Central America Human Capital Review Promoting more and better investments in human capital ii Central America Human Capital Review  | Promoting more and better investments in human capital Acknowledgments Central America Human Capital Review  | Promoting more and better investments in human capital ACKNOWLEDGMENTS This report was prepared by a World Bank team led by Monica Yanez-Pagans and Aylin Isik-Dikmelik that included Emmanuel Vazquez, Ana Sofia Martinez, Juan Bedoya, Polly Jones, Juliana Chen Peraza, Gonzalo Zunino, Wendy de Leon, and Paola Polanco. The report benefited from substantial contributions from Igor Kheyfets and Andres Ham, and from comments and feedback from peer reviewers Jamele Rigolini, Margaret Grosh, and Dhushyanth Raju. The team also benefited from consultations with the human development teams in Guatemala, El Salvador, Costa Rica, and the Dominican Republic, including Gianluca Cafagna, Miriam Montenegro, Julieta Trias, Federica Secci, Gaston Blanco, and Suhas Parandekar. The team would like to thank former representatives of the Ministries of Education from Guatemala, Dominican Republic, and Panama for providing access to the administrative data used to conduct the analysis on school dropouts included in Chapter 2. In addition, a team from the Instituto Centroamericano de Estudios Fiscales (ICEFI) led the compilation of the social spending database for Guatemala and Costa Rica used in this report to document public investments in the provision of social services. This report also greatly benefited from the validation of findings and comments received from Luisa Muller and Ancell Shecker for Guatemala and the Dominican Republic, respectively. Also, special thanks to Aya Alphs for editing the report, to Magdalena Lizardo and Lisselotte Galvez for compiling some of the labor market inputs from the Dominican Republic to Angelica Herrera for the excellent administrative support, to Fernando Paredes for supporting the dialogue with ICEFI, and to Alejandro Espinosa for designing the cover page and providing the graphic design for the report. This report was prepared under the guidance of Michel Kerf (Country Director for Central America), Luis Benveniste (Regional Director for Human Development, Latin America and the Caribbean Region), Emanuela Di Gropello (Practice Manager, Education Global Practice, Latin America and the Caribbean Region), Pablo Gottret (Practice Manager, Social Protection and Jobs Global Practice, Latin America and Caribbean Region), Rita Almeida (Practice Manager, Education Global Practice, Europe and Central Asia Region), and Marina Bassi (Human Development Program Leader for Central America). iii Contents Central America Human Capital Review  | Promoting more and better investments in human capital CONTENTS Acknowledgments iii List of Acronyms x Executive Summary 1 Introduction 10 CHAPTE R 1 Human Capital Accumulation and Utilization in UMI Countries in Central America 12 Evolution and current state of human capital in Central America 13 Outcomes by life cycle stages 20 Public social spending on human capital and social protection programs 26 Impact of the COVID-19 pandemic on human capital 29 CHAPTE R 2 Improving Human Capital Accumulation through Better Education Outcomes 32 Low and unequal access in preprimary and secondary education 33 High school dropout rates among teenagers 35 A learning crisis 39 Constraints to better education outcomes 44 Impacts of the COVID-19 pandemic on education outcomes and governments’ mitigation strategies 46 CHAPTE R 3 Deploying and Enhancing Human Capital for Better Employment Outcomes 55 NEETs and the trends and challenges in the school to work transition 56 Constraints to employability and utilization of human capital for better jobs 62 Impacts of the COVID-19 pandemic on labor market outcomes 70 iv Contents Central America Human Capital Review  | Promoting more and better investments in human capital CHAPTE R 4 Policies to Promote More and Better Investments in Human Capital 77 Improving the adaptiveness of education systems and labor markets to protect human capital and increase resilience and preparedness to more frequent shocks 79 Improving the efficiency and effectiveness of social public spending to accelerate/ promote human capital accumulation 80 Strengthening institutions for the delivery of better and more equal social services to foster human capital accumulation for all 82 Annex: Additional Information on Chapter 3 89 BOX E S Box 1. Human Capital Index (HCI) 17 Box 2. Learning Poverty 42 Box 3. Implementation of Distance Learning during School Closures in the Dominican Republic 49 Box 4. The Long-Term Vision of the Role of Education Technologies in Accelerating Learning and Reducing Learning Gaps in Panama 51 Box 5. Summary of the Key Learning Constraints and Outcomes for School-Aged Children 53 Box 6. Social Protection and Labor Market Policy Responses to the COVID-19 Pandemic 74 Box 7. Key Labor Market Constraints and Outcomes for Youth and Active Population 76 FIGUR E S Figure 1: Human Development across the Life Cycle 11 Figure 2: Gross Domestic Product Growth (Annual %), 2010 2021 14 Figure 3: Poverty Rate (% Population Earning $6.85 a Day or Less in 2017 PPP), 2010–2022 14 Figure 4: HCI Score by GDP per Capita (2017 PPP), circa 2020 14 Figure 5: HCI of UMIs in Central America, circa 2020 14 Figure 6: Subcomponents of the HCI in UMI Countries in Central America, 2020 15 Figure 7: Basic and Full Utilization-Adjusted HCI in Central America, circa 2020 15 Figure 8: HCI in UMI countries in Central America, 2010 and 2020 16 Figure 9: Contribution of Each Component of the HCI to the Observed Change between 2010 and 2020 16 v Contents Central America Human Capital Review  | Promoting more and better investments in human capital Figure 10: Disaggregated HCI by Income Quintiles, 2020 16 Figure 11: Pre-COVID-19 Subnational HCIs 18 Figure 12: Subnational Human Capital Index and Poverty Rates in UMI Countries in Central America, 2020 19 Figure 13: Life Cycle Stages Covered in This Human Capital Review 20 Figure 14: Net Enrollment in Early Childhood Development (ECD) Programs (%) by GDP per Capita, circa 2019 21 Figure 15: Prevalence of Stunting (% of Children under Five) by GDP per Capita, circa 2019 21 Figure 17: People Using Basic Drinking Water Services (% of population) by GDP per Capita, circa 2019 22 Figure 18: People Using Basic Sanitation Services (% of population) by GDP per Capita, circa 2019 22 Figure 16: Mortality Rate of Children Under-Five (per 1,000 live births) by GDP per Capita, circa 2019 22 Figure 19: Preprimary Net Attendance Rate for Children 3 to 5 Years Old, circa 2020 23 Figure 20: Secondary School Net Enrollment Rate by GDP per Capita (%), circa 2019 23 Figure 21: Adolescent Fertility Rate (Births per 1,000 Women Aged 15–19), by GDP per Capita, circa 2019 24 Figure 22: Contraceptive Prevalence, Any Method (% of Married Women Aged 15–49) by GDP per Capita, circa 2019 24 Figure 23: Gross Enrollment in Tertiary Education (%) by GDP per Capita, circa 2019 24 Figure 24: Share of Youth Not in Education, Employment, or Training (% of Youth Population) by GDP per Capita, circa 2019 24 Figure 25: Adult Mortality Rate, Male (per 1,000 Male Adults) by GDP per Capita, circa 2019 25 Figure 26: Hospital Beds (per 1,000 People) by GDP per Capita, circa 2019 25 Figure 27: Prevalence of Overweight Adults in Central America (%), circa 2019 25 Figure 29: Social Expenditure of Central Government as % of GDP in Central America, circa 2019 26 Figure 30: HCI, per Capita Social Expenditures, and Efficiency Frontier in Latin America, circa 2020 26 Figure 28: Death by Non-Communicable Diseases (% of Total Deaths) by GDP per Capita, circa 2019 26 Figure 31: Performance Indicators for Social Assistance Programs in Central America by Income Quintiles, circa 2019 27 Figure 32: Intentional Homicides (per 100,000 people) and GDP per Capita, circa 2019 28 Figure 33: Emigration Ratio in Central America, circa 2020 28 Figure 34: Percentage of Households Which Reported a Reduction in Income during the COVID-19 Pandemic in Central America, 2020 and 2021 29 Figure 35: Percentage of Households Which Reported Running out of Food during the Pandemic in Central America, 2020 and 2021 29 Figure 36: School Attendance Rate in Central American UMI Countries during the Pandemic in Central America, 2020 and 2021 30 Figure 37: School Disengagement Rate in Central American UMI Countries during the Pandemic in Central America, 2020 and 2021 30 Figure 38: Social Public Expenditures as % of GDP and as % of Total Expenditure in Central America, 2015–2021 31 Figure 39: Net Enrollment Rates by Education Level in Central America (%) , 2010–2020 34 vi Contents Central America Human Capital Review  | Promoting more and better investments in human capital Figure 40: Net Attendance Rates by Income Quintiles in Central America (%), 2010–2020 36 Figure 41: Net Enrollment Rate in Secondary Education in UMI Countries in Central America (%), circa 2019 37 Figure 42: Dropout Rates for Students by Education Level in Central America (%), 2019 37 Figure 43: Out-of-School Teenagers by Gender and Country in Central America (%), 2010–2020 38 Figure 44: Reported Reasons for Dropping out of School among 13- to 18-Year-Olds in UMI Countries in Central America (%), circa 2010 and 2020 38 Figure 45: Schooling Attainment for Teenage and Adult Mothers in Latin America, circa 2019 39 Figure 46: Repetition Rates in Primary School (% of Total Enrollment), circa 2019 39 Figure 47: Percentage of Third Grade Students below the Minimum Proficiency Level in Reading in Central America (%), 2019 40 Figure 48: Learning Gains in Reading and Math for Third Grade Students in Latin America, 2013 and 2019 40 Figure 49: Learning gap in Grade 6 ERCE Score Differences between the Lowest and Highest Income Quintile in Central America, 2019 40 Figure 50: Learning gap in Grade 6 ERCE Results for Indigenous and Non-Indigenous Students in Central America, 2019 40 Figure 51: Learning Poverty in UMI Countries in Central America, circa 2018 41 Figure 52: Learning Poverty in the Dominican Republic by Province, 2017 44 Figure 53: Learning Poverty in Panama by Province, 2018 44 Figure 54: Learning Poverty in Guatemala by Department, 2018 44 Figure 55: Dropout Rates for Students in Public Schools, Grades 1 to 12, by Learning Quintiles, Panama (%), 2019 45 Figure 56: Dropout Rates for Students in Grades 6 and 9 by Learning Quintiles, Guatemala (%), 2019 45 Figure 57: Associated Factors Affecting Grade 3 ERCE Results in Reading in Latin America, 2019 45 Figure 58: Variance in Grade 3 ERCE Results in Math Explained by Differences within Classrooms in Latin America, 2019 45 Figure 59: School Closures in Number of Weeks for Central America, March 2020 to March 2022 46 Figure 60: Simulated Learning Losses due to COVID-19 in Central America (in Learning- Adjusted Years of Schooling), 2022 47 Figure 61: Simulated Learning Losses Due to COVID-19 School Closures: Third Graders below MPL in Reading in Central America, (%), 2019 47 Figure 62: Effect on Average Annual Income per Student in Central America, 2019 47 Figure 63: Change in Dropout Rates by Education Level between 2019 and 2021 in Central America, (%) 48 Figure B3.1: Percentage of Students Aged 4-17 Who Watched Distance Learning Programs in Gran Santo Domingo and Santiago, 2020–2021 49 Figure 64: Enrollment by Sector and Education Level in the Dominican Republic, 2017/2018 to 2021/2022 50 Figure 65: Enrollment by Sector and Education Level in Guatemala, 2019 to 2022 50 Figure 66: Education Level of NEETs (15 to 24 years old) in Central America, 2019 57 Figure 67: Inactive vs. Unemployed NEETs (15 to 24 years old) in Central America, 2019 57 Figure 68: NEET Rate (%) in Central America, Total and by Gender, 2011 57 vii Contents Central America Human Capital Review  | Promoting more and better investments in human capital Figure 69: NEET Rate (%) in Central America, Total and by Gender, 2019 57 Figure 70: School to Work Transition by Age in Central America, 2019 58 Figure 71: School to Work Transition by Age in Central America, Female, 2019 59 Figure 72: Guatemala School to Work Transition by Age, Male, 2011 and 2019 60 Figure 73: Labor Force Participation by Age Group in Central America, 2019 60 Figure 74: Informality by Age Group in Central America, 2019 60 Figure 75: Labor Force Participation by Gender in Central America, 2019 61 Figure 76: Informality by Gender in Central America, 2019 61 Figure 77: Average Total Wage by Age Group in Central America 61 Figure 78: Average Total Wage by Gender in Central America, 2019 61 Figure 79: Labor Force Participation by Age of Youngest Child (≤ 14 Years Old) in Household and by Gender for Central America, 2019 61 Figure 80: Per Capita Opportunity Cost of Labor Income for Teenage Mothers in Latin America, 2021 62 Figure 81: Labor Income According to Age at Which Women Became Mothers (USD per year) in Latin America, 2021 62 Figure 82: Informality Rate in Central America, 2019 63 Figure 83: Employment Statistics of Bottom 40 Percent of the income distribution in Central America, 2019 63 Figure 84: Education Level of Working-Age Population in Central America, 2019 64 Figure 85: Employment Rate by Education Level in Central America, 2019 64 Figure 86: Pre-COVID-19 Full Utilization-Adjusted Human Capital Index by Province in Panama, 2019 65 Figure 87: Pre-COVID-19 Full Utilization-Adjusted Human Capital Index by Province in the Dominican Republic, 2019 65 Figure 88: Educational Attainment across Generational Cohorts (Ages 25–64) in Central America, 2019 66 Figure 89: Educational Attainment for Women across Cohorts (Ages 25–64) in Central America, 2019 67 Figure 90: Skill Wage Premium by Gender in Central America (%), circa 2010-2020 68 Figure 91: Changes in the Task/Skill Content of Jobs, 2011–2019 69 Figure 92: Task/Skill Content of Jobs with Respect to Average Worker by Age Group in Central America, 2019 69 Figure 93: Task/Skill Content of Jobs with Respect to Average Worker by Age Group in Central America, 2019 70 Figure 94: Change in NEET Rate during the Pandemic by Gender in Central America, 2020 and 2021 71 Figure 95: Difference in Labor Market Indicators by Education and Age Group in Costa Rica, 2020/2021 71 Figure 96: Composition of NEETs by Gender in Costa Rica, 2020–2021 71 Figure 97: Share of Employees Reporting Income Reduction in Their Job during the Pandemic by Sector in Costa Rica, 2019–2020 72 Figure 98: Difference in Labor Market Indicators in Panama by Education and Age Group, 2019–2020 72 Figure 100: Job Loss in the Dominican Republic by Sector (Quarter on Quarter Difference), 2019–2020 73 viii Contents Central America Human Capital Review  | Promoting more and better investments in human capital Figure 99: Share of Job Loss in the Dominican Republic by Educational Level and Gender, 2019–2020 73 Figure A.1: School to Work Transition by Age, circa 2011 90 Figure A.2: School to Work Transition for Females by Age, 2011 91 Figure A.3: Employment Statistics of Top 60 Percent of Income, 2019 91 Figure A.3: Composition of Employment by Level of Education, by Economic Sector 92 Figure A.4 Task Profile of the Employed, Guatemala, 2019 93 Figure A.5: Task Profile of the Employed in Dominican Republic by Gender and Educational Level, 2019 93 Figure A.6: Task/Skill Content of Jobs by Sector, Costa Rica, 2019 94 TABL E S Table 1: Programs Implemented during the Pandemic that Might Have Contributed to Student Retention 52 Table 2: Policy Areas for Improving the Adaptiveness of Education Systems and Labor Markets to Protect Human Capital, and Increase Resilience to More Frequent Shocks 83 Table 3: Policy Areas for Improving the Efficacy of Social Public Spending to Accelerate/ Promote Human Capital Accumulation 84 Table 4: Policy Areas for Strengthening Institutions in the Delivery of Better and More Equal Social Services to Foster Human Capital Accumulation 86 Table A.1: Determinants of NEET Status, 2019 89 ix List of Acronyms Central America Human Capital Review  | Promoting more and better investments in human capital LIST OF ACRONYMS AI Artificial Intelligence CA Central America CCT Conditional Cash Transfer ECD Early Childhood Development ERCE Estudio Regional Comparativo y Explicativo (Regional Comparative and Explanatory Study) GDP Gross Domestic Product GVC Global Value Chain HCI Human Capital Index LAC Latin America and the Caribbean LAYS Learning-Adjusted Years of Schooling MPL Minimum Proficiency Level NCD Non-Communicable Disease NEET Not in Employment, Education, or Training OECD Organization for Economic Co-operation and Development PPP Purchasing Power Parity SDG Sustainable Development Goals SES Socioeconomic Status TVET Technical and Vocational Education and Training UMI UMI UNESCO United Nations Educational, Scientific and Cultural Organization USD United States Dollar x EXECUTIVE SUMMARY Central America Human Capital Review  | Promoting more and better investments in human capital 1 Executive Summary Central America Human Capital Review  | Promoting more and better investments in human capital H uman capital consists of the knowledge, skills, and health that individuals acquire throughout their lives, and is a powerful force for inclusive economic growth and poverty reduction. Investing in nutrition, quality healthcare, education, and job and skill development is crucial for building human capital. This, in turn, is essential for sustainable growth, poverty reduction, and shared prosperity in countries. However, policy makers may struggle to advocate for human capital investments as the associated benefits can take a long time to materialize. The economic returns of investments in young children’s human capital, for example, are not evident until they enter the workforce. To maximize their fiscal resources and develop their population’s economic potential, governments need to prioritize expenditures that deliver measurable improvements in human development outcomes and contribute to the future of their communities and nations. This Human Capital Review aims to provide analytical foundations in the support of policies that improve human capital outcomes for the following four UMI countries in Central America: Costa Rica, Guatemala, Panama, and the Dominican Republic. The objective of this report is to identify the key constraints to human capital growth and under- stand how education and labor market policies can foster a resilient recovery, promote inclusive growth, and contribute to poverty reduction in these countries. The review also estimates the impact of the COVID-19 pandemic on human capital outcomes using a multi-sectoral approach. The analysis compares human capital outcomes in the decade before the COVID-19 pandemic (2010–2019) against trends during the pandemic (2020–2021). Lastly, the report focuses on these four countries, which are the only UMI in Central America to take advantage of new data collected during the pandemic, which allowed to quantify some of the impacts of COVID-19 and understand some of their long-term implications for human development outcomes. Even though the development of human capital occurs at all ages and involves multiple sectors, this report focuses on selected education and labor market outcomes for school-age children (6-14 years old) and youth (15-24 years old). Human capital accumulation is a dynamic process in which investments build upon each other and accumulate unevenly across different stages of the life-cycle. The most rapid growth occurs during early childhood (under 6 years old), school-age (6–14 years old), and youth (15–24 years old). This report emphasizes school-age children and youth, two life-cycle stages that have significant impacts on productivity later in life. It specifically examines learning outcomes, school dropouts, and youth employment outcomes as these have all been identified as major contributors to the low human capital levels in the four countries selected for this review. The focus on school-age children and youth should not overshadow the significance of early childhood development, which plays a crucial role in preparing children for success in school and future productivity as adults. Early childhood development programs additionally play a vital role in compensating for inequalities in human capital accumulation, which begin very early in life. The report consists of four chapters. Chapter 1 provides a regional context for the human capital outcomes in the four selected countries in Central America. Chapter 2 examines the structural challenges to human capital accumulation during school years and the pandemic’s effects on both student learning outcomes and school dropout rates. Chapter 3 explores structural challenges in labor markets, focusing on the transition to employment for young adults, worker productivity, and the alignment of workers’ skills with labor demand. This chapter also estimates the pandemic’s impacts on youth employment outcomes, including employment rates, labor income, and the share of youth not in education, employment, or training (NEET). Based on the findings from the diagnostic analysis in the previous chapters, Chapter 4 presents recommendations to accelerate human capital accumulation and promote a sustainable and inclusive recovery. 2 Executive Summary Central America Human Capital Review  | Promoting more and better investments in human capital Human Capital Trends in Central America Over the past decade, all four UMI economies in Central America analyzed in this review experienced growth with differing poverty reduction trends. Economic growth in the four countries surpassed the 2 percent Latin American average during this period, with slightly higher growth rates in Panama and the Dominican Republic. There was a rapid rebound in 2021 after pandemic-induced setbacks, and growth in these countries seems to be resuming long-term trends. In the past decade, the poverty rate decreased in the Dominican Republic from 44 percent to 20 percent and in Panama from 22 percent to 12 percent.1 However, poverty rates in Costa Rica and Guatemala remained stagnant during this period at around 14 percent and 55 percent, respectively. With over half the population in poverty, Guatemala has the third-highest poverty rate in the Latin American region. Except for Costa Rica, all countries face significant shortfalls in human capital levels, mainly due to poor education outcomes. The Human Capital Index (HCI) estimates the amount of human capital a child born today can acquire by the age of 18 considering the risks of poor health and education in their country. Apart from Costa Rica, all countries have lower HCI levels than expected given their country’s respective income levels. On average, a child born in the Dominican Republic, Panama, and Guatemala can expect to be only half as productive as he or she could be due to the low quality of health and education services in these countries. In contrast, a child born in Costa Rica can expect to be 63 percent as productive as he or she could be with the current education and health services available. Poor education outcomes, which are reflected in the unequal and limited access to schools and low learning levels, are common factors reducing human capital levels in all countries. In Guatemala, health outcomes including high child stunting rates and relatively low adult survival rates present an additional constraint to human capital accumulation. Over the last decade, human capital accumulation has stagnated or even slightly deteriorated in the four countries in contrast with the global trend of improvement in human capital accumulation. In all four countries, the primary factor hindering HCI growth is a lack of progress in education outcomes, including expected years of schooling (Panama and Guatemala) and harmonized test scores (Dominican Republic and Costa Rica). Health improvements, on the other hand, positively contributed to the HCI in Panama, Guatemala, and Costa Rica. Moreover, disparities in human capital between populations of different socioeconomic backgrounds and different subnational levels within each country are very large. The HCI score for Guatemala’s highest income quintile, for exam- ple, is almost 20 percentage points higher than those in the poorest quintile. This gap can be attributed to unequal access to health and education services, which are poor in quality. At the subnational level, a strong negative correlation exists between human capital and poverty levels, meaning that areas with higher poverty levels experience lower human capital. For example, a child born today in the province of Panama will be 50 percent as productive as he or she could be, which is 13 percentage points higher than a child born in one of the indigenous provinces in Panama – a level comparable to that in low-income countries experiencing high levels of fragility and conflict. Violence and emigration pose additional challenges to human capital accumulation in some countries. Guatemala ranks as the eleventh most dangerous country in the world. In addition to hindering productive economic activity and growth, violence discourages individuals from investing time and money in human capital accumulation. In Central Amer- ica, emigration is a phenomenon related to violence as well as low living standards, lack of opportunities, extreme weather events, and food insecurity. As of 2020, a significant portion of the population in the Dominican Republic and Guatemala had emigrates (15 percent and 8 percent, respectively). Although emigration rates in Panama and Costa Rica are smaller, they remain larger than those in countries with similar income levels. Human capital accumulates over the life cycle. Early years investments having strong long-term implications for individ- uals’ productivity later in life. The early childhood, school-age, and youth life stages are when human capital accumulation is most rapid, followed by the prime working age and old age, where human capital accumulation tends to stagnate or decline. Therefore, ensuring that children are healthy and well-nourished from an early age is crucial for their survival and development to their full potential. 1 This refers to the poverty rates estimated using the international poverty line of $6.85 a day (measured in 2017 PPP). 3 Executive Summary Central America Human Capital Review  | Promoting more and better investments in human capital The challenges to ensuring good human capital outcomes during early years vary across countries, but a common issue is the limited access to high-quality early childhood development programs. Growth and development in early years lay the foundation for future learning, productivity, and lifelong success. Early childhood development programs also play a critical role in compensating for human capital accumulation inequalities. Access to such programs is quite low in all four countries examined in this review when compared to countries with similar income levels. Despite the high payoffs of investing in quality early interventions, these four countries do not invest enough in young children due to a combination of budget constraints and challenges in delivering comprehensive health, nutrition, and early learning services. A significant challenge among children under the age of five is the high prevalence of stunting, particularly in Guate- mala. Ensuring children are healthy and well-nourished during early years is crucial for their survival and for developing their full potential. The most critical time for good nutrition is from pregnancy until a child’s second birthday. Improving children’s food quality and feeding practices during these years is key to preventing malnutrition and fostering human capital accumulation. In Guatemala, 44 percent of children under five are stunted, which is the sixth highest rate worldwide. This statistic suggests that caregivers in Guatemala face considerable constraints in securing nutritious, safe, affordable, and age-appropriate food for their children. Another challenge among children under the age of five is the high mortality rate, and access to clean drinking water and basic sanitation services. In the Dominican Republic, children under the age of five experience relatively high mortality rates, which are twice the average for Latin American countries. Access to water is also fundamental to children’s nutrition, however Panama struggles to provide access to clean drinking water to its population. Additionally, all countries except Costa Rica have poor access to basic sanitation services despite all being UMI countries. While primary education access is universal in Costa Rica, the Dominican Republic, Panama, and Guatemala, enrolling children in preprimary education remains a challenge. In 2020, a mere 60 percent of three- to six-year-olds in Costa Rica, the Dominican Republic, and Panama had access to preprimary education. The situation is even more dire in Guatemala, where only 10 percent of children in the same age group had access to preprimary schooling. Limited access to quality early childhood development programs coupled with inadequate preprimary school opportunities lead to children entering primary school unprepared to learn effectively. High secondary school dropout rates pose a significant constraint to human capital accumulation among the youth, especially in Guatemala and Panama. In 2019, Guatemala and Panama had net secondary enrollment rates of 44 and 64 percent, respectively, which are much lower than the average for countries with similar income levels (80 percent). Guatemala’s secondary enrollment rate even ranks the lowest in Latin America. Several factors contribute to the low enroll- ment and high dropout rates in secondary, including teenage pregnancy. In all four countries – Guatemala, Panama, the Dominican Republic, and Costa Rica – teenage pregnancy rates are higher than those in countries with similar income levels and are a main contributing factor to high dropout rates. The Dominican Republic ranks first in Latin America for teenage pregnancy. This issue often results in young girls discontinuing their education, which, in turn, negatively impacts their future economic opportunities and overall human capital accumulation. Apart from teenage pregnancy, other factors that contribute to high dropout rates include financial constraints and lack of interest in studying, which is probably a reflection of the low learning and skills acquired during their time at school. In Panama and Guatemala, the lack of enrollment in higher education hampers the development of critical skills among young people, creating challenges as they transition into the workforce. Tertiary education enrollment rates (defined by the ratio of higher education enrollment to the population aged 18–24) are lower than expected in the coun- tries examined in this review except for Costa Rica. Furthermore, the high proportion of young adults not enrolled in tertiary education have not necessarily been able to transition into the labor market. The percentage of youth who are not in education, employment, or training (NEET), is notably high in the Dominican Republic and Guatemala. Health outcomes in adulthood present an additional challenge to human capital accumulation in all four countries. Depending on the specific country, adults face different issues including limited access to high quality health services, high overweight rates, and high rates of non-communicable diseases (NCDs). In terms of health services, the ratio of hospital beds per 1,000 people in the four countries is lower than expected for countries with similar levels of income. Guatemala has the lowest ratio in Latin America at 0.4 beds per 1,000 people while Costa Rica’s ratio is 1.1, which is less than half of the regional average. Additionally, in Costa Rica and the Dominican Republic, approximately 60 percent of adults are 4 Executive Summary Central America Human Capital Review  | Promoting more and better investments in human capital overweight, posing a significant risk to workforce productivity. Finally, the high incidence of NCDs in Costa Rica further affects productivity and life expectancy, as well as healthy aging among the older population. Inadequate human development outcomes across different life stages reflect low and inefficient levels of public social spending. Except for Costa Rica, the countries examined in this review make insufficient and inefficient investments in the provision of social services. International evidence shows that targeted poverty cash transfer programs, especially conditional cash transfers (CCTs) have the potential to boost human capital and reduce disparities between different populations. The coverage of CCTs for the poorest quintile in these four countries is, however, below the Latin American average. The COVID-19 pandemic significantly hindered human capital accumulation across all stages of the life cycle. Despite efforts to expand and adapt social protection programs to alleviate the pandemic’s impacts, economic contraction led to increased poverty rates. Some of the containment measures in the four countries were among Latin America’s most severe, leading to job losses, reduced income, and food insecurity. These factors put a strain on the ability of households to preserve of their human capital levels. Panama suffered one of the worst economic downturns in Latin America because of the pandemic, while CCTs somewhat buffered Guatemala’s economic downturn. Early childhood development services were particularly affected by the pandemic, with a significant decrease in the provision of essential services including nutrition and healthcare. This reduction was driven by factors such as social distancing measures, limited personnel, and resource reallocation towards pandemic control. Because essential cognitive and physical development occurs in early childhood, the inability to provide these services had far-reaching implications. In terms of education, despite the concerted efforts of governments to maintain continuity in educational services during the pandemic, the transition to online learning posed a unique set of challenges. In addition to connectivity issues, not all households had access to the necessary technological devices for online learning, which further exacerbated learning gaps. Moreover, families were often ill-equipped to support home-based learning due to a lack of resources or parents’ unfamiliarity with the curriculum. Lastly, the pandemic disproportionately affected young people as there were significant job losses for this population group. Many businesses and industries that typically employ young people such as hospitality and retail were severely affected by lockdown measures. This, in turn, lead to widespread layoffs and a significant increase in the share of NEETs. The disproportionate increase in NEET rates among young people is not only indicative of the hardship they faced during the pandemic, but also suggests potential long-term consequences. Prolonged periods of unemployment or disconnection from education can hinder young people’s career prospects, wage growth, and overall life trajectory. Foundational Learning as a Building Block for Human Capital Accumulation While significant improvements have been made in increasing access to education, there are still gaps in access at the preprimary and secondary levels. Overall preprimary and secondary education enrollment rates are significantly lower than expected for UMI countries. Guatemala’s lower and upper secondary enrollment rates are for example, 65 percent and 37 percent, respectively. Furthermore, the trajectory of enrollment rates does not seem to be improving apart from in Costa Rica where there has been some progress in increasing preprimary and secondary school enrollment rates since 2010. Lastly, school enrollment rates in preprimary and secondary school are highly correlated with socioeconomic status, and students from poor and vulnerable households lag behind their more affluent peers in all four countries. High dropout rates, especially in secondary education, pose a significant hurdle to fostering human capital accu- mulation among youth. Evidence suggests that teenagers who leave school prematurely will likely earn less and face more socioeconomic challenges than their peers who complete more years of schooling. Since 2019, the secondary net enrollment rates of Guatemala and Panama have been far below those of countries with similar income levels. Alarmingly, Guatemala has the lowest secondary net enrollment rate in Latin America. Low enrollment rates in secondary education are closely related to high dropout rates. The two main reasons for dropping out of secondary school in these four countries are lack of interest in studying and financial constraints. Teenage pregnancy also plays a role in school dropout for girls, leading to lower average years of education compared to women who become mothers later in life. When measured by births per 1,000 women aged 15–19, the teenage pregnancy rate of the Dominican Republic ranks first in Latin America (92 births) followed by Panama (80 births). In comparison, the averages in Latin America and in UMI countries are 50 and 41 births, respectively. 5 Executive Summary Central America Human Capital Review  | Promoting more and better investments in human capital Learning outcomes in the four countries are worryingly low, undermining human capital development efforts. The latest regional student learning assessments (Estudio Regional Comparativo y Explicativo, ERCE) conducted in 2019, show that learning outcomes are remarkably low in all the countries except for Costa Rica. This is concerning because it shows students are not acquiring the foundational skills needed to advance their educational trajectories. Despite performing above the Latin American average, Costa Rican students still lag behind the average for OECD countries. In Panama and Guatemala, high dropout rates are common among students with low learning outcomes, highlighting the fact that students who struggle to succeed in school are more likely to drop out. Furthermore, between 2013 and 2019, learning levels have largely remained stagnant in the four countries. Learning poverty, which is a metric that combines learning levels and the out-of-school population at age 10, clearly shows that the education systems in these four countries are failing to ensure that all children enroll in school and develop foundational skills. Except for Costa Rica, all countries have high learning poverty rates, exceeding the averages for both UMI countries and Latin America. Constraints to better education outcomes Several factors play a significant role in enhancing learning outcomes for students. Factors that contribute to increased learning include student characteristics such as: prior enrollment in preschool, longer hours of study outside of the class- room, high attendance rates, and lower grade repetition rates. Additionally, parents’ involvement and high expectations of their children’s learning potential also play an important in helping students achieve better learning outcomes. Lastly, teachers who prepare their lessons in advance, provide student support, and have high expectations for their pupils’ learning potential also contribute significantly to improved learning outcomes. Limited access to quality early childhood programs, poor training of teachers, insufficient educational invest- ments, and poor management, hinder improved learning outcomes. Children who miss preschool often begin school ill-prepared to learn. Moreover, factors such as malnutrition, illness, low parental investment, and the harsh realities of poverty all negatively impact early childhood learning. Even schools with resources struggle to level the playing field for disadvantaged children. The educational systems in these four countries also struggle to attract skilled individuals to teaching roles and adequately support their professional development. Both pre-service and in-service teacher training programs often fail to provide teachers with adequate subject knowledge and pedagogical skills. Investments in attract- ing, training, and retaining quality teachers are vital to improving learning outcomes. Except for Costa Rica, public investment in education that could address some of these challenges remains below international benchmarks in the countries examined in this review. The most substantial learning achievement gaps exist within classrooms, underlining the critical need for targeted instruction. It is common for classrooms in these countries to have students with vastly different learning levels. Some students, for example, may be reading full texts and solving intricate math problems while others are just starting to recog- nize letters or grasp basic numerical concepts. This predicament underscores the importance of using structured teaching methods and targeted instruction to accelerate learning and reduce inequalities. The school closures resulting from the COVID-19 pandemic in the four countries were among the longest in the world. Projections presented in this report suggest that learning losses from these school closures are potentially substantial. The economic impact of learning losses can be quantified in terms of students’ lifetime earnings losses considering schooling, life expectancy, and labor market factors. Under an optimistic scenario, the findings of this review suggest that the average student in Costa Rica could lose an equivalent of 16 percent of projected lifetime earnings, while students in other countries face losses ranging from 4 percent in the Dominican Republic to 7 to 8 percent in Panama and Guatemala. If these learning losses are not addressed, they will likely lead to higher dropout rates in the future. Somewhat surprisingly, school dropout rates do not seem to have dramatically increased during the pandemic. However, once the 2022 school enrolment data becomes available, the long-term impacts of the pandemic on school dropout rates will be clearer. To support the learning recovery agenda, more effective and efficient public investments in education need to be made. Public investments in education should prioritize: (i) consolidating the curriculum to focus on foundational and transferable skills; (ii) assessing students’ learning losses; (iii) accelerating learning recovery and reducing educational gaps between different populations; and (iv) strengthening management information systems by creating early warning systems 6 Executive Summary Central America Human Capital Review  | Promoting more and better investments in human capital that monitor school dropouts. More efforts should also be placed on the use of formative learning assessments to improve pedagogical practices and to increase students’ learning at relatively low cost. Lastly, investing in developing and implementing regulatory frameworks to integrate education technologies that support learning and reduce inequalities in the education sector will also be critical. Deploying and Enhancing Human Capital for Better Employment Outcomes Despite progress in Costa Rica and Panama during the past decade, the share of NEETs remains high in all four countries. When youth are out of school and out of work, they stop accumulating and using their human capital. In 2019, both the Dominican Republic and Guatemala had NEET rates (24 percent and 28 percent, respectively) surpassing the Latin American average of 21 percent. Conversely, the share of NEETs in Costa Rica and Panama (16 and 17 percent, respectively) falls below the regional average. While each country’s NEET demographic differs slightly, a common thread is that women make up most NEETs. This reflects the additional hurdles they face, including caregiving duties and teenage pregnancy, which often result in school dropouts. NEET trends reveal both challenges and progress in the school to work transition. Costa Rica and Panama have managed to decrease their shares of NEETs over the last decade because adolescents are both staying in school longer and finding employment. Conversely, in Guatemala and the Dominican Republic, there were observed gains in education, but they were outweighed by lower labor market participation among the youth. In Costa Rica, Panama, and the Dominican Republic, women have made progress in education and employment outcomes. Costa Rica, in particular, has the highest share of 20-year-old women pursuing education (60 percent), which represents an increase from 2010. In Guatemala, however, education gains among young women do not appear to have translated into better labor market outcomes. In all four countries, young and female workers have poorer labor market outcomes. Except for Guatemala, youth labor force participation is about half that of adults. Even when employed, young workers earn less and are more likely to be in informal employment. Similarly, women are less likely to participate in the labor market compared to men, with particularly large disparities in Panama (35 percentage points) and Guatemala (45 percentage points). This is in part explained by the high incidence of teenage pregnancy that often leads to school dropout and fewer high-paid employment opportunities. Young mothers are more likely to take on lower-paying or informal jobs, resulting in lower income levels. Constraints to employability and utilization of human capital for better jobs Labor markets in the four countries face structural challenges to varying degrees. These center around three primary issues: the scarcity of quality jobs (demand side), the multiple obstacles faced by the workforce (supply side), and mismatches between the available jobs and workers’ skills. Except for Costa Rica, labor markets are predominantly informal in the countries studied in this review, ranging from half of the labor market in the Dominican Republic and Panama, to a staggering 80 percent in Guatemala. Across these four countries, the struggle to attain fundamental learning and build human capital remains a persistent issue, particularly affecting those in the lower income bracket. Workers in the bottom 40 percent of the income distribution are more likely to be informally employed and less likely to receive a salary, except in the Dominican Republic. Inadequate investments in human capital result in a limited supply of skills and difficulty in adapting to the evolving world of work. Lack of human capital results in a workforce that is low-skilled, which influences overall productivity and quality of employment opportunities. Higher educated workers (with 14 years or more education) are in short supply, accounting for only 20 percent of the working age population in any country, and only 5 percent in Guatemala. There is a growing demand for ‘future skills’ around the world, which includes critical thinking, creativity, problem solving. These skills, however, are in short supply among the UMI countries in Central America. While other structural challenges are at play, the lack of human capital and relevant skills contribute to low productivity and limited job creation. Human capital is not being fully utilized because of challenges to both labor supply and demand, which further exacerbates the existing disparities within countries. The lack of participation in the labor market and the lack of better employment opportunities mean that existing human capital is underutilized. In terms of its utilization of human capital, 7 Executive Summary Central America Human Capital Review  | Promoting more and better investments in human capital Costa Rica slightly exceeds the average for UMI and Latin American countries because of higher employment rates, more availability of better jobs, and a lower mismatch between the labor market and workers’ skills. Panama, Guatemala, and the Dominican Republic, on the other hand, fall short of the Latin American average. The slightly lower availability of good employment opportunities relative to the participation in the labor market suggests the existence of a limited supply of good jobs in most countries. Some provinces in Panama and the Dominican Republic have even lower utilization rates of human capital than those observed in fragile and conflict-affected countries. The observed increases in educational attainment in the past decade have not translated into better skills, jobs, or wages. Women have made more substantial educational gains than men as the share of women with 14 or more years of education has more than doubled in the last decade in all of the countries in this report with the exception of Guatemala. Despite this, the overall supply of skilled labor, however, remains limited. The proportion of workers with no more than 8 years of education ranges from approximately 25 percent in Panama to 60 percent in Guatemala. Additionally, wage premiums for skills have not increased, and women have even experienced declining returns, except for highly-educated women in Costa Rica. The lack of opportunities for more educated workers, particularly young ones, leads them to seek work in other countries and results in important human capital losses. A labor force whose skills are misaligned with the jobs of the future partially accounts for the lack of good jobs and leaves some workers vulnerable to displacement. Recently, there have been signs of a shift in employment towards future jobs in both Costa Rica and the Dominican Republic. In all four countries, those with a higher level of education are more likely to secure these future-oriented jobs. However, the limited supply of relevant skills potentially hampers the generation of such jobs, intensifying demand-side structural constraints. Guatemala holds the lowest share of employ- ment in “future skills” and the highest share of “past skills” relative to the average employment across the other three countries. Youth and informal workers are at greater risk of technological displacement and low growth in earnings as they are largely concentrated in jobs of the past. The COVID-19 pandemic hit labor markets hard, although the extent to which it affected different populations varied in each country. The pandemic disproportionately impacted youth employment, causing the share of NEETs to increase and thereby reverse progress that had been made in the past. In Costa Rica, more educated youth and women, however, experienced fewer job losses while those with less education were hit hardest. In Panama, the nega- tive impacts of the pandemic extended beyond job losses, as about half of those employed in 2021 reported income losses. Compared to Costa Rica and Panama, the Dominican Republic experienced a somewhat less severe impact on its labor market, although impacts varied among sectors. Governments developed measures to mitigate the impacts of the pandemic on labor markets, but challenges to their implementation were substantial. Pre-existing challenges included different levels of institutional capacity, limited coverage of labor market programs, and a lack of focus on priority groups, constrained effectiveness of mitigation efforts. While each of the four countries possess some elements of an effective employment support system, the capacity varied widely. Costa Rica modernized its main training agency, the National Institute of Learning, in the immediate aftermath of the pandemic, to better respond to shifting labor market needs. In the Dominican Republic, labor market programs have had success in developing partnerships with the private sector to implement technical and vocational education (TVET) programs, and some training programs have been designed to target vulnerable youth. However, the training system in Dominican Republic remains fragmented, and there is a lack of programs to encourage labor market integration. Except for Costa Rica, the coverage of labor market programs is low in all countries. Given the impacts of COVID-19 and the limited capacity of labor market programs, youth, less skilled workers, and future entrants to the workforce face important constraints in improving their labor market outcomes. Evidence suggests that, in general, it takes about ten years for youth entering the labor market during a downturn to catch up (in employment and earnings) with others. Other research points specifically to the pandemic’s damaging effects (“scarring”) on employment outcomes including lower labor market participation, increased unemployment, and a higher likelihood of working informally. Those with less than a college education are impacted by these negative outcomes to a greater degree. In addition to this, future entrants to the labor market will have suffered learning losses during to the pandemic, meaning their labor market outcomes will be even further impaired. 8 Executive Summary Central America Human Capital Review  | Promoting more and better investments in human capital Policies to Promote More and Better Investments in Human Capital The report presents a range of policy recommendations to tackle the structural constraints to human capital accumu- lation among school-age children and youth. These recommendations are grouped into three thematic areas: (i) improving the adaptiveness of education systems and labor markets; (ii) improving the efficiency and effectiveness of public social spending; and (iii) strengthening institutions to deliver better and more equal services. Each recommendation encompasses short-term and medium- to long-term policy recommendations. Given the exposure to natural disasters and other shocks in Central America, it is essential to improve the adaptiveness of the education system and labor markets. In terms of education, adaptiveness entails consolidating curricula, focusing on foundational learning and prerequisites for future learning, transferable competencies, and socioemotional skills. This approach also involves expanding school connectivity, increasing access to technological devices, strengthening teacher training, and leveraging educational platforms to support distance or hybrid learning methods during shocks. Moreover, education systems also need to improve their administrative information systems to: (i) identify students who are at higher risk of dropping out during crises, (ii) enhance student learning assessments to evaluate learning losses during shocks, and to guide the development of remedial programs. For labor markets, adaptiveness means evaluating the scope of labor market programs, utilizing technology to provide employment-supporting training, and fortifying national labor market information systems. Increasing public social spending and enhancing its efficiency and efficacy is paramount to improve human capital accumulation and reduce disparities. For the education sector, increased efficiency implies fortifying national systems to monitor students’ learning trajectories and utilizing the data to enhance learning outcomes. This also involves improving instruction through structured pedagogy, offering teacher coaching and peer mentoring programs to boost instructional effectiveness, teaching to the right level, and leveraging education technologies when appropriate. Other measures include increasing the attractiveness of pre-service teacher training programs with an emphasis on mandatory induction programs during the initial years of teaching; creating teacher career advancement paths and linking them to teacher training programs; using merit-based salary increase schemes for teachers with accompanying measures of quality assurance and improved working conditions; developing strategies to strengthen school management capacity of school directors and to improve school autonomy; and finally, using equity considerations in the allocation of public education spending. For labor markets, increased efficiency implies better utilization and enhancement of human capital, which can be achieved by developing effective mechanisms for funding TVET; strengthening the connections between education systems and labor markets; improving the coverage and effectiveness of labor market/training programs; and creating strategies to tackle teenage pregnancy. Institutional capacity is also vital in establishing education systems that can accelerate learning, reduce dropouts, support labor market inclusion among youth, and create more and better jobs. For the education sector, this involves introducing effective systems of institutional empowerment and accountability for the delivery of education services at all levels; developing institutional capacity and regulatory frameworks for integrating technologies into the education sector with a long-term vision; and strengthening regulatory frameworks to push for the professionalization of the teaching career. For labor markets, some of the key elements to improve institutional capacity include improving access to TVET, especially for vulnerable youth; strengthening employment services; and improving the inclusivity and dynamism of the social protection delivery system. 9 Introduction Central America Human Capital Review  | Promoting more and better investments in human capital INTRODUCTION H uman capital, the knowledge, skills, and health that people need to achieve their potential, is a powerful force propelling economies to achieve inclusive growth and poverty reduction. Human capital consists of the knowl- edge, skills, and health that people invest in and accumulate throughout their lives (Figure 1). Investing in people through nutrition, quality health care and education, jobs and skills helps develop human capital. Human capital is a central driver of sustainable growth and a key contributor to poverty reduction and shared prosperity (World Bank, 2018b). Yet, policy makers sometimes find it hard to make the case for human capital investments as the benefits of investing in people can take a long time to materialize. That is, the benefits of investing in the human capital of young children will not deliver economic returns until those children grow up and join the workforce. Therefore, to make the most of their fiscal resources and the power of their people, governments need to invest better and prioritize expenditures that deliver measurable improvements in human development outcomes, thereby enabling every person to achieve their potential so that they can fully contribute in the future to their communities and countries. This Human Capital Review provides analytical foundations to advocate for policies that promote better human capital outcomes in the four UMI countries in Central America: Costa Rica, Guatemala, Panama, and the Domin- ican Republic. The objective of the review is to identify the constraints to human capital growth and to understand how investments and education and skills/labor market policies can accelerate inclusive and resilient recovery, growth, and poverty reduction in these countries.2 The review focuses on key structural issues using a multi-sectoral approach, but it also presents estimates on the impact of the COVID-19 pandemic on key human capital outcomes. Not only did countries experience the initial hit from COVID-19, but there is a risk that human development outcomes erode and structural issues are exacerbated. Countries in Central America are quite diverse in terms of its level of income and human development outcomes. This report focussed on the four UMI countries in Central America to take advantage of new data collected during the pandemic, which allows to quantify the impacts of the COVID-19 pandemic and to understand some its longer-term implications for human development outcomes. The analysis covers the decade preceding COVID-19 (2010-2019) and trends during the pandemic (2020-2021). Within a broad life cycle approach to human development, the Human Capital Review focuses on two critical stages: school-age children (6-14 years old) and youth (15-24 years old) transitioning to the labor markets. While recognizing that human development takes place at all ages with contributions from multiple sectors, low learning outcomes are the main contributors to the low levels of human capital in these four UMI countries in Central America, creating an import- ant constraint for human capital accumulation. In turn, the poor foundational learning and high dropouts translate into a lack of relevant skills among the youth to transition to the labor market and subsequently in poor labor market outcomes. Consequently, this report is focussed on two key education outcomes – foundational learning and school dropouts – and in two key labor market outcomes – school to work transition and employability among the youth. The focus on these outcomes in this report does not imply that early child development is not a fundamental factor contributing to prepare children to succeed at school, follow a proper learning trajectory, and become a productive adult later in life. Indeed, early childhood development programs have a critical role to play to compensate for inequalities in human capital accumulation, which starts very early in life. 2 This review draws on data from household and labor force surveys, administrative data, reviews of the impact evaluations, literature on the effects of developnent interventions on human development outcomes, and simulations of the COVID impact on selected outcomes. 10 Introduction Central America Human Capital Review  | Promoting more and better investments in human capital Figure 1: Human Development across the Life Cycle Human Capital Accumulation Takes Place at all Ages Education Health & Nutrition Social Protection & Jobs Early Youth Childhood School Age and Young Active Later Development Children Adults Years Years Transport Water and Sanitation Electricity Digital connectivity Information and communication technologies Governance Resilience Public Private Partnerships This report is organized around four chapters. Chapter 1 contains a sub-regional landscape analysis of the human capital outcomes for the period between 2010 and 2020, including recent trends on public social spending and the impact of COVID-19 on some human development outcomes. Chapter 2 contains a diagnostic of the structural challenges for human capital accumulation during the school years focused on the education sector. The two key school outcomes that we focus on are learning and school dropout3. Chapter 3 contains a diagnostic of the structural challenges in labor markets (from both demand and supply perspectives) and the main constraints for young adults to transition to the labor market, to increase the productivity of the current labor force, and to improve the alignment of skills and abilities with demand. Based on the diagnostic presented in the first three chapters, Chapter 4 outlines priority policy reforms for the education sector and the labor markets to accelerate human capital accumulation and pronate a sustainable and inclusive recovery. 3 Global evidence suggests that teenagers who drop out prematurely from the school system will, on average, earn less and experience more social and economic challenges than their peers with more years of completed education. 11 CHAPTER 1 HUMAN CAPITAL ACCUMULATION AND UTILIZATION IN UMI COUNTRIES IN CENTRAL AMERICA HOSPITAL ESCUELA UNIVERSIDAD ESCUELA 12 Central America Human Capital Review  | Promoting more and better investments in human capital Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital C hapter 1 presents a landscape analysis of the state of human capital and its recent evolution in four UMI countries in Central America – Costa Rica, Guatemala, the Dominican Republic, and Panama. Results from this analysis show that while these four countries have experienced growth and some poverty reduction in the past decade, there are persistent shortfalls in human capital levels, mainly related to education outcomes. Moreover, human capital growth has largely stagnated in the past decade, contrasting with the evolution of human capital at the global level. This report highlights the key constraints in improving human development outcomes amongst school-age children, youth, and working-age adults (i.e., active years). These outcomes partly reflect insufficient and ineffective public investments in social services compared to international standards as well as inefficiencies in the use of these investments. Lastly, some of the measures taken to control the spread of COVID-19 in these countries were among the most drastic in Latin America, resulting in significant job loss, income reduction, food insecurity, and school closures that further hindered the ability of households to protect their levels of human capital. Evolution and current state of human capital in Central America Between 2010 and 2019, upper-middle-income (UMI) countries in Central America have experienced growth and some poverty reduction. Over most of the past ten years, economic growth in Costa Rica, Guatemala, Panama, and the Dominican Republic exceeded the 2.2 percent average observed in Latin America (Figure 2). During the past decade, the poverty rate in these countries as measured by the international poverty line of $6.85 a day (2017 purchasing power parities or PPP), reduced significantly in the Dominican Republic from 44 percent to 20 percent, and from 22 percent to 12 percent in Panama. The poverty rate in Costa Rica and Guatemala stagnated during this period at around 14 percent and 55 percent, respectively (Figure 3). The Human Capital Index (HCI) is a measure introduced by the World Bank in 2018 . It quantifies the amount of human capital that a child born today can expect to attain by age 18, given the risks of poor health and education currently prevailing in the country where they live (Box 1). Essentially, it measures the potential productivity of the next generation’s workers compared to a benchmark of complete education and full health. The HCI is estimated based on three subcomponents – child survival, education, and health. The index is scaled from 0 to 1, with 1 being the best possible outcome. A country in which a child born today can expect to achieve both full health and complete high-quality education would score a value of 1 on the HCI. Conversely, a country in which a child achieves neither of these would score a value of 0. The human capital indexes of the four countries are low compared to countries with similar levels of income. Except for Costa Rica, all UMI countries in Central America have lower levels of HCI than would be expected for countries with their level of income (Figure 4). Children born before the COVID-19 pandemic in Guatemala had an average HCI of 0.46 (Figure 5). This means that the GDP per worker in Guatemala could be more than twice as high in the future if the country offers access to high quality education and health services to its children. Similarly, children born before the COVID-19 pandemic in Panama and the Dominican Republic had an average HCI of 0.50. In contrast, the HCI of children born before the COVID-19 pandemic in Costa Rica was 0.63, implying potential increases in GDP per worker of a third if access and quality of education and health services can be improved to high-income countries benchmark. 13 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Figure 2: Gross Domestic Product Growth (Annual %), Figure 3: Poverty Rate (% Population Earning $6.85 a 2010 2021 Day or Less in 2017 PPP), 2010–2022 20 60 15 50 10 40 GDP Growth Annual %) 5 Poverty rate %) 0 30 -5 20 -10 -15 10 -20 0 2010 2012 2014 2016 2018 2020 2010 2012 2014 2016 2018 2020 2022 Costa Rica Dominican Republic Guatemala Panama Costa Rica Dominican Republic Guatemala Panama Source: Macro Poverty Outlook published October 2022. Values for 2022 are Source: World Development Indicators. estimated. The low levels of human capital in the four countries included in this review are mainly explained by poor education outcomes. The common factor pulling down the levels of human capital in Guatemala, Costa Rica, Panama, and the Dominican Republic are the low education outcomes, as measured by the expected years of learning-adjusted school, which capture both access and quality of education (Figure 6). Yet, the constraints to improve the HCI vary slightly across the four countries. While in Panama, the Dominican Republic, and Guatemala, improving education outcomes is the main issue to focus on, an additional dimension that is important in Guatemala is the high stunting rates for children under five and the relatively low adult survival rates. Moreover, the existing human capital available in countries is not being fully utilized in the labor market. In many countries, a sizeable fraction of the population is not employed, or are in jobs in which they cannot fully use their skills and cognitive abilities to increase their productivity. The HCI can be adjusted to construct two metrics – one that allows understanding the extent to which Figure 4: HCI Score by GDP per Capita (2017 PPP), Figure 5: HCI of UMIs in Central America, circa 2020 circa 2020 0.9 a) Total HCI 0.70 0.8 0.63 Costa Rica 0.60 0.60 Human Capital Index, 2020 0.7 0.50 0.50 Human Capital Index, 2020 0.50 0.46 0.6 Panama 0.40 0.5 Dominican Republic 0.30 0.4 Guatemala 0.20 0.3 0.10 0.2 6.0 7.0 8.0 9.0 10.0 11.0 12.0 0.00 Guatemala Panama Dominican Costa Rica Rest of LAC Natural log of GDP per capita PPP $), 2020 Republic Source: The HCI database and World Development Indicators. Source: The HCI database. 14 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Figure 6: Subcomponents of the HCI in UMI Countries in Central America, 2020 b) Child Survival subcomponent of HCI c) School subcomponent of HCI 1.0 0.991 0.971 0.974 0.985 0.987 1.0 1.0 1.0 0.9 0.9 0.9 0.9 0.8 0.8 0.8 0.8 0.7 0.7 0.67 0.65 0.7 0.7 0.6 0.6 0.55 0.54 0.55 0.6 0.6 0.5 0.5 Costa Rica Dominican Guatemala Panama Rest of LAC Costa Rica Dominican Guatemala Panama Rest of LAC Republic Republic d) Health subcomponent of HCI 1.0 0.95 0.94 0.93 0.93 1.0 0.9 0.88 0.9 0.8 0.8 0.7 0.7 0.6 0.6 0.5 Costa Rica Dominican Guatemala Panama Rest of LAC Republic Source: World Bank HCI dataset. human capital is being fully utilized (basic utilization HCI) using the fraction of the working age population who are employed, and one that takes into consideration the income gains4 that could be generated from moving workers to jobs where they can better use their human capital to increase productivity (“better employment”).5 The two utilization-adjusted HCIs for Costa Rica exceed the average for countries in Latin America as well as that from UMIs countries. In contrast, the utilization HCIs in Panama, Guatemala, and the Dominican Republic fall short of the average in the countries in Latin America. In all the four countries, the full utilization-adjusted HCI is slightly lower than the basic utilization-adjusted HCI, suggesting the lack of good jobs (Figure 7). Figure 7: Basic and Full Utilization-Adjusted HCI in Central America, circa 2020 0.5 Utilization-Adjusted Human Capital Index, 2020 0.4 0.4 0.3 0.3 0.2 0.2 0.1 0.1 0.0 Costa Rica Dominican Republic Guatemala Panama Rest of UMI Rest of LAC Full Basic Source: The HCI database. 4 Ensuring a better match between workers skills/human capital and the corresponding employment. When a worker who has higher human capital is working in a job that requires a lower set of skills, this impacts not only his or her earnings, but also a lost opportunity for the economy for a better utilization of existing human capital. 5 Pennings, 2020. 15 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Figure 8: HCI in UMI countries in Central America, Figure 9: Contribution of Each Component of the HCI 2010 and 2020 to the Observed Change between 2010 and 2020 -0.3 0.2 0.70 Panama -1.2 -1.4 0.237 0.2 0.65 Guatemala 1.9 0.3 Costa Rica -1.3 2.1 0.411 Rest of LAC Dominican 0.60 Republic 1.1 0.4 2020 HCI -0.4 1.3 0.054 0.55 Costa Rica 3.1 0.1 Dominican Republic 2010-2020 0.2 0.50 Panama 2017-2020 -1.3 4.0 0.081 Guatemala 0.2 0.45 -3 -2 -1 0 1 2 3 4 5 6 0.40 Fraction of Children Under 5 Expected Years of School 0.40 0.45 0.50 0.55 0.60 0.65 0.70 Not Stunted Learning-Adjusted Years of School 2010 HCI Harmonized Test Scores Probability of Survival to Age 5 Survival Rate from Age 15-60 Source: World Bank HCI Dataset. Source: World Bank (2021). The lack of progress in the education outcomes is largely Figure 10: Disaggregated HCI by Income Quintiles, responsible for the stagnation in human capital accu- 2020 mulation over the past decade. Figure 8 shows that HCI Dominican Republic scores increased only marginally in Costa Rica, Guatemala, and the Dominican Republic (1 to 2 percentage points) (Figure 8). Panama’s HCI even deteriorated over the past decade. This stagnation contrasts with the global evolution .2 .4 .6 .8 1 of human capital, which shows HCI scores increasing by Guatemala 3.5 percentage points on average over the past decade. For all four countries, the main factor pulling down the HCI is the lack of progress in the education outcomes. In Panama and Guatemala, the challenging component is expected .2 .4 .6 .8 1 years of schooling while in the Dominican Republic and Costa Rica it is harmonized test scores (Figure 9). Poorest 2nd 3rd 4th Richest Conversely, changes in expected years of schooling contrib- quintile quintile quintile quintile quintile uted positively to human capital accumulation in Costa Rica and to a lesser extent in the Dominican Republic. Source: World Bank, 2020, HCI. Poverty quintiles are based on 2014 data from The same was true of harmonized test scores in Guatemala, the INE (Instituto Nacional Electoral) for Guatemala and the 2013 Demographic and Health Survey (DHS) for the Dominican Republic. Information for Costa while in Panama they contributed slightly to the deterio- Rica and Panama is not available. ration in human capital accumulation. Within country disparities in human capital levels across the socioeconomic distribution are very large. Except for the Dominican Republic, the three other UMI countries in this review are among the ten most unequal countries in Latin America as measured by the Gini index, which is based on income data. The large levels of income inequality can be explained to some degree by unequal access to and limited quality of health and education services. This translates into large disparities in human capital levels between lower- and higher-income groups (Figure 10). For example, the HCI score for the highest income quintile in Guatemala is almost 20 percentage points higher than that of the poorest quintile group. These differences are also clearly seen in the Dominican Republic, even if to a lesser extent. 16 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Box 1. Human Capital Index (HCI) The HCI measures the expected productivity of children born today at the age of 18 assuming educational opportunities and health risks remain the same. This is meant to highlight how improvements in current health and education outcomes shape the productivity of the next generation of workers. The HCI captures key stages of a child’s trajectory from birth to adulthood. In the early years, there is a risk that a child will not survive to his or her fifth birthday. If the child does reach school-age, there is a further risk that he or she will not start school, let alone complete the full cycle of 14 years of schooling (preschool to grade 12). The time children spend in school may translate to uneven learning, as a result of a variety of factors including the quality of teachers and schools. When she turns 18, she carries with her the lasting effects of poor health and nutrition during childhood that limit her physical and cognitive abilities as she moves into adulthood. The HCI is estimated using three components: • Component 1 (Child survival): Child survival from birth to school-age is measured using under-five mortality rates. • Component 2 (Education): The quantity of education is measured as the number of school years a child can expect to complete by the age of 18 given the prevailing pattern of enrollment rates. The quality of education is measured using harmonized international test scores of students at the end of primary education. These two measures are combined to a measure of expected learning-adjusted years of schooling as proposed in Filmer et al. (2018). • Component 3 (Health): The overall health environment is captured by two proxies: (i) adult survival rates, defined as the fraction of 15-year-olds who survive until the age of 60, and (ii) the rate of stunting for children under the age five. Adult survival rates reflect the range of fatal and nonfatal health outcomes that a child born today would experience as an adult if current conditions remain the same. Stunting broadly represents the risks to good health that children born today are likely to experience between their prenatal and early childhood years, which have important consequences for health and well-being in adulthood. These three components are combined as follows to construct the HCI: Human Capital Index Child Health Education Health Productivity of a person Survival rate of a Learning-adjusted years of Adult survival rate born today when children under-five school 18 years old and not stunted 17 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Box 1. Human Capital Index (HCI) (continuation) Where components are defined as: Child survival = _____________ 1 − under 5 mortality rate ​  ​     0.08*​(​expected years of school*​___________ 1  ​   Education  =  ​ ;​ e​ ​   − ​    )​ 14​ harmonized test scores 625 (​0.65*​(adult survival rate−1)​+0.35*​(not stunted rate−1)​)​ ___________________________ ​ = ​ Health ​ e​  ​          ​ ​ 2 The resulting index ranges between 0 and 1. A country in which a child born today can expect to achieve full health (no stunting and a 100 percent adult survival rate) and full education potential (14 years of high-quality school by age 18) would score a value of 1. Therefore, a score of 0.70 indicates that the future productivity of a child born today is 30 percent below what could have been achieved with a complete education and full health. The index can also be used as a means of understanding how much a country can increase its income if investing in human capital. For instance, if a country has a score of 0.50, the gross domestic product (GDP) per worker could be twice as high if the country reached the benchmark of complete education and full health. Source: World Bank (2021a). Disparities in human capital within countries at the subnational level are also conspicuous, with low levels of human capital corresponding to areas where poverty is widespread. The subnational HCI show large inequalities in accessing high-quality health and education services (Figure 11). In Panama, for instance, the HCI of a child born before the COVID- 19 pandemic in the province of Panama is 0.5, In comparison, the HCI of a child born in the indigenous province of Kuna Yala is 0.37, which is comparable to that of low-income countries with high levels of fragility and conflict like Congo and Sierra Leone. A similar pattern of low human capital levels that correspond to areas with high poverty can be observed in Guatemala and to a lesser extent in the Dominican Republic (Figure 12). In Guatemala, the HCI differs by 14 percentage points between its poorest and richest provinces despite a narrower gap in poverty rates (45 percent versus 5 percent poverty rates). On the other hand, the difference between the highest- and lowest-poverty provinces in the Dominican Republic where poverty rates are 85 percent and 25 percent, respectively, is only 4 percentage points. Figure 11: Pre-COVID-19 Subnational HCIs b) By Department in Guatemala, 2019 a) By Province in Panama, 2016 [.315,.352] [.307,.315] [.249,.307] No data 35.3 37.3 40.1 42.5 45.7 50.8 Source: World Bank estimates based on: 2020 provincial survival rates for chil- Source: World Bank estimates based on: 2014–15 department survival rates and dren 0–5 from the Ministry of Health; provincial reading score averages of third stunting rates for children 0–5 from the National Institute of Statistics; depart- graders on the 2016 national student assessment (Prueba CRECER); provincial ment math score averages of third graders on the 2019 national student assess- enrollment rates in preschool, primary, and secondary school and the number ment; 2014 department enrollment rates in preschool, primary, and secondary of education years at each level from the Ministry of Education; adult survival school and the number of education years at each level from the National Institute rates from the 2018 Vital Statistics provided by the National Statistics Office; of Statistics: and adult survival rates from the 2018 Vital Statistics provided by and childhood stunting rates based on the 2008 Encuesta de Niveles de Vida. the National Statistics Office. 18 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Figure 11: Pre-COVID-19 Subnational HCIs (continuation) c) By Province in the Dominican Republic, 2019 [0.43,0.47] [0.47,0.48] [0.48,0.51] Source: World Bank estimates based on: 2020 provincial survival rates from the National Statistics Office; the provincial math score averages of 9th graders on the 2019 national student assessment (Pruebas Diagnosticas); 2020 provincial enrollment rates in preschool, primary, and secondary school and the number of education years at each level from the Ministry of Education; regional averages for childhood stunting and adult survival rates from the 2019 Encuesta Nacional de Hogares de Propósitos Múltiples, con la metodología de Encuestas de Indicadores Múltiples por Conglomerados, sobre la situación de niños, niñas, adolescentes y mujeres en la República Dominicana (ENHOGAR-MICS survey). Figure 12: Subnational Human Capital Index and Poverty Rates in UMI Countries in Central America, 2020 a) Panama b) Guatemala 80 Ngäbe Buglé 50 Alta Verapaz Quiché 70 Chiquimula 40 Huehuetenango Poverty rate % of population) Poverty rate % of population) 60 Jalapa Baja Verapaz 50 Totonicapán San Marcos 30 Petén Izabal 40 Kuna Yala Sololá Zacapa Jutiapa Suchitepéquez 30 Bocas del Toro 20 Chimaltenango Darién Santa Rosa Quezaltenango 20 Veraguas Retalhuleu El Progreso Coclé 10 Escuintla 10 Chiriquí Guatemala Colón Herrera Sacatepéquez Los Santos Panamá 0 0 0.35 0.40 0.45 0.50 0.55 0.35 0.40 0.45 0.50 HCI HCI c) Dominican Republic 90 Independencia 80 Pedernales Baoruco Monte Plata Poverty rate % of population) 70 Monte Azua Hato Mayor Cristi El Seibo 60 San Juan Valverde La Altagracia Dajabón 50 San Pedro de 40 Duarte Puerto Plata Macorís La Vega S. Domingo San Cristóbal 30 Santiago Hato Mayor 20 Monseñor Nouel 10 0 0.42 0.44 0.46 0.48 0.50 0.52 HCI Source: Authors’ elaboration based on the subnational HCI and poverty rates. Poverty rates for Panama are based on the Household Survey 2019 (using the 550 PPP household per capita poverty line). Poverty rates for Guatemala come from the Multidimensional Poverty Index based on the Population Census 2018, and those for the Dominican Republic come from the Poverty Map 2014. 19 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Outcomes by life cycle stages Low levels of human capital are reflected in a wide range of human development outcomes along the life cycle. Life-cycle human capital accumulation is a dynamic process. An individual’s human capital grows in a cumulative fashion, building on earlier investments It accumulates unevenly across the life cycle with the most rapid growth occurring during school-age and youth. These two life-cycle stages are the most significant in terms of productivity later in life and are there- fore the focus of this report (Figure 13). This section highlights some of the key constraints to increasing human capital accumulation at different stages of the life cycle in the four UMI countries in Central America. While an in-depth analysis of all human development outcomes is beyond the scope of this report, we provide a landscape analysis to some of the key issues in human capital accumulation. A more detailed discussion of education outcomes and the transition to the labor market is covered in Chapters 2 and 3, respectively. Figure 13: Life Cycle Stages Covered in This Human Capital Review Education Health & Nutrition Social Protection & Jobs Early Youth Childhood School Age and Young Active Later Development Children Adults Years Years Transport Water and Sanitation Electricity Digital connectivity Information and communication technologies Governance Resilience Public Private Partnerships 20 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Early childhood development The challenges in ensuring good human capital outcomes during the early years differ across countries, but all share limited access to high-quality early childhood development programs. Growth and development in the early years provide the foundation for future learning, productivity, and lifetime success.6 Early childhood development programs are also critical to compensate for the previously mentioned inequalities in human capital accumulation that exist between socioeconomic groups and regions, which are known to start very early in life. Less than 3 percent of 0- to 3-year-olds have access to an early childhood development program in Costa Rica, whereas in Guatemala and Panama this number is roughly 1 percent. In all four countries, access to early childhood development programs is very low compared to the levels of access observed in countries with similar levels of income (Figure 14). UMI countries in Central America do not invest enough in young children. In addition to budget constraints and an insufficient understanding of the high payoffs of quality early childhood interventions, these countries also face challenges in delivering health and nutrition services for young children. Figure 14: Net Enrollment in Early Childhood Figure 15: Prevalence of Stunting (% of Children under Development (ECD) Programs (%) by GDP per Capita, Five) by GDP per Capita, circa 2019 circa 2019 100 60 90 Net enrollment in ECD % of children) 50 80 % of children ≤ 5 years old) Prevalence of stunting Guatemala 70 40 60 50 30 40 20 30 Panama 20 10 Costa Rica Dominican Republic 10 Costa Rica Guatemala Panama 0 0 6 7 8 9 10 11 12 7 8 9 10 11 12 -10 Log of GDP per capita PPP $) Log of GDP per capita PPP $) Source: World Development Indicators and UNESCO. Net enrollment in early Source: World Development Indicators. childhood educational development programs is the ratio between the enrollment in early childhood educational development programs and the school-age popu- lation of early childhood educational development programs, multiplied by 100. Another challenge among children under-five, is the high prevalence of stunting, which is particularly prominent in Guatemala where childhood stunting is among the highest in the world. Ensuring that children are healthy and well-nourished during their early years is critical for survival and for developing their full potential. The most critical time for good nutrition is from pregnancy until a child’s second birthday.7 Improving the quality of children’s food and feeding practices during these years is the fundamental to both preventing malnutrition and fostering human capital accumulation. In Guatemala, 44 percent of children under-five are stunted, which is 11 percent higher than the average for UMI coun- tries and 10 percent higher than the Latin American average (Figure 15). This Figure makes Guatemala the country with the sixth highest stunting rate in the world and suggests that parents in Guatemala face important constraints in securing nutritious, safe, affordable, and age-appropriate food for their children. Furthermore, children under-five in the Dominican Republic and Guatemala have relatively high mortality rates, reflecting the limited access to quality health services. In 2019, the mortality rates among children under five in the Dominican Republic and Guatemala were 34 deaths per 1,000 live births and 25 deaths per 1,000 live births, respec- tively compared to 17 deaths per 1,000 live births on average in Latin America (Figure 16). At the global level, infectious diseases (e.g., pneumonia, diarrhea, and malaria) and pre-term birth complications (e.g., birth asphyxia, birth trauma, 6 Devercelli et al., 2022. 7 Development Initiatives, 2020. 21 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital and congenital anomalies) remain the leading causes of Figure 16: Mortality Rate of Children Under-Five (per death for children under five. However, access to basic 1,000 live births) by GDP per Capita, circa 2019 lifesaving interventions such as the presence of skilled 140 assistants at birth, postnatal care, breastfeeding, adequate 120 nutrition, vaccinations, and treatment for common child- Mortality rate, under-5 per 1,000 live births) hood diseases can play an important role in saving many 100 young lives. 80 60 Good nutrition is dependent on access to water, which 40 is necessary to the healthy growth and development of a Dominica Republic 20 Guatemala child during their early years. In the past, early childhood Panama Costa Rica interventions involving water have been centered on ensur- 0 6 7 8 9 10 11 12 ing a hygienic living environment through access to clean -20 water. The importance of water in nutrition outcomes is, Log of GDP per capita PPP $) however, much broader. A reliable water supply is needed Source: World Development Indicators and Health Nutrition and Population to grow food to feed families, secure livelihoods, and Statistics. provide income for other nutrition inputs. In the case of UMI countries in Central America, all countries except for Panama have access to basic drinking water services that is expected for their level of income (Figure 17). In terms of access to basic sanitation services, all countries except for Costa Rica have access to basic sanitation services that is lower than expected for their levels of income (Figure 18).8 Figure 17: People Using Basic Drinking Water Services Figure 18: People Using Basic Sanitation Services (% (% of population) by GDP per Capita, circa 2019 of population) by GDP per Capita, circa 2019 120 120 People using at least basic sanitation People using at least basic drinking Costa Rica 100 Costa Rica water services % of population) Guatemala 100 services % of population) Dominican Panama Dominican 80 Republic Republic 80 Panama 60 60 Guatemala 40 40 20 20 0 0 7 8 9 10 11 12 6 7 8 9 10 11 12 -20 Log of GDP per capita PPP $) Log of GDP per capita PPP $) Source: World Development Indicators. Source: World Development Indicators. School-age children Access to primary education is universal in the four countries studied for this review, but challenges persist in enroll- ing children in preprimary education. In 2020, only 60 percent of 3- to 6-year-olds had access to preprimary education in Costa Rica, the Dominican Republic, and Panama (Figure 19). This trend remained stable over the last decade in the Dominican Republic and Panama. Preprimary attendance in Costa Rica, however, rapidly increased over the same period, having started at a mere 35 percent in 2010. In Guatemala, low access to preprimary education is much more extreme. Only 14 percent of 3- to 5-year-olds have access to preprimary school, and attendance rates have been deteriorating signifi- cantly over the past decade. The low access to quality early childhood development programs combined with the low level 8 Basic sanitation services refers to facilities that are not shared with other households including flush to piped sewer systems, septic tanks, pit latrines with or without slabs, and toilets. 22 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Figure 19: Preprimary Net Attendance Rate for Figure 20: Secondary School Net Enrollment Rate by Children 3 to 5 Years Old, circa 2020 GDP per Capita (%), circa 2019 100 120 100 School enrollment, secondary % net) 80 67.1 66.1 64.5 67.6 65.4 64.3 68.0 63.2 64.7 Costa Rica 80 60 64.1 65.0 60.4 60.3 63.6 62.6 63.6 63.2 62.3 59.6 56.7 Republica 60 Dominicana Panama 40 48.2 43.5 42.6 43.4 38.6 39.6 35.1 35.8 40 Guatemala 20 23.8 14.0 20 0 0 10 11 12 13 14 15 16 17 18 19 20 20 20 20 20 20 20 20 20 20 20 20 6.5 7.5 8.5 9.5 10.5 11.5 12.5 Costa Rica Dominican Republic Guatemala Panama Log of GDP per capita 2017 PPP $) Source: SEDLAC (CEDLAS and The World Bank). Source: World Development Indicators. of access to preprimary school translates into children who arrive unprepared to primary school.9 This means that even in the case where access to good primary schools improves, unprepared learners will not be able to learn much as they could, thus compromising their human capital accumulation. Youth High dropout rates in secondary school are an important constraint in human capital accumulation among youth, particularly in Guatemala and Panama. The factors that contribute to low enrollment in secondary school are discussed in more detail in Chapter 2, but they include high adolescent fertility rates and high dropout rates during the transition from primary to secondary school because of lack of finances and lack of motivation. In terms of dropout rates in the transition from primary to secondary school, Guatemala and Panama had respective net secondary school enrollment rates of 44 and 64 percent in 2019, which are well behind those observed in countries with similar income levels (Figure 20). Guatemala’s secondary school enrollment rate is the lowest in Latin America. Global evidence suggests that teenagers who drop out prematurely from the school system will, on average, earn less and experience more social and economic challenges than their peers with more years of completed education.10 All four countries have higher teenage pregnancy rates than expected when compared with countries with similar levels of income. The Dominican Republic has the highest rate in Latin America with 92 births per 1,000 women aged 15–19 and is followed by Panama whose rate is 80 births per 1,000 women aged 15-19. In comparison, the Latin American average is only 50 births per 1,000 women aged 15-19 and in UMI countries, the average is 41 births per 1,000 women aged 15-19 (Figure 21). This is despite the high prevalence in the use of contraceptives by women in all the four countries, except for Panama (Figure 22). High teenage pregnancy leads not only to lower levels of human capital accumulation but also incurs high economic and social costs as it negatively effects education achievement, labor market participation, income, and health spending.11, 12 Enrollment in tertiary education is low in most of the four countries, which limits the ability of youth to acquire relevant skills for a successful transition to the labor market. Except for Costa Rica, gross enrollment rate in tertiary education (defined as the ratio between higher education enrollment and the population aged 18–24) is low in all countries given their income levels. School enrollment rates in tertiary education in Guatemala, Panama, and the Dominican Republic 9 World Bank, 2018a. 10 Patrinos and Psacharopoulos, 2004; Oreopoulos and Salvanes, 2011; Bentaouet-Kattan and Székely, 2015; Cardenas, De Hoyos, and Székely, 2015. 11 The legal framework protects the right to education of pregnant and parenting teenagers in Costa Rica, the Dominican Republic and Guatemala. Panama does not explicitly have this protection in its legal framework (Her Atlas, UNESCO, 2019). However, the decision to drop out often has other causes including childcare respon- sibilities, financial needs, and stigma, among others. 12 UNPFA, 2022. 23 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Figure 21: Adolescent Fertility Rate (Births per 1,000 Figure 22: Contraceptive Prevalence, Any Method (% Women Aged 15–19), by GDP per Capita, circa 2019 of Married Women Aged 15–49) by GDP per Capita, circa 2019 200 100 90 Contraceptive prevalence, any method births per 1,000 women ages 15-19) % of married women ages 15-49) 150 80 70 Costa Rica Adolescent fertility rate 60 Guatemala Dominican Republic 100 Dominica Republic Guatema Panama 50 Panama 40 50 Costa Rica 30 20 0 6 7 8 9 10 11 12 13 10 0 -50 7 8 9 10 11 12 Log of GDP per capita PPP $) Log of GDP per capita PPP $) Source: World Bank, World Development Indicators. Source: World Bank, World Development Indicators. are 44 percent, 64 percent, and 71 percent, respectively, compared to 79 percent and 81 percent on average among UMI countries and in Latin America (Figure 23). In addition to this, the large share of young adults who are not enrolled in tertiary education have not necessarily been able to successfully transition to the labor market. The share of youth who are not in education, employment, or training (NEET) is high, especially in the Dominican Republic and Guatemala (Figure 24). The issue of the NEETs will be analyzed in more detail in Chapter 3. Figure 23: Gross Enrollment in Tertiary Education (%) Figure 24: Share of Youth Not in Education, by GDP per Capita, circa 2019 Employment, or Training (% of Youth Population) by GDP per Capita, circa 2019 120 80 70 Share of youth not in education, School enrollment, tertiary % gross) 100 60 employment or training Costa Rica 80 50 Dominican Republic 60 Panama 40 30 Guatemala Dominican 40 Guatemala Republic 20 Panama Costa Rica 20 10 0 0 6 7 8 9 10 11 12 6.5 7.5 8.5 9.5 10.5 11.5 12.5 -10 Log of GDP per capita PPP $) Log of GDP per capita PPP $) Source: World Development Indicators. Source: World Development Indicators. Active years stage Poor health outcomes later in life suggest that there are important challenges in access to health services in the four countries. Guatemala and the Dominican Republic have relatively high mortality rates among its adult men with 199 and 196 deaths per 1,000 adult men, respectively, which is 15 percent higher than the Latin American average (Figure 25). Mortality rates are correlated with access to adequate health services, which are fundamental in ensuring that adults remain 24 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Figure 25: Adult Mortality Rate, Male (per 1,000 Male Figure 26: Hospital Beds (per 1,000 People) by GDP Adults) by GDP per Capita, circa 2019 per Capita, circa 2019 600 14 500 12 Hospital beds per 1,000 people) deaths per 1,000 male adults) Adult mortality rate, male 10 400 8 300 6 Dominican 200 Guatemala Republic 4 Panama 100 Costa Rica Dominican Panama 2 Republic Costa Rica Guatemala 0 0 7 8 9 10 11 12 6 7 8 9 10 11 12 Log of GDP per capita PPP $) Log of GDP per capita PPP $) Source: World Development Indicators. Source: World Development Indicators. healthy and productive throughout their lifespan.13 In most countries, hospitals account for the largest part of overall fixed investment, making hospital beds an indicator of the resources available for delivering services to inpatients. In all four countries, the number of hospital beds per 1,000 people is low compared to countries with similar levels of income (Figure 25). Guatemala ranks at the bottom among all countries in Latin America for this health outcome with only 0.4 beds per 1,000 people. Costa Rica has 1.1 beds per 1,000 people, which is still less than half the average number of hospital beds among countries in the region. In Costa Rica and the Dominican Republic, a large share of the adult population is overweight, which contrib- utes to low levels of productivity. Approximately 60 percent of adults in Costa Rica and the Dominican Republic are overweight, which is much higher than the average rate observed in Latin America and other UMI countries (Figure 27). Various economic impacts have been associated with the overweight of adults including direct medical costs, productivity costs, and human capital costs.14 In terms of human capital costs, being overweight or obese increases the number of days students are absent from school, therefor, resulting in lower educational attainment. Also, there is a link between health and productivity, which can operate directly through nutrition and the physical ability to work, or indirectly through social stigma, discrimination, or mental health.15 Figure 27: Prevalence of Overweight Adults in Central America (%), circa 2019 64 62 61.6 61.2 % of total adult population) 60 Overweight population 58.8 58 56.3 55.9 56 55.2 54 52 Costa Rica Dominican Republic Guatemala Panama Rest of UMI Rest of LAC Source: World Development Indicators. 13 According to the 2019 Global Burden of Disease report published by the Institute for Health Metrics and Evaluation, around 80 percent of deaths are caused by noncommunicable diseases in LAC. 14 Hammond and Levine 2010. 15 Mazhar and Rehman, 2021. Cawley, 2004. 25 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Except for Costa Rica, deaths among the population Figure 28: Death by Non-Communicable Diseases (% caused by non-communicable diseases (NCDs) is lower of Total Deaths) by GDP per Capita, circa 2019 than expected for UMI countries. NCDs represent 82 100 percent of total deaths in Costa Rica, which is higher than 90 Costa Rica the average in Latin America (77 percent) (Figure 28). 80 Death by non-communicable However, in Guatemala, the Dominican Republic, and Dominican Panama disease % of total deaths) 70 Republic Panama, the share of total deaths caused by NCDs among 60 Guatemala the population is 62 percent, 72 percent, and 78 percent, 50 respectively. In general, the high prevalence of NCDs in 40 a country has negative effects on people throughout their 30 life cycle. NCDs decrease workers’ productivity, reduce 20 the quality of life, and are one of the main causes of low 10 life expectancy, especially among men. While providing 0 comprehensive treatment is expensive and contributes 7 8 9 10 11 12 to unnecessary public spending in the health system, the Log of GDP per capita PPP $) prevention of NCDs is relatively cost effective. Source: World Development Indicators. Public social spending on human capital and social protection programs Poor human development outcomes in UMI countries in Central America reflect inefficient spending and low levels of public investment in the delivery of social services. Public social spending in Guatemala, Panama and the Dominican Republic is only six to seven percent of the GDP, which is much lower than the average of countries in Latin America (11 percent) (Figure 29). Moreover, the limited public social spending in these countries is spent inefficiently compared to other countries in Latin America (Figure 30). This means that countries in the region that spend similar amounts of money per capita in the delivery of social services achieve much better human capital outcomes than those in Guatemala, Panama, and the Dominican Republic. In the case of education, for instance, the allocation of public spending in these three countries has traditionally been focused on current expenditures rather than teacher development policies, education technology and connectivity, or school management, which are all known to be effective at improving learning outcomes. The reasons behind the inefficiency in spending vary by sector and country and will be examined in more detail in the next chapters. Figure 29: Social Expenditure of Central Government Figure 30: HCI, per Capita Social Expenditures, and as % of GDP in Central America, circa 2019 Efficiency Frontier in Latin America, circa 2020 14 0.65 CHL CRI 12 11.67 11.24 MEX TTO LCA 0.60 COL ARG URY ECU ATG Social expenditure % GDP) 10 Human Capital Index KNA GRD 0.55 SLV DMA BRA 8 7.09 JAM PRY VCT 5.96 6.38 6 0.50 NIC DOM PAN GUY HND 4 0.45 GTM 2 HTI 0 0.40 Guatemala Panama Dominican Costa Rica Rest of LAC 0 1000 2000 3000 4000 5000 Republic Per capita social expenditure 2017 PPP) Source: CEPALSTAT. Social public expenditure is computed as the sum of central Source: HCI database, CEPALSTAT, and World Development Indicators. government expenditure on health, education, and social protection. 26 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Figure 31: Performance Indicators for Social Assistance Programs in Central America by Income Quintiles, circa 2019 a) Coverage of Social Assistance Programs 100 90 80 Beneficiary households 70 % of population) 60 50 40 30 20 10 0 Costa Rica Dominican Republic Guatemala Panama b) Beneficiary Incidence of Social Assistance Programs 35 30 25 Beneficiary households % of population) 20 15 10 5 0 Costa Rica Dominican Republic Guatemala Panama c) Coverage of Conditional Cash Transfer Programs 50 45 40 Coverage % of population) 35 30 25 20 15 10 5 0 Costa Rica Dominican Republic Guatemala Panama d) Beneficiary Incidence of CCTs 90 80 70 Beneficiary households 60 % of population) 50 40 30 20 10 0 Costa Rica Dominican Republic Guatemala Panama 1st quintile 2nd quintile 3rd quintile 4th quintile 5th quintile Source: ASPIRE performance indicators using latest available year before the onset of the COVID-19 pandemic: 2019 for Costa Rica, the Dominican Republic, and Panama, and 2014 for Guatemala. For Latin America and the Caribbean, a weighted average was used for the 2010–2019 period using the latest available survey year by country. Note: Beneficiary incidence is the number of individuals who live in a household where at least one member received the transfer / total number of direct and indirect beneficiaries. 27 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Social protection programs could be leveraged to improve human capital and address disparities, but there is room for improvement. International evidence suggests that poverty targeted cash transfer programs, especially conditional cash transfer (CCT) programs, have the most potential among social assistance programs to improve human capital accumulation and to mitigate existing disparities.16 The coverage of .social assistance programs in UMI countries in Central America is large for the bottom quintile, exceeding the 60 percent average among countries in Latin America (Figure 31). However, the share of social assistance beneficiaries that are from the bottom income quintile among all beneficiaries (a proxy for targeting accuracy), ranges between 25 and 30 percent, which is well below the regional average of 45 percent. In the case of CCT programs, targeting accuracy varies widely from 80 percent in the bottom income quintile in Panama to less than 30 percent in the case of the Dominican Republic and Guatemala. Costa Rica’s performance is better, but still falls short of the Latin America average. In the four countries, the coverage of CCT programs in the bottom income quintile falls short of the Latin America average of slightly below 50 percent. Some of the UMI countries in Central America are characterized by significant levels of violence, which hinder human capital accumulation and economic development. The homicide rate of Guatemala is a staggering 25 inten- tional homicides per 100,000 people, which is almost four times the global average (6.5), exceeds the Latin America average (18.2) (Figure 32). This ranks Guatemala as the eleventh most dangerous country in the world in terms of the level of violence experienced by its population. The overall homicide rates in Costa Rica (11.2), Panama (10.2), and the Dominican Republic (9.6) are also higher than the average in UMI countries (8.4), but lower than the average for the rest of the countries in Latin America. Violence has many adverse implications for the broader economy, both in the short and long term, hindering productivity, economic activity, and long-term growth. From a human capital perspective, violence dissuades individuals from investing time and money in education. It may, for example, deter some people from attending night school out of fear of becoming a victim of violent crime, or worse, induce some individuals to turn to a life of crime instead of completing their education.17 Figure 32: Intentional Homicides (per 100,000 Figure 33: Emigration Ratio in Central America, circa people) and GDP per Capita, circa 2019 2020 50 30 45 25 24.5 40 Intentional homicides 35 20.4 per 100,000 people) 20 Emigration Ratio 30 25 Guatemala 15 14.8 20 Dominican 10 15 Republic 7.6 10 Costa Rica Panama 5 3.2 5 2.9 0 0 6.5 7.5 8.5 9.5 10.5 11.5 Costa Dominican Guatemala Panama Rest Rest Log of GDP per capita PPP $) Rica Republic of UMI of LAC Source: World Development Indicators. Source: World Development Indicators and United Nations Department of Economic and Social Affairs. High levels of migration in some of the countries also pose an important constraint to human capital accumulation. The emigration ratio, which measures the number of migrants divided by the total population, shows that a large share of the population in the Dominican Republic and Guatemala emigrates (14.8 percent and 7.6 percent, in 2020, respectively) (Figure 33). The emigration ratio for Guatemala is almost two and a half times the average in the rest of Latin America (6.4 percent) and more than five times the average among UMI countries (2.9 percent). Emigration ratios in Panama and Costa Rica are much smaller at 3.2 percent and 2.9 percent, respectively, but these are still larger than those observed in countries 16 World Bank, 2019a. 17 International evidence shows that violence has a detrimental effect on both school attendance and standardized test scores and that it increases the dropout rates of students substantially (see Bunivic and Morrison,1999 and Foureaux and L. Menezes, 2021). 28 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital with similar levels of income. The flow of migrants from one country to another can have very different effects on a country based on how this affects both its level and composition of human capital.18 In Central America, emigration is generally linked to low living standards, lack of opportunities, and is exacerbated by extreme weather events, food insecurity, and violence.19 Impact of the COVID-19 pandemic on human capital The COVID-19 pandemic had a significant impact on the four UMI countries studied in this review. To contain the spread of the virus, governments implemented measures for social distancing, minimized non-essential economic activi- ties, and suspended in-person education services soon after the pandemic began. Some of the measures taken to control the spread of the disease where among the most drastic in Latin America. For instance, in Panama, Guatemala, and the Dominica Republic strict curfews were implemented nationwide during the peak of the pandemic. Despite the expansion of social protection programs and other emergency interventions for mitigating its socioeconomic impacts, economic growth shrank significantly, translating into higher poverty rates. In 2020, economic growth reduced substantially in all four countries with annual growth rates dropping to -2 percent in Guatemala, -4 percent in Costa Rica, -7 percent in the Dominican Republic, and -18 percent in Panama.20 The more severe impact observed in Panama was due to its high dependence on sectors heavily affected by mobility restrictions such as tourism, construction, and retail. In the case of Guatemala, remittances extensively cushioned the impact of COVID-19 on the economy. This sharp reduction in economic growth has shown to correspond with an increase in poverty rates in the Dominican Republic and Costa Rica, the two countries for which there is recent poverty data available. The data suggest poverty rates estimated using the international poverty line of $6.85 per day (2017 PPP) increased from 14 to 20 percent in Costa Rica and from 20 to 24 percent in the Dominican Republic. The pandemic caused significant job loss and income reduction, hindering the ability of households to maintain or increase their levels of human capital. Many workers became unemployed or left the labor force in all four countries.21 Those who remained employed suffered a decline in working hours and labor income, which dropped for 54 percent of households in the Dominican Republic and 58 percent of households in Guatemala (Figure 34). This decline in income, in turn, limited household access to food and nutrition. About 47 percent of households in Guatemala reported running out of food due to a lack of money or other resources in the early stages of the pandemic, whereas in the Dominican Republic and Costa Rica the percentage was 44 and 30, respectively (Figure 35). Figure 34: Percentage of Households Which Reported Figure 35: Percentage of Households Which Reported a Reduction in Income during the COVID-19 Pandemic Running out of Food during the Pandemic in Central in Central America, 2020 and 2021 America, 2020 and 2021 % of households that report a reduction % of HH tha ran out of food due to lack o money last 30 days) of total income during quarantine 46.3 69.9 43.6 % of households 59.2 34.0 47.7 45.1 29.7 27.9 % of households 23.0 Dominican Republic Guatemala Costa Rica Dominican Republic Guatemala May 2020 June - July 2021 May 2020 June - July 2021 Source: World Bank, High Frequency Phone Surveys, Washington DC. Data for Source: World Bank, High Frequency Phone Surveys, Washington DC. the first wave of COVID-19 (between May 21st and June 1st, 2020). 18 For a discussion on the effects of migration on human capital formation see, for instance, Corraco and Lazarova, 2012. 19 Mejia-Mentilla et al., 2023. 20 See Figure 2 and Figure 3 in Chapter 1. 21 World Bank, 2021b. 29 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Although governments made important efforts to continue providing education services, considerable challenges were faced. The pandemic led to nationwide lockdowns and school closures in all four countries. Although the timing of the pandemic was similar around the world, the duration of school closures disproportionately affected the UMI countries in Central America, with Panama’s closures ranking among the longest in the world.22 To mitigate learning losses, all coun- tries moved education service delivery to remote learning or, when possible, a hybrid modality. Compared to pre-pandemic values, school attendance was significantly reduced when remote learning was in place. Average attendance rates, which were measured by the share of students who attended some form of education (either in-person or remote), were 33 percent lower in Guatemala during the pandemic compared to pre-pandemic values (Figure 36). For the Dominican Republic and Costa Rica, attendance rates during the pandemic were 14 percent and 10 percent lower compared to pre-pandemic values. Remote learning also seems to have created a significant disincentive for engagement in school activities for a large part of students. The share of students attending school (either in-person or remotely) who were not studying at home or engaging in any learning activity during the pandemic was 45 percent in Guatemala, 21 percent in Costa Rica, and 18 percent in the Dominican Republic, compared to the average of 23 percent in Latin America (Figure 37). The impact of these school closures on learning outcomes and school dropout are discussed in more detail in Chapter 2.23 Figure 36: School Attendance Rate in Central American UMI Countries during the Pandemic in Central America, 2020 and 2021 100% 97% 86% 80% 60% 40% 20% 0% Dom. Rep. Guatemala LAC Costa Rica -20% -12% -40% Before pandemic After pandemic Percent difference Source: World Bank, High Frequency Phone Surveys, Washington DC. Data for the first wave corresponds to the period between February 2020 and May/July 2021. Figure 37: School Disengagement Rate in Central American UMI Countries during the Pandemic in Central America, 2020 and 2021 60% 250% 150% Disengagement Rate 40% Percent Difference 50% 23% 20% -50% 0% -150% Costa Rica Guatemala Dom. Rep. LAC 2021 2020 Percent difference Source: World Bank, High Frequency Phone Surveys, Washington DC. Data for the first wave corresponds to the period between February 2020 and May/July 2021. 22 World Bank, 2022b. 23 Countries faced many challenges during the provision of remote learning. For a case study on how the implementation of remote learning worked in the Dominican Republic, see World Bank, 2022a. 30 Human Capital Accumulation and Utilization in UMI Countries in Central America Central America Human Capital Review  | Promoting more and better investments in human capital Figure 38: Social Public Expenditures as % of GDP and as % of Total Expenditure in Central America, 2015–2021 Panel A: Costa Rica as a % of GDP Panel B: Costa Rica as a % of Public Expenditure 8% 35% Public Expenditure as % of GDP 7% 30% % of Total expenditure Public Expenditure as 6% 3.0% 2.9% 2.9% 2.9% 25% 12.4% 12.1% 2.7% 2.7% 2.8% 12.1% 11.2% 5% 10.3% 11.3% 10.4% 20% 4% 0.7% 0.7% 0.7% 0.6% 0.6% 0.6% 0.7% 15% 2.9% 2.9% 2.8% 2.5% 2.3% 2.4% 2.7% 3% 2% 3.7% 10% 3.6% 3.6% 3.6% 3.5% 3.6% 3.4% 15.0% 15.0% 14.9% 14.6% 13.3% 14.0% 12.5% 1% 5% 0% 0% 2015 2016 2017 2018 2019 2020 2021 2015 2016 2017 2018 2019 2020 2021 Panel C: Guatemala as a % of GDP Panel D: Guatemala as a % of Public Expenditure 8% 50% Public Expenditure as % of GDP 7% % of Total expenditure Public Expenditure as 6% 2.5% 40% 10.1% 9.4% 10.2% 10.1% 10.3% 15.4% 1.4% 1.3% 5% 1.3% 1.4% 1.4% 30% 8.9% 8.3% 8.5% 9.4% 12.3% 4% 1.2% 1.3% 1.4% 1.7% 9.0% 1.2% 1.1% 20% 3% 2% 23.1% 23.1% 3.0% 3.0% 3.1% 3.2% 3.3% 3.1% 10% 22.6% 21.9% 20.3% 21.9% 1% 0% 0% 2016 2017 2018 2019 2020 2021 2016 2017 2018 2019 2020 2021 Education Health Social Protection Source: Authors’ elaboration based on central government expenditure data from the Instituto Centroamericano de Estudios Fiscales (ICEFI) and GDP for Guatemala from IMF (World Economic Outlook Database, April 2022). Despite the significant worsening of human capital outcomes stemming from the pandemic, overall government spending on the delivery of social services has remained stagnant or even reduced in some countries. During the first year of the pandemic, social public expenditure, which encompasses educational, health, and social protection, modestly increased in Costa Rica both as a percentage of GDP and as a percentage of total public expenditure. Most of it, however, was invested in programs to help households mitigate the socioeconomic impacts of the pandemic (Figure 38). Although spending of the social expenditure in Costa Rica quickly returned to longer-term trends in 2021, some of the education budget was still being used to support the delivery of health services, including the COVID-19 vaccines. The patterns of public spending during the pandemic in Guatemala were slightly different. In 2020, the public spending allocated to the provision of social assistance programs almost doubled from 1.4 percent to 2.5 percent, while public investments in education and health remained stagnant. Public spending allocated to education, on the other hand, decreased in 2021, but its share in the total expenditure increased, suggesting an interest in supporting the learning recovery agenda. In 2021, there was also a large reallocation of public investments in Guatemala from social assistance programs to the provision of health services. In summary, the pandemic led to a sharp decline in human capital at the three critical stages of the life cycle examined in this review. This was due to sharp reductions in critical inputs to child development such as early nutrition, access to health services, and school closures. The decrease in household incomes and the lockdown measures made it difficult for households to access markets leading to an increase in food insecurity in most countries. The lengthy school closures are expected to translate into learning losses and to larger inequalities. The pandemic’s effects on school dropouts and students’ return to school after the reopening of schools is also a concern that will be discussed in more detail in Chapter 2. Youth were disproportionately affected in terms of job loss, and most did not have the option to continue education given the school closures. Whether or not these negative impacts will translate into a permanent reduction of levels of human capital will depend on their magnitude and, perhaps more importantly, the rate at which human capital accumulates in the future. Youth under the age of 25 today were most affected by the erosion of human capital and will make up 90 percent of the prime-age workforce in 2050.24 24 World Bank, 2022c. 31 CHAPTER 2 IMPROVING HUMAN CAPITAL ACCUMULATION THROUGH BETTER EDUCATION OUTCOMES HOS ESCUELA UNIVERSIDAD ESCUELA GUARDERÍA 32 Central America Human Capital Review  | Promoting more and better investments in human capital Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital C hapter 2 presents a diagnostic of the challenges in building foundational learning for all, with a focus on school-age children. Without foundational learning, students often struggle to complete school and, consequently, face difficulties as young adults in the transition to the labor market. The results of this report show that despite the considerable improvements in access to education over the last decade, challenges persist to enroll children in preprimary education. For example, students enrolled in preprimary and secondary education disproportionately come from better-off households, perpetuating already significant income inequalities in Costa Rica, the Dominican Republic, Guatemala, and Panama. Additionally, secondary education dropout rates in some of the countries in this report are among the highest in Latin America and represent an important constraint to human capital accumulation among youth. Among girls, teenage pregnancy is an important contributor to school dropout.25 Overall, learning outcomes are very low and have largely stagnated between 2013 and 2019. Although poorer and predominantly indigenous areas have lower learning outcomes, the largest differences were observed within classrooms. The school closures resulting from COVID-19, which were among the longest in the world, further eroded learning losses in the four UMIs countries in Central America. If these learning losses are not mitigated, they will likely lead to significant decreases in future earnings and productivity for an entire generation. Low and unequal access in preprimary and secondary education Since 2010, the UMI countries in Central America have made considerable progress in access to education, but challenges persist in enrolling children in preprimary school. Although access to primary education is almost universal, enrollment in preprimary education remains overall very low (Figure 39). On average, in 2020, only 6 in 10 children were enrolled in a preprimary school in Costa Rica, the Dominican Republic, and Panama. The enrollment rates are even worse in Guatemala where only 1 in 10 children were enrolled in preprimary school in 2014 (the latest year for which there is data available to estimate preprimary education access in Guatemala). Costa Rica is the only country that made important gains in preprimary net enrollment rates during the last decade, growing from 35 to 57 percent. Panama showed a modest increase in preprimary enrollment rates from 60 percent to 65 percent, while the Dominican Republic and Guatemala show stagnating or declining preprimary net enrollment rates. Low enrollment rates in lower and upper secondary education also create important obstacles in human capital accu- mulation. Except for Costa Rica where the enrollment rate in secondary education is roughly 94 percent, the other three countries face challenges to retain students in the education system. The Dominican Republic and Panama have average secondary enrollment rates of 83 percent and 72 percent, respectively, while Guatemala’s enrollment rates in lower and upper secondary are a mere 65 and 37 percent, respectively. During the past decade, Costa Rica was the only country that was able to make progress in both lower and upper secondary enrollment, while lower secondary enrollment in Guatemala is on a declining trend, falling from 81 percent in 2010 to 65 percent in 2020. 25 Household surveys from Costa Rica and Panama reveal that the most common self-reported cause of dropping out amongst youth between the ages of 12 and 17 was that they were not interested in studying, followed by a lack of economic resources. Poor, rural, and indigenous populations have the highest rates of dropout (Adelman & Székely, 2017). 33 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 39: Net Enrollment Rates by Education Level in Central America (%) , 2010–2020 a) Costa Rica 100.0 99.9 99.5 97.7 98.1 98.5 99.9 99.9 99.9 96.8 100 93.5 93.7 94.1 94.6 96.2 98.1 96.1 91.7 91.7 91.4 93.7 91.7 88.4 90.1 85.8 84.7 80 80.3 80.3 82.8 79.7 Net enrollment rate %) 59.6 62.3 56.7 60 48.2 43.5 42.6 43.4 38.6 39.6 35.1 35.8 40 24.7 23.9 24.9 24.1 25.0 23.8 25.3 25.4 25.8 20 21.4 23.3 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Pre-primary Primary Lower secondary Upper secondary Tertiary b) Dominican Republic 95.4 95.3 100 94.1 93.8 92.0 95.5 95.4 94.7 94.3 96.0 95.9 93.3 93.5 97.9 98.1 98.0 94.3 94.4 91.9 92.1 89.3 91.1 67.6 76.7 76.7 80 73.8 72.5 72.7 74.4 Net enrollment rate %) 75.7 75.4 78.0 73.7 67.1 66.1 74.8 65.4 68.0 64.5 64.3 40 60 24.4 24.0 23.8 26.0 23.3 20.4 22.7 20 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Pre-primary Primary Lower secondary Upper secondary Tertiary c) Guatemala 100 95.5 92.5 90.2 88.3 88.0 87.3 88.3 89.4 89.3 89.4 81.1 77.1 80 72.9 72.6 72.2 71.6 69.0 Net enrollment rate %) 66.9 67.2 65.3 60 38.4 39.0 41.7 41.4 40.7 41.2 41.6 41.7 40.8 37.0 40 23.8 14.0 20 9.1 7.0 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Pre-primary Primary Lower secondary Upper secondary Tertiary Source: Enrollment rates in primary, lower secondary and upper secondary are from UNESCO Institute for Statistics, Enrollment rates in preprimary and tertiary are attendance rated from Socio-Economic Database for Latin America and the Caribbean (SEDLAC) . Enrollment in preprimary is defined as the share of 3- to 5-years-olds attending any educational institution. Children attending in preprimary and secondary education disproportionately come from wealthier households, perpetuating income inequalities. Inequality in access to education can perpetuate income inequalities by limiting oppor- tunities for upward socioeconomic mobility. Without equitable access to quality education, individuals from disadvantaged backgrounds may struggle to acquire the skills and knowledge necessary to secure well-paying jobs, leading to entrenched 34 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 39: Net Enrollment Rates by Education Level in Central America (%) , 2010–2020 (continuation) d) Panama 96.8 95.9 95.2 100 93.6 91.3 89.8 86.8 89.5 89.5 91.6 88.7 90.4 87.8 87.5 80 63.6 64.1 65.0 Net enrollment rate %) 60.3 62.6 63.6 63.2 63.2 64.7 60.4 60 67.3 66.6 66.8 61.5 63.5 56.4 40 24.7 27.5 21.3 22.4 23.1 24.1 23.3 24.2 19.1 20.3 20 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Pre-primary Primary Lower secondary Upper secondary Tertiary Source: Enrollment rates in primary, lower secondary and upper secondary are from UNESCO Institute for Statistics, Enrollment rates in preprimary and tertiary are from Socio-Economic Database for Latin America and the Caribbean (SEDLAC) . Enrollment in preprimary is defined as the share of 3- to 5-years-olds attending any educational institution. cycles of poverty and widening disparities in income. The gaps in access to preprimary and secondary education by socio- economic status are substantial in many of the four countries covered in this report (Figure 40). The country with the largest gap is Panama, where 89 percent of children from the highest income group attend in preprimary school compared to 52 percent among those from the lowest income group. While preprimary attendance among upper-income groups has risen in Panama over the past decade, it has stagnated among the poorest. Trends are less clear for the Dominican Republic, although the most recent data (2016) show clear gaps between upper- and lower-income groups. Costa Rica is the only country where attendance in preprimary school is relatively high and where there has been a noticeable decrease in the socioeconomic gap in access over the past ten years. In terms of secondary school, attendance has increased in both Costa Rica and Panama, and the gap in access has narrowed, to a greater extent in the former from 34 to 9 percentage points between 2010 and 2020. Although data for the Dominican Republic is not up to date, it suggests some narrowing of the income gap at the secondary level, which is partially attributable to the lack of progress among the wealthier population. High school dropout rates among teenagers The high dropout rate in secondary education is another factor impeding the accumulation of human capital among youth, particularly in Guatemala and Panama. In 2019, Guatemala and Panama had net secondary enrollment rates of 44 and 64 percent, respectively, which are well behind those observed in countries with similar levels of income. Guatemala’s secondary enrollment rate is even ranked the lowest among all countries in Latin America (Figure 41). Secondary enrollment rates in the Dominican Republic (71 percent) are also lower than the Latin American average (78 percent). A staggering 60 percent of teenagers in Guatemala are out of school, most of whom drop out during the transition from primary to lower secondary education (Figure 42). In Panama, 40 percent of teenagers are out of school and have been found to drop out at a similar rate in both lower and upper secondary education. Global evidence suggests that teenagers who drop out prematurely from the school system will, on average, earn less and experience more social and economic challenges than their peers with more years of completed education.26 The number of dropouts has been increasing in Guatemala over the past decade and is disproportionately composed of girls. The share of girls who are out of school has risen from 23 percent in 2010 to 37 percent in 2020, while the share of out-of-school boys rose from 15 percent to 33 percent over the same period (Figure 43). This is worrisome as economies are strengthened by having an educated female population. Investing in the education of young girls also reduces gender inequalities by providing them with more opportunities to earn higher incomes, participate in the decisions that most 26 Patrinos and Psacharopoulos, 2004; Oreopoulos and Salvanes, 2011; Bentaouet-Kattan and Székely, 2015; Cardenas, De Hoyos, and Székely, 2015. 35 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 40: Net Attendance Rates by Income Quintiles in Central America (%), 2010–2020 a) Costa Rica Pre-primary Secondary 97 97 96 98 98 99 98 99 99 100 100 93 94 90 90 81 80 76 73 80 90 71 68 70 70 67 66 67 70 81 Net enrollment rate Net enrollment rate 61 62 76 77 60 60 71 71 69 65 50 60 50 59 60 60 55 52 40 40 30 37 39 30 20 32 33 32 28 28 20 10 24 10 0 0 13 14 15 16 17 18 19 20 15 16 17 18 19 20 10 11 12 10 11 12 13 14 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 (b) Dominican Republic 100 100 90 88 89 89 90 88 88 90 87 90 85 77 79 80 75 80 69 73 70 70 Net enrollment rate Net enrollment rate 60 60 64 50 50 65 63 60 56 56 55 55 56 40 52 51 40 50 50 30 43 30 20 20 10 10 0 0 10 11 12 13 14 15 16 10 11 12 13 14 15 16 20 20 20 20 20 20 20 20 20 20 20 20 20 20 (c) Panana 100 100 96 95 94 97 98 97 97 98 93 92 90 88 86 88 88 89 90 84 85 83 80 81 80 80 70 70 Net enrollment rate Net enrollment rate 60 60 67 70 63 66 50 50 59 62 61 53 56 56 55 40 52 51 53 51 52 53 52 40 49 49 30 30 20 20 10 10 0 0 10 11 12 13 14 15 16 17 18 19 10 11 12 13 14 15 16 17 18 19 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Poorest Q2 Q3 Q4 Richest Source: SEDLAC (CEDLAS and The World Bank). Net attendance rates in preprimary are estimated as the share of 3 to 5-year-olds attending any educational level. affect them, and build better futures for themselves and their families. Moreover, educating girls also contributes to more stable, resilient societies that give all individuals the opportunity to fulfil their potential.27 Lastly, there is also evidence that girls who receive an education are less likely to marry young, and are more likely to lead healthy, productive lives, which, in turn, has been shown to have a positive impact on child survival rates.28 27 UNICEF, 2021. 28 Pritchett and Sandefur, 2020. 36 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 41: Net Enrollment Rate in Secondary Education in UMI Countries in Central America (%), circa 2019 90 82.4 80.3 80 77.6 70.6 70 63.8 60 50 43.8 40 30 20 10 0 Costa Rica Dominican Guatemala Panama Rest of UMI Rest of LAC Republic Source: World Development Indicators. UMI follows World Bank’s 2022 classification. Figure 42: Dropout Rates for Students by Education Level in Central America (%), 2019 18 16.8 16 14 12 10.9 9.5 10 8.4 Dropout rate %) 8 6.7 6.4 5.7 6 4 2.7 2 0 2019 2019 2019 Dominican Republic Guatemala Panama Primary Lower secondary Upper secondary Secondary Source: Authors’ elaboration using student-level administrative data on enrollment from the Ministers of Education of the Dominican Republic, Guatemala, and Panama. Data for Panama only covers the public sector and is based on Sistema de Estadística MEDUCA (SIDE). The dropout rate is computed as the proportion of students enrolled in a particular year who did not enroll in school the following year. Teenagers between the ages of 13 and 18 report dropping out from school mainly because of either a lack of motivation or financial reasons. In Costa Rica, the Dominican Republic, and Panama, the main reason for dropping out is lack of interest in studying (21, 40, and 29 percent, respectively) (Figure 44). The lack of interest or motivation in studying may arise because of several different reasons, for example, (i) not having acquired foundational skills during primary education, which, in turn, makes it difficult to continue learning, or (ii) a secondary school curriculum and/or teaching instruction methods that are not engaging or are not useful for the transition to the labor market. In contrast to the other three countries, almost half the students in Guatemala report dropping out of school for financial reasons. The decision to drop out of school for financial reasons is often complex and multifaceted. It’s not simply about not having enough money for tuition, school material or school uniforms, but rather includes the opportunity cost of attending school, personal and family obligations, aspirations, etc. Teenage pregnancy contributes substantially to school dropout among young women and leads to lower education attainment in comparison with other women. Less than half of teenage mothers complete secondary education, and only 7 percent complete tertiary education in contrast to around 19 percent of women who waited until adulthood to become mothers (Figure 45).29 These numbers suggest that teenage pregnancy also seriously impedes human capital accumulation. 29 UNFPA, 2022. 37 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 43: Out-of-School Teenagers by Gender and Country in Central America (%), 2010–2020 Costa Rica Dominican Republic 40 40 35 35 30 30 25 25 20 20 15.7 15 15 9.6 10.8 10 8.1 8.0 6.2 6.3 9.1 10 7.7 7.8 6.7 8.8 5.8 8.3 5.5 5.0 4.4 4.1 3.2 3.0 5 2.6 5 0 0 10 11 12 13 14 15 16 17 18 19 20 21 10 11 12 13 14 15 16 17 18 19 20 21 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Guatemala 40 36.8 36.5 34.4 36.1 35.3 35 30.7 31.3 31.3 31.8 30 26.6 32.7 33.2 25 22.9 27.7 30.2 30.3 20 23.7 23.8 24.3 25.1 15 19.2 10 15.0 5 0 10 11 12 13 14 15 16 17 18 19 20 21 20 20 20 20 20 20 20 20 20 20 20 20 Female Male Source: World Development Indicators. Out-of-school teenagers are defined as youth belonging to lower secondary school age group who are not enrolled in school. Figure 44: Reported Reasons for Dropping out of School among 13- to 18-Year-Olds in UMI Countries in Central America (%), circa 2010 and 2020 Costa Rica Dominican Republic Other 38.9% Other 9.0% 19.2% 17.5% Can't afford studies 18.9% Family obligations 15.9% 15.4% 14.7% Lack of interest 20.5% Lack of interest 39.9% 34.5% 41.1% Finds studying difficult 11.9% Lack of money 11.6% 12.5% 9.8% Work 8.9% Physical or mental disability 15.0% 16.3% 6.6% Pregnancy or marriage 1.0% Work 8.6% 2.1% 10.3% 0% 10% 20% 30% 40% 50% 0% 10% 20% 30% 40% 50% 2020 2010 2020 2010 Guatemala Panama Lack of money 49.5% 29.0% 30.8% Lack of interest 39.0% Lack of interest 17.6% 28.1% Lack of money 24.5% 24.0% Housework 7.2% 10.9% 4.7% Pregnancy 12.8% 4.1% Work 19.6% 6.4% 1.6% Work Pregnancy 12.6% 0.7% 11.3% Other 35.4% Other 10.1% 20.3% 0% 10% 20% 30% 40% 50% 60% 0% 10% 20% 30% 40% 50% 2014 2006 2019 2010 Source: Authors’ elaboration based on Guatemala’s 2006 and 2014 Living Conditions Surveys, Costa Rica’s 2010 and 2020 National Household Survey (ENAHO), Panama’s 2010 and 2019 Household Surveys (EH), and the Dominican Republic’s 2010 and 2020. 38 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital In Panama, 11 percent of girls reported pregnancy as the cause of dropping out from secondary school in 2020, represent- ing a 2 percent increase since 2010. Teenage pregnancy is lower among girls in Costa Rica (2 percent) and Guatemala (3 percent) but is still a statistically significant cause of dropout. The Dominican Republic has the highest adolescent fertility rate among all countries in Latin America, but there is no available microdata that might give insight into how this contrib- utes to school dropout among girls. Repetition rates in primary education are high in most countries included in this report and may contribute to the high dropout rates in secondary education. Except for Costa Rica, the share of primary school students who repeat a grade is very large for UMI countries. In Guatemala, this Figure is 9 percent, while in the Dominican Republic and Panama it is 5 percent and 3 percent, respectively (Figure 46). The net effect of grade repetition on dropouts, however, is not necessarily clear as it may encourage student learning and prevent school dropout or discourage students and promote school dropout.30 It is known, on the other hand, that high grade repetition rates require significant investments in education, which may be problematic in Central America, where public investments in human capital are already very low. Figure 45: Schooling Attainment for Teenage and Figure 46: Repetition Rates in Primary School (% of Adult Mothers in Latin America, circa 2019 Total Enrollment), circa 2019 100 30 90 32.7 25 80 47.3 Repeaters, primary, total 70 20 % of total enrollment) % of mothers 60 15 50 40 47.6 10 Guatemala 30 45.7 Dominican Republic 5 20 Costa Rica Panama 10 7 18.9 0 6 7 8 9 10 11 12 13 0 Teenage mothers Adult mothers -5 Tertiary Secondary Primary Log of GDP per capita PPP $) Source: UNFPA LACRO, based on the Modelo de Impactos Laborales, Educa- Source: World Development Indicators. GDP per capita in 2017 PPP $. tivos, en la Nómina y Asistenciales (MILENA) studies in Argentina, Colombia, Ecuador, Guatemala, Guyana, Mexico, Paraguay, Peru, Panama, and the Domin- ican Republic. A learning crisis Learning outcomes in the UMI countries in Central America are very low and threaten efforts to build human capi- tal. The pre-pandemic results in 2019 of the Estudio Regional Comparativo y Explicativo (Regional Comparative and Explanatory Study, ERCE), a regional student learning assessments, show that learning outcomes in the four countries were very low, with the exception of Costa Rica (Figure 47).31 In the Dominican Republic, a disconcerting 70 percent of third graders did not reach the minimum reading proficiency level for their age. This means that these children were not only unable to read and understand a simple, age-appropriate text, but that they lacked the skills needed to acquire the foundational learning that school is meant to provide. In Guatemala and Panama, an average 60 percent of third graders do not reach the minimum reading proficiency level. In contrast, in Costa Rica, only 20 percent of third graders do not reach the minimum reading proficiency level. While the students’ learning levels in Costa Rica are much higher than the average in Latin America, these are still quite low compared to the average in the OECD countries.32 30 Behrman and Anil, 1991. 31 The regional student learning assessments implemented in Latin America and the Caribbean are called ERCE. ERCE measures standardized learning outcomes of students in 3rd and 6th grades covering reading, mathematics, and natural sciences. The latest round was in 2019 and implemented in 16 countries. The results were published by UNESCO in 2021. 32 The results from the 2019 Programme for International Student Assessment (PISA) show that 34 percent of teenagers in Costa Rica do not reach the minimum learning levels in reading, math, and sciences compared to an average of 13 percent among OECD countries. 39 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Between 2013 and 2019, learning levels in the Central American subregion have largely stagnated. The only country that shows statistically significant gains in third grade reading and math learning outcomes is the Dominican Republic, however these gains are very small in magnitude and had a very low baseline (Figure 48). Panama and Guatemala test scores in the regional learning assessments in both academic disciplines for third and sixth graders decreased between 2013 and 2019, and those for Costa Rica remained flat during this period. Figure 47: Percentage of Third Grade Students below Figure 48: Learning Gains in Reading and Math for the Minimum Proficiency Level in Reading in Central Third Grade Students in Latin America, 2013 and 2019 America (%), 2019 90 84 84 30 83 Peru 80 Dom. Rep. 73 20 Brazil 70 Score difference in mathematics 61 Improvement in 59 math and Nicaragua Paraguay 60 10 worsening in Improvement in 2019 vs. 2013) 50 46 % of students reading reading and 40 0 30 25 Guatemala Panama -10 20 Improvement in 10 -20 Worsening in reading and reading and worsening in 0 math math Dominican Guatemala Panama Costa Rica Argentina -30 Republic -40 -30 -20 -10 0 10 20 30 40 Grade 3 Grade 6 Score difference in reading 2019 vs. 2013) Source: ERCE, 2019. Source: World Bank (2022) based on 2013 TERCE and 2019 ERCE regional standardized learning assessments. Countries appearing in the graph are only to those with statistically representative differences in learning outcomes. Poor and indigenous students often fail to thrive in school and face significantly more challenges in acquiring foun- dational learning and accumulating human capital. The socioeconomic background of students is highly correlated with student learning outcomes. Poorer students in the four countries studies consistently rank among the lowest performers in the regional student learning assessments. The learning gaps by socioeconomic status among UMI countries in Central America are, moreover, much larger than the average observed in Latin America. This suggests that their education systems suffer greater inequalities and are less effective in promoting foundational learning for all (Figure 49). Similar inequalities in student learning outcomes can be observed between indigenous students and their non-indigenous peers, with indigenous students consistently behind in their learning levels (Figure 50). Figure 49: Learning gap in Grade 6 ERCE Score Figure 50: Learning gap in Grade 6 ERCE Results for Differences between the Lowest and Highest Income Indigenous and Non-Indigenous Students in Central Quintile in Central America, 2019 America, 2019 Latin America -100 Latin America -55 Dominica Republic -91 Costa Rica -94 Costa Rica -111 Panama -91 Guatemala -126 Panama Guatemala -47 -133 -160 -140 -120 -100 -80 -60 -40 -20 0 -160 -140 -120 -100 -80 -60 -40 -20 0 Source: ERCE, 2019. Statistics reported are the average scores for Reading, Math, Source: ERCE, 2019. Statistics reported are the average scores for Reading, Math, and Science. and Science. 40 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 51: Learning Poverty in UMI Countries in Central America, circa 2018 90 80.7 80 70 67.3 66.6 60 54.4 50 40 34.2 32.5 30 20 10 0 Costa Rica Dominican Guatemala Panama Rest of UMI Rest of LAC Republic Source: World Development Indicators. Harmonized Learning Test Scores are based on TERCE, 2013. Learning poverty is a major contributor to human capital deficits and undermines sustainable growth and poverty reduction. Learning poverty is a combined measure of learning and the number of children out-of-school by the age of 10.33 High rates of learning poverty are an early sign that education systems are failing to ensure that all children enroll in school and develop critical foundational skills (Box 2). This makes it much harder for children to acquire the technical and higher-order skills needed to thrive in increasingly demanding labor markets. It also impedes countries from developing the human capital needed for sustained economic growth. Before the pandemic, 80 percent of children in the Dominican Republic were either out of school or could not read or understand a simple text by the age of 10 (i.e., were learning poor) (Figure 51). In Guatemala and Panama, almost 70 percent of children were learning poor before the pandemic, while in Costa Rica only 30 percent children were learning poor. Except for Costa Rica, these levels of learning poverty are very high and much larger than the average for UMI countries (34 percent) and for the Latin America region (54 percent). The contribution of learning and the out-of-school population to the learning poverty levels in each of the four countries varies. Costa Rica and the Dominican Republic have lower levels of students who are not in school (0.1 percent and 4 percent, respectively). However, the Dominican Republic has a very low percentage of students in school who are proficient at reading (20 percent) compared to Costa Rica (68 percent). While both Guatemala and Panama have similar percentages of students in school that are proficient in reading (37 percent and 38.5 percent respectively) and similar levels of students out of school (11 percent and 13 percent, respectively). Learning poverty varies substantially within countries, with poorer and predominantly indigenous provinces and departments having the largest learning poverty rates. Access and quality of education varies substantially at the subna- tional levels in all four UMI countries in Central America. Overall, in the Dominican Republic, Guatemala, and Panama, poorer areas are also associated with lower access and lower quality of education services (Figure 52). In Guatemala and Panama, access and quality of education is also significantly lower in areas with a large share of indigenous populations (Figure 53 and Figure 54). Low learning outcomes also an contribute significantly to student dropout rates. In Panama, the dropout rate in public schools is overwhelmingly concentrated among those in the bottom quintile of the learning distribution. Nearly a third of these students drop out compared to 2 percent in the top quintile (Figure 55). In Guatemala, 20 percent of students in the bottom quintile drop out compared to 5 percent in the top of quintile (Figure 56). These numbers highlight the fact that without foundational learning, students often fail to thrive in school. 33 World Bank et al., 2022. 41 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Box 2. Learning Poverty The learning poverty indicator measures the share of children who are unable to read and understand a simple text by the age of 10, adjusted by the share of children who are out-of-school and therefore considered as not proficient in reading. This metric was developed by the World Bank and the UNESCO Institute for Statistics in 2019 to measure of the capacity of education systems to provide universal access to education and to deliver foundational learning. High rates of learning poverty are an early sign that education systems are failing to ensure that children develop critical foundational skills. All foundational skills are important, but learning poverty is focused on reading because: (i) reading at age 10 is an easily understood measure of learning; (ii) reading is a gateway to learning in every other subject; and (iii) reading proficiency can serve as a proxy for foundational learning in other subjects and for the quality of the system. The learning poverty indicator also helps illustrate progress toward the broader goal of Sustainable Development Goal (SDG) 4 to ensure inclusive and equitable quality education for all. It highlights progress towards SDG 4.1.1 (b), which specifies that all children at the end of primary reach at least a minimum proficiency level (MPL) in reading, and SDG 4.1.4, which is linked to the share of primary-school-aged children out of school. The high rate of learning poverty was a rising problem before the pandemic and will likely deepen in the coming years. Additionally, disruptions to schooling caused by the pandemic are expected to disproportionately affect the Latin American region. According to the latest simulations, this region will experience the largest relative increase in learning poverty (26.7 percentage points) in the world and the number of children who are learning poor will reach 79 percent by 2022. a) Learning Poverty in Costa Rica who are out 0.1% of school Percentage children at the end of who are in enrolled 99.9% primary education in school Percentage of children at the end 68% 32% of primary education who are proficient who are not proficient who are in enrolled in at reading at reading school Percentage 60% 40% children at the end who are not learning poor who are learning poor of primary education b) Learning Poverty in the Dominican Republic who are out Percentage of school children at the end of primary 96% who are in enrolled 4% in school education Percentage of children at the end of primary 20% 80% who are proficient who are not proficient education who are at reading at reading in enrolled in school Percentage children at the 18% 82% end of primary who are not who are learning poor education learning poor 42 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Box 2. Learning Poverty (continuation) c) Learning Poverty in Guatemala Percentage children at the who are in enrolled 89% end of primary in school education 11% who are out Percentage of school of children at the 37% 63% end of primary who are proficient who are not proficient education who at reading at reading are in enrolled in school Percentage children at the 33% 67% end of primary who are not who are learning poor education learning poor d) Learning Poverty in Panama Percentage children at the who are in enrolled 87% end of primary in school education 13% who are out Percentage of school of children at the end of 38.5% 61.5% who are proficient who are not proficient primary at reading at reading education who are in enrolled in school Percentage children at the 34% 66% end of primary who are not who are learning poor education learning poor • Proficient at reading is a metric of students who can read at least at the minimum proficiency level for his age, as defined by the Global Alliance to Monitor Learning (GAML) in the context of the Sustainable Development Goal 4.1.1. • Learning poverty is a metric of the lack of minimum reading skills by 10-years-old students adjusted for those who are out-of-school, capturing therefore education access and quality in one single measure. Source: World Bank, 2019b; Azevedo et al., 2021; Azevedo et al., 2022. 43 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 52: Learning Poverty in the Dominican Republic Figure 53: Learning Poverty in Panama by Province, by Province, 2017 2018 [0.43,0.47] [0.47,0.48] [0.48,0.51] [63,90] [54,61] [35,47] Source: Authors’ elaboration based on the percentage of third graders who scored Source: Authors’ elaboration based on the percentage of third graders who scored below level 2 on the 2017 national reading assessments and the percentage of below level 2 on the 2018 CRECER reading test and the percentage of 10-year- 10-year-old children who did not attend school according to the 2010 Population old children who did not attend school according to the 2010 Population Census Census data from IPUMS. data from IPUMS. Figure 54: Learning Poverty in Guatemala by Constraints to better education Department, 2018 outcomes The latest regional student assessments highlight some of the key factors that contribute to better learning outcomes. Students who achieve higher learning outcomes are those who: (i) have parents with high expectations of their learning potential, (ii) have benefited from preschool, (iii) spend more hours studying, (iv) have parents that are involved in their education, (v) have high attendance rates, and (vi) have low repetition rates (Figure 57). It has been shown that teachers can also positively influence learning outcomes of their students by: (i) preparing their lessons ahead of time, (ii) supporting students in their learning process, and (iii) having high expectations of students’ 35.3 37.3 40.1 42.5 45.7 50.8 learning potential. Source: Authors’ elaboration based on the percentage of third graders with ‘unsatis- factory’ and ‘needs to improve’ performance levels on the 2014 DIGEDUCA test From an education policy perspective, increasing and 10-year-old children who did not attend school according to the XII National preschool enrollment, teacher effectiveness, better Population Census and VII National Housing Census collected in 2018 by the National Institute of Statistics. investments in education, and school management capacity can contribute to achieve better learning outcomes. First, children who do not attend preschool are often unprepared when they begin primary school, putting them on a lower learning trajectory. Second, teachers often lack the skills or motivation to be effective. Education systems are not always able to attract applicants with strong backgrounds to the teaching profession nor provide incentives that would allow teachers to grow. Addi- tionally, pre-service and in-service training programs tend to be very weak and result in teachers having insufficient subject knowledge and/or pedagogical skills. Third, public 44 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 55: Dropout Rates for Students in Public Figure 56: Dropout Rates for Students in Grades 6 and Schools, Grades 1 to 12, by Learning Quintiles, 9 by Learning Quintiles, Guatemala (%), 2019 Panama (%), 2019 35 30 31.22 30 25 25 20 19.35 % of students % of students 20 15.53 15 14.97 15 10 9.91 10 6.89 5.04 5 3.74 5 2.73 1.95 0 0 2019 2019 Quintile 1 [GPA <3] Quintile 2 [GPA 3 to 3.5] Quintile 3 [GPA 3.5 to 4] Quintile 4 [GPA 4 to 4.5] Quintile 5 [GPA 4.5 to 5] Source: Authors’ elaboration using student-level administrative data on enrollment from MINEDUC for school years 2019 and 2020. The dropout rate for each year is computed as the proportion of students enrolled in the level who did not enroll in any school in the following year. Quintiles of school learning are computed using the national student standardized assessments for grade 6 (2013), and grade 9 (2019). investments in education are limited and are often not used efficiently, which results in a lack of education inputs or inputs that do not reach schools when needed. Lastly, poor management and governance at school significantly impact learning. Although effective school leadership does not translate into learning directly, it helps improving teachers’ quality and ensuring an effective use of resources. Targeted instruction will be necessary to accelerate learning and narrow learning gaps. The analysis of the regional student learning assessments implemented in 2019 show that, in Guatemala and Panama, nearly half the variance in third grade math scores is explained by differences within classrooms while the other half is explained by differences between schools (Figure 58). In the Dominican Republic, differences within classrooms explain 45 percent of the variance in learning Figure 57: Associated Factors Affecting Grade 3 ERCE Figure 58: Variance in Grade 3 ERCE Results in Math Results in Reading in Latin America, 2019 Explained by Differences within Classrooms in Latin America, 2019 Educational 87 60 69 expectations of parents 46 59 50.7 42 49.1 Factors associated with students and their families Socioeconomic 44 50 level of the family 43 44.7 44.3 50 Variance %) 51 Attended preschool 18 0 40 30 34.9 Days spent studying 51 59 per week 52 30 30 Parental involvement 22 17 in learning 8 27 20 0 Absence from school -11 -20 -24 10 -72 Repetition -67 -66 -89 -100 -50 0 50 100 0 Costa Rica Dominican Panama Guatemala Latin Republic America Panama Guatemala Dominican Republic Costa Rica Source: ERCE, 2019. Reported values correspond only to regression coefficients Source: ERCE, 2019. that are statistically significant at standards levels. 45 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital outcomes grade 3 math scores, while in Costa Rica only a third of the of the variance is explained by within classroom differences. These results indicate that in all four countries, it is common to have classrooms with students who have vastly diverse levels of learning. This highlights the important role of structured pedagogy and targeted instruction to accelerate learning and to reduce inequalities in the delivery of education services.34 Impacts of the COVID-19 pandemic on education outcomes and governments’ mitigation strategies School closures in Central America resulting from the COVID-19 pandemic were among the longest in the world and have caused an unprecedented disruption in the delivery of education services. Between March 2020 and March 2022, schools in Panama and Costa Rica were fully closed for an average of 55 and 43 weeks, respectively (Figure 59). These numbers place these two countries among the top in the global ranking of the number of weeks schools were fully closed during the pandemic. In the Dominican Republic and Guatemala, schools were fully closed for 33 weeks, which is close to the average of 30 weeks among countries in Latin America. School closures in all four countries were, however, much longer than the 24-week average observed among UMI countries at the global level. Figure 59: School Closures in Number of Weeks for Central America, March 2020 to March 2022 100 86 87 90 82 80 70 62 Number of weeks 60 55 53 55 53 50 43 41 39 40 33 33 32 32 30 29 30 22 24 21 20 20 10 - Dominican Costa Rica Guatemala Panama Latin America Upper Middle Global Republic Income Partially opened Fully closed Fully or partially opened Source: UNESCO Global monitoring of school closures caused by COVID-19 Dashboards. Estimates are weighted by the number of students from preprimary to upper secondary education in each country. Simulations using observed data on the length of school closures project large learning losses that will further erode the weak foundations of the education systems in the Central American subregion. Using different assumptions related to the effectiveness of remote learning, the expected learning losses in the four countries in terms of learning-adjusted years of schooling (LAYS) range from 1.4 (Dominican Republic) to 2.3 (Costa Rica) in the pessimistic scenario, and from 0.8 (Dominican Republic) to 1.3 (Costa Rica) in the optimistic scenario (Figure 60).35 The expected learning losses are much larger in absolute terms in Costa Rica compared to the other countries because its education system ranks among the top in Latin America, meaning that school closures have a much larger negative impact on students’ learning. The share of third graders below minimum proficiency level (MPL) in reading is projected to increase from 59 to 79 percent in Panama, from 25 to 35 percent in Costa Rica, from 61 to 76 percent in Guatemala, and from 73 to 82 percent in the Dominican Republic. This is in contrast with the expected average increase from 37 to 50 percent among countries in Latin America (Figure 61). The few studies that measure actual learning losses in Latin America do not share the same level of statistical rigor, but all suggest learning losses are real. Two robust studies from Brazil show that the COVID-19 pandemic has caused pronounced learning losses. The first study used standardized test scores to compare learning before and after school closures 34 Abdul Latif Jameel Poverty Action Lab (J-PAL). 2019. 35 The scenarios used for the simulations are described in detail in World Bank, UNICEF, and UNESCO (2022). 46 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 60: Simulated Learning Losses due to COVID-19 in Central America (in Learning-Adjusted Years of Schooling), 2022 10.00 9.0 9.00 8.00 7.8 7.71 Learning-Adjusted Years 7.00 6.3 6.5 6.6 6.85 6.72 of Schooling LAYS) 6.00 5.78 5.99 5.28 5.58 5.00 5.22 4.00 4.38 4.39 3.00 2.00 1.00 0.00 Guatemala Panama Dominican Republic Costa Rica Latin America Baseline Optimistic Intermediate Pessimistic Source: World Bank, UNICEF, and UNESCO, 2022, which uses data from the 2019 ERCE, UNESCO’s calendar, and UNICEF’s monthly monitoring of school closures. and found significant learning losses for all assessed grades in math and reading. The learning losses are larger for primary education students. Even if the learning returns to its pre-COVID-19 trajectory (2011–2019), the study found that it would still take between four and fifteen years to recover learning losses.36 The second study concluded that remote learning caused setbacks in learning equivalent to 75 percent of in-person learning. Other studies for Colombia and Mexico show significant learning losses as well, particularly among disadvantaged groups. Lastly, there are studies that show that younger students and early grades have also been disproportionally affected. The evidence presented in all these studies indicate that similar learning losses can be expected in the four UMI countries in Central America. Figure 61: Simulated Learning Losses Due to Figure 62: Effect on Average Annual Income per COVID-19 School Closures: Third Graders below MPL Student in Central America, 2019 in Reading in Central America, (%), 2019 90 Dominican 80 Republic Guatemala Panama Costa Rica 80 73 73 71 0 70 61 59 60 -5 % of third graders -4 Change in annual income %) -5 -6 50 46 -10 -7 -8 -9 -10 40 37 32 30 -15 -13 25 -16 -16 20 -20 10 -21 -25 0 -24 Costa Rica Dominican Guatemala Panama Latin America Republic -30 Baseline Optimista Optimistic Intermediate Pessimistic Source: World Bank, UNICEF, and UNESCO, 2022, using data from the 2019 Source: World Bank, UNICEF, and UNESCO, 2022. ERCE, UNESCO’s calendar, and UNICEF’s monthly monitoring of school closures. If learning losses due to school closures are not mitigated, there will be significant decreases in future earnings and productivity for an entire generation of students. Learning loss in the four countries can also be quantified in terms of the students’ lifetime earnings using information on returns to schooling, life expectancy, and labor market variables. This exercise suggests that the average student in Costa Rica today would lose a minimum of $1,140 (2017 PPP) in average annual earnings, which is equivalent to $14,191 in lifetime earnings or a 16 percent loss in projected lifetime earnings (Figure 62). 36 World Bank, UNICEF, and UNESCO, 2022. 47 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital The average student in the Dominican Republic today could lose $565 (2017 PPP) in average annual earnings, which is the equivalent of $9,178 in lifetime earnings or a 4 percent loss in projected lifetime earnings. In the optimistic scenario in Panama and Guatemala, the average student would lose the equivalent of 8 and 7 percent of projected lifetime earnings, respectively. These results suggest that in addition to short-term learning losses, UMI countries in Central America will likely suffer significant long-term losses in human capital accumulation if learning losses are not mitigated. While the four countries developed strategies to miti- Figure 63: Change in Dropout Rates by Education gate learning losses and continued delivering education Level between 2019 and 2021 in Central America, (%) services remotely during school closures, the imple- a) Dominican Republic mentation challenges were significant. All four coun- 20 tries developed multimodal strategies for the provision of 14.54 15 distance education services. These included developing 10 Percentage change partnerships with the private sector to improve internet 5 1.08 1.13 connectivity and accessibility to digital devices as well as 0 the development of printed and digital educational mate- -5 -1.36 -1.36 -2.17 -1.21 -2.00 rial to accompany lessons delivered by radio and TV. Most -10 of the countries made extensive use of social networks plat- -15 -20 -16.31 forms to reach students and teachers including WhatsApp Initial Primary Secondary and short message services (SMS) via cell phones. More- 2019-2018 2020-2019 2021-2020 over, most governments scaled up support to teachers and b) Guatemala encouraged the involvement of parents and caregivers, as their participation in children’s learning during lockdown 20 15 was essential. Despite these efforts, the delivery of remote 10 Percentage change distance learning services faced many limitations, especially 3.4 3.4 4.0 2.7 4.0 5 2.0 amongst the most vulnerable countries and groups (see 0 Box 3 for a summary of the experience in the Dominican -5 -0.3 -0.4 Republic). These included: (i) limited connectivity in the -10 region; (ii) limited access to devices needed for distance -15 learning, especially amongst vulnerable groups; (iii) chal- -20 Pre-primary Primary Lower Upper lenges in implementing specific response strategies in a secondary secondary timely manner; (iv) lack of teacher preparedness for remote 2020-2019 2021-2020 learning; and (v) institutional constraints. c) Panama 20 The impact of the pandemic on dropout rates was lower 15 than anticipated, and the respective magnitudes and 10 Percentage change 1.5 5 recovery patterns were very different across countries. 0 Detailed administrative data on enrollment in Guatemala, -5 -3 -3 the Dominican Republic, and Panama show that changes -10 in dropout rates varied significantly by education level -15 in each of the countries (Figure 63). In the Dominican Primary -20 Lower secondary Upper secondary Republic, there was a nearly 15 percent increase in the 2020-2019 dropout rate of students in preprimary education (three- to six-year-olds) during the first year of the pandemic. Source: Authors’ elaboration using student-level administrative data on enrollment The recovery was very quick, however, with dropout rates from the Dominican Republic MINERD for 2017/2018 to 2021/2022 school years; Guatemala MINEDUC for 2019 to 2022; and Panama MEDUCA for returning to the pre-pandemic baseline numbers in the school years 2019 to 2021 from the Sistema de Integración de Datos Estadísticos following year. Dropout rates among students in Domin- (SIDE). Data for Panama only covers the public sector. ican primary and secondary schools do not seem to have been impacted during the pandemic. In Guatemala, dropout rates in preprimary and primary schools during the first year of the pandemic were largely unaffected but increased by approximately 3 percent during the second year of the pandemic. In contrast, dropout rates among students in lower secondary and upper secondary schools show an increase of 4 percent during the first year of the pandemic and have not yet returned to pre-pandemic rates. However, throughout the two years of the 48 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Box 3. Implementation of Distance Learning during School Closures in the Dominican Republic To mitigate learning losses, the Dominican Republic implemented a variety of distance education methods to reach students across the country. These included the distribution of printed booklets for home study, the broadcasting of educational programs on television and radio, the delivery of digital devices to students and teachers, and the development of train- ing programs for school principals and teachers on pedagogical and administrative issues, communication tools, and the emotional well-being of their students. An assessment of the distance learning strategies imple- Figure B3.1: Percentage of Students Aged 4-17 Who mented during the pandemic provides interesting insights Watched Distance Learning Programs in Gran Santo into how well they functioned for students and their fami- Domingo and Santiago, 2020–2021 lies as well as for teachers and school principals. Teachers 70% were shown to be strongly committed to remote teaching, 62% 60% and 80 percent of students managed to have access to the 52% educational materials. Additionally, 20 percent of primary 50% students and 40 percent of secondary students had a digi- 40% 39% 35% 33% tal device with which they could connect to their teacher 30% 24% 25% from home. With respect to school principals, 90 percent 28% 30% 32% 20% 24% 25% reported that all or most of their teachers had a digital device and received training remotely. 10% 0% Nov Dec Jan Feb Mar Apr Despite these efforts, the implementation of distance 2020 2021 education presented many challenges. Sixty percent of Pre-Primary & Primary students Secondary students primary school parents reported spending a great deal of time supporting their children. Additionally, access to Source: World Bank, 2021. The statistics were computed using television broad- casting data. educational materials was not consistent. Although most parents reported having access to the booklets, they did not receive all the booklets that were developed to support the distance learning strategy. Student engagement was also difficult to maintain. Television ratings indicate that both the number of viewers and the average viewing time of educa- tional programs declined progressively during the school year. Moreover, 60 percent of parents reported that their children studied fewer than three hours daily, and nearly all parents and school principals believe that children learned less or much less than they would have if they had attended school in person. pandemic, younger children in preprimary and primary show larger dropout rates than those from secondary education.37 In Panama, the pandemic seems to have had very limited impact on dropout rates. During the first year of the pandemic (the only year for which data is available), dropout rates increased modestly for students in primary, while dropout rates for students in lower and upper secondary decreased by 3 percent. The enrollment data for the 2022 school year is still not available but will be important in better understanding the longer-term impacts of COVID-19 on dropouts in Panama. Although dropout rates did not increase as much as anticipated during the pandemic, there was an important crowding out of students from private to public schools. The enrollment shares by sector and education level in the Dominican Republic show that there was a very large shift of students mainly in preprimary and primary education from private to public schools (Figure 64). In Guatemala, there was similar evidence of a large shift of students in preprimary and lower secondary from private to public schools (Figure 65). There are at least three possible explanations behind these trends. First, the pandemic may have created incentives for families to move their children from private to public schools due to 37 Ham et al., 2022. 49 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital the lack of distance learning programs for younger students or the lack of perceived value added of the distance learning for younger students. Second, the economic crisis increased financial pressure on households, potentially making private school tuition fees too difficult to pay. Third, some of the existing social protection programs or new programs created to mitigate learning losses might have created incentives to enroll children in public schools. Figure 64: Enrollment by Sector and Education Level in the Dominican Republic, 2017/2018 to 2021/2022 100% COVID-19 83.6% 82.8% 82.3% 81.5% 84.2% 83.4% 81.0% 80% 77.6% 77.7% 77.3% 71.9% COVID-19 COVID-19 60% 53.5% 53.3% 51.6% 54.7% 40% 46.5% 46.7% 48.4% 45.3% 20% 28.1% 22.4% 22.3% 22.7% 19.0% 18.5% 16.4% 17.2% 17.7% 15.8% 16.6% 0% 2017-2018 2018-2019 2019-2020 2020-2021 2021-2022 2017-2018 2018-2019 2019-2020 2020-2021 2021-2022 2017-2018 2018-2019 2019-2020 2020-2021 2021-2022 PREPRIMARY PRIMARY SECONDARY Public Private Source: Authors’ elaboration using student-level administrative data on enrollment from MINERD. Figure 65: Enrollment by Sector and Education Level in Guatemala, 2019 to 2022 100% 90.2% 90.5% 90% 83.8% 83.5% 87.3% 87.1% 88.3% 88.6% COVID-19 80% 70.0% 70.4% 68.6% 69.4% 70% COVID-19 COVID-19 COVID-19 60% 50% 43.6% 42.6% 44.5% 43.5% 40% 30% 36.2% 37.4% 34.4% 34.8% 22.9% 22.5% 23.7% 22.7% 20% 15.9% 16.2% 12.6% 12.8% 11.6% 11.3% 9.6% 9.3% 19.2% 18.9% 20.1% 20.7% 10% 5.6% 5.6% 6.2% 6.5% 1.0% 1.1% 1.5% 1.5% 1.4% 0% 1.4% 2019 2020 2021 2022 2019 2020 2021 2022 2019 2020 2021 2022 2019 2020 2021 2022 Public Private Cooperative Municipal Source: Authors’ elaboration using student-level administrative data on enrollment from MINEDUC for school years 2019 to 2022. Cooperate schools are secondary schools in which parents have a large role in its management and mainly serving rural communities. Municipal schools also mainly serve rural areas and are focused on providing students with technical and vocational training. In the Dominican Republic, the government invested heavily in purchasing digital equipment for teachers and students in public schools to improve connectivity and the delivery of distance education services during the pandemic. More than one million students from primary and secondary education from public schools received a tablet or netbook (around 50 percent of the total enrollment in public schools), while all teachers from the public education sector benefited from a laptop (table 1). These huge investments seem to have created lots of incentives for students to stay enrolled in public schools or to transfer from private to public schools. 50 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Box 4. The Long-Term Vision of the Role of Education Technologies in Accelerating Learning and Reducing Learning Gaps in Panama On April 6, 2022, Panama approved a far-reaching law called the Digital Equity Law (Ley de Equidad Digital), which was built upon lessons learned during the pandemic. This law establishes the general guidelines for the formulation, develop- ment, and implementation of public policies aimed at increasing the equity of the education system. The law proposes the implementation of pertinent teaching-learning models, including flexible modalities for education services facilitated by technology, which guarantee that students to acquire digital competencies and skills throughout their educational trajectories. The main objectives of this law are the following: • Implement a large-scale technological and digital transformation to accelerate learning among students in all grades. • Progressively provide access to necessary technological infrastructure including energy, connectivity, digital devices, digital content, and education platforms to support pedagogy and accelerate learning among students. • Consolidate progress made during the pandemic towards integrating digital education platforms and content in public schools. • Promote innovations that facilitate pre-service and in-service teacher training to guarantee the pertinence of curricula and the mastery of the new teaching-learning models that are aligned with the technological and digital transformation of the education sector. • Implement a national curriculum for all education levels across public and private schools that includes strengthened digital competencies needed for the development of 21st century skills. • Create a continuous evaluation system that focuses on improving decision making, the efficacy of education plat- forms, learning outcomes, and that measures the impact of this education public policy. The Digital Equity Law is an important regional example of how to build on the lessons learned during the pandemic by focusing on the long term and the sustainability of education policies. This law will be critical in articulating efforts and building public-private partnerships to make sure education technologies translate into better and more learning for students, particularly for those from disadvantaged backgrounds. The law is expected to have an impact on 935,522 students in Panama from preprimary to upper secondary education. Source: Digital Equity Law, Official Gazette of the Republic of Panama, April 6, 2022. In Guatemala, two pre-existing programs might have created incentives for students in preprimary and primary education to stay enrolled or to enroll in public schools during the pandemic. The first one was the student health insurance program, which was created in 2020 to benefit all students enrolled in public preprimary and primary schools. The main objective of this program is to improve access to primary health services among vulnerable children.38 The insurance policy allows students to access private in-network health providers, including doctors, pharmacies, hospitals, and funeral services. The first phase of the implementation of this program was implemented at the peak of the pandemic between 38 This insurance program is fully paid by the government, and it covers health expenses due to accidents (up to GTQ 15,000 or USD 1,945, annually) and selected common health conditions and medicines (up to GTQ 300 or USD 39, annually) at a monthly cost of GTQ 9.50 per student/month (USD 1.20). 51 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Table 1: Programs Implemented during the Pandemic that Might Have Contributed to Student Retention Dominican Republic Guatemala Panama Program Distribution of laptops, tablets, Student Health Insurance School Feeding Programa de Asistencia notebooks, and interactive boards Social Educativa (PASE-U) to students, teachers, and schools Objective Improve connectivity to access Improve access to primary Improve nutrition for Prevent student absenteeism, distance learning. health services among vulnerable students repetition, and dropouts vulnerable students. Covers health expenses due to accidents (up to USD 1,945, annually), selected common health conditions, and medicines (up to USD 39, annually). Beneficiaries Teachers and students serving in/ Students enrolled in public Students enrolled in public Students enrolled in public enrolled in public primary, lower preprimary and primary preprimary and primary and private primary, lower secondary, and upper secondary schools. schools. secondary, and upper schools. (Students: 2,532,062) (Students: 2,532,062) secondary schools. (Teachers: 1,491 laptops; Primary (Students: 760,617) students: 758,783 tablets; Secondary students: 251,431 netbooks; Schools: 7,725 interactive digital boards) Size of the annual N/A Annual cost to the government Preprimary (ages 4 to 5) and Primary : USD 270 benefit in 2021 (in is USD 14.4 (USD 1.20 primary (ages 6 to 11) : USD Lower secondary : USD 360 USD) monthly, per student) 140 (USD 0.78 per student per Upper secondary : USD 450 school day) Paid at the beginning of the school year and to be used to purchase uniforms and school supplies Year created 2021 2020 2017 2010 Coverage and size of benefits Originally named Beca increased in 2019 and Universal, and since 2020, 2021. An additional increase PASE-U planned for 2022. Created in response Yes No No No to COVID-19 Source: Authors’ elaboration. April and December 2020, reaching 1,191,660 students from 11,670 public schools located in 138 urban areas and depart- mental capital municipalities. The second phase was implemented between January and December 2021, and reached an additional 456,360 students. The second program is the school feeding program. This program was well established before the pandemic and was the largest instrument protecting poor and vulnerable children from negative income shocks and food insecurity in Guatemala. During the closure of schools, the quick adaptation of the program from cooked meals to the distribution of uncooked food items allowed the continuity of meals at home, while also giving earning opportunities to local farmers who were the main providers of food. In October 2021, the government modified the program, expanding the coverage to children in early child education (ages 0 to 3) and in secondary education (ages 12 to 17). Additionally, the benefit for students in preprimary (ages 4 to 5) and primary (ages 6 to 11) was increased from GTQ 4.00 (USD 0.52) to GTQ 6.00 (USD 0.78) per student per school day. 52 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Box 5. Summary of the Key Learning Constraints and Outcomes for School-Aged Children Constraints Outcomes Impacts of the COVID-19 Pandemic • School closures • Learning losses • Low effectiveness of distance learning • School dropouts during the pandemic • Adaptation and creation of social measures/programs linked to the delivery of • Temporary crowding out of students from private to public schools social public education services Structural Issues • Limited provision of quality, public preprimary education • Low and unequal access to preprimary education • Financial constraints in accessing quality education services and investing in quality preprimary education • Lack of access to quality preprimary education to start school ready to learn. • High repetition rates in primary education • Lack of accumulation of foundational skills • Low primary education completion rates • Low learning trajectories • Low learning outcomes • Limited use of structured pedagogy to facilitate instruction in the classrooms. • High learning poverty rates • Low standards for entry into the teaching career, poor quality training programs detached from the realities of the classrooms, unattractive teaching career incentives, and weak in-service support for teachers. • Constraints for parents to get involved in their children’s education. • Low capacity among school management and directors and low school autonomy • Low learning trajectories among poor and vulnerable students • Large learning gaps at the subnational level by socioeconomic and • Lack of minimum conditions for enabling learning in schools that serve poor and indigenous status vulnerable students. • Inequitable public education spending • Students with vastly diverse levels of learning in the same classroom • Large learning gaps within classrooms • Limited use of technologies and structured pedagogy for the delivery of education services and for teaching at the right level • Low learning trajectories • High school dropouts in lower and upper secondary education • Teenage pregnancy In Panama, before the pandemic, the government had a well-established conditional cash transfer program called Programa de Asistencia Social Educativa (PASE-U, formerly known as Universal Scholarship program). This program aims to prevent student absenteeism, repetition, and school dropout. Beneficiary students include students enrolled in public and private primary, lower secondary, and upper secondary schools in the country. The program is very generous in terms of the size of its benefits, paying primary students USD 270 per year, lower secondary students USD 360 per year, and upper secondary students USD 450 per year. The money is expected to be used to purchase food, personal hygiene items, medication, school supplies, uniforms, books, or technological tools to support the learning process. The fact that this program equally benefits students from private and public schools might explain why the large crowding out of students from the private to the public sector is not observed in Panama. Recovering and enhancing learning and bringing back students who have dropped out will require strong political will as well as a more effective and efficient public investments in the education sector. To recover learning losses and 53 Improving Human Capital Accumulation through Better Education Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital bring back students who have dropped out during the pandemic, countries would need to focus in at least four critical priorities: (i) curricular consolidation with a focus on foundational and transferable skills; (ii) the assessment of students’ current learning levels to properly diagnose learning losses; (iii) interventions that can help accelerate learning recovery; and (iv) develop strong early warning systems to monitor school dropouts. The curricular consolidation would require teaching practices to be reoriented towards the acquisition of foundational skills. Learning assessments with a focus on formative assessments should be prioritizes as they have the potential to improve students’ learning at a relatively low cost. Pre-ex- isting programs and learning recovery programs developed during the pandemic that produced positive results should be scaled up. Recovering and enhancing learning will also require a strong focus on access to devices and connectivity, which are still very low in the Central American subregion. This would need to be combined with institutional strengthening for the management and integration of education technologies in the education sector (see Box 4 for an example of how Panama is thinking about these issues). In terms of monitoring dropouts, it will be critical to strengthen the capacity of the national education management information systems to develop early warning systems and low-cost interventions to try to address dropouts. 54 CHAPTER 3 DEPLOYING AND ENHANCING HUMAN CAPITAL FOR BETTER EMPLOYMENT OUTCOMES HOSPITAL ESCUELA UNIVERSIDA Central America Human Capital Review  | Promoting more and better investments in human capital 55 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital C hapter 3 presents the challenges encountered in the utilization and enhancement of the human capital, focusing on the school to work transition of young adults and employability. It also aims to map the structural constraints in the labor market, with a particular focus on supply-side constraints and the skills mismatch between supply and demand. The results of this report show that despite improvements over the last decade, a significant share of youth remain out of school and out of employment, reflecting structural challenges in school to work transition, especially for women. In addition, despite increases in educational attainment, supply of relevant skills, particularly among youth remains limited, which, in turn, contributes to limited availability of quality jobs, and poor labor market outcomes. The limited supply of skills is partly driven by inadequate investments in human capital, which is also reflected in poor foundational learning outcomes. The COVID-19 pandemic deepened the employability issues especially for youth and lower-skilled workers who are at risk of long-term scarring in their employment outcomes. Finally, the institutional capacity to address employability challenges remain limited in most countries. This underscores the need for concerted policy action in the short-term to ensure employment among youth and lower-skilled workers, while continuing investments in institutional capacity to better deploy and enhance human capital in the medium and long term. The phenomenon of youth (ages 15–24) who are neither in education, employment nor training, (NEETs) is a critical concern to the accumulation and utilization of human capital in the UMI countries in Central America. Youth is a critical period when the emphasis of human capital starts to shift from accumulation to utilization. When youth are out of school and out of work, they not only stop accumulating human capital but also do not utilize their existing human capital. This has long-lasting negative effects on productivity throughout an individual’s lifetime by lowering wages and employment opportunities, which consequently hampers overall economic growth. Trends over the past decade show that the share of NEETs in Costa Rica, the Dominican Republic, Guatemala, and Panama have been influenced to varying degrees by: (i) persistently high dropout rates especially in secondary school and an increased number of out of school teenagers, and (ii) limited labor market opportunities. It is important to note that these trends differ dramatically between young men and women, highlighting the gender disparities and the need for a differentiated approach. The pandemic further exacerbated existing challenges in the utilization and enhancement of human capital among youth, women, and low-skilled adults, underscoring the need for concerted policy action. NEETs and the trends and challenges in the school to work transition The profile of NEETs varies across the four countries but there are common characteristics, such as education and inactivity. For instance, most NEETs have completed secondary education, except in Guatemala where approximately half have completed no more than primary schooling (Figure 66). Another common characteristic is the fact the majority are inactive rather than unemployed, which means that they are not employed but also are not searching for jobs (i.e., are not participating in the labor market) (Figure 67). A likely contributing factor is that most NEETs live in households with children, which often implicates additional responsibilities at home and therefore less time to seek and maintain employment. Inactivity may also be fueled by the duration of unemployment for youth in Guatemala (8 months) and in the Dominican Republic (4 months), which potentially discourages young people from seeking employment. A persistent common characteristic is that women constitute most NEETs in all four countries, which reflects the additional constraints young women face including childcare responsibilities and teenage pregnancy. Women repre- sent nearly 60 percent of NEETs in Costa Rica and more than 80 percent in Guatemala. As mentioned in the Chapter 2, dropouts disproportionately affect girls in Guatemala and 10 percent of girls drop out from secondary school in Panama because of pregnancy, reflecting some of the additional challenges underlying the high NEET rates among young women. 56 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Multivariate regression results largely confirm these characteristics over time and across three of the four countries (Costa Rica, the Dominican Republic, and Panama) (table A.1). Youth that live in households with young children (0 to 3 years old), are married, or are female have a higher probability of being a NEET, while youth with more years of education and who live in a household where the head of household has more years of education and/or is employed have a lower probability of being a NEET. Figure 66: Education Level of NEETs (15 to 24 years Figure 67: Inactive vs. Unemployed NEETs (15 to 24 old) in Central America, 2019 years old) in Central America, 2019 100 90 Guatemala 80 Percentage of NEETs 70 63 55 Costa Rica 60 75 50 95 40 Dominican Republic 30 20 Panama 10 0 0% 20% 40% 60% 80% 100% Guatemala Dominican Panama Costa Rica Percentage of NEETs Republic 2019 Primary or less Secondary Tertiary Unemployed Inactive Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT), Guatemala 2019 (ENEI), Panama 2019 (EML) using SEDLAC harmonization. Data for Costa Rica 2019 (ECE) provided by the NSO and standardized for this report. The tertiary level includes complete and incomplete education levels. Despite improvements in the past decade, the share of NEETs in the four countries remain substantial. In 2011, the share of NEETs in all countries except Guatemala was more or less on par with the Latin America average of 20 percent (Figure 68). By 2019, the pre-pandemic share of NEETs in both Panama and Costa Rica had decreased to 16 and 17 percent, respectively, which was below the Latin American average of 21 percent. This decrease can be largely attributed to improvements in opportunities for young women (Figure 69). On the other hand, the share of NEETs in the Dominican Republic and Guatemala (24 and 28 percent, respectively) both exceed the Latin American average and represent an overall increase in NEETs since 2011. Similarly, to Panama and Costa Rica, the change in NEET rate was mostly driven by trends for women, however in these countries, the trends indicate worsening opportunities for women. Figure 68: NEET Rate (%) in Central America, Total and Figure 69: NEET Rate (%) in Central America, Total and by Gender, 2011 by Gender, 2019 50 46 45 42 45 40 40 35 Percentage of youth 30 Percentage of youth 35 32 30 27 28 30 28 25 24 25 25 21 21 21 25 23 20 20 16 14 16 17 15 15 15 12 11 13 10 9 10 7 5 5 0 0 Total NEETs Male NEETs Female NEETs Total NEETs Male NEETs Female NEETs Panama Costa Rica Dominican Republic Guatemala Panama Costa Rica Dominican Republic Guatemala Source: Authors’ elaboration based on Dominican Republic 2010 (ECNFT), Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT), Guatemala 2011 (ENEI), Panama 2011 (EML), and Costa Rica 2011 (ECE) Guatemala 2019 (ENEI), Panama 2019 (EML) using SEDLAC harmonization. using SEDLAC harmonization. Data for Costa Rica 2019 (ECE) provided by the NSO. 57 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital The persistence of high NEET rates reflects sustained challenges in the school to work transition across the four countries, though there are signs of improvement in some countries. In the Dominican Republic, Panama, and Costa Rica, both girls and boys are staying longer in school (Figure 70). Nearly all 15-year-old youth are now in school, an improvement compared to 2010, particularly for girls (Figure 71, Figure A.2). In Costa Rica and Panama, adolescents also have had slight gains in employment compared to the beginning of the decade (Figure 71, Figure A.1). On the other hand, the school to work transition continues to be challenging in Guatemala and Dominican Republic. In both countries, advances in education are outweighed by modest (Dominican Republic) or no gains (Guatemala) for adolescents and young adults on employment. Figure 70: School to Work Transition by Age in Central America, 2019 a) Costa Rica b) Dominican Republic 100% 100% 90% 90% 80% 80% 70% 70% Youth ages 10 to 35) 60% 60% Youth ages 10 to 35) 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 15 17 19 21 24 26 28 30 32 34 10 12 14 16 18 20 22 24 26 28 30 32 34 Age Age c) Guatemala d) Panama 100% 100% 90% 90% 80% 80% 70% 70% Youth ages 10 to 35) Youth ages 10 to 35) 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 1 3 5 7 9 11 13 15 17 19 21 23 25 10 12 14 16 18 20 22 24 26 28 30 32 34 Age Age Work only School only School and work Not in school or work Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT), Guatemala 2019 (ENEI), Panama 2019 (EML) using SEDLAC harmonization. Data for Costa Rica 2019 (ECE) provided by the NSO. Information for Costa Rica is available from age 15 onwards. In three of the four countries, the improvements in education in the school to work transition were more concen- trated among women than men. In the Dominican Republic, Panama, and Costa Rica, despite a steep decline in school enrollment during late adolescence, the share of young women who remain in school is still substantially higher than at the beginning of the decade. There was also a more noticeable increase in employment among young women. While both trends contributed to reducing the overall share of NEETs, this reduction has not translated into significant changes in NEET rates for women; young women still constitute most NEETs. At the same time, there are country variations. For 58 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 71: School to Work Transition by Age in Central America, Female, 2019 a) Costa Rica b) Dominican Republic 100% 100% 90% 90% 80% 80% 70% 70% Youth ages 10 to 35) Youth ages 10 to 35) 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 15 16 17 18 19 20 21 22 24 25 26 27 28 29 30 31 32 33 34 10 12 14 16 18 20 22 24 26 28 30 32 34 Age Age c) Guatemala d) Panama 100% 100% 90% 90% 80% 80% 70% 70% Youth ages 10 to 35) Youth ages 10 to 35) 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 10 12 14 16 18 20 22 24 26 28 30 32 34 10 12 14 16 18 20 22 24 26 28 30 32 34 Age Age Work only School only School and work Not in school or work Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT), Guatemala 2019 (ENEI), Panama 2019 (EML) using SEDLAC harmonization. Data for Costa Rica 2019 (ECE) provided by the NSO. example, in Panama, both those solely attending school and those combining work and school underlie the adolescent gains in education. Costa Rica stands out for the highest shares of those pursuing education after reaching 18: among 20-year-old youth, 60 percent of women and half of men, are students, both up from 2010 levels. In contrast, youth in Guatemala have advanced less in both schooling and labor market opportunities. Sixty percent of 15-year-old girls are now in school (Figure 71), which is a slight improvement compared to 2010, but still substantially below the other three countries (Figure A.2). Moreover, education gains do not appear to have translated into better labor market outcomes for young women, again in contrast to the other countries. Boys are also staying in school longer and fewer need to work while studying. At the same time, 25 percent of 15-year-old boys report work as their sole activity, reflecting a very early transition to work at the expense of their education, and higher than in 2010 (Figure 72). In all four countries, youth face poorer labor market outcomes. Youth, in general, exhibit worse labor market outcomes in terms of labor force participation as it is about half that of adults in all countries, except Guatemala where the gap is slightly smaller (Figure 73). Even when employed, young workers face more precarious conditions. In all four countries, they earn 50 to 60 percent of what adult workers earn, and with the exception of the Dominican Republic, they are more likely to be informal (Figure 77 and Figure 74). 59 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 72: Guatemala School to Work Transition by Age, Male, 2011 and 2019 a) 2011 b) 2019 100% 100% 90% 90% 80% 80% 70% 70% Youth ages 10 to 35) Youth ages 10 to 35) 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 10 12 14 16 18 20 22 24 26 28 30 32 34 10 12 14 16 18 20 22 24 26 28 30 32 34 Age Age Work only School only School and work Not in school or work Source: Authors’ elaboration based on Guatemala 2011 and 2019 (ENEI) Q2 using SEDLAC harmonization. Women are also at a disadvantage as their labor force participation is far less than that of men. The gender gap in labor force participation is prominent at 35 percentage points (pp) in Panama and 45 pp in Guatemala (Figure 75). Fertility and childcare responsibilities are significant factors contributing to higher levels of inactivity for women as they limit the accumulation and the use of human capital. The labor force participation of women is particularly low when they have very young children in the household. In Costa Rica it falls below 40 percent and in Guatemala to 30 percent (Figure 79). While women are more likely to be informal only in Costa Rica, in all 4 countries they earn 80 to 90 percent of their male counterparts (Figure 76, Figure 78). An additional challenge for young women is the high prevalence of teenage pregnancy and associated dropout rates, which have a long-lasting negative impact on earnings. According to a recent study by United Nations Population Fund (UNFPA) the average per capita foregone labor income for teenage mothers in a dozen Latin American countries Figure 73: Labor Force Participation by Age Group in Figure 74: Informality by Age Group in Central Central America, 2019 America, 2019 90 100 80 90 70 80 % of informal workers 60 70 Participation rate 60 50 50 40 40 30 30 20 20 10 10 0 0 Costa Rica Dominican Panama Guatemala Costa Rica Dominican Panama Guatemala Republic Republic Youth Adult Youth Adult Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT), Guatemala 2019 (ENEI) using SEDLAC harmonization. Data for Panama 2019 (EML) and Costa Rica 2019 (ECE) provided by the National Statistics Offices. Note: The term “youth” encompasses 15- to 24-year-olds while “adult” includes 25- to 64-year-olds. Legal definition of informality is used. More specially this refers to those employed (including not-salaried, employer, salaried, self-employed) without social security, as a share of total employed. 60 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 75: Labor Force Participation by Gender in Figure 76: Informality by Gender in Central America, Central America, 2019 2019 100 90 90 80 80 70 % of informal workers 70 60 Participation rate 60 50 50 40 40 30 30 20 20 10 10 0 0 Guatemala Dominican Costa Rica Panama Costa Rica Dominican Panama Guatemala Republic Republic Women Men Female Male Figure 77: Average Total Wage by Age Group in Figure 78: Average Total Wage by Gender in Central Central America America, 2019 1200 1200 1000 1000 Hourly income PPP 2005) Hourly income PPP 2005) 800 800 600 600 400 400 200 200 0 0 Dominican Costa Rica Guatemala Panama Dominican Costa Rica Guatemala Panama Republic Republic 15 to 24 25 to 40 41 to 64 Female Male Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT), Guatemala 2019 (ENEI) using SEDLAC harmonization. Data for Panama 2019 (EML) and Costa Rica 2019 (ECE) provided by the National Statistics Offices. Note: The term “youth” encompasses 15- to 24-year-olds while “adult” includes 25- to 64-year-olds. Legal definition of informality is used. More specially this refers to those employed (including not-salaried, employer, salaried, self-employed) without social security, as a share of total employed. Figure 79: Labor Force Participation by Age of Youngest Child (≤ 14 Years Old) in Household and by Gender for Central America, 2019 Costa Rica Guatemala 0.9 1 0.9 0.8 Labor force participation rate 0.8 Labor force participation rate 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Age of youngest child in the household ≤14 years old) Age of youngest child in the household ≤14 years old) Male Female Male Female Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT), Guatemala 2019 (ENEI), Panama 2019 (EML) using SEDLAC harmonization. Data for Costa Rica 2019 (ECE) provided by the NSO. 61 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 80: Per Capita Opportunity Cost of Labor Income for Teenage Mothers in Latin America, 2021 3,500 3,336 3,000 2,500 In 2021 US Dollars 2,000 1,500 1,357 1,073 1,174 1,121 1,100 954 1,000 826 854 707 757 680 500 360 0 Argentina Bolivia Colombia Ecuador Guatemala Guyana Honduras Mexico Panama Paraguay Peru Dominican Republic Per capita Opportunity cost-Labor Income Average per capita opportunity cost of labor income Source: UNFPA LACRO (2021) based on MILENA studies undertaken in 10 countries. Comparison of labor income of teenage mothers with those who became mothers as adults, controlling for other factors (e.g., education, etc.). Blue denotes countries covered in this report. (including Panama, the Dominican Republic, and Guate- Figure 81: Labor Income According to Age at Which mala) is $1,100/ year. The estimate for Panama is by far Women Became Mothers (USD per year) in Latin the highest at slightly over $3,000 Figure 80. Findings America, 2021 from the same study showed that the average earnings gap 10000 between women who became mothers as teenagers and 9000 Annual Income in US Dollars) those as adults was 45 percent (Figure 81). While the gap 8000 7000 is slightly less in the Dominican Republic and Guatemala, 6000 it is higher in the case of Panama. In sum, early moth- 5000 ers have less chance of entering the formal labor market, 4000 3000 tend to access worse and/or informal jobs, and thus end 2000 up with lower levels of income. In turn, this has a strong 1000 adverse impact on the economic autonomy of women who 0 Dominican Republic Guatemala Panama were adolescent mothers. Teenage Mothers Adult Mothers Source: UNFPA LACRO, based on the MILENA study of selected countries. Constraints to employability and utilization of human capital for better jobs At the same time, labor markets across these countries face certain structural challenges. The structural challenges in all four countries fall into the following three categories: (i) demand-side constraints, which translate to a limited availability of quality jobs, (ii) supply-side constraints, which translate to a labor force facing multiple barriers, including lack of relevant skills, and (iii) skills mismatches/market frictions between existing jobs and workers. The structural constraints to limited job creation, especially of good jobs, include limited competition and access to finance; lack of innovation, infrastructure, and integration into GVCs; constraints to trade; and crime/insecurity, as analyzed in detail by Ulku & Zaourak (2021). A detailed treatment of the demand side challenges falls outside the scope of this report, however the constraints related to labor supply (in particular role of education and skills) and the mismatches between demand and supply have been analyzed, as they constitute critical constraints to employability and utilization of human capital. Labor market challenges are particularly evident among those in the bottom of the income distribution. For instance, informality is prevalent across three of the four countries, with Costa Rica being the exception (Figure 82), however this along with other labor market variations translate into somewhat different challenges for workers in the bottom 40 percent of the income distribution (Figure 83 and Figure A.3). In all countries, there is an employment gap between the bottom 40 percent of the income distribution and the top 60 percent, which ranges from 10 pp (Panama and Guatemala) to 30 pp (Costa Rica). Except for Guatemala, the unemployment rate for the bottom 40 percent is 2 to 3 times higher than 62 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital for the better off. Additionally, the bottom 40 percent have limited access to quality jobs in most instances. For instance, they are instead more likely to be informally employed and less likely to be in wage employment compared to workers in top 60 percent in all countries, though less so for the Dominican Republic. In addition, in Panama and Guatemala, 15 to 20 percent of workers in the bottom 40 percent are unpaid workers further indicating their precarious (low quality) employment. Figure 82: Informality Rate in Central America, 2019 Figure 83: Employment Statistics of Bottom 40 Percent of the income distribution in Central America, 2019 90 20 100% 80 18 90% 19% 16 70 Unemployment rate 80% 44% 14 51% 60 70% 50 12 68% Percentaje 10 Rate 60% 40 50% 8 30 6 40% 81% 20 4 30% 56% 10 49% 2 20% 0 0 32% 10% Costa Rica Dominican Guatemala Panama 0% Republic GTM 19 DOM 19 PAN 2019 CRI 19 Total employment rate Share of informal workers Informal as % of employed Formal as % of employed Unemployment rate Source: Authors’ elaboration based on Dominican Republic 2019, Guatemala 2019 using SEDLAC harmonization. Data for Panama 2019 and Costa Rica 2019 provided by the National Statistics Offices. Note: Legal definition of informality is used. More specially this refers to those employed (including not-salaried, employer, salaried, self-employed) without social security, as a share of total employed. Employability challenges also partly reflect the inadequate investments in human capital, which manifest themselves in a limited supply of skills and worse employment outcomes. The low levels of human capital translate into a low-skilled labor force, which, in turn, has implications for the productivity and quality of employment for both workers and the economy. The challenges to acquire foundational learning and accumulate human capital persist across all four countries, with disproportionate impacts for those at the bottom of the income distribution (See chapter 2). Furthermore, a changing world of work means increased demand for advanced skills or “skills of the future.” which include critical thinking, creativ- ity, problem solving, ability to work in groups and/or supervise, adaptability, and digital skills. 39 The demand for these skills has likely been reinforced by the COVID-19 pandemic and the move to digital forms of work, however skills of the future appear to be in short supply in Central America, as presented in more detail below. While a multitude of structural challenges limit job creation,40 low levels of human capital and lack of relevant skills are also critical contributing factors to low productivity limited job creation. Limited supply of skills, in particular, skills of the future, results in skills mismatches and present additional constraints for employers to create such jobs. For instance, while workers with at least 14 years of education41 have the highest employment rate across all countries, they account for no more than around 20 percent of the population in any country and only 5 percent in Guatemala (Figure 84, Figure 85) reflecting the limited supply of skills even without delving into their relevance. In fact, both demand and supply side challenges are reflected in the lack of full utilization of existing human capital. Both lack of participation in the labor market (basic utilization-adjusted HCI) as well as lack of better employment/jobs (full utilization-adjusted HCI) mean that existing human capital is not being utilized to its full potential, contributing to 39 This is in addition to foundational skills on literacy and numeracy. 40 These include, less flexible labor market regulations, lack of innovation, constraints to trade, lack of integration into global value chains, etc. (Ulku and Zaourak, 2021) 41 Using years of education as a proxy for skills. 63 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 84: Education Level of Working-Age Population Figure 85: Employment Rate by Education Level in in Central America, 2019 Central America, 2019 70 90 80 60 70 50 60 40 50 30 40 30 20 20 10 10 0 0 Guatemala Costa Rica Dominican Panama Guatemala Costa Rica Dominican Panama Republic Republic Low Medium High None Low Medium High Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT), Guatemala 2019 (ENEI), and Panama 2019 (EML) using SEDLAC harmonization. Data for Costa Rica 2019 (ECE) provided by the NSO. deepening inequalities.42, 43 Except for Costa Rica, the utilization-adjusted indices in the other three countries fall short of the Latin America average, highlighting larger labor market mismatches and lower availability of good jobs. In all coun- tries, the full utilization-adjusted HCI index is slightly lower than the basic utilization-adjusted HCI, showing that limited availability of good jobs is a significant constraint for better utilization of existing human capital. While the rest of Latin America shows the same pattern, other UMI countries do not. Stark disparities in the utilization of existing human capital within countries present an additional challenge. Analysis indicates that some provinces in Panama and the Dominican Republic have lower UHCIs than those of fragile and conflict countries in Africa (Figure 86, Figure 87. This is a result of both geographic disparities in the levels of human capital and the limited availability and heterogenous spatial distribution of jobs and good jobs. The overall educational attainment of the working-age population in the four countries has increased, reflecting the gains in access for younger generations, in particular, for women. Younger cohorts have progressively higher levels of education across all four countries, and the youngest cohorts have the highest educational attainment in the working-age population (Figure 87). However, the extent of educational gain across cohorts among the four countries differed; for instance, in contrast to the other countries, Guatemala made only a few gains for the younger generation on the share of those with 14 plus years of education (Figure 88). The gains in educational attainment over time (i.e., observed in younger cohorts) have been larger for women compared to men. In particular, except for Panama, the share of women with 14 plus years of education more than doubled. Furthermore, in Guatemala, the young women have made substantial gains in secondary education, with the share of women with secondary education (9 to 13 years) increasing by 8 times (from only about 50 percent in the earlier cohorts to more than 40 percent in the recent generation Figure 91). However, the gains in educational attainment among the working-age population did not generally translate into better skills and better jobs. The overall supply of skilled labor remains limited: the share of employed with no more than 8 years of education is still significant, ranging from a low of about 25 percent in Panama to 60 percent in Guatemala, with differences across sectors. Overall, agricultural employment is dominated by workers with no more than 8 years of employment. On the other hand, services have the highest share of workers with 14 or more years of education, but even there they account for no more than between a quarter to a third of workers and substantially less in Guatemala (Figure A.3). 42 Utilization-adjusted human capital indices (UHCIs) adjust the HCI for labor-market underutilization of human capital (Pennings 2020). The basic UHCI measures utilization as the proportion of the working-age population in paid employment or self-employed in order to capture the gains from employing all potential workers. The full UHCI introduces the concept of “better employment” (i.e., the ratio of employment in “better jobs” to total employment), which is defined as paid non-agricultural employees, plus employers. It incorporates both the employment effect from the basic UHCI measure and the potential gains that could be made from moving workers to jobs where they can better use their human capital to increase productivity. 43 It should be noted that the UHCIs are designed to complement the main HCI, and not to replace it. This is because they have different purposes: the HCI is an index of supply of a factor of production (in the future), whereas the UHCIs are a hybrid between an index of factor supply (capturing investment in human capital), and a productivity index (capturing how efficiently that human capital is used in production). Incomplete utilization focuses on private gains to worker productivity and cannot capture broader gains in other dimensions (women’s empowerment, technological advancement, etc.). 64 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 86: Pre-COVID-19 Full Utilization-Adjusted Figure 87: Pre-COVID-19 Full Utilization-Adjusted Human Capital Index by Province in Panama, 2019 Human Capital Index by Province in the Dominican Republic, 2019 UHCI Full .315,.352] .307,.315] .249,.307] UHCI Full [.218,.249] .312,.328] .291,.312] .278,.291] [.263,.278] No data Source: World Bank, 2019. Estimates made using previously estimated HCI data Source: World Bank, 2019. Estimates made using previously estimated HCI data and utilization rates. The employment and better employment rates are from and utilization rates. The employment and better employment rates are from Panama 2019 EML. Dominican Republic 2019 ECNFT. Skill wage premiums have not increased for workers over the past decade, and in most cases, women even experienced declining returns (Figure 89). Skill wage premium reflects the higher wages earned (in percentage terms) by workers with tertiary education compared to workers that did not complete primary education controlling for other factors (such as age, gender, employment type, education, etc). In all four countries the skill wage premiums have been stagnant at best which have important implications. The lack of opportunities for more educated (mostly young) workers can provide an impetus for international migration to maximize returns on skills. For instance, migrants from Guatemala tend to be more educated than their peers at home.44 The skill wage premium trends are more worrying or women, with observed declining returns in most cases. For instance, in 2010, women with tertiary education in Dominican Republic were expected to earn around 53 percent more than women with a primary education or less, however, this declined to only 45 percent in 2019 (Figure 89). Higher educated women in Costa Rica were a notable exception. There are multiple factors that might explain stagnant and declining skill wage premiums. A likely contributing factor is the lack of growth in jobs that demand advanced skills, meaning that the supply of skills is outpacing their demand (see below for further analysis of relevant skills). This mismatch may reflect the limited institutional capacity to monitor labor demand and develop effective and relevant workforce policies that consider the realities of the labor market. The quality of education may also be contributing to this phenomenon of stagnant and declining skill wage premiums in the sense that higher educational attainment may not necessarily reflect more or relevant skills. There may also be a labor market mismatch between the specific skills demanded and offered due a lack of relevant school curricula and training programs, for example, technical and vocational education and training (TVET). The stagnant skill premium also partly reflects the predominance of jobs intensive in the “skills of the past”. Following the classification system proposed by Acemoglu and Autor (2011), the task content of jobs can be measured in the following five skill categories: non-routine analytical, non-routine interpersonal, routine cognitive, routine manual, and non-rou- tine manual. Jobs intensive in routine skills (i.e., routine manual, routine cognitive), are more likely to be automated and disappear with technological change while jobs intensive in non-routine, especially analytical and interpersonal skills (e.g., such as critical thinking, creativity, problem solving, ability to work in groups and/or supervise, adaptability, leadership, etc.) are likely to benefit from technological change and have been increasing in prevalence in the advanced economies across the globe. Analysis indicates that a significant share of the available jobs in these countries are intensive in skills of 44 Arayavechkit et al., 2022; Mejia-Mantilla et al., 2023. 65 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 88: Educational Attainment across Generational Cohorts (Ages 25–64) in Central America, 2019 a) Costa Rica b) Dominican Republic 100% 100% 90% 90% Working-age population ages 15 to 64) Working-age population ages 15 to 64) 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 58 63 68 73 78 83 88 94 58 63 68 73 78 83 88 94 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 54 59 64 69 74 79 84 89 54 59 64 69 74 79 84 89 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 Cohort Cohort c) Guatemala d) Panama 100% 100% 90% 90% Working-age population ages 15 to 64) Working-age population ages 15 to 64) 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 58 63 68 73 78 83 88 94 58 63 68 73 78 83 88 94 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 54 59 64 69 74 79 84 89 54 59 64 69 74 79 84 89 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 Cohort Cohort 0 to 8 years 9 to 13 years 14+ years Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT), Guatemala 2019 (ENEI), and Panama 2019 (EML) using SEDLAC harmonization. Data for Costa Rica 2019 (ECE) provided by the NSO. the past, and this share has been growing over the past decade (especially in Guatemala -routine cognitive, Figure 91).45 The same is true in Costa Rica (routine manual), although jobs with skills of the future have gained more prominence over the last five years (in particular, nonroutine cognitive and interpersonal). Similarly, there are signs of a recent shift in the Dominican Republic towards a higher task content of nonroutine cognitive analytical and interpersonal tasks (future skills) and a decrease in manual tasks, perhaps driven by the increased share of college educated workers and pointing to the emergence of new higher value-added jobs. 45 This type of analysis could not be undertaken in Panama due to lack of detailed occupations data required to calculate the task content of jobs. 66 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 89: Educational Attainment for Women across Cohorts (Ages 25–64) in Central America, 2019 a) Costa Rica b) Dominican Republic 100% 100% 90% 90% Working-age population ages 15 to 64) Working-age population ages 15 to 64) 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 58 63 68 73 78 83 88 94 58 63 68 73 78 83 88 94 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 -19 54 59 64 69 74 79 84 89 54 59 64 69 74 79 84 89 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 Cohort Cohort c) Guatemala d) Panama 100% 100% 90% 90% Working-age population ages 15 to 64) Working-age population ages 15 to 64) 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 8 3 8 3 8 3 8 4 8 3 8 3 8 3 8 4 95 96 96 97 97 98 98 99 95 96 96 97 97 98 98 99 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 54 59 64 69 74 79 84 89 54 59 64 69 74 79 84 89 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 Cohort Cohort 0 to 8 years 9 to 13 years 14+ years Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT), Guatemala 2019 (ENEI), and Panama 2019 (EML) using SEDLAC harmonization. Data for Costa Rica 2019 (ECE) provided by the NSO. At the same time, a labor force out of step with “jobs of the future” partly accounts for absence of such jobs. Increased educational attainment (in particular, tertiary education) can be a significant driver in the transition to jobs intensive in future skills, especially in cases where there is a demand for a better educated workforce as has been observed in developed countries in Europe.46 On the other hand, the limited supply of relevant skills, as has been the case in all four countries to varying degrees, likely constrains the generation of jobs of the future, interacting with and exacerbat- ing a range of structural constraints on the demand side (as discussed briefly previously). The limitation on the supply of relevant skills is most challenging for Guatemala given its unfavorable productive structure, with the lowest share of employment in “future skills” and highest share of the “past skills” compared to average task content of employment across the other three countries. 46 Ridao-Cano and Bodewig, 2018. 67 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 90: Skill Wage Premium by Gender in Central America (%), circa 2010-2020 a) Costa Rica, 2011–2020 b) Dominican Republic, 2010–2020 100 100 90 90 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 11 12 13 14 15 16 17 18 19 20 10 10 11 12 13 14 15 16 17 18 19 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 c) Guatemala, 2011–2019 d) Panama, 2011–2019 100 100 90 80 80 70 60 60 50 40 40 30 20 20 10 0 -20 11 12 13 14 15 16 17 18 19 11 12 13 14 15 16 17 18 19 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Skill wage premium-women Skill wage premium-men 95% Confidence Interval 20 Source: Authors’ elaboration based on Costa Rica 2020 (ECE), Dominican Republic (ECNFT), Guatemala (ENEI), and Panama (EML) using SEDLAC harmonization. Data for Costa Rica 2020 (ECE) provided by the NSO. Note: Skill wage premium reflect higher wages earned (in percentage terms) by workers with tertiary education compared to workers that did not complete primary educa- tion and is estimated using controls (such as age, gender, employment type, education, etc.). The results for the Dominican Republic and Costa Rica for 2020 represent the drastic shift in employment due to the COVID-19 pandemic, capturing the fact that only essential workers and those that could work from home continued to work, likely biasing the results. In all four countries, the highly educated are more likely to be employed in the jobs of the future. The task/skill content of the job seems to be closely linked with the level of education. In the three countries, jobs using non-routine cognitive (both analytical and interpersonal) skills are most commonly held by workers with complete tertiary education (Figure 92). Education, not gender, seems to be mainly driving the differences in employment in jobs of the future, as evidenced by the initially puzzling gender gap observed in Guatemala: women are more likely to be employed in jobs of the future compared to men, but this disappears when educational level is controlled (Figure A.4). This possibly reflects the fact that most of the employed women in Guatemala are highly educated (and thus more likely to be employed in jobs of the future), while women with lower levels of education have much lower rates of employment and labor force participation. However, there are some differences among countries in the degree to which education is the driving factor in the employment of jobs in the future. For example, data suggests polarization in task content for women in the Dominican Republic: females of all education levels are more likely to have jobs relatively intensive in nonroutine skills (“future skills”), while this is true only for highly educated men (Figure A.5). 68 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 91: Changes in the Task/Skill Content of Jobs, 2011–2019 a) Costa Rica b) Dominican Republic c) Guatemala 0.15 0.25 0.3 0.2 0.2 0.1 0.15 0.1 0.05 0.1 Changes in the Task Changes in the Task 0 Changes in the Task 0.05 0 -0.1 0 -0.2 -0.05 -0.05 -0.3 -0.1 -0.1 -0.4 -0.15 -0.2 -0.5 -0.15 11 12 13 14 15 16 17 18 19 11 12 13 14 15 16 17 18 19 11 12 13 14 15 16 17 18 19 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 NRCA NRCI RC RM NRM Source: Authors’ elaboration based on Costa Rica (ECE), Dominican Republic (ECNFT), and Guatemala (ENEI) using SEDLAC harmonization and The Occupational Information Network (O*NET). Note: NRCA: Nonroutine Cognitive Analytical, NRCI: Nonroutine Cognitive Interpersonal, RC: Routine Cognitive, RM: Routine Manual, NRM: Nonroutine Manual. Figure 92: Task/Skill Content of Jobs with Respect to Average Worker by Age Group in Central America, 2019 a) Costa Rica b) Dominican Republic 2 2 1.5 1.5 1 1 Task/Skill Content of Jobs Task/Skill Content of Jobs 0.5 0.5 0 0 -0.5 -0.5 -1 -1 -1.5 -1.5 -2 Never Incomplete Incomplete Complete Incomplete Complete -2 Never Incomplete Incomplete Complete Incomplete Complete attended Primary Secondary Secondary Tertiary Tertiary attended Primary Secondary Secondary Tertiary Tertiary c) Guatemala 2.5 2 1.5 NRCA 1 NRCI Task/Skill Content of Jobs 0.5 RC 0 RM NRM -0.5 -1 -1.5 -2 Never Incomplete Incomplete Complete Incomplete Complete attended Primary Secondary Secondary Tertiary Tertiary Source: Authors’ elaboration based on Costa Rica (ECE), Dominican Republic (ECNFT), and Guatemala (ENEI) using SEDLAC harmonization and the Occupational Information Network (O*NET). Note: NRCA: Nonroutine Cognitive Analytical, NRCI: Nonroutine Cognitive Interpersonal, RC: Routine Cognitive, RM: Routine Manual, NRM: Nonroutine Manual. 69 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 93: Task/Skill Content of Jobs with Respect to Average Worker by Age Group in Central America, 2019 a) Costa Rica b) Dominican Republic 0.3 0.2 0.15 0.2 0.1 0.1 Task/Skill Content of Jobs Task/Skill Content of Jobs 0.05 0 0 -0.05 -0.1 -0.1 -0.2 -0.15 -0.2 -0.3 -0.25 -0.4 [15,24] [25,40] [41,64] [65+] -0.3 [15,24] [25,40] [41,64] [65+] c) Guatemala 0.3 0.2 0.1 NRCA Task/Skill Content of Jobs 0 NRCI RC -0.1 RM -0.2 NRM -0.3 -0.4 -0.5 [15,24] [25,40] [41,64] [65+] Source: Authors’ elaboration based on Costa Rica (ECE), Dominican Republic (ECNFT), and Guatemala (ENEI) using SEDLAC harmonization and the Occupational Information Network (O*NET). Note: NRCA: Nonroutine Cognitive Analytical, NRCI: Nonroutine Cognitive Interpersonal, RC: Routine Cognitive, RM: Routine Manual, NRM: Nonroutine Manual. Some workers, particularly informal workers and youth are more at risk of technological displacement because they are concentrated in jobs of the past. Both informal and young workers tend to be employed in the jobs of the past, which are more at risk of displacement and/or low earnings growth as technological change accelerates. Young workers perform fewer non-routine cognitive tasks (complementary to technological progress), which require skills of the future (Figure 93). Jobs intensive in routine skills seem to be concentrated to a greater extent in some sectors across the four countries. For example, in Costa Rica, jobs in agriculture, mining, accommodation and similar services, transportation and manufacturing seem to be at greater risk than sectors with more jobs intensive in nonroutine skills such as information and communication, finance, and professional and technical services (Figure A.6). Impacts of the COVID-19 pandemic on labor market outcomes The COVID-19 pandemic has had a significant impact on labor markets in all four countries, with some groups were more severely affected than others. Evidence confirms the disproportionate labor market impact of the pandemic on youth, lower-skilled workers, and women around the world.47 The NEET rate increased in all four countries since the pandemic, reversing earlier gains (Figure 94). In Costa Rica, the NEET rate increased from 17 percent in the first trimester of 2020 (pre-pandemic) to 25 percent in the first trimester of 2021, driven more by men, as was the case for Panama and 47 High-frequency phone surveys show that more than 25 percent of young workers (18 to 24 years old) in Latin America were left without work in the first two months of the pandemic compared to 14 percent of workers aged 25 to 64 years old (World Bank, 2021b). 70 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Figure 94: Change in NEET Rate during the Pandemic by Gender in Central America, 2020 and 2021 a) Dominican Republic b) Panama c) Costa Rica 45% 60% 80% 40% 70% 50% 35% 60% 30% 40% 50% NEET Rate 25% NEET Rate NEET Rate 30% 40% 20% 30% 15% 20% 10% 20% 10% 5% 10% 0% 0% 0% Total Female Male Total Female Male Total Female Male NEETs NEETs NEETs NEETs NEETs NEETs NEETs NEETs NEETs 2019 2020 Percentage change right axis) Source: Authors’ elaboration based on Costa Rica 2020 (ECE), Dominican Republic 2019–2020 (ECNFT), and Panama (EML) 2019–2020, using SEDLAC harmoni- zation. Data for Costa Rica 2021 (ECE) provided by the NSO. the Dominican Republic as well.48 However, the countries differed on the extent to which education and/or gender did or did not shield workers from the impact. In Costa Rica, there were variations in post COVID-19 trends, with certain workers bearing the brunt of the impacts. Higher education levels corresponded with limited employment losses among youth (Figure 95). The same was true for women, but not for men. The population with lower levels of education, on the other hand, were the most affected, which deepened segmentation in the labor market. Inactivity became more common among young men than before, while share of unemployed increased among young women because of job losses (Figure 96). There were differing reasons for inactivity among young men relative to young women with the latter more likely to cite family obligations as the main reason. Figure 95: Difference in Labor Market Indicators by Figure 96: Composition of NEETs by Gender in Costa Education and Age Group in Costa Rica, 2020/2021 Rica, 2020–2021 20 100 16 90 15 14 80 11 45.3 53.7 55.6 54.7 Changes 2021-2020 in percentage 10 70 57.8 65.6 Percentage of NEETs 6 7 60 5 5 50 0 0 40 -2 30 -5 20 -10 10 Total Low Medium High Total Low Medium High 0 Youths Adults 2020 2021 2020 2021 2020 2021 Participation rate Employment rate Unemployment rate Unemployed among NEETs Inactives among NEETs Source: Authors’ elaboration based on Costa Rica Labor Force Survey (ECE), Quarter 1 of 2020 and 2021. 48 Isik-Dikmelik et al. (2023). Strengthening SPL Systems in Central America for an Inclusive and Resilient Recovery. World Bank: Washington D.C. 71 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital In Panama, negative impacts were not limited to job losses. About half of the employed population in 2021 reported a decline in labor income because of the pandemic. The highest share of employees reporting a loss in income worked in manufacturing (Figure 97). In contrast to Costa Rica, more education did not serve to protect either women or youth from the deterioration in employment (Figure 98). Populations with lower levels of education were hit the hardest. Among youth, adverse impacts were more significant among the new entrants. The share of NEETs who were unemployed as opposed to inactive rose from 50 percent in 2019 to 60 percent, which was mostly due to changes in labor market statuses among women NEETs (Figure 94). Figure 97: Share of Employees Reporting Income Figure 98: Difference in Labor Market Indicators in Reduction in Their Job during the Pandemic by Sector Panama by Education and Age Group, 2019–2020 in Costa Rica, 2019–2020 30 120.00% 26.7 25 19.3 100.00% 20 15.2 15 11.1 80.00% 33.99% 9.1 10.1 41.85% 10 6.9 53.70% 5 60.00% 1.0 0 -5 -4.1 40.00% -5.4 66.01% -10 -8.5 -8.0 -7.2 58.15% -11.0 -11.3 20.00% 46.30% -15 -12.7 -12.2 -14.5 -20 High Medium Low High Medium Low 0.00% Agriculture Manufacturing Services Youth Adult Reduction No reduction Participation rate Employment rate Unemployment rate Source: Authors’ elaboration based on Costa Rica Labor Force Survey (ECE), Source: Authors’ elaboration based on Panama Labor Market Survey (EML), Quarter 1 of 2020 and 2021. 2019–2020. Compared to Costa Rica and Panama, the impact of the pandemic on overall labor market indicators was somewhat less severe in the Dominican Republic. While overall employment was less affected, there were wide differences among sectors. Declines in employment were driven by jobs losses in services and for that reason had a greater impact on workers with medium to high education levels. Areas in the services sector that experienced sharp employment decreases include education, hotels and restaurants, financial intermediation, and real estate, renting and business activities (Figure 100). The poor performance in services probably contributed to the fact that 60 percent of all job losses were among women regardless of their education level (Figure 99). While the governments of all four countries have taken measures to contain the impact of the COVID-19 pandemic, the low coverage of labor market programs and the varying degrees of institutional capacity for effective labor market inclusion remain a challenge. Governments have responded swiftly with a range of measures to mitigate the impacts of COVID-19 including social assistance programs, social insurance programs and labor market policies and regulations (Box 6). As documented above, the unprecedented nature of the shock with the confinement measures had deep impacts especially on employment outcomes for the more vulnerable groups in the labor market, particularly for women, and low skilled. The response included programs intending to reach at least some of these populations but faced challenges,49 including due to varying institutional capacity as outlined below. It also presented valuable lessons to draw on to address the longer-term constraints; and at times provided an impetus for innovation through use of technology. Effective employment support requires robust institutional capacity to be able to identify the needs of jobseekers, workers, and employers, and to design and implement packages of programs tailored to needs. This will involve 49 Isik-Dikmelik et al. (2023). Strengthening SPL Systems in Central America for an Inclusive and Resilient Recovery. World Bank: Washington D.C. 72 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital effective delivery of employment services (e.g., job search Figure 99: Share of Job Loss in the Dominican assistance, labor market intermediation, hiring subsidies) Republic by Educational Level and Gender, and a robust job training system (including TVET) that is 2019–2020 linked with both the formal education system and market 120% needs. The importance of the latter is increasing in line with the advances in the coverage and graduation of pre-univer- 100% sity education as discussed in Chapter 2. The availability and use of labor market information, ideally through a 35% 30% 80% 48% strong labor market information ecosystem will be critical in developing effective employment services and a strong job training system. This ecosystem could improve labor 60% market outcomes by: (i) identifying the needs of the market in terms of skills and occupations, (ii) taking advantage 40% 70% 65% of technology to improve job matches (e.g., AI-assisted 52% job portals), (iii) identifying relevant trainings, and (iv) 20% providing a wide range of labor market analytics to inform different actors including students, training providers, and 0% Low Medium High employers. Women Men While all four countries have some of the basic elements Source: Authors’ elaboration based on Dominican Republic Labor Market Survey required for employment support system, their capacity (ECNFT), 2019–2020. Low education level refers to 0-8 years, medium to 9-13 years, and high to 14 or more years of schooling. varies widely. Similarly, the labor market information ecosystem is at different stages of development across these countries. Each country supports professional training through the formal education system in secondary schooling and post-secondary institutions as well as through professional training agencies. The coverage of labor market programs is low in these countries, except for Costa Rica. Figure 100: Job Loss in the Dominican Republic by Sector (Quarter on Quarter Difference), 2019–2020 Sector 2019Q1-2020Q1 2019Q2-2020Q2 2019Q3-2020Q3 2019Q4-2020Q4 Agriculture, hunting, forestry and Fishing -3.6% -8.6% -1.9% -1.8% Mining and quarrying 82.0% 17.8% 32.8% -1.3% Manufacturing 0.4% -0.1% -3.6% -4.8% Electricity, gas and water supply 7.9% 3.1% 13.1% 5.5% Construction -11.3% -13.7% -1.3% 4.7% Wholesale and retail trade 0.1% -11.0% -4.9% -1.2% Hotels and restaurants 10.6% -19.7% -29.0% -28.8% Transport, storage and communications 3.2% 5.0% 10.0% 5.8% Financial intermediation -10.4% -29.7% -20.7% -9.1% Real estate, renting and business activities -7.1% -9.4% -15.9% -20.5% Public administration and defence 9.3% 11.5% -7.4% 1.2% Education 0.5% 3.5% -5.6% -15.7% Health and social work 6.3% -4.0% -9.2% -11.0% Other community, social and personal service activities -11.1% -15.1% -10.7% -6.3% Activities of private households as employers 0.0% -31.1% -16.9% -17.4% Extraterritorial organizations and bodies 114.0% -70.4% -22.3% 3.8% Source: Authors’ elaboration based on Dominican Republic Labor Market Survey (ECNFT), 2019–2020. Low education level refers to 0-8 years, medium to 9-13 years, and high to 14 or more years of schooling. 73 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Box 6. Social Protection and Labor Market Policy Responses to the COVID-19 Pandemic Governments implemented social protection and labor market measures to protect human capital and contain labor market impacts, starting with cash transfers. All four countries deployed cash transfers mainly to reach new beneficiaries, including informal workers. The scale and duration of benefits varied in each country. The largest scale cash transfer programs were in Panama and Guatemala, where coverage reached almost 80 percent of the population. Social assistance measures in nearly all countries included in-kind support to assist food security (e.g., school feeding programs) and utility waivers. Social insurance measures were limited due to narrow coverage and the lack of unem- ployment insurance. Social Assistance Social Insurance Labor Market Policies and Labor Regulations SS contributio ns waiver/subsidy Utility and financial obligation Healthcare insurance support Wage subsidies /soft loans In Kind /food voucher Training measures Shorter work time Labor regulations School feeding Paid sick leave Cash Transfer Pensions waivers Costa Rica • • • • • • • • • Guatemala • • • • • • Panama • • • • • • • Dominican Republic • • • • • • • • • Labor market measures centered on access to credit for micro, small, and medium enterprises (SMEs) (formal and informal) and wage subsidies or similar support for formal workers. Costa Rica implemented a guarantee scheme to encourage commercial lending to SMEs by guaranteeing potential losses on new loans and complementary measures such as adapting the conditions of existing credits through debt restructuring. Panama created the Panama Agro Solidario program to support small and medium agricultural and livestock producers with zero-interest loans up to US$100,000. Wage subsidies were also deployed. Guatemala created a wage subsidy fund called Fondo para la Protección del Empleo to support formal workers in the private sector whose contract had been suspended. The Fondo de Asistencia Solidaria al Empleados (FASE) was created in the Dominican Republic to support approximately 906,000 laid-off workers and was later expanded to assist workers kept on the payroll as well. Training was also deployed and, in some cases, was coupled with social assistance support. Costa Rica launched the Protect Skills Plan, a scholarship program for online training promoted by the government for 50,000 beneficiaries of the Bono Proteger program. Panama partnered with the virtual learning platform Coursera to provide free access to training programs for more than 250,000 independent workers, entrepreneurs, and students. Free online training was also offered in Guatemala and Honduras through the Capacitate para el Empleo online portal. Source: Isik-Dikmelik et. al (2023). 74 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital In some countries, labor market programs (training, in particular) do not focus on the vulnerable and unemployed. Costa Rica has been modernizing its main training agency, the Instituto Nacional de Aprendizaje (National Institute of Learn- ing), to better fit the changing needs of the labor market. The modernization efforts are in line with international evidence and include: (i) outsourcing training to accredited, private providers, (ii) prioritizing vulnerable students and workers (e.g., economically disadvantaged, women, etc.), and (iii) monitoring labor demand. Labor market programs and the training system in the Dominican Republic also have some strong points. INFOTEP, the country’s training agency, contracts with private providers and has accumulated experience in working with youth. Nearly 20 percent of students in the Dominican Republic are following the technical modality in secondary schools, with the goal to increase the share to about a third by 2024. In addition, the country is working on a reform to improve post-secondary technical education provided through the Ministry of Education. At the same time, there is room for improvement. Shortfalls include a fragmented training system; constraints in resources restricting the reach of the programs, which may stem from limitations in funding mechanism and related misperceptions; and lack of programs to promote labor market insertion (wage subsidies for youth, for example). The varying but largely limited capacity of labor market programs and the severe impacts of COVID-19 do not bode well for youth (in particular, new entrants to the labor market), and less-skilled workers who will likely face long-term effects underscoring the need for concerted policy action. Existing literature indicates that it takes about ten years for youth that enter the labor market during a downturn to catch up with their peers that joined during periods of economic stability or prosperity. This is particularly true for workers with lower levels of education.50 Data from Latin America show a greater long-term effect on labor market outcomes including lower labor force participation, higher unemployment, a greater likelihood of working informally, and long-term effects on earnings.51 Again, outcomes are worse for those with less than college education. College educated workers, on the other hand, seem to face fewer and shorter-term impacts. As outlined in Chapter 2, post-COVID-19 entrants to the labor market will also be at a disadvantage as they will have suffered learning losses ranging from 4 percent in the Dominican Republic to 16 percent in Costa Rica, even in the most optimistic scenario, further adversely affecting their labor market outcomes. In sum, the COVID-19 crisis exacerbated pre-existing labor market challenges in skills, employability, and institutional capacity for employment support. New chal- lenges also surfaced for younger generations in terms of human capital accumulation and utilization. All this underscores the importance of concerted policy action in the short and long term to ensure the protection, enhancement, and better utilization of human capital. 50 For instance, evidence from Mexico indicates that for new entrants with secondary education, wages take 9 years to recover and that overall, they are less engaged in the labor market (Moreno and Sousa, 2021). 51 De Silva et al., 2021. 75 Deploying and Enhancing Human Capital for Better Employment Outcomes Central America Human Capital Review  | Promoting more and better investments in human capital Box 7. Key Labor Market Constraints and Outcomes for Youth and Active Population Constraints Outcomes Impacts of the COVID-19 Pandemic • Cessation of non-essential production, especially high contact jobs • Large job losses, especially for women, youth, and lower-skilled workers • Interruption of labor market programs/services including TVET • Increased inactivity among youth (increased NEET rate) • Limited adaptiveness of LM/TVET system • Interruption of accumulation of human capital or deterioration of existing human capital amongst vulnerable populations • Risk of long-term effects for new LM entrants and vulnerable workers Structural Issues • Limited availability of quality jobs, especially for youth and poorer • Limited utilization of existing human capital populations • Low supply of skills • Lack of Foundational skills acquired at school • Poor labor market outcomes for youth • Low participation in the labor force (high levels of inactivity) • High share of informal jobs • Lower earnings • Limited institutional capacity for labor market inclusion • Low supply of relevant skills/future skills • Limited link between education and labor market/training system • Low employability of youth • Gaps in coverage and mix of labor market programs to address diverse constraints. • Lack of information on (and changes in) skills in demand • Limited effectiveness of TVET • Limited effectiveness of employment services 76 CHAPTER 4 POLICIES TO PROMOTE MORE AND BETTER INVESTMENTS IN HUMAN CAPITAL HOSPITAL ESCUELA HOSP Central America Human Capital Review  | Promoting more and better investments in human capital 77 Policies to Promote More and Better Investments in Human Capital Central America Human Capital Review  | Promoting more and better investments in human capital T his chapter presents a range of policy recommendations to improve human capital accumulation and utilization by addressing the structural constraints and the impacts of the COVID-19 pandemic discussed in chapters 2 and 3. The policy recommendations are given for the short and medium to long term and are organized around three thematic areas: (i) improving the adaptiveness of education systems and labor markets to protect human capital and increase resilience and preparedness to more frequent shocks; (ii) improving the efficiency and effectiveness of social public spending to accelerate/promote human capital accumulation; and (iii) strengthening institutions for the delivery of better and more equal social services to foster human capital accumulation for all. While the UMI countries in Central America have experienced growth and some poverty reduction in the past decade, there are persistent shortfalls in human capital levels, mainly related to education outcomes. Except for Costa Rica, human capital levels as measured by the human capital index (HCI) are low compared to their middle-income peers and the average in Latin America. Moreover, over the last decade, human capital levels have largely stagnated or only increased marginally (2 percentage points in Costa Rica and Guatemala, 1 percent in the Dominican Republic, and Panama shows a reduction in its HCI). This performance contrasts with the global evolution of human capital levels over the past decade, which have improved in most countries with HCI scores increasing on average by 3.5 percentage points. In the four UMI countries in Central America, the main factor pulling down the HCI in the last decade is the lack of progress in the education outcomes. This report highlights the key constraints to improve human development outcomes across the life cycle, with a focus on school-age children and the youth and working age adults. Within the school-age stage, we focus on foundational learning and school dropout outcomes, and within the youth and working age adults we focus on employability and the transition to the labor market outcomes. Without foundational learning, students often fail to thrive later in school or when they join the workforce. They don’t acquire the human capital they need to power their careers and economies once they leave school, or the skills that will help them become engaged citizens and nurture healthy, prosperous families. Human capital is a central driver of sustainable growth and a key contributor to poverty reduction and shared prosperity. Therefore, human capital deficits undermine sustainable growth and poverty reduction. Yet, policy makers sometimes find it hard to make the case for human capital investments as the benefits of investing in people can take a long time to materialize. COVID-19 has impacted people’s lives around the world, hitting the UMI countries in Central America particularly hard. Some of the measures taken to control the spread of the disease in the four countries where among the most drastic in Latin America, causing a sharp reduction in economic growth, an increase in poverty, significant job losses, income reduction, and food insecurity, disproportionately affecting school-aged children and the youth. Among school-age children, school closures are expected to translate into large learning losses and larger inequalities in education outcomes. In Guatemala, school closures also seem to have translated into larger school dropouts, despite the country being already an outlier in the subregion in terms of its high level of school dropout in secondary school. Among the youth, job losses affected them disproportionately and translated into an increase in the share of NEETs, and precarity signaling risk of long-term scarring. Accelerating the rate at which human capital accumulates for school-age children and the youth is more pressing than ever so that countries in the subregion can address the structural constraints to build more inclusive human capital and achieve their full productivity. Whether the negative impact on human capital accumulation for these groups translates into a permanent reduction of the levels of human capital would depend on the size of the impact itself but, perhaps more importantly, on the rate at which human capital accumulates thereafter. Therefore, accelerating human capital accumulation will be key to support the generation of children and youth affected by the pandemic, so that all of them can 78 Policies to Promote More and Better Investments in Human Capital Central America Human Capital Review  | Promoting more and better investments in human capital at least achieve the foundational learning needed to continue developing skills and learning and to subsequently transition to the labor market. A durable recovery from COVID-19 will require governments to build inclusive education systems where all children can learn and thrive, ensuring strong linkages between education systems and labor markets for better and more inclusive employment outcomes, and building a more adaptive social protection system to increase resilience and preparedness to more frequent shocks. This chapter presents a range of policy recommendations to improve human capital accumulation and utilization by addressing the structural constraints and the impacts of the COVID-19 pandemic discussed in chapters 2 and 3. Recommendations have been made by area of policy and are presented for both the short and medium/long terms . The short-term policies are intended to accelerate human capital recovery and accumulation by focusing on: (i) learning losses, (ii) dropouts, (iii) social protection programs to promote improvements in health/nutrition and access to education for the most vulnerable. The medium/long-term policies focus on: (i) improving learning outcomes, (ii) reducing dropout rates, (iii) addressing high rates of teenage pregnancy, (iv) strengthening the effectiveness and adaptiveness of labor market/TVET systems, (v) investing in the social protection systems to improve shock responsiveness during crises and to promote more equitable outcomes, and (vi) improving institutional capacity for the delivery of social services. These policy recommen- dations have been summarized in Table 2, Table 3, and Table 4 at the end of the chapter. Improving the adaptiveness of education systems and labor markets to protect human capital and increase resilience and preparedness to more frequent shocks The COVID-19 pandemic reinforced the need to improve the adaptiveness of the education systems and labor markets, which is particularly important in Central America where there is increased exposure to natural disasters and other shocks. Effective, adaptive, and resilient education systems and labor markets that prioritize vulnerable populations would help mitigate the negative impacts of shocks on human capital by increasing the resiliency of households to shocks, helping students continue learning during the closure of schools, and helping affected workers to return to employment and self-reliance during the recovery. Considering the increased exposure of the Central American countries to natural disasters and other shocks, improving the adaptiveness of the education systems and labor markets is imperative for better preparedness and increased resilience of the vulnerable in the sub-region. From the school-age perspective, in the short term, this means conducting standardized student learning assessments to estimate learning losses, developing more online teaching materials that are integrated with the teaching training programs,  strengthening the development of early warning systems and administrative records to inform the adaptation of education programs aimed at protecting children and youth during crises. In the medium/long term, this implies expanding connectivity in schools, access to technological devices, teacher training, and the use of educational platforms to support distance or hybrid learning methods during health or environmentally-related shocks. It also implies working on the consolidation of school curricula by prioritizing foundational learning, prerequisites for future learning, transferable competencies, and socioemotional skills. From the youth and active population perspective, in the short term, this means assessing the coverage and the mix of existing labor market programs and establishing systems for their digital delivery and training. This will ensure continuity and support for both youth and active populations during crises. In the medium/long-term, this means strengthening labor market information systems by exploiting administrative and real-time data to improve the understanding and monitoring of skills demanded by the labor markets, and to respond to changing demand during crises. From a cross-cutting perspective, in the short term, it is important to incorporate triggers and emergency protocols into social assistance programs to allow for rapid scalability and potential adjustments to eligibility during crises. This ensures that support reaches those who need it most in a timely manner. Additionally, conducting stress tests of the social protection system can help identify potential gaps that may limit opportunities for vulnerable individuals and prompt necessary reforms. Looking towards the medium and long term, improving the coverage of social registries and education administrative data and keeping them up-to-date is crucial. These systems play a vital role in quickly identifying and reaching the vulnerable and affected populations. By enhancing data collection, management, and analysis, institutions can swiftly and accurately identify those in need, allowing for more targeted and efficient delivery of social assistance and support. 79 Policies to Promote More and Better Investments in Human Capital Central America Human Capital Review  | Promoting more and better investments in human capital Improving the efficiency and effectiveness of social public spending to accelerate/promote human capital accumulation Improving the efficiency and its effectiveness of social public spending is essential for enhancing human capital levels. Given the persistently insufficient and inefficient public investments allocated to social services in most countries, it will be crucial for governments to prioritize the enhancement of social public spending efficiency to bolster human capital levels. While this holds true across all countries, this is particularly important for the four UMI countries in Central America, which face important fiscal constraints to grapple with the daunting task of supporting the post-COVID-19 recovery agenda. While this might take different forms in different countries, there are some commonalities as to how this can be achieved. From the school-age perspective, in the short-term, it is crucial to utilize the results from student standardized learning assessments as a tool to support schools and teachers to improve students’ learning outcomes. This can be combined with the development of school learning monitoring systems with formative purposes that allowing teachers to continuously assess and monitor students’ learning outcomes at the classroom level. Coordinated interventions should be implemented to enhance instruction through structured pedagogy, accompanied by the development of teaching and learning materials such as lesson plans and teacher guides that can be used to assist educators. Furthermore, offering permanent teacher coaching and peer mentoring programs, as well as establishing teacher support systems, can greatly improve the effective- ness of instruction. Additionally, programs should be developed to lower constraints for parents to actively participate in their children’s education. Targeted instruction that aligns with the individual learning levels of each student and utilizes appropriate educational technologies (and providing training to teachers on their effective use), is essential. Introducing minimum qualification levels for school principals, encompassing pedagogical management, monitoring, school leadership, teacher management, and objective setting, can contribute to the overall improvement of education quality. To facilitate this, classroom observation tools should be designed to help school principals understand classroom dynamics and provide constructive feedback to teachers for their ongoing development. Lastly, it is crucial to support schools in designing and implementing annual school improvement plans that involve all stakeholders, ensuring a collaborative approach to enhanc- ing educational outcomes. From the school-age perspective, in the medium/long-term, it is crucial to implement a series of key strategies to enhance education quality and equity. Strengthening national systems for student monitoring learning is essential. This involves developing robust systems to track students’ learning trajectories and utilizing these data to provide targeted support that leads to improved learning outcomes. Developing a comprehensive training and performance evaluation system for school principals, linked to school quality improvements, is essential for effective school management. Strategies should be implemented to strengthen the school management capacity of school directors and enhance school autonomy. This will empower schools to make informed decisions and improve overall educational outcomes. Efforts should be made to improve equity considerations in the allocation of public education spending. This includes ensuring that schools serving poor and vulnerable students have the minimum enabling conditions for learning such as adequate resources, infrastructure, and support services. By implementing these strategies, countries can foster a more equitable and quality education system that supports the holistic development and success of all students. To attract and retain high-quality teachers, pre-service teacher training programs should be made more appealing. Emphasis should be placed on providing support during the initial years of practice and should include mandatory induction programs. Creating flexible teacher career pathways is important, offering opportunities for promotion in both leadership and teaching tracks based on individual interests, abilities, and the needs of schools. Introducing a teacher practice eval- uation system at all levels and establishing teacher career advancement paths in teaching and management can help link performance evaluation to teacher training programs. To motivate teachers and ensure teaching quality, a well-phased, merit-based salary increase scheme should be introduced. This should be accompanied by measures for quality assurance and improved working conditions. From the youth and active population perspective, in the short term, it is necessary to implement several key measures to strengthen TVET systems. First, revising the TVET funding formula and linking it to institutional performance can incentivize better outcomes and ensure that resources are allocated efficiently. Additionally, introducing mechanisms to support income-generating activities of TVET institutions can enhance their financial sustainability and provide oppor- tunities for self-sufficiency. To prepare young people for the “jobs of the future”, it is crucial to expand alternative and 80 Policies to Promote More and Better Investments in Human Capital Central America Human Capital Review  | Promoting more and better investments in human capital non-school educational programs that allow them to certify skills and occupations relevant to emerging industries. More- over, considering the expansion of short-cycle programs tailored to meet the specific demands of employers can bridge the skills gap and facilitate smoother transitions into the labor market. A diverse mix of labor market programs should also be expanded to cater to different needs. This includes offering job search assistance, job intermediation services, training and skills certification programs, as well as hiring subsidies to facilitate employment opportunities. By providing a comprehen- sive range of support, individuals can access the appropriate resources to overcome constraints and achieve successful labor market outcomes. To address the challenges faced by vulnerable youth and the active population, it is essential to pilot and evaluate effective active labor market policies. These policies should be specifically targeted to these groups, aiming to improve their employment prospects and reduce inequalities. By identifying successful interventions, governments can scale up impactful initiatives and enhance outcomes for the most vulnerable populations. From the youth and active population perspective, in the medium/long-term, it is crucial to implement various strat- egies to enhance TVET systems by promoting private sector participation and addressing the needs of different target groups. Introducing incentive policies that encourage greater private sector involvement and increase the participation of youth and adults in TVET can contribute to a more dynamic and responsive training ecosystem. To ensure the relevance of acquired skills and better job matches, it is important to promote private-sector-led training initiatives. Collaborating with businesses can help align training programs with industry needs, leading to improved employability and a smoother transition into the workforce. Articulating education and employment policies jointly, under the leadership of Ministries of Education and Ministries of Labor, with active involvement from the private sector, is crucial. This collaborative approach ensures that policies are comprehensive, align with labor market demands, and promote effective skill development and employment opportunities. Moreover, establishing profiling mechanisms that recognize the diverse constraints faced by vulnerable youth and the active population is essential. By understanding the unique challenges individuals face, tailored support can be provided to address their specific needs, leading to more effective outcomes. To overcome additional constraints that may hinder participation, compensation or cross-referral protocols can be introduced. This could alleviate practical challenges such as childcare or transportation, which would promote attendance and completion of training and employment programs. Establishing or supporting quality after-care school programs is crucial in minimizing the time available for young people to engage in risky behavior. By offering structured activities, mentorship, and support during non-school hours, these programs contribute to the overall well-being and positive development of young individuals. By implementing these strategies, countries can foster private sector engagement, ensure skill relevance and job matching, align education and employment policies, provide targeted support to vulnerable groups, remove constraints to participation, and promote the holistic development of youth and adults in the TVET sector. Lastly, the impact of teenage pregnancies on education trajectories and lifetime earnings should be addressed through the development of public information campaigns that increase awareness and promote responsible sexual behavior. It is important to highlight the consequences of early parenthood and encourage young people to prioritize their education and career prospects. Expanding and ensuring access to sexual and reproductive care for adolescents, particularly in high-risk areas, is crucial. By providing comprehensive and youth-friendly services, including contraception, counseling, and support, young people can make informed choices regarding their reproductive health. This contributes to their overall well-being and empowers them to pursue their educational and career aspirations. From a cross-cutting perspective, in the short term, it is essential to continue investing in and expanding well-targeted social protection programs such as CCT programs. These initiatives provide direct support to vulnerable populations and help alleviate poverty. Additionally, incorporating behavioral approaches and nudges into social assistance programs can incentivize individuals to utilize essential services like healthcare, nutrition, and early childhood development. By providing incentives and promoting the use of these services, institutions can further support human capital accumulation and enhance the overall well-being of beneficiaries. Looking towards the medium and long term, it is important to consider consolidating and harmonizing social assistance programs. This entails eliminating fragmentation and moving away from less progressive programs. By streamlining and integrating various programs, institutions can achieve greater efficiency, improve targeting accuracy, and ensure that resources are allocated effectively to those who need them most. This consolidation also helps avoid duplication of efforts and ensures that support reaches the intended beneficiaries more comprehensively. 81 Policies to Promote More and Better Investments in Human Capital Central America Human Capital Review  | Promoting more and better investments in human capital Strengthening institutions for the delivery of better and more equal social services to foster human capital accumulation for all Institutional capacity plays a vital role in establishing effective education systems that accelerate learning, reduce dropout rates, promote labor market inclusion for youth, and facilitate the creation of high-quality jobs. Strength- ening institutions is a complex and time-consuming process, unique to each country based on their specific challenges and constraints. By prioritizing institutional strengthening, countries can build a solid foundation for sustainable and inclusive development in both education and labor market sectors. While strengthening institutional capacity might take different forms in different countries based on their own challenges and constraints, there are some of the common elements in the education sector and labor markets that are described below. From the school-age population perspective, in the short term, it is essential to introduce an effective system of institutional empowerment and accountability at all levels. This entails establishing mechanisms that promote transparency, inclusivity, and responsiveness in decision-making processes, ensuring that the voices and concerns of all stakeholders are heard and considered. By doing so, institutions can become more accountable and responsive to the needs and aspirations of the school-age population. Looking ahead to the medium and long term, it is vital to focus on strengthening the institutional capacity and regulatory framework necessary for the integration of educational technologies into the education sector. This involves investing in infrastructure, providing teachers with appropriate training, and developing robust curricula that incorporate digital tools and resources. By embracing educational technologies, institutions can enhance learning experiences, expand access to quality education, and prepare young people for the demands of the future. Furthermore, it is crucial to strengthen the regulatory framework to drive reforms in the teaching career. This includes revisiting and updating policies and practices to ensure that teaching is an attractive and respected profession. By providing professional development opportunities, career advancement pathways, and competitive compensation, institutions can attract and retain high-quality educators who are dedicated to nurturing the potential of the youth. From the perspective of the youth and active population, in the short term, promoting and strengthening private sector participation in TVET delivery is essential. Collaborating with businesses can enhance the relevance of training programs and improve job prospects for young individuals. Additionally, establishing an explicit focus on reaching and serving poor and vulnerable students ensures that they have equal access to quality education and training opportunities. Developing and strengthening evaluation and tracking systems is vital in assessing the quality of TVET programs delivered. By monitoring outcomes and continuously improving program effectiveness, institutions can provide high-quality education that meets the needs of the labor market. To improve labor intermediation, leveraging technological advancements such as artificial intelligence (AI) can facilitate better matching between job seekers and employers, optimizing the labor market outcomes for the youth and active population. Looking towards the medium to long term, introducing incentives to promote completion among vulnerable students is essential. By addressing the constraints students face and by offering support throughout their educa- tional journey, institutions can increase the students’ chances of successfully completing their training programs. Developing a performance management system to improve the delivery of employment services is also crucial. This involves establishing clear performance indicators, setting targets, and regularly assessing and improving the effectiveness of employment support programs. By monitoring performance, institutions can make data-driven decisions to enhance outcomes for job seekers. Strengthening the functioning of local employment offices is vital to better support the inclusion of youth and vulnerable populations in the labor market. This can be achieved through improvements in business processes, enhancing the profes- sionalism of employment counselors, and fostering collaboration with other stakeholders. By providing comprehensive and tailored employment services, local employment offices can effectively assist individuals in accessing job opportunities and achieving successful labor market integration. From a cross-cutting perspective, in the short term, it is essential to improve the up-to-datedness of social registry data by conducting cross-checks with other administrative databases and promoting interoperability. This ensures that the information used for social protection programs remains accurate and reliable. Additionally, promoting the adoption of digital payments facilitates efficient and secure transactions, improving the delivery of social assistance and reducing administrative burdens. Looking towards the medium and long term, introducing open and continuous registration for social protection programs is key. This enables individuals to access support whenever needed, without restrictive 82 Policies to Promote More and Better Investments in Human Capital Central America Human Capital Review  | Promoting more and better investments in human capital registration periods. Moreover, developing an integrated social information system that ideally includes information on the supply of programs creates a comprehensive database that facilitates better coordination and decision-making. This system allows for a holistic view of social protection efforts and enables institutions to identify gaps, avoid duplication, and optimize resource allocation. Table 2: Policy Areas for Improving the Adaptiveness of Education Systems and Labor Markets to Protect Human Capital, and Increase Resilience to More Frequent Shocks Outcomes  Policy Areas   Short Term  Medium/Long Term  School-Age Children • Learning losses during the Strengthen the adaptiveness of • Conduct standardized student • Expand connectivity in schools, access pandemic (expected) the education system learning assessments to estimate to technological devices, teacher • School dropouts during the learning losses. training, and use of educational pandemic •  Develop further online teaching platforms to support distance or materials and improve  its integration hybrid learning methods during • Temporary crowding out of students with the teaching training programs. shocks. from private to public schools •  Strengthen the development of early • Consolidate curricula, prioritizing warning systems. foundational learning, prerequisites for future learning, transferable •  Strengthen administrative records to competencies, and socioemotional inform the adaptation for education skills and social programs aimed at protecting children and the youth during crises Youth and Active Population • Large job losses, especially for Strengthen the adaptiveness of • Assess the coverage and the mix of • Strengthen or establish labor market women, youth and lower skilled the labor market /TVET system labor market programs to identify information systems exploiting during the pandemic gaps. administrative and real-time data • Limited adaptiveness of LM/TVET • Establish/strengthen the digital to improve the understanding and system  delivery for training /labor market monitoring of skills demanded by the programs to ensure continuity labor markets, and to respond flexibly • Interruption of accumulation of and support to youth and active to changing demand during crises  human capital or deterioration of existing human capital population • Risk of long-term scarring for new LM entrants and vulnerable workers Cross-Cutting • Limited shock responsiveness of Strengthen the shock- • Incorporate triggers and emergency • Improve coverage and dynamism the SP system responsiveness of the social protocols (through secondary of social registries, and education • Limited dynamism and coverage of protection system to protect legislation) for social assistance administrative data, which allow a SP delivery system human capital and increase programs allowing for rapid rapid identification of the vulnerable/ resilience of the poor and scalability and potential adjustments affected during crises. vulnerable  to eligibility during crises. • Undertake stress test of the SP system to identify potential gaps limiting opportunities 83 Policies to Promote More and Better Investments in Human Capital Central America Human Capital Review  | Promoting more and better investments in human capital Table 3: Policy Areas for Improving the Efficacy of Social Public Spending to Accelerate/Promote Human Capital Accumulation Outcomes  Policy Areas  Short Term  Medium/Long Term  School-Age Children • Low learning outcomes Improve national systems for • Use results from student learning • Strengthen national systems for student • High learning poverty rates monitoring education outcomes  assessments to support schools and learning monitoring. teachers improve their students’ learning • Develop systems to track students’ outcomes. learning trajectories and use them to •  Develop school learning monitoring support the improvement in learning systems with formative purposes for outcomes teacher to continuously assess learning levels at the classroom level • High repetition rates in Improve effectiveness of • Coordinated interventions for improving • Increase attractiveness of pre-service primary education teaching to foster students’ instruction through structured pedagogy; teacher training programs; emphasize • Low primary completion rates learning  develop teaching and learning materials support during initial years of practice (lesson plans, teacher guides) to support including mandatory induction. • School dropouts in lower and teachers. •  Develop flexible teacher career pathways upper secondary education • Offer permanent teacher coaching including possibilities for promotion • Large learning gaps within and peer mentoring programs and both in leadership and teaching tracks, classrooms  teachers support systems to improve the based on interest and ability, as well as effectiveness of instruction. school needs. • Develop programs to lower constraints •  Introduce teacher practice evaluation for parents to get involved in their system at all levels, through creating children’s education. teacher career advancement paths in • Targeted instruction that is appropriate both teaching and management areas, to the learning levels of each student as well as link this to teacher training leveraging on the use of education programs. technologies when appropriate and •  Introduce well-phased, merit- training teachers on their use based salary increase scheme with accompanying measures of quality assurance and improved working conditions • Low school directors’ Strengthen school autonomy • Introduce minimum qualification • Develop a solid training and performance managerial capacity  and management  levels for school principals related to evaluation system for school principals pedagogical management, monitoring, linked to school quality improvements. school leadership, teacher management, • Develop strategies to strengthen school and objectives setting. management capacity of school directors • Design classroom observation tools to and to improve school autonomy help school principals understand what is going on inside the classrooms and to providing feedback to teachers with formative purposes.  • Support schools in the design and implementation of annual school improvement plans involving all stakeholders • Low levels of human capital Improve equity of the public • N/A • Improve equity considerations in the • Large disparities in human education funding  allocation of public education spending. capital levels across • Improve minimum enabling conditions socioeconomic  distribution   for learning in schools serving poor and vulnerable students 84 Policies to Promote More and Better Investments in Human Capital Central America Human Capital Review  | Promoting more and better investments in human capital Outcomes  Policy Areas  Short Term  Medium/Long Term  Youth and Active Population • Low supply of relevant skills/ Develop effective mechanisms • Revise TVET funding formula, linking it to • Introduce incentive policies for greater future skills for funding TVET  institutional performance. private sector participation (including • Limited effectiveness of TVET  • Introduce mechanisms to support income PPP) and increased participation of youth generating activities of TVET institutions. and adults in TVET •  Low employability of youth Strengthen the links between • Expand alternative and non-school • Promote private sector led training to • Low supply of relevant skills/ education system and labor educational programs for young people ensure relevance of acquired skills and future skills market/training system to better that allow them to certify skills and better job matches. deploy human capital  occupations for “jobs of the future.”  • Define and articulate education and • Lack of information on (and changes in) skills in demand • Consider leveraging/expanding short employment policies jointly, led by cycle programs to meet specific demands Ministries of Education and Ministries of of the employers Labor, with active participation of private sector • Increased inactivity among Improve coverage and • Expand mix of labor market programs to • Establish (or strengthen) profiling youth (increased NEET rate) effectiveness of labor market/ ensure availability of a diverse set (from mechanisms to recognize the diverse • Interruption of accumulation training system  job search assistance, job intermediation, constraints of the vulnerable youth of human capital or to training/skills certification, and hiring and active population and provide deterioration of existing subsidies) to meet different needs. differentiated support. human capital, for vulnerable • Pilot and evaluate effective active labor • Introduce compensation or establish youth market policies targeted to vulnerable cross-referral protocols to address • Risk of long-term scarring youth and active population  other constraints (e.g., childcare, for new LM entrants and transportation), to promote attendance vulnerable workers and completion of labor market interventions  • Poor labor market outcomes for youth • Gaps in coverage and mix of labor market programs to address diverse constraints  • High school dropouts in Develop strategies to tackle • Develop public information campaigns • Establish and/or support quality after lower and upper secondary teenage pregnancy  to increase awareness about the impact care school programs to minimize time education  of teenage pregnancies on education available to engage in risky behavior trajectories and lifetime earnings. • Expand and ensure access to sexual and reproductive care for adolescents in high-risk areas Cross-Cutting • Low coverage and benefit Improve coverage and • Continue to invest in and expand the • Consider consolidating/harmonizing levels for CCT programs  effectiveness of social assistance well-targeted social protection programs/ social assistance programs to eliminate • Large disparities in human programs to accelerate/promote CCTs. fragmentation, shifting away from less capital levels across human capital accumulation of •  Incorporate behavioral approaches/ progressive programs; and improving socioeconomic distribution the poor and vulnerable  nudges to social assistance programs to targeting. (with lower levels of human provide incentives to use services (health, •   capital for those at the bottom nutrition, ECD etc.) to further support of the income distribution)   human capital accumulation 85 Policies to Promote More and Better Investments in Human Capital Central America Human Capital Review  | Promoting more and better investments in human capital Table 4: Policy Areas for Strengthening Institutions in the Delivery of Better and More Equal Social Services to Foster Human Capital Accumulation Outcomes  Policy Areas  Short Term  Medium/Long Term  School-Age Children • Low institutional capacity for the Improve institutional capacity for • Introduce effective system of • Strengthening the institutional capacity delivery of education services  the delivery of education services   institutional empowerment and and regulatory framework needed for the accountability at all levels. integration of educational technologies into the education sector with a long-term vision. • Strengthening the regulatory framework to push forward reforms to the teaching career Youth and Active Population • Limited effectiveness of TVET Increase access to TVET especially • Promote/strengthen private sector • Introduce incentives to promote • Low employability of vulnerable among the vulnerable youth  participation in TVET delivery. completion among vulnerable students   youth  • Establish (strengthen) an explicit focus on reaching/serving poor and vulnerable students • Limited effectiveness of Improve employment services  • Develop/strengthen evaluation • Develop performance management employment services and tracking systems to assess the system to improve delivery of • Poor labor market outcomes for quality of the programs delivered. employment services.  youth and vulnerable • Improve labor intermediation, • Strengthen the functioning of the local leveraging advances in employment offices (improvements in technology (i.e., A.I., incorporated business processes, professionalization in job matching portals) of employment counselors etc.) to better support the labor market inclusion/ transition of the youth and vulnerable Cross-Cutting • Limited dynamism and coverage Improve inclusivity and dynamism • Improve the up to datedness of • Introduce open and continuous of social protection delivery of social protection delivery systems  social registry data through cross- registration for social protection system  checks with other administrative programs. data bases; and promote • Develop an integrated social information interoperability. system, ideally also with information on • Promote adoption of digital supply of programs payments   86 Policies to Promote More and Better Investments in Human Capital Central America Human Capital Review  | Promoting more and better investments in human capital References Abdul Latif Jameel Poverty Action Lab (J-PAL). 2019. “Tailoring instruction to students’ learning levels to increase learn- ing.”J-PAL Policy Insights. Last modified January 2019. https://doi.org/10.31485/pi.2522.2019. Adelman, M., and M. Székely. 2017. “An Overview of School Dropout in Central America: Unresolved Issues and New Challenges for Education Progress.” European Journal of Educational Research 6 (3). https://pdf.eu-jer.com/ EU-JER_6_3_235_Adelman_etal.pdf. Arayavechkit et al. (editors). 2022. Remittances and Development: The Experience of Northern Central America. World Bank Group, Washington DC. Azevedo, et al. 2021. “The State of the Global Education Crisis: A Path to Recovery.” World Bank Group, Washington DC. http://documents.worldbank.org/curated/en/416991638768297704/The-State-of-the-Global-Education-Crisis-A-Path- to-Recovery. Azevedo, et al. 2022. “The State of Global Learning Poverty: 2022 Update.” World Bank Group, Washington DC. https:// www.worldbank.org/en/topic/education/publication/state-of-global-learning-poverty. Behrman, J., and D. Anil. 1991. “School Repetition, Dropouts, and the Rates of Return to Schooling: The Case of Indo- nesia.” Oxford Bulletin of Economics and Statistics 53 (4): 467-480. Bentaouet-Kattan, R., and M. Székely. 2015. “Patterns, Consequences and Possible Causes of Dropout in Upper Secondary Education in Mexico.” Education Research International 2015, Article ID 676472. Bruns, B., and J. Luque. 2015. “Great Teachers: How to Raise Student Learning in Latin America and the Caribbean.” World Bank Group, Washington, DC. https://openknowledge.worldbank.org/handle/10986/20488 License: CC BY 3.0 IGO. Bunivic, M., and A. Morrison. 1999. “Violence as An Obstacle to Development.” Social Development Division of the Sustainable Development Department. Interamerican Development Bank, Washington DC. https://publications.iadb. org/publications/english/document/Violence-as-an-Obstacle-to-Development.pdf. Cardenas, Mauricio, Rafael de Hoyos, and Miguel Székely. 2015. “Out of School and Out of Work Youth in Latin America: A Persistent Problem in a Decade of Prosperity.” Economia 16 (1). Cawley, J. 2015. An Economy of Scales: A Selective Review of Obesity’s Economic Causes, Consequences, and Solutions. Journal of Health Economics 43 (2015): 244–268. Corraco, M., and E. Lazarova. 2012. “Migration, Human Capital Formation, and Growth: An Empirical Investigation.” World Development 40 (5): 938-955. Development Initiatives. 2020. 2020 Global Nutrition Report: Action on Equity to End Malnutrition. Bristol, UK. Devercelli, A., M. Bendini, A. Hasan, M. Le, and A. Sherrilee. 2022. Learning During the Early Years: What Is It? Why Does it Matter? And How Do We Promote It? World Bank Group, Washington, DC. Foureaux, M., and L. Menezes. 2021. Violence and Human Capital Investment. Journal of Labor Economics 39 (3): 787-823. Ham, A., E. Vazquez, and M. Yanez-Pagans. 2022. The Effects of Differential Exposure to COVID-19 on Education Outcomes in Guatemala. World Bank Working Paper. Mimeo. Hammond R. A., and R. Levine. 2010. The Economic Impact of Obesity in the United States. Diabetes, Metabolic Syndrome and Obesity 30 (3): 285-95. Isik-Dikmelik et al. (2023). Strengthening SPL Systems in Central America for an Inclusive and Resilient Recovery. World Bank: Washington D.C. Mazhar, U., and F. Rehman. 2021. “Productivity, Obesity, and Human Capital: Panel Data Evidence.” Economics & Human Biology 44 (101096). 10.1016/j.ehb.2021.101096. Oreopoulos, Philip, and Kjell Salvanes. 2011. “Priceless: The Nonpecuniary Benefits of Schooling.” Journal of Economic Perspectives 25 (1): 159-184. Patrinos, Harry, and George Psacharopoulos. 2004. “Returns to Investment in Education: A Further Update.” Education Economics 12 (2): 111-134. Pennings, Steven. 2020. “The Utilization-Adjusted Human Capital Index.” Policy Research Working Paper 9375. World Bank, Washington, DC. Pritchett, L., and J. Sandefur. 2020. “Girls’ Schooling and Women’s Literacy: Schooling Targets Alone Won’t Reach Learning Goals.” International Journal of Educational Development 78. https://doi.org/10.1016/j.ijedudev.2020.102242. Ridao-Cano, Cristobal, and Christian Bodewig. 2018. Growing United: Upgrading Europe’s Convergence Machine. World Bank Group, Washington, DC. https://documentsinternal.worldbank.org/search/29696188. Ulku, Hulya and Gabriel Zaourak. 2021. “Unleashing Central America’s Growth Potential: Synthesis Report.” World Bank, Washington, DC. 87 Policies to Promote More and Better Investments in Human Capital Central America Human Capital Review  | Promoting more and better investments in human capital UNESCO (United Nations Educational, Scientific and Cultural Organization). 2021. Los aprendizajes fundamentales en América Latina y el Caribe. Evaluación de logros de los estudiantes. Estudio Regional Comparativo y Explicativo (ERCE 2019). Laboratorio Latinoamericano de la Evaluación de la Calidad de la Educación. Oficina Regional de Educación para América Latina y el Caribe (OREALC). UNFPA (United Nations Population Fund). 2022. Consecuencias socioeconómicas del embarazo en la adolescencia en América Latina y el Caribe – Segundo Informe regional. Implementación de la Metodología MILENA 1.0 en 10 países. UN IGME (United Nations Inter-Agency Group for Child Mortality Estimation). 2021. “Levels & Trends in Child Mortality: Report 2021.” United Nations Children’s Fund, New York. United Nations Children’s Fund. 2021. “The State of the World’s Children 2021: On My Mind-Promoting, Protecting, and Caring for Children’s Mental Health.” New York. World Bank. 2018a. World Development Report 2018: Learning to Realize Education’s Promise. World Bank, Washington, DC. https://openknowledge.worldbank.org/handle/10986/28340. World Bank. 2018b. The Human Capital Project. World Bank, Washington, DC. https://openknowledge.worldbank.org/ handle/10986/30498 License: CC BY 3.0 IGO. World Bank. 2019a. The Contributions of Social Protection and Jobs to Human Capital Formation. World Bank, Washing- ton DC. https://collaboration.worldbank.org/content/sites/collaboration-for-development/en/groups/cop-on-service- delivery-systems-in-south-asia/groups/human-capital/documents.entry.html/2019/05/29/the_contributionof-I8w7. html. World Bank. 2019b. Ending Learning Poverty: What Will It Take? World Bank Group, Washington, DC. https://open- knowledge.worldbank.org/handle/10986/32553. World Bank. 2021a. Education for All by Safeguarding Health: A Look at the Performance of Public-Sector Distance Learning in the Dominican Republic in Times of COVID-19. World Bank Group, Washington, DC. https://docu- mentsinternal.worldbank.org/Search/33833467. World Bank. 2021b. The Human Capital Index 2020 Update: Human Capital in the Time of COVID-19. World Bank, Washington, DC. Doi:10.1596/978-1-4648-1552-2. World Bank. 2021c. Jobs Interrupted: The Effects of COVID-19 in the Latin America Labor Markets. Poverty Global Practice Policy Note 2. Washington DC. https://openknowledge.worldbank.org/bitstream/handle/10986/35866/Jobs- Interrupted-The-Effects-of-COVID-19-in-the-Latin America-Labor-Markets.pdf?sequence=5. World Bank. 2022a. Collapse & Recovery: How COVID-19-10 Eroded Human Capital and What To Do About It? HD Flagship Report. World Bank, Washington DC. World Bank. 2022b. “Educación para Todos Preservando la Salud: Una Mirada al Funcionamiento de la Educación Pública a Distancia en República Dominicana en tiempos de COVID-19”. Education Global Practice, Washington DC. World Bank. 2022c. The Learning Crisis in Latin America and The Caribbean and the COVID-19 Pandemic: Sobering Results of a Deepening Trend. Regional Note from the Education Latin America and Caribbean Team. https://open- knowledge.worldbank.org/handle/10986/37924. World Bank, UNESCO, UNICEF, USAID, FCDO, Bill & Melinda Gates Foundation. 2022. The State of Global Learning Poverty: 2022 Update. World Bank, Washington DC. World Bank, UNICEF, and UNESCO. 2022. Two Years After: Saving a Generation. World Bank, Washington, DC. https:// openknowledge.worldbank.org/handle/10986/37586. 88 Annex: Additional Information on Chapter 3 Central America Human Capital Review  | Promoting more and better investments in human capital ANNEX: ADDITIONAL INFORMATION ON CHAPTER 3 Table A.1: Determinants of NEET Status, 2019 Panama Dominican Republic Costa Rica Age 0.023*** 0.010*** 0.017*** (0.000) (0.000) (0.000) Years of education -0.017*** -0.014*** -0.030*** (0.000) (0.000) (0.000) Male (=1) -0.101*** -0.087*** -0.072*** (0.001) (0.001) (0.001) Married (=1) 0.084*** 0.076*** 0.026*** (0.001) (0.001) (0.001) 0- to 3-year-old child in the household 0.054*** 0.081*** 0.060*** (0.001) (0.001) (0.001) Head of household, years of education -0.005*** -0.001*** -0.006*** (0.000) (0.000) (0.000) Head of household, employed (=1) -0.017*** -0.052*** 0.007*** (0.001) (0.001) (0.001) Urban (=1) 0.005*** -0.023*** 0.007*** (0.001) (0.001) (0.001) Observations 714,968 1,832,545 682,129 Standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1 Note: Region fixed effects included. This table corresponds to the marginal effects of a probit model. The data comes from Costa Rica 2019 (ECE), Dominican Republic 2019 (ECNFT), and Panama 2019 (EML) using SEDLAC harmonization. 89 Annex: Additional Information on Chapter 3 Central America Human Capital Review  | Promoting more and better investments in human capital Figure A.1: School to Work Transition by Age, circa 2011 a) Costa Rica, Ages 15 to 35 b) Dominican Republic, Ages 10 to 35 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 15 16 17 18 19 20 21 22 24 25 26 27 28 29 30 31 32 33 34 35 10 12 14 16 18 20 22 24 26 28 30 32 34 c) Guatemala, Ages 10 to 35 d) Panama, Ages 10 to 35 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 10 12 14 16 18 20 22 24 26 28 30 32 34 10 12 14 16 18 20 22 24 26 28 30 32 34 Not in school or work School and work School only Work only Source: Authors’ elaboration based on Costa Rica 2011 (ECE), Dominican Republic 2010 (ECNFT), Guatemala 2011 (ENEI), Panama 2011 (EML) using SEDLAC harmonization. 90 Annex: Additional Information on Chapter 3 Central America Human Capital Review  | Promoting more and better investments in human capital Figure A.2: School to Work Transition for Females by Age, 2011 a) Costa Rica, Ages 15 to 35 b) Dominican Republic, Ages 10 to 35 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 15 17 19 21 24 26 28 30 32 34 10 12 14 16 18 20 22 24 26 28 30 32 34 c) Guatemala, Ages 10 to 35 d) Panama, Ages 10 to 35 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 10 12 14 16 18 20 22 24 26 28 30 32 34 10 12 14 16 18 20 22 24 26 28 30 32 34 Not in school or work School and work School only Work only Source: Authors’ elaboration based on Costa Rica 2019 (ECE), Dominican Republic 2019 (ECNFT), Guatemala 2019 (ENEI), Panama 2019 (EML) using SEDLAC harmonization. Figure A.3: Employment Statistics of Top 60 Percent of Income, 2019 80 6 70 5 60 4 50 40 3 Total employment rate 30 Share of informal workers 2 20 Unemployment rate 1 10 0 0 Costa Rica Dominican Republic Guatemala Panama Source: Authors’ elaboration based on Dominican Republic 2019, Guatemala 2019 using SEDLAC harmonization. Data for Panama 2019 and Costa Rica 2019 provided by the National Statistics Offices. 91 Annex: Additional Information on Chapter 3 Central America Human Capital Review  | Promoting more and better investments in human capital Figure A.3: Composition of Employment by Level of Education, by Economic Sector a) Total Employment b) Agriculture Panama Panama Dominican Dominican Republic Republic Costa Rica Costa Rica Guatemala Guatemala 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% c) Industry d) Services Panama Panama Dominican Dominican Republic Republic Costa Rica Costa Rica Guatemala Guatemala 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Low Medium High Not report Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT), Guatemala 2019 (ENEI), Panama 2019 (EML) using SEDLAC harmonization. Data for Costa Rica 2019 (ECE) provided by the NSO. Note: Education level: low 0-8 years, medium 9-13 years, 14 or more years of education. 92 Annex: Additional Information on Chapter 3 Central America Human Capital Review  | Promoting more and better investments in human capital Figure A.4 Task Profile of the Employed, Guatemala, 2019 a) By Gender b) By Gender and Educational Level 0.6 2.5 0.4 2 1.5 0.2 1 0 0.5 -0.2 0 -0.4 -0.5 -1 -0.6 -1.5 -0.8 -2 -1 Female Male -2.5 Male - Medium Male -High Female - Low Female - Medium Female - High NRCA NRCI RC RM NRM Source: Authors’ elaboration based on Guatemala 2019 (ENEI) using SEDLAC harmonization, and the Occupational Information Network (O*NET). Note: Panel B results presented with respect to male workers with low level of education. NRCA: Nonroutine Cognitive Analytical, NRCI: Nonroutine Cognitive Inter- personal, RC: Routine Cognitive, RM: Routine Manual, NRM: Nonroutine Manual. Figure A.5: Task Profile of the Employed in Dominican Republic by Gender and Educational Level, 2019 2.5 2 1.5 1 0.5 0 -0.5 -1 -1.5 Male - Medium Male -High Female - Low Female - Medium Female - High NRCA NRCI RC RM NRM Source: Authors’ elaboration based on Dominican Republic 2019 (ECNFT) using SEDLAC harmonization and The Occupational Information Network (O*NET). Note: With respect to male workers with low level of education. NRCA: Nonroutine Cognitive Analytical, NRCI: Nonroutine Cognitive Interpersonal, RC: Routine Cognitive, RM: Routine Manual, NRM: Nonroutine Manual. 93 -1.5 -1 -0.5 0 0.5 1 1.5 Agriculture, forestry and fishing Mining and quarrying Manufacturing Annex: Additional Information on Chapter 3 Electricity, gas, steam and air conditioning supply Water supply Construction Wholesale and retail trade; repair of motor vehicles and motorcycles NRCA Transportation and storage Accommodation and food service activities NRCI Information and communication Financial and insurance activities 94 RC Real estate activities Central America Human Capital Review  | Promoting more and better investments in human capital Figure A.6: Task/Skill Content of Jobs by Sector, Costa Rica, 2019 Professional, scientific and technical activities RM Administrative and support service activities Public administration and defence; compulsory social security NRM Education Human health and social work activities Arts, entertainment and recreation Other service activities Activities of households as employers; undifferentiated goods - and services-producing activities of households for own use Activities of extraterritorial organizations and bodies Total Source: Authors’ elaboration based on Costa Rica 2019 (ECE) provided by the NSO and standardized by the team, and The Occupational Information Network (O*NET). Note: NRCA: Nonroutine Cognitive Analytical, NRCI: Nonroutine Cognitive Interpersonal, RC: Routine Cognitive, RM: Routine Manual, NRM: Nonroutine Manual.