6 education 0.6 Rural Areas Urban Areas Source: authors' calculations. In addition, the authors perform back-of-the-envelope calculations to approximate the potential benefits of receiving additional BEEP transfers on employment, salaries, and delayed parenthood. This analysis is relevant in a context where the returns to education are not small and where the fertility rate among adolescent girls and women ages 15-19 is one of the highest in the Caribbean region (22 percent). For example, we combine the estimated effects of additional cash transfers on high school completion with estimates of delay in having a first child by education level obtained from the National Demographic and Health Survey, ENDESA (2013). Doing so indicates that 10 additional US dollars per year transferred to the group of girls that received both school transfers translate into a 0.6 month delay in having their first child, both in urban and rural areas, relative to those who only completed primary education, and around 0.4 months relative to those with an incomplete secondary education (see figure above). Source: Hernandez et al (2022) Access to early childhood care services for adolescent mothers is also vital to them remaining in school or returning after giving birth. Providing childcare services is another key way for adolescent mothers to remain in school after giving birth or return to school and continue their education in case of dropping out, especially for those who come from poor households. After the birth of the child, the main reasons for teenagers to drop out of school are: i) not having someone to leave their baby with (37.1 percent), ii) it being their first child (14.4 percent), and iii) having to work (6.6 percent). Not having someone to leave their baby with is by far the main reason for dropping out of school among adolescents who belong to households in extreme poverty (Figure 44). Incidentally, childcare services can also reduce gender gaps around labor participation: this is discussed in the next deep dive. 125 Avanza (previously BEEP) was suddenly discontinued in July 2016 by the Ministry of Education and restarted in early 2019, this time financed directly by Supérate with modifications in the transferred amounts. 126 Households of students in the first group also received Aprende (previously ILAE) during their childrens’ primary education years. In addition, the Aprende-only group were still eligible to continue receiving this transfer during high school if they were not covered by Avanza. 69 Figure 43. Reasons for not returning to school after pregnancy among 15–19-year-olds, by poverty status Total 14.4 6.6 37.1 4.7 4.0 5.9 20.4 It was her first pregnancy Not poor nor vulnerable 5.8 9.6 45.8 13.1 5.0 13.9 TOTAL She had to work Not poor but vulnerable 12.6 7.7 33.5 4.2 4.3 7.2 22.7 Moderate poor 18.9 6.2 37.1 6.1 2.7 6.5 18.1 She had no one to leave the child Extreme poor 9.9 3.4 47.6 4.8 3.3 0.0 23.6 with She did not want to be separated Attended school during Did not attend school from his son or daughter during pregnancy Total 7.5 9.7 24.1 6.0 3.8 7.1 32.4 Her son or daughter was sick Not poor nor vulnerable 47.7 37.8 14.5 Her husband was opposed to her Not poor but vulnerable 4.4 9.3 20.4 5.7 13.5 31.1 attending Moderate poor 13.1 12.6 25.1 6.7 2.3 32.7 For health problems Extreme poor 3.5 7.0 24.4 7.9 6.6 0.0 43.9 Due to family problem Total 19.2 4.5 46.3 3.9 4.2 5.1 12.4 She changed residence pregnancy Not poor nor vulnerable 8.9 14.6 44.7 0.0 0.0 21.3 For tiredness Not poor but vulnerable 18.1 6.5 42.3 3.1 6.1 2.9 16.9 Moderate poor 22.6 2.1 44.7 5.7 3.9 9.2 8.8 She finished her studies Extreme poor 16.2 70.1 1.9 3.9 Other 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Source: Authors based on ENHOGAR 2018. Supérate is currently planning to address issues related to job awareness, training and job placement, and develop care services through a network of community caregivers that facilitate the return to school of adolescent mothers. The articulated and coordinated work among Supérate, the Ministry of Public Health, the Ministry of Education, INAIPI (Annex 6), CONANI and INFOTEP creates the opportunity to strengthen programs and projects in progress that promote the economic autonomy of women and keep adolescents in school. This implies the reallocation and protection of financial resources; the training of human resources; and the development by the authorities of a common agenda aimed at addressing the key elements that affect the problem of adolescent pregnancy and early unions. There are currently 122 Comprehensive Early Childhood Care Centers (CAIPI) and 521 Child and Family Care Centers (CAFI) in the DR.127 An assessment of these programs including an evaluation of their impacts is desirable to inform their potential scaling-up. The provision of community care services in under-served municipalities where there is a higher school dropout associated with pregnancy is also urgently needed; this has been prioritized by GANA in implementing its Policy to reduce early unions and teenage pregnancy.128 There is also a need to improve qualifications and infrastructure in schools to meet the needs of pregnant adolescents and young mothers, including in terms of extra-curricular support strategies.129 Coordination is needed between Supérate and the Ministry of Public Health, MINERD, INAIPI, the 127 INAIPI Memories, 2020. 128 The Cabinet for Children and Adolescents (GANA) is a coordination body created in November 2020, chaired by the first lady Raquel Arbaje and coordinated technically by CONANI, which has a mandate to: (i) implement the policy of prevention and attention to early unions and pregnancy in adolescents, and (ii) harmonize the key public plans around these problems and their intersectoral implementation in the following prioritized municipalities: Santo Domingo de Guzmán, Santo Domingo Este, Santo Domingo Norte, Santo Domingo Oeste, Los Alcarrizos, Boca Chica, Azua, Santiago de los Caballeros, San Cristóbal, Bajos de Haina, San Juan de la Maguana, Las Matas de Farfán, El Cercado, Higüey, Puerto Plata, San Francisco de Macorís, Baní, San Pedro de Macorís, Barahona, Dajabón. 129Although progress has been made in ensuring that pregnant adolescents can remain in or return to school, some obstacles remain. For instance, the impossibility of continuing in school if the adolescent is in a polytechnic school, or the non-recognition of art courses unless they have completed two years, along with the difficulty of completing the internship required to complete the intermediate level (interview with key stakeholders). 70 National Council for Children and Adolescents (CONANI), and the National Institute of Professional Technical Training (INFOTEP).130 Retaining girls in school can also reduce teenage pregnancies through the provision of SRH information. Since 2018, MINERD and MSP have been designing a strategy for the prevention of adolescent pregnancy and for psychosocial and pedagogical support. The strategy was designed by MINERD’s Directorate of Orientation and Psychology as part of the PAPSE II Project to provide comprehensive education on gender and sexuality, with content appropriate to the student's age. The strategy rests on two types of interventions: i) inclusion of comprehensive sexual education in the school curriculum at all levels and with age-appropriate content, and ii) development of activities to strengthen the curriculum and involve all school actors, including the teaching, counseling and psychology personnel and families. These activities include theater, sexual and reproductive health fairs, and conferences with opinion leaders who promote a culture of peace, gender equality and equity, and positive masculinities, among others. But the provision of SRH information could be improved. A broader training process is required for teachers and counselors. There are supporting materials in place that define the conceptual framework, methodological approach, and working guidelines to support the teaching of SRH. However, these materials need to be adapted to the needs and skills of teaching staff. Natural science and human/religious studies staff are often assigned to introduce SRH in their course. But unlike counselors and psychologists, they often lack the skills needed to teach comprehensive sexuality education properly. Furthermore, progress in the implementation of comprehensive education in sexuality has been slow and has met with resistance from conservative groups and faith-based organizations. A constructive dialogue is required between the different actors so that scientifically based information can be made available to people (students, families and teachers) via the educational process, in turn generating values favorable to preventing early pregnancy. Another example of good practice is associating cash transfers with measures to boost girls’ agency, for example through building soft skills and promoting learning about nutrition and reproductive health (World Bank, 2012). Overall, the three types of interventions described (i.e., programs that provide girls with life skills and reproductive health knowledge; programs that expand girls’ economic opportunities; and programs designed to ensure that girls remain in school or that enable them to return to school) are not intended to be an exhaustive list. To improve girls’ educational attainment, additional interventions may also be needed. The three types of interventions were selected because there is evidence that they help improve SRH knowledge and delay child marriage and early childbearing. The programs and interventions are also not mutually exclusive; implemented together, they can complement each other. While some programs are better than others in achieving the desired goals, all three categories of programs may have significant benefits of many kinds. With different interventions targeting different groups of girls (e.g., those in school or with the potential to return to school, and those who dropped out and may not be able to return), all three categories should be considered when formulating a strategy to improve opportunities for adolescent girls. 130 The National Institute of Professional Technical Training (INFOTEP) is the institution responsible for professional technical training in the Dominican Republic. The National Council for Children and Adolescents (CONANI) is the governing institution of the National System for the Protection of the Rights of Children and Adolescents, whose mission is to guarantee the fundamental rights of children and adolescents in the Dominican Republic. The National Institute for Comprehensive Early Childhood Care (INAIPI) is a decentralized institution at the functional and territorial level, responsible at the national level for managing the provision of quality comprehensive care services to boys and girls aged 0 to 5 and their families (see Annex 6). 71 Access to Healthcare and Contraception Services Apart from the interventions mentioned in the previous section, health care for adolescents needs to be expanded to all municipalities to guarantee accessibility. As already noted, there is an unmet demand for comprehensive health care services among poor teenagers: either out of ignorance as to where these services are offered, or the remoteness or inexistence of centers in their communities. Despite the existence of the National Program for Comprehensive Health Care for Adolescents (PRONAISA), the availability of healthcare services for adolescent girls remains limited. In the current care model, PRONAISA performs stewardship, management and regulation functions. Service provision is actually carried out by Regional Health Services (SRS), which are decentralized entities of the National Health Services (SNS), through networks of services articulated according to level of complexity: For adolescents, the First Level of Attention are the health care centers (CPN) responsible for health promotion services, disease prevention and basic care, including sexual and reproductive health. Of 109 existing health care centers for adolescents in 2019, only 79 were effectively operating, representing 7.8 percent of the 1,395 health centers managed by the National Health Service. Of the 23 centers that were not in operation, 10 were in provinces with a high share of girls aged 15-19 years who were or had been pregnant: including Independencia and Pedernales, the two provinces with the highest percentages of pregnant adolescents and whose only centers were not working (PRONAISA, 2019). So clearly care model needs to be expanded to all municipalities to guarantee accessibility for teenagers. It is essential to ensure that teenage girls have access to healthcare services adapted to their needs. A primary care unit (UNAP) that operates in a CPN must serve between 2,500 to 3,500 inhabitants (500 to 700 families) of each territory and operate in close coordination with schools and communities. However, this requires trained human resources, financial resources, and effective management, monitoring and supervision mechanisms. To date, training strategies for human resources have not lasted; there is high staff turnover, especially when there is a new national administration.131 In-service training is necessary since medical residencies do not provide training in adolescent-friendly health services or long-term contraceptive methods, which are the recommended contraceptive methods for adolescents. In addition, services tend to only be available during school hours. It is also necessary to earmark a budget line to provide comprehensive care services to adolescents: especially the 16-19 age group, who are excluded in the Basic Health Plan. Healthcare for adolescents is financed with resources from the budget and the Dominican Social Security System (SDSS), but assessing the availability of resources is difficult and there are gaps in programming. It is not possible to distinguish the resources effectively directed to adolescent care, since administrative reports are not broken down by age of patients. The provision of healthcare for adolescents belongs to the Collective Health Program which has a protected budget, but distribution between subprograms is variable depending on emergencies, which may be detrimental to the budget directed to adolescents. In addition, within the SDSS Basic Health Plan, comprehensive care for children and adolescents covers the population aged 0 to 15 years; the needs of adolescents aged 16-19 are not recognized. Finally, except for vaccinations, promotion and prevention services tend to be in little demand among the insured and there is some confusion as to whether these services should be financed by the MSP or SDSS. The Family Health Insurance program should include SRH services for adolescent girls. Within SENASA (National Health Insurance), the financing for health promotion and disease prevention activities has focused on chronic diseases that mostly affect adults (such as hypertension and diabetes) and the provision of vaccination services for children. The need for adolescent health services, including SRH, 131 Although in the 2010-2015 period, the MSP reports having trained 2,300 civil servants, 1,240 young multipliers and 320 multiplier families, from 2016-2019 training activity was considerably reduced. 72 tends to be neglected. The transfer of resources from SENASA to the UNAPs is not linked to indicators of inputs and health outcomes for the adolescent population. Low-cost digital modalities to deliver SRH services should also be expanded to increase their reach. A key lesson of the pandemic is the need to incorporate telemedicine to ensure access to comprehensive healthcare services for adolescents, conditioning it to a first face-to-face approach that allows for building up the clinical history and trust between providers and adolescents. The pandemic forced innovation to reach teenagers and maintain their attention, including a switch to virtual programs in those communities where there is internet access. To avoid the closure of projects such as the Girls Club due to social distancing measures, virtual sessions were held. Likewise, the use of technological platforms and networks facilitated the training of health personnel on the implementation of guidelines for the care of adolescents. In particular, the PlaneAPP application provided information on comprehensive genderuality education and the georeferencing of comprehensive adolescent health care services. The quality of many comprehensive health centers needs to be improved. Of the 79 centers in operation in 2019, only three centers correctly identified the structure of the interview that must be carried out during a medical checkup with an adolescent. This, despite the fact that 80 percent had indicated that they knew the interview protocols. Also, more than half of the centers did not have or needed to repair key instruments for consultations, as well as equipment and furniture.132 Within provinces, 50 percent or more of the centers in operation did not have a simple stretcher, curtain, or screen. The levels of deficiency evidenced were higher in the more complex care centers (level II and III). Quality would also be enhanced if services were provided with extended session schedules so that they can be accessed more easily by adolescent girls. Local assessments of these services need to be carried with representation from adolescents, and gaps in territorial coverage especially in areas with high rates of teenage pregnancies must be closed. An annual medical visit by all adolescents for preventive purposes should be a conditionality to access social protection programs. This would enhance demand for adolescent care, especially among vulnerable populations. Just as there is a consultation for Healthy Children, a consultation for Healthy Adolescents is essential.133 The UNAP to which adolescents belong could issue the certification and integrate their medical history into the system and contribute to the reduction of adolescent pregnancies. This could also help to identify, in coordination with schools and community organizations the girls and adolescents with the greatest exposure to risks, and refer them as needed to the local Child and Adolescent Protection Boards. The UNAP must have duly trained personnel to accompany groups in situations of greater vulnerability and their families, including for referrals. The conditional cash transfer Comer Es Primero (renamed ALIMENTATE under the new Supérate strategy) requires annual health checkups for pregnant women and under-fives, but not for teenagers. This opens up the possibility of establishing a new conditionality in the Supérate program so that adolescents and preteens are encouraged to access comprehensive health services. But the way that Supérate, schools, UNAPs and community workers interact needs to be reorganized to avoid creating a demand for health care services for adolescents (including for genderual and reproductive health services) without an appropriate and competent supply to satisfy it. 132Such as Fetal Heart Rate Monitor, Orchidometer, Otoscope, Ophthalmoscope, X-ray Viewer. 133Included in the PDSS Service Catalog 1.3. Comprehensive treatment of boys and girls (aged 5 – 15), leaving out the group of 15 to 19 years. Also, included in section 1.8 School health, assuming that it refers to the group from 6 to 18 years old, also leaving out those who are 18 and up to 19 years old. 73 Another important recommendation is to expand the services provided for adolescent girls at comprehensive healthcare facilities. The statistical health information systems must report on early adolescence (10 to 15 years) and not just for those between 15 and 19. There is also a need to promote new medical specialties that focus on adequate care in pediatrics, obstetrics, and gynecology for younger girls (10-14) given that pregnancies often take place in this age bracket.134 Training is needed for general practitioners, nurses, assistants, and promoters in the UNAPs, CPNs and in Health Management Directorates. Capacity-building is also needed for Regional Health Services (SRS) and their dependencies to allow them to accompany and monitor the UNAPs, CPNs and hospitals. The role of the Committee for the Assured Availability of Contraceptive Supplies must be strengthened, and budgets for comprehensive adolescent health care services and contraceptive methods must be secured. As part of comprehensive healthcare services for adolescent girls, a particular emphasis is required on the detection of situations of abuse. More generally, training of judges, prosecutors, police and legislators is needed, as well as awareness and citizen oversight on the rights of children and adolescents. This will help to recognize the harmful practice of marriage at an early age, and the situations of genderual abuse, rape, and human trafficking that may affect adolescent girls. This training should incorporate paths for dealing with situations of violation of rights (including violence in its different manifestations) and action protocols to avoid the re-victimization of girls, boys, and adolescents. In addition, apart from informing adolescents about services provided by the first level care centers in matters of genderual and reproductive health, schools must inform them of the role of local social protection networks and local protection boards in surveillance and receiving complaints in cases of violence and abuse. The Dominican Republic has arrangements for the purchase and availability of contraceptive methods for adolescent girls, but there is resistance to using some of the most cost-effective methods of contraception. Since 2005, the country has a Committee for the Assured Availability of Contraceptive Supplies (DAIA). A young DAIA Committee acts as a space for inter-sectoral articulation (government, NGOs and international cooperation), and develops an advocacy plan for the promotion of genderual and reproductive rights.135 The aim is to help prevent early pregnancies and ensure the supply of post-obstetric care as needed. Yet there is still resistance from health service providers to giving SRH advice and counseling, and to providing IUDs and subdermal implant, both highly cost-effective methods of contraception. Provider training for the provision of long-term family planning services is currently being strengthened. Although there is an agreement between DIMIA and SENASA to proceed with the implementation of LARC methods in post-obstetric care before leaving the hospital, implementation has not been fully consistent. To summarize, having a poor education, low aspirations, and limited economic opportunities appear to increase the risk of teenage pregnancy at the individual level, while the main societal factors are patriarchal social norms and a culture of child marriage. Programs that (1) provide girls with life skills and reproductive health knowledge; (2) expand girls’ economic opportunities; and (3) ensure that girls remain in school or that enable them to return to school, can improve SRH knowledge and delay child marriage and early childbearing, especially when linked to conditional cash transfers. Better access to information and services on sexual and reproductive health, as well as greater labor-market opportunities for women, have also been identified as key preventive factors of teenage pregnancy. 3.3 FEMALE LABOR FORCE PARTICIPATION AND WAGE GAPS 134 This is due to a phase of menarche at an increasingly younger age. That is, the first menstruation of the woman, corresponding to her first ovulation, at an age less than 10 years. 135 Created in 2018, with the support of UNFPA, NGOs and the MSP. In it, young people associated with community-based organizations participate and are empowered at the individual level. 74 Women in Dominican Republic had fewer jobs and receive lower wages than men. As mentioned in the previous sections, closing such disparities could go a long way in reducing poverty in the country, which increasingly affects women. This section identifies with greater precision the size and determinants of the gender wage gap across the wage distribution to establish which segments are most affected and prioritize interventions. Women’s participation in labor markets is low and wide relative to men but has been equalizing ever since the turn of the century; the fallout from the COVID-19 pandemic has reversed this equalizing trend. The second part of this section seeks to understand the main factors behind the asymmetric participation of women on labor markets. It then proposes potential solutions to reduce the exposure of women to economic shocks going forward. 3.3.1: SIZE OF THE GENDER WAGE GAP AND ITS DRIVERS 1. Women in the Dominican Republic earn less across the whole wage distribution Conventional analysis on the wage gap between men and women at the mean of the distribution shows that the gap is stubbornly rigid. The difference between hourly earnings across sexes has remained stable since 2000, hovering between 11 and 20 percent (see section 3.3.2). However, when accounting for the fact that employed women tend to have higher levels of education than men, the gender earnings gap was about to about 23 percent in 2018. This gap has also been stable since 2000 (World Bank, 2021). Is the wage gender gap similar for everyone? To obtain wage gaps by gender across the entire population, we estimate quantile regressions using a copula-based methodology proposed by Arellano and Bonhomme (2017) that corrects for selection into employment. This is necessary because women do not select themselves randomly to participate in the labor market; therefore, we otherwise only observe wages among employed women (a self-selected sample). The selection effect is vital in measuring gender differences accurately, since any analysis based solely on working women may yield biased estimates of the gender wage gap. To assess the wage gap across the entire population, this report uses data from the ENCFT 2019, which contains variables that can explain women's labor participation and the difference in wages between men and women. The sample is restricted to individuals between 18 and 59 years old, where 49% are women and 51% are men. At the corresponding quantile, we define the gender wage gap as the differences between the log hourly wage for men and women. The independent variables for the wage equation are experience, experience squared, levels of educational attainment (5 categories: less than elementary education, elementary education, less than high school, high school, less than undergraduate, and undergraduate education), and regions.136 The selection equation contains the same covariates, but we add the number of children younger than six and marital status. These last two variables are assumed to affect employment probability but not the wage itself (see Mulligan and Rubinstein 2008; Maasoumi and Wang 2019; and Huber and Mellace, 2014). After controlling for sample selection, women in the Dominican Republic earn less than men across the whole wage distribution, especially among low earners. Regardless of their salary, all Dominican women earn less than men. Yet the difference in wages between men and women is greater for low-earning individuals than high-earning individuals. The gender wage gap is higher than traditionally documented. Figure 45 shows the gender wage gap with and without controlling for selection. According to our estimates, in 2019, without correction, men's earnings were on average 14 percent higher than women's earnings across the entire population of wage earners and hover between 2 percent to 22 percent: except for the top ventile of the income distribution, 136 The variable experience is approximated using age and education. 75 where the difference is negative and close to zero. Once the selection is considered, the wage gap increases even further by about 12 percent on average across the whole distribution of wages, which can be interpreted as men having a salary approximately 27 percent higher than women. Nonetheless, the wage gap goes up to 30 percent on average at the bottom 40th percentile. These numbers suggest that the entry of women into the labor market could increase the overall observed gender wage gap. This is explained by the positive selection of women into employment (i.e., the wages for women who do not work are lower than for women who work) resulting in an underestimate of the gender wage gap. The magnitude of the gender wage gap among the lower quantiles is greater, for both the uncorrected and corrected gender wage gap. This underscores the severity of the difference in pay between men and women among low-earning workers. Figure 44. gender wage gap in the Dominican Republic Source: Authors’ estimations using ENCFT data. 2. Unobservable characteristics explain most of the wage difference between men and women, particularly among the low-paid. This part examines what could explain the gender wage gap (after correcting it for selection) between observable or unobservable characteristics. For this purpose, we apply the Machado and Mata (2005) technique to simulate the entire distribution of the wages that female workers would have earned if all women worked and had men's distribution of characteristics. This last step allows to decompose the gender wage gap into i) the size of the gap that can be explained by observable characteristics (composition effect) and ii) the size of the gap that responds to the compensation for these characteristics (structural effect). The structural effect is the wage difference that can be attributed to differences in return to workers' characteristics. It is calculated as the difference between the distribution of wages for men and a counterfactual distribution. The composition effect is the proportion of the wage gap that can be explained by the difference in the distribution of observable workers' characteristics. It is defined as the difference between the counterfactual distribution and the distribution of wages for women. Figure 45. Decomposition of the corrected gender wage gap 76 0.5 Ln Wage Men- Ln Wage Women 0.4 0.3 0.2 0.1 0 -0.1 -0.2 0.05 0.15 0.25 0.35 0.45 0.55 0.65 0.75 0.85 0.95 quantiles Composition Effect Structural Effect Total Effects Source: Authors’ estimations using ENCFT data. Observable characteristics of workers, such as years of education and experience, explain little of the corrected gender wage gap (composition effect). If anything, the composition effect plays a more critical role in explaining the gender wage gap for high earners. By contrast, differences in the returns to labor characteristics and skills play a significant role in the wage gap between women and men (structural effect) (figure 46). The structural effect may capture unmeasured productivity, union premium, bargaining power, among other factors. Another important element that could explain the importance of the structural effect would be social norms like traditional views concerning the role of women in society. These attitudes are experienced and internalized very early in life and can determine choices of occupation or industry, yielding lower wages in the future. A matching comparison approach like the Ñopo decomposition of the wage gap can provide further insights into the gender wage gap in the Dominican Republic (Annex 7). Matching technics are used to answer the question of how much of the gender wage gap can be explained by the existing difference in observable characteristics and how much by unobservable characteristics such as discrimination. Figures 47 shows the results for the Ñopo decompositions on a set of observable characteristics. Each bar represents one decomposition and observable variables are sequentially added. The overall wage gap is represented by the red dashed line and the contribution of each component to the wage gap’s components is given a different color within each bar. Differences in characteristics like age, household composition, and occupational category between males and females explain almost a quarter of the wage gap (4.6 percent points). The rest is of the gap is explained by non-observable characteristics (Δ0), and the contribution of the characteristics that are not shared by women and men (components ΔF and ΔM) is almost nonexistent (see Annex 7 for detailed results of the Ñopo decompositions). Sectoral gender segregation explains part of the wage gap in the Dominican Republic. As figure 17 noted, women concentrate in lower-paid sectors like domestic employment or personal services (the ΔF component in figure 47). By contrast, jobs in construction, transport, agriculture, and mining are dominated by men and tend to be better paid (the ΔM component).137 Such asymmetry is reflected in the second bar of figure 47 and contributes to widening the wage gap. 137 By 2019, the share of males in each of the mentioned sectors was above 90 percent. 77 Promoting productivity in female-dominated sectors is therefore vital to improve wage earnings. Wage differentials across firms are a more important driver of wage inequality than individual characteristics such as gender, age, and area of residence (World Bank, 2021). Almost a third of wage inequality among formal workers in the Dominican Republic is explained by inequality in the average wage paid by firms. Inequality in wages paid across sectors of economic activity and across firms with different productivity levels explain almost quarter of the individual wage inequality. Given that wage and productivity levels are highly correlated, greater productivity dispersion across firms would be related to higher wage inequality. Education could improve the wage gender gap but will not close it. Dominican women show observable characteristics like higher levels of education that in principle would make them better candidates for higher-paid jobs. However, as the third bar in figure 47 shows, most of the gap in hourly wages for women (which were 20 percent lower than for men in 2019) remains even if women display similar levels of education to men. In other words, the gender differences in wages are largely explained by unobservable factors. Focusing on economic sectors where the share of women and men is similar reports the same outcome. 138 Most of the wage gender gap is explained by unobservable characteristics such as discrimination. A recent study on the feminization of poverty in the Dominican Republic arrives at a similar conclusion (MEPYD, 2023). Figure 46. Wage gap decomposition in Dominican Republic (Total employment for population aged 20- 55) – 2019 25 20 20.3 4.6 4.9 15 1.4 Wage gap 10 20.2 19.6 15.1 13.1 5 0 -5 Age + Urban + Informality + + Economic sector (2) + Education level (3) + Job tenure (4) Children + Other earner within HH + Type of employment (1) Unboservables (Δ0) Characteristics that only men have (ΔM) Characteristics that only women have (ΔF) Observable differences (ΔX) Total wage gap Source: Author’s calculations using ENFT. So, what did we learn from the above decompositions? (1) the average gap is not necessarily a good approximation of the full wage distribution; (2) after correcting for sample selection the wage gap is higher than under conventional estimates and more pronounced among the poor; (3) most of what drives the difference in wages between women and men are intangible factors like discrimination and not associated with schooling or experience, which mainly influence the gap among the upper segments of the population; and (4) women predominate in low-earning sectors. In closing, it is important to stress that the large difference in wages due to unobservable characteristics is understood as discrimination, but it may also reflect the impact of unmeasured characteristics such as family 138Manufacturing, mining, electricity and water supply, commerce, hotels, restaurants, financial series, public administration and defense, real estate, social and health services, and education. 78 backgrounds or obligations. Social norms and the absence of early childhood education may in turn explain lower skill levels yielding lower salaries in adulthood. Therefore, targeting the wage structure may not be enough, and it is necessary to consider premarket factors that are influenced by how women conduct a job search and the type of jobs that they tend to seek. 3.3.2. UNDERSTANDING THE GENDER GAP IN LABOR FORCE PARTICIPATION AND EMPLOYMENT The compound effect of higher wages for women and higher rates of labor participation can make a dent on poverty. Earnings for women in the Dominican Republic are persistently low when compared to those of men. This section sets out to explore if the employment participation of women has increased over time, and if such participation weathered the economic crisis during 2020. 3. Increasing female labor force participation came to a halt with the COVID-19 pandemic. The gender gap in job participation had been closing until the pandemic. As noted in section 3.2, ever since 2009, the entry of women into the labor force has outpaced the participation of men, effectively closing the gap. While in 2009 just 47 and 36 percent of women participated in the labor market and had a job, respectively, those figures climbed up to about 53 and 48 percent in 2019 (see figure 48). This helped reduce the gender gap significantly, since men’s outcomes were stable during this period (Winkler and Montenegro, 2021). Figure 47. Evolution of employment and participation rates, by gender (2000 - 2021). 90 80 78.4 75.7 75.3 72.7 70 60 Percentage 52.6 51.2 50 47.8 45.0 40 30 20 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 LFP Male LFP Female Employment rate Male Employment rate Female Source: Own elaboration based on figures from the Banco Central - ENFT (2000-2016) and ECNFT (2017-2020) Note: In 2015 and 2016, both the ENFT and ENCFT were collected simultaneously to assess their comparability. In 2017, the Central Bank of the Dominican Republic discontinued the ENFT and kept the ENCFT as the official source of information to estimate poverty. 79 4. Domestic and care activities are among the main reasons for not working or searching for a job among inactive women and many existing jobs for women are informal and of lower quality. Most inactive women declare domestic and care activities as the second main reason for staying out of the labor market. During 2021 inactive women reported domestic and care activities as the second reason for not working (only after still studying). For men this factor was almost inexistent. And for women currently inactive but with past work experience, household chores and care activities are the main reason behind for not returning to work or searching for a job (Figure 49 b.). Caring responsibilities also increase the odds of getting part-time or lower-paid jobs for women. Figure 48. Reasons for not working or searching for a job among economically inactive population in 2021 a. Inactive population b. Inactive population with work experience 70 61.9 60 52.0 49.5 60 50 50 38.8 40 Percent Percent 40 34.8 30 25.5 30 20.4 18.9 20.8 20 16.2 20 14.5 6.9 9.2 7.7 10 10 5.5 6.1 2.7 2.7 2.7 3.3 0 0 Education Household Retired Health Other Education Household Retired Health Other work and problems reasons work and problems reasons child care child care Female Male Female Male Source: Author’s calculations based on ECNFT. Note: The population that has worked in the past but is currently inactive represents 45 percent of the population out of the labor force. The allocation of care activities (children and elderly) within the households is far from equal in the Dominican Republic. According to the ENHOGAR survey for 2016, women spend twice as much time as men on these activities. Women spend on average four to six hours per day on unpaid care activities, while men dedicate between two and three hours. The gap is higher among the poor and is negatively correlated with hours at work in the labor market (Figure 50). The closure of schools during the pandemic exacerbated this trend, as children had to be looked after and women largely absorbed this task. Childcare activities place an extra burden on women seeking to enter or stay in the labor market. Figure 49. Average hours worked, and care activities by socio-economic group and gender (2016) Men Women 80 90 100 Average weekly hours Average weekly hours 70 80 60 37.5 38.2 50 40.0 40.9 44.8 40.6 44.4 45.1 45.8 45.9 40 30 20 43.7 39.6 36.4 36.2 31.0 10 17.1 18.0 18.0 18.9 20.9 0 Poorest II III IV Richest -10 Poorest II III IV Richest Care (Children and elderly) Employment Care (Children and elderly) Employment Source: ENHOGAR 2016. Social norms and stereotypes also deter women from joining the labor force. play also a role determining stereotypes. As noted in section 1.3 around 40 percent of Dominicans agreed with the statement that women should only work when their male partner does not make enough money. These type of views lend to perceive unpaid care work as a female prerogative and highlight the limitations of standard labor supply arguments, which neglect the role of social norms on women’s ability to enter and remain in the labor market (Ferrant, G, et al, 2014). Despite higher insertion into the labor market, working longer hours and pursuing higher education in greater numbers, significant wage gaps persist between men and women. The difference between hourly earnings across sexes held relatively stable in the last two decades, hovering between 11 and 20 percent (Figure 51) higher earnings for men than women. The sector of work, years of experience and discrimination at work are some of the factors driving the gender wage gap historically. Domestic and social services have higher concentration of women and are historically lower paid and characterized by a disproportionate percentage of informal workers (Figure 52). The shares of women employed by sector have remained similar over the years. Informality entails lower earnings but also less access to social benefits, health insurance or paid sick leave. Figure 50. Hourly gender wage gap (2002-2021) Figure 51. Share of (informal) women employed by economic sector (2019) 0.5 Domestic services Social services Education 0.4 Other services Hotels and Restaurants Finance Hourly wage gap 0.3 Commerce Public administration Public services 0.2 Real state Industry Mining Transport 0.1 Agriculture Construction Fishing 0.0 0 20 40 60 80 100 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Share of women within sector Share of informal workers (women) Source: Author’s calculations based on ECNFT. Source: Author’s calculations based on ECNFT. Note: The gap is derived as the difference of median hourly wages as a proportion of male income. 81 Box 10. What happened to women participation and employment during the pandemic crisis? To understand the drivers of changes in labor participation, the team analyzed longitudinal data from the National Labor Force Survey (the Encuesta Nacional Continua de Fuerza de Trabajo, ENCFT) in the second quarter of both 2019 and 2020, which contain a wealth of information on labor market dynamics. The period of analysis coincides with the strictest months of the lockdown in the Dominican Republic, which led to a decline in employment of 5.8 percent. The advent of the pandemic led to a sharp decline in job participation. The Dominican Government implemented severe confinement measures and a close-to-total shutdown of the economy aside from essential sectors. As a result, around 270,000 jobs were lost by the end of 2020, and the employment rate fell from 61 to 57 percent. GDP contracted by an estimated 6.7 percent over the same period. In 2020, four economic sectors accounted for almost three quarters of employment losses: hotels and restaurants (-17 percent), domestic services (-17 percent), other services (-11 percent), and commerce (-4.3 percent). The pandemic’s impact on labor markets hit women harder than men. In Q2-2020, the female participation rate fell by around 8 percentage points, while male participation did so by 5 percentage points. As a result, the pre-pandemic gains for women in terms of labor force participation were reversed during the pandemic, falling to 2009 levels. By 2021, participation and employment rates only partially recovered for both women and men and gaps remained almost unaltered (Figure 48). Impacts were largest among young urban women. Job losses and declines in labor income for those women who kept their jobs were larger in urban areas as in rural areas, mainly in the Ozama region. Young women (25 –34 years old) were also disproportionately affected. Four highly female-intensive sectors explain 41 percent of all employment losses (88 percent of female employment losses) between Q2 2019 and Q2 2020.139 Higher levels of education helped shield the participation of women in the labor market during the pandemic. Most of the reduction in women’s employment during the peak of the pandemic in 2020 occurred among those with less tha n tertiary education (figure 53). Panel regression analysis confirmed that regardless of their socio-demographic or job characteristics, Dominican women were 11 percent more likely to lose their jobs than men at the peak of the employment crisis during the second quarter of 2020. Such magnitude prevailed even after controlling for other confounding factors (Table A8.4).140 In contrast, as stated in the results in Table A8.3, there is no evidence for an asymmetric impact for women on wages or hours worked. Figure 52. Decomposition of changes in the employed population Q2 2019-2020 for the longitudinal sample, by gender, social characteristics, and location. 139 Commerce, Hotels and restaurants, other services, and domestic services. 140 Following Cuevas, et al (2021). This report uses the methodological approach to assess the impact of COVID 19 on Labor Market in Dominican Republic. Taking advantage of the rich longitudinal characteristics of the ECNFT (2019-2020) we reproduced its econometric specification to estimate the determinants of job and income losses. The detailed results of the MCO estimation and descriptive statistics can be found in ANNEX 13. We begin with a simple specification using only a gender indicator to predict job loss and gradually add control variables to identify which ones contribute to change the magnitude of the coefficient associated with gender. This estimation does not change significantly after controlling for education, work location, interview month, job characteristics, and the presence of young children. 82 Source Based on data from ECNFT 2019-2020 (second quarter). Educational attainment is defined as high skilled for those with 11 or more years of education (i.e., complete secondary education in Dominican Republic). Formality is defined as: the sum of informal and formal employment in the informal sector, informal domestic service employees, and unpaid family workers who work in the formal sector. Informal sector is defined as: workers whose company or business does not meet the criteria to be registered in the National Taxpayers Registry (RNC), does not record commercial transactions in auditable accounting books, or does not have legal organization or organized accounting. Informal employment is defined as: workers who do not have access to social security due to their work relationship. Note: The sum of the bars represents the total change (decrease) in employment in percent points between 2019 and 2020 for the longitudinal sample. For females the sum of the bar is -12.4 p.p., for males it is -9.4 p.p. The drivers of impacts on job losses were then assessed using standard regression methods. The analysis begins with a simple specification using different age groups and the level of education to predict job losses. More control variables are gradually added, including work location (urban/rural), interview month, job characteristics (formal/informal, essential/non-essential, home-based, sector), and the presence of young children in the household (see Table A8.5). The team estimated separate ordinary least squares (OLS) regression models for men and women. Figure 54 shows the results for our preferred and most complete specification on job and individual characteristics that may help explain job loss.141 Positive coefficients imply increasing chances of job loss. Why did women bear the brunt of the impact on labor markets during the pandemic? First, because several sectors with historically high participation of women were among the most affected during the pandemic. For instance, 9 out of every 10 individuals employed in domestic services are women, and 17 percent of the jobs in this sector were lost between 2019 and 2020. Second, because informal jobs are overrepresented among women and are more precarious. Despite registering lower levels of informal jobs (50 percent) than men (58 percent) prior to the crisis, female workers on the informal sector were 8 percent more likely to lose their job (Figure 54). They can easily vanish without a contractual obligation that protects employees, and government support schemes during the pandemic targeted formal jobs. And finally, because women became the primary caregivers at home during lockdowns and when children returned home. According to the World Bank High Frequency Survey conducted in mid-2021, 30 percent of adult women reported that domestic activities increased during the pandemic crisis and 45 percent reported increased time spent on care activities. The multivariate regression analysis confirmed that indeed informality and the presence of school-aged children at home are associated with an increase in job losses among, but not at all for males. Incidentally, women (and men) in occupations more amenable to work from home were less likely to lose their jobs as did those working in an essential sector. (figure 54) Figure 53. Change in the probability of job loss between 2019 – 2020 (Q2) 83 0.1 0.076** 0.056 0.064* 0.040 0.1 0.026 0.023 Coefficient 0.0 -0.1 -0.026 -0.042 -0.1 -0.2 -0.134*** -0.149*** -0.2 High skilled Work from home Urban Informal Presence of children (broadband) Male Female Source: Based on data from ECNFT 2019-2020 (second quarter). Note: symbols***, **, * indicate that the corresponding regression coefficient is statistically significant at the 1%, 5%, 10%, and 15% levels. The model includes gender, age, area, and economic sector of the individual as controls. Unlike the widening gap in participation and employment rates between men and women, the wage gap did not widen during the crisis. According to the ECNFT survey, the gender gap on participation rates increased from 25.8 to 26.4 percent between 2019 and 2020, due to a faster recovery in the employment rate for men. 42 percent of the jobs that that were lost by men in Q2 were recovered in Q4 by them, while only 7 percent of female employment was recovered. By contrast, hourly wage gaps between men and women within the bottom 40 percent of the population remained unaltered, with males consistently earning 20 percent more than their female counterparts in 2019 and 2020. The wage gap between women and men even decreased from 10 to 5 percent among workers in the top 60 percent of the income distribution in the same period. 3.3.3: RECOMMENDATIONS TO IMPROVE FEMALE LABOR FORCE PARTICIPATION AND TACKLE THE GENDER WAGE GAP This section documents and discusses selected policy options to increase female labor market participation and reduce the wage gaps by gender. Each policy recommendation is associated with the key factors that drive the wage gaps between men and women and job losses among women during the pandemic. Such factors were identified in the deep-dive analysis of labor market participation and wage gaps. As noted above, informality and the presence of children at home seem to have driven job losses during the pandemic. The presence of women in low-earning sectors and discrimination explain, to a large extent, the wage gap among most of the population. As a preview, specific policy recommendations for promoting jobs for women, include: (a) increase the construction (or adaptation) of classrooms to ensure the universalization of the Extended School Day program and expand the offer of care services to address the disproportionate responsibility of women for care activities, especially following the pandemic; (b) increase the budget for the PREPARA (Basic Education Program for Adults and Distance High School) program; include soft skills into the curriculum and create an agreement with INFOTEP to facilitate vocational training for the PREPARA graduates to become more employable; and (c) incentivize formality by maintaining and expanding quotas under Law 488-08 and extending the Social Security Contributory Regime. And for closing the wage gender gap: (d) mandate firms and public institutions to establish grievance redress mechanisms (including the establishment of public dashboards with the scale of minimum wage levels and the contact details to access the ombudsman services provided by the Ministry of Labor); (e) level minimum wages with non-sectorized companies for activities where there is a greater relative presence of women such as hotels, bars and restaurants, and manufacturing; and (f) promote awareness among domestic service workers and employers of the recently approved regulation that establishes minimum wage and guarantees labor rights for the 84 domestic service sector; and establish a friendly online platform to affiliate and make contributions to the Social Security Subsidized Contributory Regime as part of the ongoing pilot project to incorporate domestic workers into the social security system (figure 55). Figure 54. Drivers of jobs losses and wage gaps and associated key policy recommendations The Dominican Republic has encouraged the labor participation of women through promoting business entrepreneurship, passing regulations to protect the rights of pregnant women and domestic workers, and providing services that free up time for work that would otherwise be used for domestic labor, which women perform disproportionately in the DR. To promote entrepreneurship, the Banca Solidaria Program provides access to microcredit and training for women's microenterprises. There are also affirmative provisions in public contracting to favor micro, small and medium-sized women-owned enterprises.142,143 More recently, the government has enacted policies to free up time otherwise dedicated to family care, such as the establishment of the Extended School Day, the Early Childhood Care Centers, and the recent initiative to develop a care policy within the Supérate Program. Other non-affirmative policies have an impact on women's employability and employment (such as education, training and technical-professional training program) and on women's labor income (such as the minimum wage setting policy). The scope of these policies in terms of coverage and impact is varied and there are multiple means available to improve their effectiveness. 142 Law 488-08, which establishes a Regulatory Regime for the Development and Competitiveness of Micro, Small and Medium Enterprises (MSMEs), provides those public institutions must make 20% of their purchases of goods and services from female owned MSMEs, provided that the goods and services demanded are offered by said companies. 143 The Certifícate Mujer Program of the Ministry of Industry, Commerce and MSMEs offers the following benefits to female- owned MSMEs: 1) access to exclusive purchase tenders for women owned MSMEs, access to promotional activities and access to markets; advance of 20% of the total invoice tendered to be used as an advance to working capital to guarantee compliance with the contract; preferential rates for loans through Banca Solidaria, Banco de Reservas and multiple private capital banks; technical assistance through Mipymes Centers. See MICM (2021). 85 The expansion in coverage of the Extended School Day Program (JEE) and the Comprehensive Early Childhood Care Centers (CAIPI) could increase female labor participation. The JEE Program was created in 2014 to double the school day at primary and secondary level from 4 hours to 8 hours. 144 That same year the National Education Council adopted the JEE as a State policy, based on the Constitution of the Republic and the General Education Law 66-97. Before JEE, the effective time of teaching activities was 54.3% (2.17 hours) out of the theoretical 4 hours of teaching time. In the JEE model the effective time increased to 63.3 percent (3.8 hours) out of the theoretical 6 hours dedicated to teaching time (with a recalibrated, more realistic understanding that a sizeable portion of the teaching day will be lost to other activities. According to most parents and guardians (92.3%) of the 6th grade students who participated in the 2018 Diagnostic Evaluation, the JEE Program has allowed the mother, the father, or another member of the household to work outside the home. Garganta and Zentner (2022), based on a non-experimental methodology, show a positive and significant effect on the participation rate of women with children eligible for the JEE, corresponding to an average increase of 3 percentage points (pp) in the probability of labor participation. This was particularly relevant for married mothers (5.96 pp increase), in urban areas (7.71 pp increase) and with low education (9.38 pp increase). Despite its proven success and being enshrined into law, there is an unmet demand for the JEE program. In the 2019-2020 year, 1,334,607 students and 5,298 educational centers were in the JEE, which represents an enrollment rate of 71 percent.145 During the Covid-19 pandemic, the application of the Extended School Day for the 2020-2021 school year was suspended. In 2021, MINERD carried out a needs survey in all educational districts to reassess the updated need for classrooms and improvements. It was determined that 6,181 classrooms would be required to materialize the universalization of the Extended School Day Policy. This means that universalization requires the construction of new centers and classrooms. This is a challenge for the MINERD given the current fiscal space in the Ministry. Since 2021, there was a sharp drop in the construction of new centers. Only 7 centers were built in 2021 and 124 new classrooms. These figures are well below the 2,533 new classrooms constructed per year from the 2013-17 period or the 1,390 new classrooms constructed per year during 2018-20. The CAIPIs form part of the National System for the Protection and Comprehensive Care of Early Childhood and offer full-time (10 hours a day) education and protection services to boys and girls aged 0- 5 belonging to vulnerable families. There are 139 CAIPIs in the DR serving 29,098 boys and girls.146 147 The Dominican Government has recently started a pilot program for Communities of Care, a promising initiative whose potential for scaling must be evaluated. Box 11. Towards the establishment of a National System of Care in the Dominican Republic The disproportionate burden on women of home care of dependent members is the most important barrier to women's employment, limiting their economic autonomy and the ability to generate their own income. This has contributed to the increased feminization of poverty in the Dominican Republic (section 1.2). The advent of the Covid-19 pandemic exacerbated the pre-existing gender inequalities around labor force participation and reversed the pre-pandemic gains for women on labor force participation (section 3.3). 144 The Program was established by Ordinance 01-2014 with the purpose of "achieving better learning by optimizing time and the diversity of actions for the development of educational activities with quality and equity, in addition to strengthening the school as a space for social protection of children. , girls, adolescents and young peopleâ€?. 145 MINERD (2021). Strategic Plan of the Ministry of Education 2021-2024. Page 101. 146 As of January 2022. See file:///C:/Users/lizar/Downloads/Datos%20Estad%C3%ADsticos%20a%2031%20de%20enero%202022.pdf 147 See Morrissey, T. (2017), Childcare and parent labor force participation: a review of the research literature. Rev Econ Household (2017) 15:1–24 DOI 10.1007/s11150-016-9331-3 86 As a result, the Dominican government, within the framework of the National Multiannual Plan for the Public Sector 2020-2024, proposed the creation of the National Care System. In 2021, the Dominican government launched the Communities of Care Program. Broadly speaking, the program aims to: (i) reduce gender disparities in the provision of unpaid home care, (ii) improve the employability and insertion of women into labor markets, and (iii) improve the well-being of young children, older adults, and people with disabilities. The program is organized around two main pillars to achieve its objectives: Communities of Care (CC) that will provide care services for children, elder and people with disabilities within poor and vulnerable households. And the Caregivers Network (CN) which offers training and employment in care services to foster the development of a network of caregivers while improving the prospects of employability and higher salaries. In terms of implementation, Communities of Care contemplates providing formal care services to poor and vulnerable within the anti-poverty strategy SUPÉRATE, or referring them to the main institution that provides children and elder care services in the country: the National Institute for Comprehensive Early Childhood Care (INAIPI) and the National Council for the Aging Population (CONAPE). The National Council for Disability (CONADIS) does not offer care service for people with disabilities currently, but would ideally also be integrated into the scheme. The Ministry of Economy, Planning and Development (MEPYD) is responsible for coordinating the inter-institutional working group comprised by the Ministry of Labor, the Ministry of Women, the Institute of Professional Technical Training (INFOTEP), the Single System of Beneficiaries (SIUBEN), the National Council for Children and Adolescents (CONANI) and the Strategic and Special Projects Program of the Presidency (PROPEEP). In a first stage, before alternative care offers are fully developed within the program, Communities of Care has been contemplating the possibility of identifying SUPERATE beneficiary households with unsatisfied care demand for their children in prioritized municipalities. INAIPI would then be asked to give precedence to some of these families as spots become available in currently open centers and in new centers that will start operating during the pilot phase in those referred areas.148 A team of researchers through SIUBEN proposed the implementation of an impact evaluation of the pilot. Specifically, the evaluation aims to estimate the effects of access to childcare support offered by INAIPI’s day care centers −named CAIPIS − on children and their families through a Randomized Control Trial or RCT. 149 INAIPI’s comprehensive childcare intervention through CAIPIS includes socio educational programs, early stimulation, nutrition, and health routine checkups, and the promotion of proper parenting abilities through workshops and scheduled center visits. The evaluation will be valuable to policymakers given the enormous potential of this initiative to improve children development and the well-being of families, especially, female labor force participation. Moreover, the government is aware of the importance of generating evidence to optimize the program if scaled up to a National Care System while highlighting the relevance of public childcare support for the most vulnerable population. Education and training can also contribute to increasing the employability and occupation of women. As the educational level of Dominican women increases, their participation and employment rates increase and these gaps with respect to men are closed. However, it is also necessary to reduce the high unemployment rate shown by women with primary and secondary education, as well as to increase the skills of women with less education. This implies increasing the supply of vocational technical education and training offered through the Ministry of Education and Infotep. In 2019, just 0.66 percent of self- employed women with pre-university education were attending a technical course and only 0.07 percent of self-employed women had completed technical high school.150 In 2021, only 21.6 percent of INFOTEP's training actions were aimed at people with a low level of schooling, via the Community Program. 148 Currently, CONADIS does not offer care service for people with disabilities. 149 Given the excess demand for spots in new CAIPIS, families who meet the established requirements will be randomly assigned to two groups: a treatment group, that will be offered a spot for their child; and a control group, that will not be offered a spot, but that will still be eligible for intervention in later phases of the program. 150 Technical education in the Dominican Republic has a very limited scope, in 2019 only 0.17% of the employed population had obtained a technical high school degree. 87 It is necessary to strengthen the Basic Education Program for Adults and Distance High School (PREPARA) to develop skills that increase the employability of young women and avoid a second educational dropout. PREPARA is a flexible educational offer in terms of hours for the population aged over 14 who previously dropped out of the educational system. The challenge of youth and adult education programs and vocational technical training is to encourage the return to school of dropouts, ensure educational continuity, and provide students with soft skills and abilities that allow them to enter the labor market, either as employees or self-employed. In this respect, the curriculum should add contents on soft skills for women to become more employable. PREPARA is currently outperformed by conventional Figure 55. Internal efficiency indicators of high-school programs in key performance outcomes. In adult and youth education program vis-a-vis the case of young mothers who dropped out of school, regular high school program, 2018-2019 reintegration into educational programs requires available childcare services, at least during school hours. 100 90.5 Through coordination between PREPARA and 90 SUPERATE, financial support could be offered to 78.8 80 teenage mothers so that they can continue their % enrolled students vocational studies. An educational curriculum that could 70 be developed as part of PREPARA should incorporate 60 content related to soft skills and the construction of life 50 plans, decision-making for the exercise of women’s 40 sexual and reproductive rights, the prevention of a second pregnancy, strengthening of self-esteem, and the 30 development of labor and digital skills.151 In addition, 20 13.3 PREPARA should pursue an agreement with INFOTEP 10 7.9 6.2 3.3 to provide its graduates with vocational training. The 0 experience from the provision of virtual education Promotion Repetition Drop out rate during the Covid-19 pandemic could be used to offer rate rate flexible education programs for adolescents and young PREPARA Program people. However, raising the effectiveness of the youth Regular High School Program and adult education will require more financial resources to better-train teachers and develop Source: Authors elaboration based on MINERD (2021) appropriate methodological strategies.152 Affirmative actions in public purchases in favor of women-run MSMEs have encouraged them to formalize. Law 488-08 should be maintained, and its quota increased. Law 488-08 grants entrepreneurship initiatives access to public markets by establishing that state institutions must make 20% of their purchases of goods and services from micro, small and medium enterprises and a quarter of this quota (5%) must be awarded to companies directed by women (with more than 50 percent of shareholding or social capital). From 2012 to 2020, the number of women's MSMEs participating in public purchases increased from 3,122 to 20,950. From August 2020-July 2021 the quota awarded to women's MSMEs was 5.76%, slightly higher than the amount established in the Law 488-08. Additionally, there are public initiatives that facilitate access to financing, financial, and administrative training, such as Banca Solidaria and Fundación Banreservas, which helps women capitalize, avoid over-indebtedness, improve business 151Campero (2018). 152In the 2021-2024 Institutional Strategic Plan of the Ministry of Education, in 2021 the amount of resources per student allocated to basic adult education was a third (38%) of the amount allocated to primary education for boys and girls, while that assigned to secondary education of adults was 42.7% of the amount assigned to general secondary education. See MINERD (2021). 88 management and expand business opportunities through participation in government purchases.153,154 According to Aristy (2016), women-owned companies that supply the state have significantly increased their level of income and investment in equipment and machinery, which in turn improves the growth potential of these companies.155 The inclusion of domestic service workers in the Social Security System and the implementation of the Subsidized Contributory Regime (SCR) can also incentivize formal work among women. Labor informality is widespread among self-employed and domestic service workers and is more prevalent among women than men.156 Under a subsidized contributory regime, workers (irrespective of their status) can make contributions to social security and these contributions are matched by the State. In the Dominican Republic the SCR has not come into force. As a result, self-employed workers do not have social security protection, harming those workers who are not eligible to be part of the Subsidized Social Security Regime, and do not have protection against labor risks, old age, and disability, or access to maternity subsidies in the case of women. Additionally, during the pandemic, self-employed workers and domestic service workers were at disadvantage in terms of access to emergency transfer programs. For instance, the FASE Program was aimed at workers affiliated with the Social Security Contributory Regime and the Stay-at-Home Program was aimed at households in conditions of poverty and vulnerability identified by the Single System of Beneficiaries (SIUBEN).157 Although the Government established the Pa'Ti Program aimed at protecting independent workers, its scope was limited. The implementation of the Subsidized Contributory Regime may require significant contributions from the state. This should not prevent the analysis of alternative scenarios for its implementation under conditions of sustainability. In fact, a step forward has been taken with the universalization of the Family Health Insurance in the Subsidized Regime for those people not included in the Family Health Insurance in the Contributory Regime. On September 2022, the Dominican Government established the mechanisms for the formalization of domestic workers and the implementation of the ILO Convention 189 on decent work for domestic workers through resolutions No. 14-2022 of the Ministry of Labor, No. 551-08 of the National Social Security Council and No. CNS-11-2022 of the National Salary Committee. Among the labor rights that are recognized in these resolutions are: minimum wage, registration with the Social Security Treasury, definition of working hours, improved family health insurance coverage, occupational risk coverage, survival and disability pensions, as well as an inclusion plan in the solidarity pension program. Even so, differences in labor rights remain with respect to the rest of the workers, since the domestic worker would not be entitled to advance notice payment or severance pay at the end of the employment contract. The Social Security Treasury (TSS in Spanish) has designed a pilot to implement the Social Security SCR among domestic workers. The enrollment process under this pilot will start in December 2022. But the TSS needs to establish a friendly online platform to affiliate and make contributions to the Social 153 Banca Solidaria is part of the Program for the Promotion and Support of Micro, Small and Medium Enterprises (PROMIPYME), which offers individual and collective credits to micro and small entrepreneurs. Fundación Banreservas operates as a second-tier bank that provides funds to unregulated microfinance entities of a private social nature. 154 In 2013, 47.3% of Dominican non-agricultural microenterprises were sole proprietorships where the owner was the only worker, and more than half of these microenterprises were led by women (51.4%), of which 70% operated in the family home. Similarly, more than half of microenterprises did not have an accounting record, 30% had an investment level of less than US$487, and 50% had an investment of less than US$1,220. See Fundación Reservas del País et. al. (2019), p.37. 155 See Aristy (2016), pp.40-42. 156 In 2019, the percentages of informal female employment in the categories of self-employed workers, employers or active partners in companies, and domestic service workers amounted to 98%, 58.6%, and 97.8% of total female employment in these categories, respectively. 157 The FASE program was established to protect workers and their families during Covid-19 lockdowns. The program partially maintained the income of formal workers whose private companies completely or partially stopped operations due to the lockdown. 89 Security Subsidized Contributory Regime as part of the ongoing pilot project to incorporate domestic workers into the social security system. In terms of the wage gap between men and women, we learned early in this deep dive section that: (i) the gap is higher among the poor, who tend to predominate in low-earning sectors; and (ii) most of what drives such difference in wages between women and men is intangible, such as discrimination, rather than schooling or experience, which mainly influence the gap among the richer segments of the population. These findings provide policy insights as to how to reduce the gender wage gap. On the one hand, public policies should target the differences in observable qualifications for high earners. For example, policymakers may consider policies that aim to improve human capital accumulation. The wage structure should be targeted with policies that aim at equal payment for equal jobs. On the other hand, public policies should target the wage structure for low earners. Sectors with large shares of women (including domestic workers) have a labor income below the minimum wage established for small firms. If only the cash remuneration is considered, the ENCFT of January-March 2019 shows that 58 percent of female domestic workers earned a salary equal to or less than the minimum wage established for small non-sectorized companies; this proportion was 42 percent for men employed in domestic services (Tables 14 & 15).158 Table 12. Percentage of workers with a labor income less than or equal to the minimum wage, by firm size, Q1 2019 Women Men Total Minimum wage category Minimum wage category Minimum wage category (Mar 2019) (Mar 2019) (Mar 2019) Occupational Small Medium Large Small Medium Large Small Medium Large Category/Firm firm firm firm firm firm firm firm firm firm characteristics Non-sectorized private sector employees Total 23.5 33.3 56.9 15.5 23.4 51.7 18.4 27.0 53.6 Firms with 1-10 employees 43 57.1 80.6 31.5 43.4 72.9 36.4 49.2 76.2 Firms with 11-50 9.3 15.9 44.1 10.9 19.1 53.1 10.3 18.0 50.0 employees Firms with more than 50 3.9 9.4 27.7 1.8 4.3 24.8 2.5 5.8 25.7 employees Formal sector 19.8 29.4 53.6 9.0 15.5 44.4 13.2 20.9 48.0 Informal sector 58.8 69.7 87.9 44.8 59.1 85.0 48.1 61.6 85.6 Hotels, bars and restaurant employees Total 9.9 11.9 38.4 2.1 3.9 18.3 5.6 7.4 27.2 Firms with 1-10 employees 27.0 32.1 86.0 9.4 15.9 55.2 19.6 25.2 72.9 Firms with 11-50 3.2 4.0 32.1 1.1 2.9 19.4 2.1 3.4 25.5 employees Firms with more than 50 - - - - 1.0 2.0 0.0 0.0 1.7 employees Formal sector 7.9 8.2 27.2 0.6 1.3 13.6 3.7 4.1 19.3 158 The Labor Code of the Dominican Republic establishes that the food and accommodation given to a domestic worker are estimated to be equivalent to fifty percent of the salary received in cash. 90 Informal sector 18.5 27.3 86 17.7 31.4 67.6 18.2 28.8 79.2 Export Free Zone 17.0 17.0 17.0 12.0 12.0 12.0 14.8 14.8 14.8 employees Self-employed workers 50.3 60.2 78.8 15.2 20.9 48.2 21.3 27.7 53.5 Domestic service 57.9 75.1 96.8 42.3 54.4 87.5 55.8 72.2 95.5 employees Memo: Monthly minimum wage (RD$, March 2019) Non-sectorized private 9,961 11,240 16,349 sector Hotels, bars and 7,020 7,802 10,874 restaurants Export Free Zone 10,125 10,125 10,125 Source: Authors based on ENCFT 2019 Q1. Notes: (1) Until July 2021 Res 1/2021, the National Salary Committee classified companies according to the amount of capital and/or inventories. The stratification in force from 2004 (Res 5/2004) to July 2021 was as follows: i) small company: less than RD$2.0 million; ii) medium-sized company: between RD$2.0 million and RD$4.0 million; and iii) large company: more than RD$4.0 million. In July 2021, through Res 1/2021, the National Salary Committee reclassifies companies according to the following classification: i) microenterprise: maximum 10 workers or annual gross sales of up to RD$8.0 million; ii) small firm: from 11 to 50 workers and annual gross sales greater than RD$8.0 million and up to RD$54.0 million; iii) medium-sized firm: from 51 to 150 workers and annual gross sales greater than RD$54 million and up to RD$202.0 million; and iv) large firm: 151 workers or more or annual gross sales greater than RD$202.0 million. (2) Corresponds to employed persons, excluding unpaid family workers, who worked 40 hours or more during the week and had 4 months or more in the job according to the ENCFT Q1 2019. Employed persons is classified according to the number of employees of the company where they work: micro-enterprises (1 to 10 employees), small companies (11 to 50 employees), and medium and large companies (more than 50 employees). (3) Non-sectorized firms correspond to firms in activities other than Export Free Zone or hotels, bars and restaurants. Legislation on minimum wages should avoid establishing lower minimum wages in economic activities where there is a strong presence of female wage earners. In the Dominican Republic minimum wage regulations are differentiated by economic activity. Lower minimum wages tend to exist for activities where there is a greater relative presence of female workers, like hotels, bars and restaurants and manufacturing (figure 19). The minimum wage levels established for companies in these activities have been systematically lower than those established in non-sectorized firms (that is, those that carry out activities other than those installed in free zones and in hotels, bars and restaurants)159 and should therefore be leveled with them. 159In the case of free zone companies, Law 8-90 establishes that free zone operators and companies are exempt from paying the workers' annual participation of 10% of the company's profits, normally established by the Dominican labor laws. There is also a special regime for apprenticeship contracts in free zone companies. Free zone companies and hotels are companies with a strong presence of foreign investment, where labor cost is a factor to be taken into consideration when deciding their location and said cost is also affected by exchange rate fluctuations. Expressed in US dollars, during the period 2000-2021, the monthly minimum wage for free zone companies fluctuated between a minimum of US$86.8 in 2004, when the country was affected by a strong exchange rate depreciation, and a maximum of US$203.3 in 2020; while for large hotel, bar, and restaurant companies, the monthly minimum wage has fluctuated between a minimum of US$101.0 in 2004 and a maximum of US$212.4 in 2019. In the case of large, non-sectorized companies, the minimum wage has fluctuated between the minimum of US$126 in 2004 and US$319.3 in 2019. 91 The legislation on minimum wages and the scale adjustments should be informed on a timely basis and easily observable for everyone. The Ministry of Labor should require all firms and public institutions the creation of grievance and redress mechanisms, including a public dashboard with the scale of minimum wage levels and the contact details to access the ombudsman services provided by the Ministry of Labor. Domestic workers, who are mostly women, are among the most vulnerable. The vast majority of them have a labor income below the minimum wage established for small businesses, which is against the principles of ILO Convention 189, ratified by the Dominican Government. Regulations on the minimum wage of domestic workers should thus be introduced. Domestic service workers generally face worse conditions than men in multiple domains, not just salaries. The ENCFT 2019 shows that: i) 69 percent of female domestic service workers did not receive vacations, compared to 67 percent of male domestic service workers and 16 percent of other female salaried workers; ii) 8 percent of female domestic service workers did not receive a Christmas bonus, compared to 9.8% of male domestic service workers and 4.4 percent of other female salaried workers; iii) 45 percent of female domestic workers did not receive paid sick leave, compared to 43 percent of male domestic service workers and 12 percent of other female salaried workers; iv) 83 percent of female domestic workers did not receive advance notice and severance pay prior to losing their jobs, compared to 65 percent of male domestic service workers and 33 percent of the rest of female salaried workers. Current labor legislation also contains discriminatory provisions unfavorable to employment in domestic services. The Labor Code establishes a special regime for domestic workers, considering that they are employed by a household, which is a unit that does not report profit or business for the employer or their relatives. Batista (2018) identifies the following discriminatory provisions for domestic service workers relative to other salaried workers: indeterminate working hours; no maternity protection; no right to vacations for each year of service; no right to vacation pay not taken proportional to the time worked in the event of termination; minimum wage not stipulated; non-existence of labor benefits for dismissal; accommodation and food are equivalent to 50% of salary, which does not happen in the case of other employees if they are granted. The effective implementation of the ILO Convention 189 by complying with the resolutions No. 14-2022 of the Ministry of Labor, No. 551-08 of the National Social Security Council and No. CNS-11-2022 of the National Salary Committee issued for the formalization of domestic workers can contribute to eliminate discrimination in the labor rights of domestic service workers and reduce wage gender gaps. The Dominican Republic ratified the ILO Convention 189 in 2013 through Resolution No. 104-13 which had the status of law once it was ratified by the Dominican Congress in 2015. Convention 189 is aimed at guaranteeing fundamental labor rights for domestic workers and eliminating all forms of discrimination in terms of employment and occupation, in order to ensure that domestic workers enjoy equitable employment conditions and working conditions. Although the Dominican Republic ratified the ILO convention on work in domestic service, the current legal framework and hiring practices discriminate against employment in domestic service, which is mostly carried out by women. Among the national legislation identified by Batista (2018) that collide with Resolution No. 104-13 are the Labor Code and its Application Regulations, Law 87-01 that created the Dominican Social Security System, the Regulation on Safety and Health at Work (Decree 522-06), General Migration Law (Law 285-04) and General Health Law (Law 42-01), among others. The formalization initiatives that are being implemented by the Government constitute a step forward and could eventually be incorporated as modifications to the labor code and several other laws identified in Batista (2018). There is an ongoing debate whether these incorporations are necessary since the ratification of the ILO convention has already a Law status. 93 Regardless, government needs to promote awareness among domestic service workers and employers of the recently approved regulation that establishes minimum wage and guarantees labor rights for the domestic service sector. To summarize, the presence of women in low-earning sectors and discrimination explain, to a large extent, the wage gap among most of the population. Legislation to set the minimum wage for domestic service workers should be introduced. The minimum wage should be leveled with non-sectorized companies for activities where there is a greater relative presence of women such as hotels, bars and restaurants, and manufacturing, as well as in rural areas. Informality and the presence of children at home were among the most important features explaining job losses for women during the pandemic in 2020. Affirmative provisions in public contracting in favor of micro, small and medium-sized women-owned firms can encourage formality and thereby increase job security. The provision of social security to domestic workers can also give a strong push to formalization. Programs that (1) extend school hours, (2) provide care services, and (3) increase the offer of technical and vocational skills; can all contribute to increasing the employability and occupation of women. 93 94 REFERENCES Alfani, Federica; Clementi, Fabio; Fabiani, Michele; Molini, Vasco; Valentini, Enzo. 2020. Once NEET, Always NEET? A Synthetic Panel Approach to Analyze the Moroccan Labor Market. Policy Research Working Paper;No. 9238. World Bank, Washington, DC. Apedo-Amah, Marie Christine; Avdiu, Besart; Cirera, Xavier; Cruz, Marcio; Davies, Elwyn; Grover, Arti; Iacovone, Leonardo; Kilinc, Umut; Medvedev, Denis; Maduko, Franklin Okechukwu; Poupakis, Stavros; Torres, Jesica; Tran, Trang Thu. 2020. Unmasking the Impact of COVID-19 on Businesses : Firm Level Evidence from Across the World. Policy Research Working Paper;No. 9434. World Bank, Washington, DC. Báez et al. (2017). Análisis de GeÌ?nero en la Mortalidad Materna de RepuÌ?blica Dominicana. Centro de Estudios de GeÌ?nero Instituto TecnoloÌ?gico de Santo Domingo. Bergstrom, K., & Ozler, B. (2021). Improving the well-being of adolescent girls in developing countries. World Bank Group. Baird, S, C. McIntosh, and B. Özler (2011): “Cash or Condition? Evidence from a Cash Transfer Baird, S., F. Ferreira, B. Özler, and M. Woolcock (2013): “Relative Effectiveness of Conditional and Unconditional Cash Transfers for Schooling Outcomes in Developing Countries: A Systematic Review,â€? Campbell Systematic Reviews, vol. 9(1), 1-124 Barón, J. and Mola, J. (2019). "Chapter 3: Improving Outcomes in the Education Sector" in Dominican Republic Public Expenditure Review, 2012-18, pp. 62-89 (Report No. 143651) Batista, M. (2018). La República Dominicana y el Convenio 189: Ratificación, entrada en vigor y acciones pendientes. Perspectiva 03-18. Fundación Friedrich Ebert chromeextension://efaidnbmnnnibpcajpcglclefindmkaj/viewer.html?pdfurl=http%3A%2F%2Fli brary.fes.de%2Fpdf-files%2Fbueros%2Ffescaribe%2F14844.pdf&clen=387116 Campero, C. (2018). Por qué es importante la educación de personas jóvenes y adultas en Latinoamérica, y cómo su limitada prioridad en las agendas debilita su impacto. Educación de Adultos y Desarrollo. No. 85. https://www.dvv-international.de/es/educacion-de-adultos-y- desarrollo/ediciones/ead-852018-papel-e-impacto-de-la-educacion-de-adultos/section-1- cumplir-un-papel/por-que-es-importante-la-educacion-de-personas-jovenes-y-adultas-en- latinoamerica-y-como-su-limitada-prioridad-en-las-agendas-debilita-su-impacto (TERCE): Informe Nacional República Domunucada, Ministry of Education, Dominican Republic, Education Quality Assessment Directorate. https://www.ministeriodeeducacion.gob.do/docs/direccion-de-evaluacion-de-la-calidad/hpPA- terce-informe-rdpdf.pdf. Castro, B., Ing, L., Park, Y., Abrams, J., & Ryan, M. (2018). Addressing non-communicable disease in Dominican Republic: barriers to hypertension and diabetes care. Annals of global health, 84(4), 625. Chávez, C., Cebotari, V., Benítez, M.J., Richardson, D., Fen Hiu, C., and Zapata, J. (2021), School-Related Violence in Latin America and the Caribbean: Building an Evidence Base for Stronger Schools. Innocenti Working Paper 2021-02. Florence: UNICEF Office of Research – Innocenti. Cruz, Consuelo (2013). Nota Técnica de Género República Dominicana. Borrador. World Bank. de Hoyos, R., Rogers, H., & Székely, M. (2016) “Out of School and Out of Work: A Diagnostic of NiNis in Latin America.â€? Background paper for “Out of School and Out of Work: Challenges and Solutions around the NiNis in Latin America.â€? Regional study, World Bank, Washington, DC. Dirección de Evaluación de la Calidad (2017). Informe Resultados Pruebas Nacionales 2017 Primera Convocatoria. https://www.ministeriodeeducacion.gob.do/docs/direccion-general- de-evaluacion-y-control-de-la-calidad-educativa/MoWJ-informe-estadistico-de-pruebas- nacionales-primera-convocatoria-2017pdf.pdf Dirección de Evaluación de la Calidad (2018). Informe Resultados Pruebas Nacionales 94 95 2018 Primera Convocatoria. https://www.ministeriodeeducacion.gob.do/docs/direccion-de- evaluacion-de-la-calidad/UtWE-informe-estadistico-de-pruebas-nacionales-2018-1- corregidopdf.pdf Dirección de Evaluación de la Calidad (2018). Resultados de la Evaluación Diagnóstica Nacional de Tercer Grado de Primaria 2017 Informe Nacional. https://www.ministeriodeeducacion.gob.do/docs/direccion-de-evaluacion-de-la-calidad/vfus- informe-nacional-de-la-evaluacion-diagnostica-de-tercer-grado-pdf.pdf Dirección de Evaluación de la Calidad (2019). Informe Resultados Pruebas Nacionales 2019 Primera Convocatoria. https://www.ministeriodeeducacion.gob.do/docs/direccion-de- evaluacion-de-la-calidad/TDGA-informe-estadistico-de-pruebas-nacionales-2019-1-pdf.pdf Dirección de Evaluación de la Calidad (2019). Resultados de la Evaluación Diagnóstica Nacional de genderto Grado de Primaria 2018 Informe Nacional. https://www.ministeriodeeducacion.gob.do/docs/direccion-de-evaluacion-de-la-calidad/AWZB- informe-nacional-evaluacion-diagnostica-genderto-grado-2018pdf.pdf Dirección de Evaluación de la Calidad (202O). Resultados De La Evaluación Diagnóstica Nacional De Tercer Grado De Secundaria 2019 Informe Nacional https://www.ministeriodeeducacion.gob.do/docs/direccion-de-evaluacion-de-la-calidad/NJXY- informe-nacional-2019-evaluacion-diagnostica-3ero-secundariapdf.pdf Disla, Natanael (2020). Las masculinidades y su impacto en el matrimonio infantil y uniones tempranas. Febrero 2020. Plan Internacional, Republica Dominicana. Garganta, S., Pinto, M. F., & Zentner, J. (2022). Jornada escolar extendida y fecundidad adolescente en la República Dominicana. NOTA TECNICA N ° IDB-TN-2559. https://publications.iadb.org/publications/spanish/document/El-efecto-de-la-doble-escolaridad- sobre-la-participacion-laboral-femenina-en-Republica-Dominicana.pdf Ferrant, G., Pesando, L. M., & Nowacka, K. (2014). Unpaid Care Work: The missing link in the analysis of gender gaps in labour outcomes. Boulogne Billancourt: OECD Development Center. Hernandez, Manuel A., J. A. Pellerano and G.E. Sanchez. (2022). Conditional Cash Transfers and High School Attainment Evidence from a Large-Scale Program in the Dominican Republic. IFPRI Discussion Paper 02109. Huber, M., & Mellace, G. (2014). Testing exclusion restrictions and additive separability in sample selection models. Empirical Economics, 47(1), 75-92. ILO (2014). DiagnoÌ?stico Igualdad de geÌ?nero en los Institutos de FormacioÌ?n Profesional AmeÌ?rica Central y RepuÌ?blica Dominicana. Instituto Tecnológico de Santo Domingo (2013). Costos del Embarazo y la Maternidad en la Adolescencia en la República Dominicana. Instituto Tecnológico de Santo Domingo (2016). Brechas de género: entre el dicho y el hecho de la autonomía de las mujeres en la República Dominicana. Ã?rea de Ciencias Sociales y Humanidades, Centro de Estudios del Género. Inter-American Dialogue, EDUCA, (2015). República Dominicana: El estado de las políticas públicas docentes: en la antesala de las transformaciones. Informes de Seguimiento PREAL. chrome- extension://efaidnbmnnnibpcajpcglclefindmkaj/viewer.html?pdfurl=http%3A%2F%2Feduca.or g.do%2Fwp-content%2Fuploads%2F2015%2F04%2FEstado-de-situaci%25C3%25B3n- web.pdf&chunk=true León, Juan (2018). Programa Internacional para la Evaluación de los Estudiantes PISA 2015 República Dominicana: Informe Nacional. https://www.ministeriodeeducacion.gob.do/docs/direccion-de-evaluacion-de-la- calidad/UtRl-informepisa2015-webpdf.pdf LoÌ?pez, A. (2020). “Reseña del libro Estar, ser y convivir en la escuela: una mirada profunda a la violencia escolar en República Dominicanaâ€?. Ciencia y Educación, 4(1), 93-95. Doi: https://doi.org/10.22206/cyed.2020.v4i1.pp93-95. 95 96 Machado, A.L. and Muller, M. (2018). Policy Research Working Paper8358“If it’s already tough, imagine for me...â€?A Qualitative Perspective on Youth Out of School and Out of Work in Brazil. Policy Research Working Paper 8358. World Bank, Washington, DC. Male and Wodon (2016). Basic Profile of Child Marriage in the Dominican Republic. World Bank Knowledge Brief. Male and Wodon (2016b). Basic Profile of Early Childbirth in the Dominican Republic. World Bank Knowledge Brief. Mateo DiÌ?az, M., and L. RodriÌ?guez-Chamussy (2016). Cashing in on Education: Women, Childcare, and Prosperity in Latin America and the Caribbean. Latin American Development Forum. Ministerio de Educación (2021). Plan Estratégico 2021-2024. chrome- extension://efaidnbmnnnibpcajpcglclefindmkaj/viewer.html?pdfurl=https%3A%2F%2Fwww.m inisteriodeeducacion.gob.do%2Ftransparencia%2Fmedia%2Fplan-estrategico-institucional- pei%2Fplanificacion-estrategica-institucional%2FtoJ-rrf-plan-estrategico-2021-2024- peipdfpdf.pdf. Ministerio de Economía, Planificación y Desarrollo - MEPYD. (2023). Feminización de la pobreza en República Dominicana: Probabilidad de Transición y Trampas de la Pobreza. Texto de Discusión #44. Ministerio de Economía, Planificación y Desarrollo - MEPYD. (2021). Análisis del desempeño económico y social de Republica Dominicana en 2021. Ministerio de Economía, Planificación y Desarrollo - MEPYD. (2020). “COVID bajo la lupa – Efectos de la COVID-19 en la pobreza monetaria, desigualdad y mercado de trabajo.â€? Morales, G., Aguilar, L., Hanna, J., Cohn-Lois, A., Ng, R., Pimentel, C., & Hedeman, K. (2018). Plan de acción género y cambio climático de República Dominicana (PAGCC-RD). Muggah and Aguirre Tobón (2018). Citizen Security in Latin America: The Hard Facts. Igarapé Institute. Mulligan, C. B., & Rubinstein, Y. (2008). Selection, investment, and women's relative wages over time. The Quarterly Journal of Economics, 123(3), 1061-1110. Maasoumi, E., & Wang, L. (2019). The gender gap between earnings distributions. Journal of Political Economy, 127(5), 2438-2504. National Academies of Sciences, Engineering, and Medicine (NASEM) 2021. Family Planning, Women's Empowerment, and Population and Societal Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. https://doi.org/10.17226/26023. Nix, S., Perez-Felkner, L., & Thomas, K. (2015). Perceived mathematical ability under challenge: a longitudinal perspective on gender segregation among STEM degree fields. Frontiers in psychology, 6, 530. Nopo (2012). New Century, Old Disparities: gender and Ethnic Earnings Gaps in Latin America and the Caribbean. Observatorio de Políticas Sociales y Desarrollo (2017). Epidemia de Violencia contra la Mujer: Un Enfoque desde la Masculinidad. Boletín 1, 4, 2017. Observatorio de Políticas Sociales y Desarrollo (2017). Epidemia de Violencia contra la Mujer: Un Enfoque desde la Masculinidad. Boletín 1, 4, 2017. Oficina Nacional de Estadística (2016). Encuesta Nacional de Hogares de Propósitos Múltiples (ENHOGAR). Informe General 2016. Oficina Nacional de Estadística (2021). Encuesta Nacional de Hogares de Propósitos Múltiples (ENHOGAR). Informe General 2021. Pacheco-Salazar, B. (2019). Estar, ser y convivir en la escuela: una mirada profunda a la violencia escolar en República Dominicana. Santo Domingo: Organización de Estados Iberoamericanos para la Educación, la Ciencia y la Cultura (OEI), Instituto Tecnológico de Santo Domingo (INTEC) e Instituto Nacional de Formación y Capacitación del Magisterio (INAFOCAM). León, Juan (2018). Programa Internacional para la Evaluación de los Estudiantes PISA 96 97 Plan Internacional (2021). Yo me crie así: Resumen del estudio sobre prácticas de crianza que promueven u obstaculizan la protección de niños, niñas y adolescentes del abuso y la violencia, así como la igualdad de género en las provincias: Elías Piña, Gran Santo Domingo y La Altagracia. Planteamientos n.7. Abril 2021. Plan Internacional, República Dominicana. 2015 República Dominicana: Informe Nacional. https://www.ministeriodeeducacion.gob.do/docs/direccion-de-evaluacion-de-la-calidad/UtRl- informepisa2015-webpdf.pdf Rubio Bañón, et al. (2016). Cultural factors and gender role in female entrepreneurship. Suma de Negocios Volume 7, Issue 15, January–June 2016, Pages 9-17. Salinas Mulder, Silvia et al. Vivencias y relatos sobre el embarazo en adolescentes. Plan Internacional y UNICEF. Panamá, 2014. Save the Children (2019). Caracterización del matrimonio infantil/uniones tempranas en cuatro comunidades de la provincia San Cristóbal, República Dominicana. Santo Domingo, 2019. Schultz, T. (2004): "School Subsidies for the Poor: Evaluating the Mexican PROGRESA Poverty Program," Journal of Development Economics, vol. 74, 199-250. Székely, M., & Karver, J. (2015). Youth out of school and out of work in Latin America: a cohort approach. World Bank Policy Research Working Paper No. 7421. Available at SSRN: https://ssrn.com/abstract=2666786 Teignier and Cuberes (2016). Aggregate Effects of gender Gaps in the Labor Market: A Quantitative Estimate. Journal of Human Capital, 2016, vol. 10, no. 1. Tineo, Jeannette del Carmen (2017). Niñas esposadas. Caracterización del matrimonio infantil forzado en las provincias de Azua, Barahona, Pedernales, Elías Piña y San Juan. Plan Internacional República Dominicana, 2017. Tineo, Jeannette del Carmen; prólogo de Lourdes Contreras (2014). Imaginarios de género en juventudes dominicanas: aportes para el debate desde la colonialidad del poder. Instituto Tecnológico de Santo Domingo INTEC, Republica Dominicana, 2014. Treviño et al. (2015). Informe de resultados TERCE: Factores Asociados. Laboratorio Latinoamericano de Evaluación de la Calidad de la Educación. UNESCO. https://unesdoc.unesco.org/ark:/48223/pf0000243533 UN (2015). Trends in Contraceptive Use Worldwide 2015. UNDP (2016). Más Mujeres, Más Democracia: Desafíos para la igualdad de género en la política, Estudio sobre la participación política y el acceso a la justicia contenciosa electoral de las mujeres en las elecciones generales de República Dominicana del 2016. UNFPA (2013). Motherhood in childhood: facing the challenge of adolescent pregnancy. UNICEF (2019). El matrimonio infantil y las uniones tempranas. Resumen del estudio de conocimientos, actitudes y prácticas en seis municipios de la República Dominicana. PROSOLI, Vicepresidencia de la República, UNICEF, 2019, RD. Wodon et al. (2017). Ending child marriage: Child Marriage Laws and their Limitations. The World Bank, Save the Children, Children´s Investment Fund Foundation, Global Partnership for Education. UNICEF (2019) Child marriage, adolescent pregnancy and school dropout in South Asia. Vargas G., Tahira (2021). Prácticas de crianza que promueven u obstaculizan la protección de niños, niñas y adolescentes del abuso y la violencia, así como la igualdad de género en las provincias: Elías Piña, Gran Santo Domingo y La Altagracia. Dimensión cuantitativa y cualitativa. Equipo Vargas y Asociados, Plan Internacional, República Dominicana. Vezza E., B. García, G. Cruces and J. Amendolaggine (2014). Youth and Employment Program, Impact Evaluation Report. 2008-2009 Cohorts. Ministry of Labor, Dominican Republic. file:///C:/Users/lizar/Downloads/PJyE-Evaluation-ENG-DIC2014-1.pdf World Health Organization - WHO (2019) Las ENT de un vistazo: Mortalidad de las enfermedades no transmisibles y prevalencia de sus factores de riesgo en la Región de las Américas. 97 98 Winkler, H. and M. Montenegro. 2021. “Dominican Republic Jobs Diagnostic.â€? World Bank, Washington, DC. Wodon, Q., C. Fèvre, C. Malé, A. Nayihouba, and H. Nguyen. 2021. Ending Violence in Schools: An Investment Case – Selected Findings, Washington, DC: The World Bank and the Global Partnership to End Violence against Children. World Bank (2018). Dominican Republic Systematic Country Diagnostic. World Bank (2017). Impacto Económico del Matrimonio Infantil y de las Uniones Tempranas. Resumen para la República Dominicana. World Bank (2016). Global Monitoring Report 2015/2016: Development Goals in an Era of Demographic Change. Washington, DC: World Bank. World Bank (2016b). Women, Business and the Law. The World Bank. World Bank (2012). Teenage Pregnancy and Opportunities in LAC. The World Bank. World Bank (2020). Women, Business and the Law. The World Bank. World Bank (2021b) Teenage Pregnancy in the Dominican Republic. Policy Notes. World Bank Group. World Bank (2022) Reducing Teen Pregnancy (English). LAC gender Notes Washington,D.C.: World Bank Group. Ziffer et al. (2018). Promoción de la Equidad de Género en la Educación Básica. Programa Multifase para la Equidad en la Educación Básica. 98 99 ANNEX 1. The legal and institutional framework for gender equality The Dominican Republic has signed all the most important international instruments for the protection and promotion of gender equality, and has passed domestic norms to incorporate these commitments into national legislation. Indeed, according to the Woman, Business and the Law assessment and ranking for 2020, the Dominican Republic only presents significant gaps in terms of divorce, parental leave, and pension calculation. Table A2.1: Women, Business and the Law in the Dominican Republic DR LAC Upper Global Middle income Mobility: Can a woman apply for a passport in the same way as a man? Can a 100 92 88 87 woman travel outside the country in the same way as a man? Can a woman travel outside her home in the same way as a man? Can a woman choose where to live in the same way as a man? Workplace: Can a woman get a job in the same way as a man? Does the law 100 77 74 78 prohibit discrimination in employment based on gender? Is there legislation on genderual harassment in employment? Are there criminal penalties or civil remedies for genderual harassment in employment? Pay: Does the law mandate equal remuneration for work of equal value? Can 100 69 68 66 women work the same night hours as men? Can women work in jobs deemed dangerous in the same way as men? Are women able to work in the same industries as men? Marriage: Is there no legal provision that requires a married woman to obey 80 89 83 78,5 her husband? Can a woman be "head of household" or "head of family" in the same way as a man? Is there legislation specifically addressing domestic violence? Can a woman obtain a judgment of divorce in the same way as a man? Does a woman have the same rights to remarry as a man? Parenthood: Is paid leave of at least 14 weeks available to mothers? Does the 60 69 54 54 government administer 100% of maternity leave benefits? Is there paid leave available to fathers? Is there paid parental leave? Is dismissal of pregnant workers prohibited? Entrepreneurship: Can a woman sign a contract in the same way as a man? 100 83 81 83 Can a woman register a business in the same way as a man? Can a woman open a bank account in the same way as a man? Does the law prohibit discrimination in access to credit based on gender? Assets: Do men and women have equal ownership rights to immovable 100 98 86 82 property? Do sons and daughters have equal rights to inherit assets from their parents? Do female and male surviving spouses have equal rights to inherit assets? Does the law grant the spouse equal administrative authority over assets during marriage? Does the law provide for the valuation of nonmonetary contributions? Pension: Are the ages at which men and women can retire with full pension 75 80 65 73 benefits equal? Are the ages at which men and women can retire with partial pension benefits equal? Is the mandatory retirement age for men and women equal? Are periods of absence from work due to childcare accounted for in pension benefits? Legal framework: International dimension Various international treaties, ratified and incorporated into national legislation, operate as the framework for the integration of gender equality into government policies in the Dominican Republic. The most relevant of these include: • The International Covenant for Civil and Political Rights, ratified in 1978, which established the commitment of the State to ensure that women and men enjoy equal political and civil rights 99 100 • The Convention on the Elimination of all Forms of Discrimination against Women (CEDAW) and its Protocol, ratified in 1994, which mandate that member states adopt the necessary measures to prevent, investigate and punish discrimination against women • The Inter-American Convention to Prevent, Sanction and Eradicate Violence against Women (Convención Belém Do Pará) was approved in 1997, and recognizes that gender-based violence constitutes a serious violation of human rights that States need to prevent, punish and eradicate160 • The Statute of Rome for the International Court signed in 1998 recognizes any genderual aggression and trafficking as crimes against humanity • The UN Convention against Organized Crime approved in 2000 establishes the obligation of States to assist and protect the victims of trafficking. • ILO Conventions 100 on equality of remuneration, 111 on discrimination, 183 on the protection of maternity, and 189 on domestic workers. The DR has not yet ratified Convention 156 on workers with family responsibilities. Other relevant international commitments include: (i) The Beijing Platform for Action161 adopted in 1995 and the new related measures adopted in 2000; (ii) The Millennium Development Goals approved in 2000 and the Sustainable Development Goals approved in 2015; (iii) The Action Framework of Dakar on education adopted in 2000; (iv) Inter-American program on the promotion of the human rights of women and gender equality and equity adopted in 2000 (OEA); (v) the Mexico Consensus162 adopted in 2004; (vi) The Quito Consensus163 adopted in 2007; and (vii) the Brasilia Consensus164 adopted in 2010. Legal framework: National dimension The main domestic norms that guarantee the right to equality between men and women include: • The Constitution (2010). Recognizes the right to equality and non- discrimination based on gender. In several of its dispositions the constitution makes references to gender equality, across relevant areas such as employment, violence or education, and a sanctions gender based and intrafamily violence in all its forms 160 The Belém do Pará Convention entered into force on March 5, 1995 and was the world’s first binding international treaty to recognize that violence against women constitutes a violation of human rights. As a legally binding treaty, the Belém do Pará Convention not only condemns violence against women as an assault on human dignity but also outlines states’ obligations to eliminate it. 161 The Beijing Platform for Action, adopted at the UN’s Fourth World Conference on Women (Beijing, China, 1995), flagged 12 key areas where urgent action was needed to ensure greater equality and opportunities for women and men, girls and boys. It also laid out concrete ways for countries to bring about change. 162 Regional Consensus adopted after the Ninth Regional Conference on Women in Latin America and the Caribbean held in Mexico City in June 2004. The governments of the region renewed their commitment to continue adopting measures aimed at the advancement of women´s rights, incorporating a gender perspective into public policies, adopting proactive policies for job creation, recognizing the economic value of domestic work, guaranteeing genderual and reproductive rights, curtailing violence against women, or addressing HIV/AIDS, among others. 163 The Consensus followed the Tenth Regional Conference on Women in Latin America and the Caribbean held in Quito, Ecuador, in August 2007. The Consensus focused on two strategic issues: i) political participation and gender parity in decision-making at all levels, and ii) the contribution of women to the economy and social protection, especially in relation to unpaid work. Representative governments established specific measures to overcome gender discrimination in political participation, employment, education, health and the economy. 164 Government leaders agreed to facilitate women's access to new technologies, promote a democratic and non- discriminatory media, improve health and genderual and reproductive rights of women, and promote international and regional cooperation for gender equity. 100 101 • Law 01-12 on the National Development Strategy establishes the need to mainstream gender equality as a transversal action line across all other sectors • Law 176-07 on National District and Municipalities, entails formal changes in the advancement and implementation of measures that favor equality in local management. It establishes equality as a principle, recognizes participatory budget processes including women, the need to include them in local government representations, etc. • Law 24-97 on intrafamily violence and violence against women. • Article 100 of the 2014 Criminal Code typifying femicide, sanctioned with 30- 40 years of prison. • Law 2011 that facilitates the access to women victims of violence to justice. • Law 137-03 on illegal trafficking of people. • Law 88-03 that creates the Casas de Acogida. • Law 87-01 that creates the Social Security System mandates the provision of childcare services for children from 45 days to 5 years old. • Law 04-99 on pre- and post-natal leave establishes a compulsory maternal leave of 6 weeks prior to delivery and 6 weeks after • Law 12 modifying the Electoral Law establishes a 33 percent quota of women in all candidates lists in national and local elections. • Law 13-2000 mandates the inclusion of one woman for the position of síndico and vice-síndico (mayor and vice mayor)165 165 Observatorio de Igualdad de Género CEPAL 101 102 ANNEX 2: A characterization of poverty by gender, 2019 Poverty Share of poor in group Share of Share of Ratio Female Group female in Male Female Male Female population / Male group Total 100 50.9 19.8 22.0 1.12 46.3 53.7 Age 0 - 15 26.3 48.4 35.0 34.5 0.99 51.9 48.1 15 - 24 17.8 50.8 16.9 22.4 1.32 42.3 57.7 25 - 40 22.8 51.2 16.5 23.9 1.45 39.6 60.4 41 - 55 17.1 51.9 11.0 12.2 1.11 45.5 54.5 56 and older 16.1 53.8 9.9 10.8 1.09 44.0 56.0 Area Rural 18.4 49.1 24.1 26.7 1.11 48.4 51.6 Urban 81.6 51.4 18.7 21.0 1.12 45.7 54.3 Labor market (15 years and older) Employer 2.0 23.8 2.0 2.1 1.07 74.9 25.1 Salaried 35.6 49.0 11.0 9.4 0.85 54.9 45.1 Self-employment 22.6 27.7 12.6 15.3 1.21 68.3 31.7 Unpaid worker 1.0 64.2 24.7 15.0 0.61 45.6 54.4 Unemployed 4.2 62.9 32.4 34.5 1.06 35.7 64.3 Education (15 years and older) Never attended 5.1 50.0 21.9 25.9 1.18 44.7 55.3 Incomplete primary 24.3 48.2 16.8 20.5 1.22 46.8 53.2 Complete primary 9.6 43.8 14.9 20.8 1.40 48.3 51.7 Incomplete secondary 18.4 49.7 17.2 25.1 1.47 41.3 58.7 Complete secondary 19.7 50.7 12.3 18.1 1.47 39.9 60.1 Complete or incomplete tertiary 22.8 62.3 5.2 8.5 1.63 27.1 72.9 Marital status Cohabitation (married or living together) 35.8 50.5 16.7 16.8 1.01 49.4 50.6 Divorced/Separated 15.8 61.4 6.3 21.9 3.50 15.2 84.8 Widow/widower 4.1 81.1 9.1 12.1 1.32 14.9 85.1 Single 44.3 41.6 17.9 21.8 1.21 53.6 46.4 Household headship (marital status and cohabitation) Share of heads * No cohabitation (married) 4.3 (6.7) 49.1 17.2 24.0 1.39 42.6 57.4 Cohabitation (married or living together) 53.2 (18.0) 48.3 28.3 29.3 1.04 50.9 49.1 Divorced/Separated 23.4 (43.9) 56.4 29.4 34.0 1.16 40.1 59.9 Widow/widower 10.0 (22.1) 56.6 28.9 32.6 1.13 40.4 59.6 Single 9.1 (9.4) 42.8 16.4 27.3 1.67 44.5 55.5 * Share of female heads in parentheses Poverty headcount ratio by household configuration (% households) Share of Male-headed Female-headed Total female heads Unipersonal 32.9 2.3 4.0 2.9 Adult couple with children and other adults 35.3 19.0 28.3 22.3 Adult couple, no children 30.3 4.9 4.7 4.9 One adult, female with children 91.4 32.3 38.2 37.7 Adult couple with children 16.4 26.6 26.8 26.7 Multiple adults, only female with children 18.8 22.5 22.8 22.6 Senior only, no children 44.4 5.6 3.4 4.6 Other adult combinations with children 84.3 21.0 36.6 34.2 Other adult combinations without children 38.6 5.1 8.8 6.5 Other typology 45.8 12.5 10.3 11.5 Total 37.1 13.8 18.4 15.5 102 103 ANNEX 3 – Prioritization of gender gaps A3.1 Methodology This section introduces the methodology applied to evaluate the standing of the Dominican Republic against other countries in relation to a series of gender gaps. First, we explain how we compare the Dominican Republic with other groups of countries that are similar in several dimensions and global averages. Then, we conclude, showing results that illustrate the most important challenges Dominican Republic is facing. Indicators for most of the main gaps identified in section 1 were sought to benchmark Dominican Republic’s performance against other blocks of countries. The basis for this selection was the Plan Nacional de Igualdad y Equidad de Genero 2019-2030 (PLANEG III), which has detailed knowledge of the specific challenges faced by the country and is organized in six pillars: Education for Equality; Integral Health for Women; Economic Autonomy; Citizenship, Democracy and Political and Social participation; gender-Based Violence Against Women; and Digital Technologies for Women’s Autonomy (See A3.1). A total of 130 indicators were selected. Box A3.1. The III National Plan for Equality and Gender Equity 2019-2030 (PLANEG III) The III National Plan for Equality and gender Equity 2019-2030 (PLANEG III), based on the National Development Strategy and the SDGs, is the main strategic policy instrument in this area. Some of the main priority areas and actions it highlights are summarized below in this box. (1) Education for equality: establish the non-religious nature of formal education, rebuild study plans with a gender approach for staff training at the senior staff level, and encourage teaching staff to regularly refamiliarize themselves with the Law for General Education and the Law for Higher Education, Science and Technology to ensure that it incorporates a gender equality approach as principle in education at all levels. (2) Health: includes the recognition of sexual and reproductive rights and ensuring universal access to sexual and reproductive health services, the reduction of maternal mortality and unsafe abortion, and the prevention and attention to STD, chronic health conditions, and GBV with a gender perspective. (3) Economic autonomy: includes the production of data on care and women in agriculture, the reduction of the feminization of poverty by increasing the capacity of women to generate their own income, ensuring equal labor market participation in the formal sector and in emerging areas, especially making intensive use of ICTs, the reduction of the wage gap and other types of discrimination, and the redistribution of care-related tasks between men and women. (4) Social and political participation: ensuring that women are candidates and elected in all democratic processes on an equal footing with men and their rights are protected in political spaces, launching campaigns that help dismantle patriarchal culture, increase women’s participation in the formulation and execution of policies linked to their rights as citizens, and incorporate the debate on the current electoral system and its impact on the rights of women. (5) Environment: increase the levels of knowledge about the interrelations between gender and the environment, widen the participation of women in decision-making mechanisms about the environment and climate change risk mitigation and adaptation, and widen the opportunities for women to participate in entrepreneurship linked to sustainable production models in rural areas. (6) GBV: Promote a legal framework related to the law that creates a system for attention to VAW that fuels changes in cultural patterns about this phenomenon, the creation of a system for prevention, detection, attention, sanction and reparation related to VAW, strengthening intersectoral and interinstitutional coordination mechanisms, create a unique registry, ensuring an adequate budget allocation, integrate general education in the educational system, and promote values of equality and no discrimination. (7) ICT: Incorporate a gender equality focus in the design, execution and evaluation of plans, programs and projects in public policy for digital transformation, science and technology, develop accessible and free content online that contributes directly to female empowerment and autonomy, increase female representation in STEM, and increase the understanding and analysis from an equality perspective of 103 104 the challenges and impacts of ICT on women and girls, and particularly with regards to violence, privacy or health. The performance of the Dominican Republic in the selected indicators will be assessed in comparison to the global average as well as Latin America and the Caribbean as a whole and Upper Middle-Income countries. Additionally, a comparison with a group of comparator or peer countries was conducted to ensure that results consider the key constraints that Dominican Republic is facing.166 Upon selection of the indicators from PLANEG III that were available in the World Development Indicators, the team defined a gap to measure the distance between the Dominican Republic’s performance and the best performers within each of the blocks of comparator countries. The results give a sense of how far the Dominican Republic is from achieving the best values achieved by the other countries. The performance gap for each group was computed as follows: |𝑖𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 í µí±?𝑒𝑠𝑡 í µí±?𝑒𝑟𝑓𝑜𝑟𝑚𝑒𝑟𝑠 𝑖𝑛 í µí±”í µí±Ÿí µí±œí µí±¢í µí±? − 𝑖𝐷𝑅 | í µí°ºí µí±Ží µí±?í µí±”í µí±Ÿí µí±œí µí±¢í µí±? = 100. |𝑖𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 í µí±?𝑒𝑠𝑡 í µí±?𝑒𝑟𝑓𝑜𝑟𝑚𝑒𝑟𝑠 𝑖𝑛 í µí±”í µí±Ÿí µí±œí µí±¢í µí±? − 𝑖𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑤𝑜𝑟𝑠𝑒 í µí±?𝑒𝑟𝑓𝑜𝑟𝑚𝑒𝑟𝑠 𝑖𝑛 í µí±”í µí±Ÿí µí±œí µí±¢í µí±? | Where the gap in any given indicator i between Dominican Republic and the best performers is based on the country performance in the 95th percentile within the distribution of each indicator. The worst performer is obtained within values lower than the 5th percentile. The gap goes from 0 to 100 where the best performers are countries with the lowest indicator value. In other words, the closer the indicator is to 100 the higher the severity of the constraint being faced. In turn, the quantiles were defined as percentile, we define four levels of priority to establish the relevance of the indicator: i) High (red color): gap is between 75 and 100, ii) Medium high (: gap is between 50 and 75, iii) Medium low: gap is between 25 and 50, and iv) Low: gap is between 0 and 25. Results for the full set of selected indicators are shown in Table 15. The Dominican Republic would appear to be among the worst performers globally, and in relation to LAC, Upper-Middle Income countries and structural peers, when it comes to child marriage and teenage pregnancies; school performance (measured through harmonized test scores); labor force participation and wage gaps in the private sector; high time and cost constraints to open a business; and low savings for retirement and access to digital technologies for saving purposes. 166 The structural peers include countries that share economic traits with the DR, such as their current account deficit, educational attainment, current GDP levels, GNI per capita, labor force with basic education, total population, and poverty levels. Based on these criteria, the group consists of Sri Lanka, Tunisia, Ecuador, Peru, Guatemala, Honduras, Romania and Fiji. (Source: Dominican Republic SCD, 2018). 104 105 Table A3.1 gender gaps for the Dominican Republic with respect to four groups of countries 105 106 Notes: *The Action Line serves to map each indicator to the gender gaps outlined in PLANEG III. Structural peers are Sri Lanka, Tunisia, Ecuador, Peru, Guatemala, Honduras, Romania and Fiji. 106 107 ANNEX 4. Risk factors associated with adolescent pregnancy Table A4.1 Probability of Late Adolescent Pregnancy Location, Location, Households, Location, Location and Households, Behavior, Location Households Households Behavior and Education and and Behavior Education Labor participation Urban -0.0439*** -0.0147*** 0.00917*** 0.00905*** 0.0144*** (0.00164) (0.00158) (0.00109) (0.00109) (0.00107) (%) of pregnant adolescents in municipality 0.996*** 0.802*** 0.362*** 0.366*** 0.338*** (0.00880) (0.00824) (0.00571) (0.00570) (0.00567) QLI I (Extreme poor) 0.257*** 0.0197*** 0.0122*** -0.0144*** (0.00340) (0.00250) (0.00251) (0.00253) QLI 2 (Moderate poor) 0.200*** 0.0290*** 0.0229*** 0.0178*** (0.00151) (0.00110) (0.00111) (0.00109) QLI (Vulnerable) 0.0719*** 0.00358*** 0.000999 -0.00348*** (0.00113) (0.000818) (0.000827) (0.000812) Father present -0.167*** -0.0264*** -0.0258*** -0.0253*** (0.00105) (0.000696) (0.000701) (0.000687) Total household members -0.0032*** 0.0128*** 0.0126*** 0.0130*** (0.000378) (0.000254) (0.000254) (0.000254) Born in Haití 0.0589*** -0.0179*** -0.0212*** -0.0708*** (0.00434) (0.00318) (0.00324) (0.00320) Age 0.00719*** 0.0147*** 0.00226*** (0.000284) (0.000399) (0.000284) Ever in a genderual relationship 0.482*** 0.477*** 0.466*** (0.00208) (0.00209) (0.00211) Exposure time 0.0522*** 0.0509*** 0.0489*** (0.000426) (0.000426) (0.000421) Positive perception of adolescent motherhood 0.138*** 0.136*** 0.130*** (0.00198) (0.00198) (0.00199) NEET 0.0968*** (0.00127) Educational deficit 0.0101*** (0.000332) Completed primary school 0.00177 (0.00178) Completed high school -0.0298*** (0.00211) Enrolled in higher education -0.0387*** (0.00238) Course or completed vocational training in the last 12 months -0.0145*** (0.00103) Development region Cibao Nordeste -0.0108*** -0.0113*** -0.0219*** -0.0221*** -0.0102*** (0.00248) (0.00242) (0.00174) (0.00173) (0.00168) Cibao Noroeste -0.00785** -0.0518*** -0.0400*** -0.0419*** -0.0321*** (0.00323) (0.00307) (0.00212) (0.00212) (0.00210) Cibao Norte -0.00719** -0.0162*** -0.0148*** -0.0156*** -0.0126*** (0.00179) (0.00169) (0.00120) (0.00121) (0.00119) Cibao Sur -0.0130*** 0.00133 -0.0182*** -0.0210*** -0.0110*** (0.00245) (0.00234) (0.00163) (0.00163) (0.00163) El Valle -0.0114*** -0.0445*** -0.0143*** -0.0155*** -0.0116*** (0.00381) (0.00372) (0.00220) (0.00220) (0.00215) Enriquillo -0.00313 -0.0367*** -0.0179*** -0.0209*** -0.0171*** (0.00336) (0.00316) (0.00197) (0.00197) (0.00202) - Higuamo 0.00656*** -0.0400*** 0.0102*** 0.00677*** 0.00989*** (0.00247) (0.00242) (0.00151) (0.00151) (0.00149) - Valdesia 0.00987*** -0.00508*** 0.00591*** 0.00380*** 0.00621*** (0.00193) (0.00184) (0.00128) (0.00129) (0.00125) Yuma -0.00184 -0.00828*** -0.00815*** -0.0115*** -0.0108*** (0.00253) (0.00242) (0.00168) (0.00169) (0.00169) Constant 0.0419*** 0.0443*** -0.250*** -0.360*** -0.179*** (0.00222) (0.00270) (0.00515) (0.00671) (0.00513) Observations 471,903 471,903 464,089 464,089 460,501 R-squared 0.038 0.135 0.599 0.600 0.610 Source: Authors’ estimates based on ENHOGAR 2018. Adapted from World Bank (2021). Embarazo en la adolescencia en República Dominicana. Nota de Política. Pages 20-22 Note: Weighted Least Square Models (WLS), robust standard errors in parentheses, *** p<0.01, ** p<0.05, * p<0.1, Primary education completion refers to eighth grade completion. Exposure refers to age minus age at first gender. Dependent variable equal to 1 if the adolescent has been pregnant and zero if otherwise. 107 108 ANNEX 5. Initiatives aimed at developing agency, aspirations, and alternative life projects in the Dominican Republic Baby Think It Over (Bebé piénsalo bien): Program aimed at recreating with 12-17 year olds who are attending high school the responsibilities associated with caring for a baby. The program also sensitizes parents and community leaders, provides information on SRH, and encourages critical reflection on the experience by adolescents. The program began in 2006 and as of May 2021 had involved 78,850 adolescents, or around 5,100 every year, as well as 34,376 parents. It has been offered in 16 of the 32 provinces of the country, covering 38 of 155 municipalities. The program does not have an evaluation; it is favorably accepted although it has been considered to be expensive by decision-makers interviewed for this report. I decide to wait (Yo decido esperar): Program aimed at adolescents aged 13-18 who belong to beneficiary households of the Progresando con Solidaridad Program (PROSOLI) with the aim of preventing adolescent pregnancy and Sexually transmitted diseases, as well as delaying Sexual intercourse. During the 2017-20 period, through the ‘I decide to wait’ program, PROSOLI gave 132,488 adolescents guidance on SRH. Also with the support of the Korea International Cooperation Agency, PROSOLI is developing a project to reduce teenage pregnancy through comprehensive training, social participation, and income generation for adolescents, vulnerable young people and their families. The Program is aimed at 7,500 girls and adolescents, including 300 who are pregnant or are already mothers. It plans to provide childcare services while adolescent mothers receive coaching. The program does not have an evaluation. Girls Club (Club de Chicas): Program developed by PROSOLI to promote alternative life plans in adolescents who are members of PROSOLI households through a differentiated training curriculum for adolescents aged 10-14 and 15-19. Since 2020, PROSOLI with the support of UNICEF and the European Union has been implementing a pilot project aimed at generating change in behavior, attitudes and practices on early relationships with 1,500 girls and adolescents. Girls Club’s methodology was developed by UNICEF with several modules: Sex, youth and law; SRH and Sexual and reproductive rights; and Life Project. The methodology allows girls to learn and express themselves in a playful and creative environment. Trained mentors work with the adolescents for 2 to 3 hours a week. At the pilot level, the Girls Club is being developed with 10 to 15 girls per club in the provinces of Higüey, Santo Domingo Norte and Barahona. The program of clubs starts with a 1–2-hour meeting to sensitize the families. Local teenage boys/families have expressed interest in learning about working on positive masculinity. Girls Dreaming Big Club (Club de Chicas Fabricando Sueños): Pilot project developed by UNFPA involving public institutions working to protect children and adolescents and promote gender equality at the local level. The project seeks to generate awareness at an early age of the adolescent body as a private space for protection, and how to detect and react to risky situations. The program also addresses issues such as savings and finances, superfluous needs vs. basic needs, and the management of expectations and pressures that the management of social expectations and pressures generates. The club begins with 60 hours of social work that involves training of agents from the school and community. There are clubs for 8-11- and 12–15-year-old girls, starting in four locations with expansion planned in two more. A club is made up of 25 girls plus mentors, and involves mothers and people from the community. The clubs promote dialogue between the family and the adolescent. They last for about a year and a half. One challenge is how to maintain the bond with girls and communication between club members using virtual platforms. The Girls Dreaming Big Clubs are part of the Strategic Plan for a Life Free of Violence for Women 2020-2024, under the coordination of the Ministry of Women. Evaluations show that the earlier the work with girls is started, the greater its effectiveness. That’s Not Your Thing (Tú no Tá Pa Eso), Strategy developed by MINERD. Throughout one week periodically, a program of activities related to comprehensive sexuality education is developed at the school, culminating with a visit by adolescents to the community UNAP or by UNAP staff to the school. Girls' science and technology club. In 2010, as part of a joint project with UN Women, the Research Center for Women's Action (CIPAF), created math, science, engineering and technology (STEM) clubs for girls in public schools in Santo Domingo. The clubs are called E-Chicas and Supermáticas. Girls receive training and guidance in the areas of science, mathematics and technology in the public schools that have an Extended School Day. They aim to close the gender gap in technology careers in the country. Later they were absorbed by the Ministry of Education, and by 2016 there were 92 clubs scattered throughout various schools. The average number of students per club is 15 girls. 108 109 ANNEX 6. National Institute for Comprehensive Early Childhood Care (INAIPI) The National Institute for Comprehensive Early Childhood Care (INAIPI) is a decentralized institution, at the functional and territorial level, attached to the Ministry of Education of the Dominican Republic (MINERD). It was created by Decree Number 102-13, on April 12, 2013. INAIPI is responsible for managing the provision of quality comprehensive care services to children from 0-5 years and their families. It therefore aims to contribute to Early Childhood Development (ECD), ensuring the provision of comprehensive quality services to early childhood, in its different modalities, directly or through contracts and agreements with other public and private organizations and institutions, including socially based and/or faith based. The INAIPI provides its services directly through 122 Comprehensive Early Childhood Care Centers (CAIPI) and 521 Child and Family Care Centers (CAFI), located in the most vulnerable areas of the country. Since 2019, CAIPI has also managed the centers of the Administrator of Safe Children's Stays (AEISS). The implementation of INAIPI happens through three management modalities: (i) Direct Management Centers: technically and administratively managed and implemented by INAIPI; (ii) Co-management Centers: managed by social organizations with permanent support and coordination from INAIPI, and (iii) Existing Experiences Modality: Organizations that have been offering early childhood services for years and need support to improve and expand their offer. The CAIPIs are directly managed by INAIPI and offer education, health, and nutrition services to children from 45 days until 5 years old and support to families, who receive training to improve their skills in comprehensive care of their children. These centers operate from 7:30 a.m. to 5:00 p.m. and they have specialized personnel to respond to the need and demand for comprehensive care services for this population. The CAFIs are mainly co-managed centers and form part of the Family and Community Based Program (PBFC), which seeks to promote good child-rearing practices, aimed at fathers, mothers and/or guardians applying strategies that promote the development of and adequate care for children. The AEISS implement the following activities: Early stimulation for children from 0 to 2 years old; initial education for children 3 and 4 years old; support and care in homes for children from 0 to 4 and their families; strengthening of good parenting practices through training workshops; support of Godfathers and Godmothers to children from 0 to 2 years old and pregnant women at high risk; and articulation with the UNAP and local hospitals, Civil Registry Offices, and social programs of the Government for the comprehensive care of boys and girls. Through its services INAIPI reached 205,323 infants and 162,850 families in 2020. In terms of family participation, INAIPI had 475 Fathers, Mothers and/or Guardians Committees involved, and 31,546 families participated in training workshops. Source: Authors based on INAIPI Minutes, 2020. 109 110 ANNEX 7. Ñopo wage gap decompositions Methodology The Ñopo (2008) method is a matching technique used to explain gender wage differences. The approach emphasizes differences in the supports of the distributions of observable characteristics, meaning the differences among those segments of the labor markets where men and women compete and provide information. It also provides insights into the distribution of unexplained gender pay differences. The methodology produces synthetic samples of individuals by matching males and females with the same observable characteristics. By matching individuals in several dimensions, we compare comparable individuals. As an example, the earnings of a female worker, 30 years old, with a complete secondary education, living in an urban area, with no children in her household, and who has worked as an employee in manufacturing for 6 months, will be compared to the earnings of a male worker with the exact same characteristics. In other words, this nonparametric technique analyzes gender differences in those observable characteristics where there is an overlap (the so-called common support region) in the spirit proposed by Blinder and Oaxaca (1973). We then estimate how much of the gap is explained by the observable characteristics. If the wage gap explained by observables is negligible then the pay difference is explained by non-observable variables. Formally, Ñopo (2008) assumes a Y denoting a variable that models individuals’ earnings and X, the n- dimensional vector of individuals' characteristics (such as age, education, marital status, etc.) that are presumably related to earnings. Let 𝐹 𝑀 ( . ) and 𝐹 𝐹 ( . ) denote the conditional cumulative distribution functions of individuals' characteristics X, conditional on being male and female respectively, and𝑑𝐹 𝑀 ( . ) . and 𝑑𝐹 𝑀 ( . ) denoting their corresponding probability measures. In the same way, let µ𝐹 (𝑆) = ∫𝑠 𝑑𝐹 𝐹 (𝑥) 𝑀 𝑀 with analogous µ (𝑆). The relationship between these variables is modeled by the functions 𝑔 (𝑋) and 𝑔𝐹 (𝑋) that represent the expected value of earnings, conditional on characteristics and gender (E[Y|M, X] and E[Y|F, X] respectively). Defining the wage gap as: ∆ = E[Y|M] - E[Y|M] And after the corresponding arithmetic operations wage gap can be expressed as: . 𝑑𝐹 𝑀 (𝑥) . 𝑑𝐹 𝑀 (𝑥) 𝑀 𝐹 ∆ = [ ∫ 𝑔𝑀 (𝑥) − ∫ 𝑔 𝑀 (𝑥) Ì…Ì…Ì…) ] 𝜇 (𝑆 Ì… 𝑆 𝐹 Ì…Ì…Ì… 𝜇 𝑀 (𝑆 𝐹 ) 𝑆 𝐹 𝜇 𝑀 (𝑆𝐹 ) . 𝑑𝐹 𝑀 𝑑𝐹 𝐹 +∫ 𝑔𝑀 (𝑥) [ − ] (𝑥) 𝑆 𝑀 ∩𝑆 𝐹 𝜇 𝑀 (𝑆𝐹 ) 𝜇 𝐹 (𝑆𝑀 ) . 𝑑𝐹 𝐹 (𝑥) +∫ [𝑔𝑀 (𝑥) − 𝑔𝐹 (𝑥)] 𝑆 𝑀 ∩𝑆 𝐹 𝜇 𝐹 (𝑆𝑀 ) . 𝑑𝐹 𝐹 (𝑥) . 𝑑𝐹 𝐹 (𝑥) 𝐹 𝑀 + [∫ 𝑔𝐹 (𝑥) − ∫ 𝑔 𝐹 (𝑥) ] 𝜇 Ì…Ì…Ì…Ì…Ì…Ì… (𝑆 ) 𝑆 𝑀 𝜇 𝐹 (𝑆𝑀 ) Ì…Ì…Ì…Ì…Ì… 𝑆 𝑀 𝜇 Ì…Ì…Ì…Ì… 𝐹 (𝑆 𝑀 ) Following the other equation terms above, the Ñopo decomposition divides the gender wage gap into four additive elements: ∆ = (∆𝑚 + ∆𝑥 + ∆0 + ∆𝑓 ) Where, • ∆ : Total wage gap • ∆𝑥 : The component of the wage gap explained by males and females having individual characteristics distributed differently over their common support 110 111 • ∆𝑓 : The component of the wage gap that exists because for some combinations of female characteristics there are no comparable males • ∆𝑚 : The component of the wage gap that exists because for some combinations of male characteristics there are no comparable females • ∆0 : The component that cannot be explained by differences in observable characteristics. The use of this matching criterion assumes that individuals with the same observable characteristics should be paid the same regardless of their gender. This analysis considers that the distribution of characteristics across women and men may not overlap completely. For instance, a woman in a specific age, with university studies, cohabiting, with children, and working in a specific economic sector on a part-time contract might not have a male counterpart with the same characteristics to compare their earnings. Under this decomposition method, the component ∆𝑓 could give us information about the part of the gap that is explained because this kind of employment for this specific characteristic is only provided by women. Symmetrically, the component ∆𝑚 can be interpreted as the part of the gap that can be explained because women have no access to a certain kind of employment as a self that is provided only for men. (An hypothetical example: In our dataset the jobs for unskilled workers, working as a full-time self-employed providing services as flooring installer in the construction sector are only provided for men). Results Table A7.1: Ñopo wage gaps decomposition 2019-2020 2019 (i) (ii) (iii) (iv) (v) (vi) (vii) (viii) (ix) + Presence + Another + + + + + of children earner in + Job Age Employme Economic Educatio Zone Formality in the Tenure nt type Sector n household household Δ 20.3 20.3 20.3 20.3 20.3 20.3 20.3 20.3 20.3 Δ0 20.7 21.6 21.7 21.9 22.0 15.1 13.1 20.2 19.6 ΔM 0.0 0.0 0.0 0.0 0.0 0.0 1.4 -0.2 0.6 ΔF 0.0 0.0 0.0 0.0 0.0 0.6 1.0 1.1 0.5 ΔX -0.4 -1.3 -1.4 -1.6 -1.6 4.6 4.9 -0.8 -0.4 Male in CS (%) 100 100 100 100 100.0 98.8 71.5 60.6 51.2 Female in CS (%) 100 100 100 100 100.0 99.3 97.1 88.6 77.7 Table A7.2: Common support for decompositions in Figure 31 (1) (2) (3) (4) Male in C. support (%) 98.8 71.5 60.6 51.2 Female in C. support (%) 99.3 97.1 88.6 77.7 111 112 ANNEX 8. Impact of COVID-19 on employment Table A8.1: Characteristics of workers pre-COVID (2019) in the longitudinal sample, by demographic groups Workers' characteristics (1) (2) (3) Broadband WFH Essential sector access All 13.6 19.0 29.7 Men 10.5 21.2 28.5 Women 18.1 15.9 30.9 15-24 11.6 17.8 33.1 25-34 13.5 17.7 29.6 35-44 14.9 18.1 31.1 45-64 13.6 21.2 32.3 65 and more 13.2 20.0 25.1 Low skilled 4.5 18.6 24.2 High skilled 39.9 20.4 49.7 Informal 21.3 17.6 31.9 Formal 4.1 20.7 21.1 Notes: Based on data from ENCFT 2019-2020 (second quarter). WFH is defined for workers in the top 25 percent of the distribution; broadband access considers households with access to internet and either cable TV or a fixed phone connection; essential sectors are defined following Decreto 134-20 and Decreto 153-20. We define educational attainment using 11 years of education as threshold (i.e. complete secondary education in DR) and formality using the national definition. Table A8.2: Change in employment, monthly labor income and weekly hours worked for the longitudinal and full samples, 2019-2020 (1) (2) Longitudinal Sample Full Sample 2019 Employment 1147060 4669826 2020 Employment 1024955 4255874 Δ Employment -10.6 -8.9 2019 Hours worked 39 40 2020 Hours worked 34 36 Δ Hours worked -12.4 -10.8 2019 Labor income 18381 20,021 2020 Labor income 15220 16,402 Δ Labor income -17.2 -18.1 Source: Author calculations based on data from ECNFT 2019-2020 (second quarter). Note: Labor income is in monthly pesos. Box A8.1 Specification of the model employment, hours and wage loss To estimate the determinants of job, and income losses, we follow: ∆𝑌𝑡,𝑖 = 𝛼 + 𝛽𝑋𝑖,𝑡−1 + í µí¼—í µí±?𝑖,𝑡−1 + 𝜀 (1) 112 113 Where, i denotes a worker and t the time, ∆Yt,i stands for the alternative indicators of labor market status change: Pr(𝐸𝑡,𝑖 = 0 | 𝐸𝑡−1,𝑖 = 1) (2) 𝑙𝑛𝐿𝑡 − 𝑙𝑛𝐿𝑡−1 (3) 𝐻𝑡 − 𝐻𝑡−1 (4) Where (2) refers to the probability of job loss in t, conditional on being employed in t-1; (3) to the change in the logarithm of monthly earnings of workers employed in t and t-1, and (4) represents the change in weekly worked hours for workers employed in t and t-1. 𝑋𝑖,𝑡−1 is a vector that includes characteristics such as being able to work from home, working in an essential sector, having broadband access at home, and working in an informal job in t-1. Zi,t−1 is a vector that includes individual characteristics such as age group, area, educational attainment, and gender. Table A8.3: Employment loss, hours worked, and labor income Notes: Based on data from ECNFT 2019-2020 (second quarter). Estimation using OLS regressions. Individual characteristics such as month of survey are omitted from this table. WFH is defined for workers in the top 25 percent of the distribution and considers individuals with broadband access (i.e. households with access to internet and either cable TV or a fixed phone connection). Essential sectors are defined following Decreto 134-20 and Decreto 153-20. We define educational attainment using a threshold of 11 years of education. Standard errors in parentheses. * p<0.05 ** p<0.01 *** p<0.001 113 114 Table A8.4: gender gaps in employment Loss Employment loss (1) (2) (3) (4) (5) (6) National Women 0.0898*** 0.116*** 0.111*** 0.109*** 0.113*** 0.110*** (0.0160) (0.0159) (0.0159) (0.0159) (0.0160) (0.0160) Observations 2347 2347 2347 2347 2347 2347 Controls Education, age and month X X X X X WFH with broadband access and essential sectors X X X X Rural X X X Informality X X Children X Standard errors in parentheses * p<0.05 ** p<0.01 *** p<0.001 114 Table A8.5: Determinants of employment loss by gender Standard errors in parentheses * p<0.05 ** p<0.01 *** p<0.001