Building Human Capital Where It Matters Homes, Neighborhoods, and Workplaces Alaka Holla, Norbert Schady, and Joana Silva Editors Building Human Capital Where It Matters This book, along with any associated content or subsequent updates, can be accessed at https://hdl.handle.net/10986/44282. https://reproducibility.worldbank.org A reproducibility package is available for this book in the Reproducible Research Repository at https://reproducibility.worldbank.org/catalog/461. Scan to go to the complete publication. Building Human Capital Where It Matters Homes, Neighborhoods, and Workplaces Alaka Holla, Norbert Schady, and Joana Silva Editors © 2026 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 29 28 27 26 This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, links/footnotes, and other information shown in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The citation of works authored by others does not mean The World Bank endorses the views expressed by those authors or the content of their works. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http:// creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Holla, Alaka, Norbert Schady, and Joana Silva, eds. 2026. Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces. World Bank. doi: 10.1596/978-1-4648-2277-3. License: Creative Commons Attribution CC BY 3.0 IGO Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. ISBN (paper): 978-1-4648-2277-3 ISBN (electronic): 978-1-4648-2316-9 DOI: 10.1596/978-1-4648-2277-3 Cover design: Nicole Hamam, Hamam Design. Further permission required for reuse. Interior design: Jihane El Khoury Roederer, Global Corporate Solutions Unit, World Bank, and Nicole Hamam, Hamam Design. The Library of Congress Control Number has been requested. Contents Foreword Acknowledgments About the Editors and Contributors Main Messages Executive Summary 1 Introduction Norbert Schady Why is human capital so important? Human capital: Stagnating in low- and lower-middle-income countries The importance of settings to human capital policy Notes References 2 Human Capital Accumulation in the Home Alaka Holla The home matters Why does the home matter? Policy recommendations Notes References 3 Human Capital Accumulation in Neighborhoods Andres Yi Chang, Patrick Hoang-Vu Eozenou, and Ildo Lautharte The neighborhood matters Why neighborhoods matter Policy implications Conclusion: Putting it all together Notes References 4 Human Capital Accumulation at Work Joana Silva The workplace matters What limits human capital accumulation at work? ix xi xiii xvii xxi 1 2 4 9 12 13 16 18 22 32 42 43 50 52 55 60 65 66 67 72 74 75 v vi Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces Policy recommendations Conclusion: Putting it all together Notes References 5 Implementing a Settings Approach in Policy Alaka Holla, Norbert Schady, and Joana Silva A settings lens to solve human development challenges Reforms aimed at settings can spur gains in human capital Tracking progress: A national and global data agenda Conclusion Notes References Boxes 2.1 3.1 4.1 4.2 4.3 4.4 4.5 5.1 5.2 5.3 Violent punishment at home: Belief versus behavior Defining neighborhoods as a geographic concept Measuring learning at work Labor income grows more slowly during periods of self-employment or during work at small firms Childcare and women’s labor force participation Shifting gender perceptions and addressing security concerns Expanding the focus of human capital policy to the workplace is critical for jobs Reducing malnutrition in Indonesia and Peru A social registry as a tool for integration A research agenda 86 99 99 101 106 109 111 118 124 125 125 24 52 79 81 90 93 97 111 114 122 Figures ES.1 ES.2 ES.3 ES.4 ES.5 ES.6 1.1 1.2 1.3 1.4 1.5 2.1 2.2 2.3 Human capital is built in the home, in the neighborhood, and in the workplace Learning has stagnated in many parts of the world Early skill deficits persist Neighborhood characteristics shape human capital in Brazil Returns to experience are lower among the self-employed than among wage workers in low- and middle-income countries Policy priorities for the home, the neighborhood, and the workplace Changes in adult height, by country income group School enrollment and the tertiary completion rate Student learning across countries Trends in labor force participation rates, prime-age adults, ages 25–54, 2004–24 Labor income as a function of on-the-job experience, men ages 18–67, by country income status Human capital investment at home across the life cycle Child nutrition and skills, by maternal educational attainment Skill deficits among children, by maternal educational attainment xxii xxiii xxv xxvii xxix xxx 5 6 6 7 8 18 20 21 Contents vii Skills in childhood, by resources and care at home 2.4 B2.1.1 Use and endorsement of violent punishment to discipline children ages 1–14 2.5 2.6 2.7 Resources build human capital across the life cycle Share of children ages 5–14 living with neither parent Selected indicators of well-being, children ages 10–17, by left-behind status, China Lifelong effects of preschool and parenting programs Neighborhoods affect human capital through distinct channels Neighborhood poverty in childhood and outcomes in adulthood, Brazil Households use local schools and local health facilities School and health care quality varies substantially across villages Village sanitation coverage is key to controlling waterborne disease in India Probability of secondary-school graduation, by gang territory, San Salvador, El Salvador How learning occurs in various contexts Most people are in jobs that offer little opportunity for learning at work Farmer productivity on small plots in Sub-Saharan Africa does not increase much with experience Returns to experience are lower among the self-employed than among wage workers 2.8 3.1 3.2 3.3 3.4 3.5 3.6 4.1 4.2 4.3 4.4 B4.2.1 Panel data confirm low returns to experience during self-employment in 4.5 China and India Returns to experience are lower for the self-employed and wage workers in small firms than for wage workers in medium and large firms Population not accumulating human capital through work Soft skills training on the job raises productivity among trained workers and has knowledge spillovers to untrained coworkers B4.3.1 Survey response: Do you agree that a preschooler suffers if the 4.6 4.7 mother is working? B4.5.1 The virtuous cycle of more human capital and better jobs 5.1 What needs to happen in the home, in the neighborhood, and in the workplace to avoid malnutrition Profile of a social registry The progression toward a fully developed data system 5.2 5.3 Map 3.1 Gang control, by neighborhood, San Salvador, El Salvador Tables 2.1 2.2 3.1 4.1 4.2 4.3 Evidence on increasing resources at home among poor families Evidence on improving care environments for children and adolescents Human capital policies for struggling neighborhoods How human capital accumulates at work Opportunities for human capital accumulation, by job type Evidence on policies that promote more learning on the job 22 25 28 29 31 37 53 54 56 57 58 60 75 78 80 81 83 84 85 88 92 97 110 114 121 60 33 35 61 74 76 87 viii Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces 4.4 4.5 5.1 5.2 Evidence on policies that remove barriers to the participation of women and youth in the labor markets Evidence on policies to create more high human capital jobs Outcome measures to track progress in human capital accumulation Input measures to track progress in human capital accumulation 95 98 119 120 Foreword Human capital—people’s health, skills, and knowledge—is the most valuable asset any society possesses. It is the foundation of economic growth, poverty reduction, and shared prosperity. No country has achieved sustained development without investing in it. Global trends in human capital outcomes show that progress in human capital is stalling if not reversing in many low- and middle-income countries. Children today are less likely than children 15 years ago to read with understanding or be able to do basic mathematics problems. Adult height—a marker of population health— has declined in many places. Most workers in low- and middle-income countries are in jobs with limited formal training or on-the-job learning opportunities. Women and youth are especially affected. Only 40 percent of women are in paid employment, and nearly one in five young people are neither working nor studying. These trends matter because differences in human capital account for roughly two-thirds of the gap in per capita income between rich and poor countries. This report argues that reversing these trends requires rethinking how human capital policy is designed and delivered. Much of the existing policy and research agenda has focused on expanding access to and improving the quality of health and education services or on specific stages of the life cycle. While essential, these approaches are not sufficient. Human capital is built not only through systems but also in specific settings—most importantly in homes, neighborhoods, and workplaces—where daily decisions, interactions, and opportunities shape outcomes over time. This report puts forward a simple but underappreciated observation: human capital is not built in sectors alone, nor only at specific stages of life. It is built— slowly, unevenly, and often invisibly—in places: through the health of a child, the quality of a classroom, the safety of a neighborhood, and the learning that takes place or fails to take place at work. In homes, nutrition, care, and early stimulation shape lifelong trajectories. In neighborhoods, the quality of schools, health services, infrastructure, safety, and social norms influence what people can become. And, in workplaces, skills are refined, or they are wasted, and learning by ix x Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces doing can either accelerate productivity or leave workers stuck in low-return activities. By adopting a settings-based lens, this report complements sectoral and life- cycle approaches and offers a more well-integrated framework for action. It shows how constraints in one setting can undermine investments made in another, and how coordinated action across homes, neighborhoods, and workplaces can unlock far greater returns. It also highlights the roles of public and private actors and the importance of more closely aligning financing, incentives, and institutions to support human capital accumulation where it actually happens. At a moment when countries face overlapping challenges from demographic shifts and rapid technological change to climate shocks and fragility, investing more effectively in human capital is not optional. It is foundational. This report offers practical insights and policy priorities to help countries move from fragmented interventions to coherent, people-centered strategies. It aims to support governments and partners in renewing progress on human capital accumulation and unlocking people’s potential. And, in doing so, it seeks to restore progress where it has stalled and expand opportunities where they have been out of reach. Mamta Murthi Vice President, People World Bank Group Acknowledgments This report would not have been possible without the generous support of a team of colleagues and collaborators. Special thanks are due to Mamta Murthi, Vice President for the People Vertical of the World Bank Group, for her leadership. Dena Ringold, Director, Strategy and Operations for the People Vertical of the World Bank Group, provided valuable guidance and support. The report also benefited from the guidance of Hana Brixi, Eliana Carranza, Halil Dundar, Indermit Gill, Jamele Rigolini, Alberto Rodriguez, Michal Rutkowski, Fadia M. Saadah, and Jaime Saavedra. The empirical analyses presented in this volume greatly benefited from the expertise and many contributions of Brian Stacy, who provided valuable inputs to chapter 1 and the publication as a whole. He led the production of the report’s reproducibility package. Francisco Bivar, Sergio Padilla, and Shuqiao Sun in the People Chief Economist Office provided outstanding research assistance. Florence Kondylis supplied valuable material related to agricultural labor markets. The report was shaped by multiple in-country consultations with government, civil society, and private sector representatives. The authors are especially grateful to Caroline Vagneron for leading these consultations and to Heba Elgazzar, René Antonio Leon Solano, and Facundo Cuevas for organizing these consultations in country and providing insights on the applicability of the report’s findings to current policy agendas. In-person consultations were held in Brazil, Colombia, Côte d’Ivoire, El Salvador, Poland, Tajikistan, and Türkiye, complemented by shorter virtual events in other countries. Several consultations were also held with academia and international institutions, including Paris School of Economics, Institut d’Études Politiques de Paris, Opportunity Insights at Harvard University, University of Oslo, Organisation for Economic Co-operation and Development, Center for Global Development, and African Economic Research Consortium–Kenya. The report benefited considerably from the comments and suggestions of peer reviewers, including Ndiame Diop, Roberta Gatti, Aart Kraay, Denis Medvedev, Martin Raiser, Martin Rama, and Hirokazu Yoshikawa, at the decision meeting xi xii Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces stage, and Pedro Carneiro, Ndiame Diop, Pascaline Dupas, Deon Filmer, Roberta Gatti, Aart Kraay, Santiago Levy, David McKenzie, and Carolina Sanchez Palermo, at the concept note stage. The analyses and interpretations also benefited from the generous insights of Gabriel Demombynes, Shanta Devarajan, Raquel Fernandez, Deon Filmer, John N. Friedman, Aart Kraay, Martin Rama, Michal Rutkowski, and Hirokazu Yoshikawa during a quality enhancement review. The authors are grateful for the insightful comments of World Bank colleagues, including Luis Benveniste, Ousmane Dione, Doerte Doemeland, Elisabeth Huybens, Sebastian Molineus, Ana María Muñoz Boudet, Renaud Seligmann, Ayat Soliman, Juan Pablo Uribe, Anna Wellenstein, Marina Wes, and Fei Yuan. The authors are very grateful to the External and Corporate Relations team of the World Bank, especially Christine Montgomery, Nandita Roy, Giannina Raffo, Margaret Allen, Erick Rabemananoro, Hellen Wagura, Joe Qian, Maria Elisa Costa, and Sidronio Araujo. Special thanks also go to Filloreta Jusufi, Maarten Lambrechts, Rochelle O’Hagan, Akash Pradhan, and Divyanshi Wadhwa for their work on creating the interactive versions of figures that appear in the report. Karen Hoyos Baza ensured all report activities happened on schedule. Nicole Hamam designed the volume, including the cover art. Nancy Morrison skillfully edited the report, which was copyedited by Robert Zimmermann. Gwenda Larsen and Ann O’Malley proofread the typeset pages, and Stephen Pazdan in the World Bank’s formal publishing program coordinated production of the volume. The authors are also grateful to Cindy Fisher and Patricia Katayama for their guidance on the publishing process. About the Editors and Contributors Editors Alaka Holla is a Lead Economist in the Office of the Chief Economist for People and the Office of the Chief Statistician of the World Bank Group. Holla is also the Program Manager of the World Bank’s Strategic Impact Evaluation Fund, a multidonor trust fund that promotes the use of randomized controlled trials, quasi- experimental methods, and evidence aggregation to inform policies in education, health, water and sanitation, and early childhood development in low- and middle- income countries. Since joining the World Bank as a Young Professional, she has worked in both research and operational settings in a diverse set of country contexts and on multilateral initiatives to promote the measurement of child development and the costing of programs. Holla’s work has appeared in peer- reviewed publications such as American Economic Review, Health Affairs, and Journal of Public Economics. She has been a coauthor of World Bank reports, including the flagship World Development Report 2015: Mind, Society, and Behavior; Citizens and Service Delivery: Assessing the Use of Social Accountability Approaches in the Human Development Sectors; and Collapse and Recovery: How the COVID-19 Pandemic Eroded Human Capital and What to Do About It. Her research has focused on the quality of health care, early childhood development and education, discrimination, and cost analysis. Holla received a PhD in economics from Brown University. Norbert Schady is Chief Economist for People at the World Bank Group. He was Principal Economic Adviser, Social Sector, at the Inter-American Development Bank (2010–21). At the World Bank, he was Senior Economist in the Development Research Group (2003–10), Economist in the Poverty Group in the Latin America and Caribbean Region (2000–03), and a Young Professional (1998–2000). He has taught at Georgetown University and Princeton University. Schady has published extensively in academic journals, including American Economic Journal: Applied Economics; American Economic Journal: Economic Policy; Journal of Development Economics; Journal of Human Resources; Quarterly Journal of Economics; and Review of Economics and Statistics. He is also the author or coauthor of numerous flagship reports, including The Early Years: Child Well-Being and the Role of Public Policy; Conditional Cash Transfers: Reducing Present and xiii xiv Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces Future Poverty; Closing the Gap in Education and Technology; and Collapse and Recovery: How the COVID-19 Pandemic Eroded Human Capital and What to Do About It. Schady’s main research areas are early childhood development, teacher quality, cash transfer programs, and the effects of economic contractions on the accumulation of human capital. He received a BA from Yale University and a PhD from Princeton University. Joana Silva is Deputy Chief Economist for People at the World Bank Group and an associate professor at Católica Lisbon School of Business and Economics. She has expertise in applied economics, labor economics, international economics, development economics, and public economics. Silva has led several World Bank lending operations and has extensive experience advising governments, in particular, on the design, implementation, and evaluation of economic reforms, social programs, and monitoring and evaluation systems. Her research has been published in leading academic journals, including American Economic Review, Econometrica, and Journal of International Economics. She has coauthored six books, including four World Bank regional flagship reports: Employment in Crisis: The Path to Better Jobs in a Post–COVID-19 Latin America; Wage Inequality in Latin America: Understanding the Past to Prepare for the Future; Inclusion and Resilience: The Way Forward for Social Safety Nets in the Middle East and North Africa; and Collapse and Recovery: How the COVID-19 Pandemic Eroded Human Capital and What to Do About It. She holds a PhD in economics from the University of Nottingham. Contributors Patrick Hoang-Vu Eozenou is a Senior Economist in the Global Engagement Unit of the Health, Nutrition, and Population Global Practice at the World Bank. He has more than 15 years of experience in development economics, with a focus on poverty, health financing, equity, and financial protection. Before joining the World Bank as a Young Professional, he was a postdoctoral fellow at the International Food Policy Research Institute, where he worked on the evaluation of nutrition and agricultural interventions in rural settings. Eozenou has led World Bank health operations in Mali, and he has led the World Bank’s engagement on joint World Bank–World Health Organization monitoring of universal health coverage and has contributed to several flagship reports. His research has been published in peer-reviewed journals, including the Journal of Development Economics and The Lancet Global Health. Eozenou holds a PhD in economics from the European University Institute. Ildo Lautharte is a Senior Economist in the Education Global Practice at the World Bank Group. His work spans both operations and research, with a focus on foundational skills and learning recovery after the COVID-19 pandemic. His research has appeared in peer-reviewed academic journals, including Economics About the Editors and Contributors xv of Education Review and the Journal of Health Economics. Lautharte has also coauthored World Bank country reports, such as the Brazil Human Capital Review and the Amazon Economic Memorandum. He received a PhD from the University of Cambridge in 2018, where he received the James Claydon Prize in Economics from St. Edmund’s College. Andres Yi Chang is an Economist in the Education and Skills Global Unit at the World Bank Group, specializing in human capital and skills development in low- and middle-income countries. He previously served in the Office of the Chief Economist for People, where he managed the Service Delivery Indicators program. His research spans learning trajectories, educational inequalities, teacher perspectives on education policy, the effects of COVID-19 and rising temperatures on learning outcomes, and health service delivery. His work has appeared in Health Affairs Scholar, International Journal of Educational Development, and Journal of Public Economics. He coauthored Collapse and Recovery: How the COVID-19 Pandemic Eroded Human Capital and What to Do About It. Earlier in his career, Yi Chang held research and policy roles at the World Bank Group Development Research Group, Yale University, the Organization of American States, the Pan American Health Organization, and Leibniz University Hannover. He holds a BA from the University of California, Berkeley, and an MA in international development and economics from Yale University. Main Messages Human capital—the health, knowledge, and skills of people—is what people need to thrive. It is what families and communities need to prosper and what individuals need to find good jobs. Building human capital is not only about what we do. It is also about where we do it. Policies usually focus on improving schools and health clinics to build more human capital. This report shows that a person’s home, neighborhood, and workplace matter just as much and deserve more attention in policy. The report begins by examining global trends in human capital development. There has been a shocking lack of progress in key outcomes. Despite rising incomes and reductions in poverty, two-thirds of low- and middle-income countries have experienced a decline in health, learning, or on-the-job skill development over the past 15 years. For example, average adult height—a marker of population health—has declined in many places. Student learning—measured by harmonized test scores—has remained stagnant in low- and lower-middle- income countries; in most countries, scores are even worse today than in 2010. Similarly, female labor force participation has remained low and stagnant in low- and middle-income countries. Part of the strategy to reverse these disappointing trends must acknowledge all the settings where human capital is built. Investment in the home, the neighborhood, and the workplace needs to be accelerated. THE HOME What happens early in life—at home—is decisive for skill development and success. A family’s resources and the choices families make about their children’s care, health, and learning can have lifelong impacts. For instance, this report demonstrates that children and adolescents whose mothers have more education perform better in tests of vocabulary and mathematics than children whose mothers have less education. These gaps emerge before the age of 5. They remain constant throughout childhood and adolescence. Resources at home are clearly important. They allow families to buy books or pay university tuition. xvii xviii Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces But resources alone are not sufficient. The care a child receives at home is vital. Care involves nurturing, reading, and playing with children. Care involves helping children navigate emotions. Care involves keeping children safe. This report demonstrates that resources do not compensate for shortfalls in care. Strengthening human capital accumulation at home calls for policies that increase resources within the home and that support families in providing the care children need to thrive. These include job programs and cash transfers for poor families. They also include providing parents with tools to create stimulating and nurturing home environments. The evidence summarized in this report from around the world shows that such programs improve educational attainment and lifelong health. They also translate into higher earnings when children join the labor force as adults. THE NEIGHBORHOOD Neighborhoods play a key role in productivity and people’s well-being. They provide access to quality schools, health care, safe streets, and job opportunities. Two families with the same income may not build the same level of human capital if they live in different neighborhoods. For example, evidence from Brazil shows that a person who grew up in a low-income household in a low-income neighborhood earns half as much in adulthood as a person who grew up in a low-income household in a high-income neighborhood. Neighborhoods matter so much for two reasons. First, people typically go to schools and clinics in their own neighborhoods. If these services are inadequate, this will certainly affect learning and health outcomes. Second, even if schools and clinics are good in a neighborhood, other problems, such as violence and pollution, can limit access and opportunity. For policy, a neighborhood lens means bringing together different sectors. Sometimes, unlikely partners, such as government departments focusing on education, the environment, and infrastructure, must act together to improve human capital outcomes. THE WORKPLACE The workplace matters for human development as well. People often think of a job as an end goal, the place where skills are put to use. Yet, learning continues in the workplace. Indeed, half the total human capital accumulated over a lifetime is acquired at work. Through training and experience, people build skills. This report shows that about 70 percent of workers in low- and middle-income countries are in small-scale agriculture, low-quality self-employment, or microfirms with no more than five workers. These jobs generally offer only limited opportunities for learning. Even with the same gain in experience, earnings increase only half as much among the self-employed as among salaried workers. Main Messages xix Countries need to invest more in policies that make work a better engine of learning. Governments and stakeholders can support employer-provided training and promote a learning culture in existing jobs. For example, in India, an on-the- job soft skills training program raised the productivity of garment workers by over 13 percent and improved the productivity of untrained coworkers by almost 12 percent. Stakeholders can also upskill labor market entrants and help them match with jobs through job platforms and formal and informal apprenticeships. This report’s review of large-scale apprenticeship programs in Colombia, Côte d’Ivoire, and Nigeria shows that they increased both skills and earnings. Countries can also increase women’s participation in the workforce by investing in childcare and ensuring a safe commute to work. Well-designed incentives and regulations that support firm growth can result in more high human capital jobs and, therefore, more learning on the job. In all these efforts, the private sector is an essential partner. Recognizing the importance of the home, the neighborhood, and the workplace expands the set of policy options to improve productivity and well-being. This report provides examples of countries that have successfully integrated investments in these key settings, including collaboration with the private sector, to solve human capital challenges, such as malnutrition and low on-the-job learning. The report also argues for a more ambitious data agenda to track progress in these settings more closely and proposes the main building blocks of this agenda. An integrated settings approach and the collection and use of more data are what is needed to build human capital where it matters. Executive Summary Human capital—health, skills, knowledge, and experience—is what people need to thrive. It is also what is needed for families and communities to prosper and for people to find good jobs. No country has ever achieved sustained periods of economic growth or significant reductions in poverty without investing in human capital. Much of policy and research on human capital has focused either on a sector (such as education, health care, or social protection), or on an age-group (such as children under age 5). This is not surprising. National ministries (or their subnational counterparts) are organized along sectoral lines. The life cycle, meanwhile, has been the standard organizing framework to analyze the accumulation of human capital for decades. This report complements the sectoral and life-cycle approaches to human capital by focusing on the settings where human capital is built and what this implies for policies to increase human capital accumulation in an economy. The report takes stock of trends in human capital across the world and presents evidence that the home, the neighborhood, and the workplace are critical and are often missing from human capital policy (refer to figure ES.1). Understanding the dynamics of human capital accumulation in these places can present opportunities to deploy public and private financing more effectively to raise the stocks and flows of human capital. The stagnation in human capital accumulation Despite its importance to development, human capital accumulation has stagnated in many low- and middle-income countries (refer to chapter 1). In some dimensions, poorer countries exhibit worse outcomes today than they did two decades ago. For example, average adult height—a widely used proxy for latent health—rose by about 1 centimeter per decade in Western Europe during the twentieth century and A reproducibility package is available for this book in the Reproducible Research Repository at https://reproducibility.worldbank.org/catalog/461. xxi xxii Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces FIGURE ES.1 Human capital is built in the home, in the neighborhood, and in the workplace HOME A family’s resources and the choices made in homes about diet, health, education, and care determine how much human capital is built for generations. Neighborhood attributes, NEIGHBORHOOD such as service quality, pollution, levels of violence, and the local economy, also shape how much human capital can be built. Skill development does WORKPLACE not stop at school. People continue to learn and build their human capital on the job. Source: Original figure for this publication. Executive Summary xxiii FIGURE ES.2 Learning has stagnated in many parts of the world HLO test scores, 2025 HLO test scores, 2010 Sources: HLO (Harmonized Learning Outcomes) Database, World Bank, https://datacatalog.worldbank.org/search / dataset/0038001; World Bank Income Groups, 2024. Note: The data displayed refer to countries with HLO data for both 2010 and 2025. Country groupings are based on the 2024 income classification. For data by region and country, refer to the interactive figures online at https://humancapital.worldbank.org/en /building-human-capital-where-it-matters. at a similar pace in China in recent decades, but, in several Sub-Saharan African countries, adults are shorter today than they were 25 years ago, indicating a deterioration in underlying health. Learning outcomes show an equally troubling pattern. On average, children in low- and lower-middle-income countries show lower achievement levels today than they did 15 years ago. The largest declines have been observed in Sub-Saharan Africa (refer to figure ES.2). Skill development at work also exhibits worrisome trends. On average, an individual acquires only about half as much human capital through work in India relative to Brazil, and an individual in Brazil only half as much as an individual in the United States. A settings lens for human capital accumulation Without significant investment in health care, education, and on-the-job learning, low- and middle-income countries will continue to fall behind. This report argues that focusing on how human capital outcomes are shaped PeruRussian FederationSaudi ArabiaSouth AfricaTürkiyeCameroonCôte d’IvoireCubaMadagascarMalaysia200250300350400450500550600650250300350400450500550600650Low and lower middle incomeUpper middle incomeHigh income xxiv Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces in the home, in the neighborhood, and in the workplace will help governments and stakeholders design more effective policies to increase human capital, which will lead to more well-paying jobs, less poverty, and higher levels of economic growth. Human capital accumulation in the home Family background shapes human capital accumulation from the start (refer to chapter 2). By the time they are 5 years old and before they have attended school, children in rural Peru whose mothers have primary educational attainment or less have roughly half the vocabulary relative to children whose mothers have completed at least secondary school. Broadly similar patterns are apparent in Ethiopia, India, and Viet Nam. The disadvantage persists throughout school age and adolescence (refer to figure ES.3). The children of mothers with lower educational attainment never catch up. These patterns reflect differing conditions within the home. Why does the home matter so much? One reason is because families vary in their access to resources. To thrive, children need nutritious food, safe and sanitary living conditions, and opportunities to learn. Many of these needs can be met only if families spend money. Healthier foods and diets tend to cost more than less healthy options, and, even if schools and health care are provided free of charge, families still need to purchase books and medicines and cover transportation costs. Homes also vary a great deal in the care environment, that is, how much time parents invest in helping children learn, how much they play with their children, their style in maintaining discipline, and how much social-emotional support they offer to children and adolescents. Children’s early development increases with the number of care or stimulation activities at home, such as an adult singing or playing with a child. Both resources and care matter, but resources cannot easily make up for low- quality care. This can be observed with data on China, where millions of children are left by their parents in the care of other relatives when the parents move from rural to urban areas in search of better jobs. These left-behind children live in homes with higher household income, but they do worse on tests of mathematics and language and exhibit higher levels of depression. In some parts of the world, there are also substantial differences by sex in the resources and care that children receive in the same household. Because resources and care both matter, policies that increase resources or improve the quality of care generally improve human capital outcomes. The availability of more resources, either through higher earnings or through cash transfers, has been shown to improve child outcomes in many settings. Parenting programs that are aimed at changing the care environment in the home can also have large positive effects on Executive Summary xxv FIGURE ES.3 Early skill deficits persist a. Ethiopia Child’s ranking for vocabulary (percentile) Child’s ranking for mathematics (percentile) 68 44 80th 65th 50th 35th 71 43 80th 65th 50th 35th 72 43 65 45 4 5 6 7 8 9 10 11 12 13 14 15 16 4 5 6 7 8 9 10 11 12 13 14 15 16 Child’s age (years) Child’s age (years) b. India Child’s ranking for vocabulary (percentile) Child’s ranking for mathematics (percentile) 64 44 65th 50th 35th 59 46 65th 50th 35th 62 44 65 43 4 5 6 7 8 9 10 11 12 13 14 15 16 4 5 6 7 8 9 10 11 12 13 14 15 16 Child’s age (years) Child’s age (years) c. Peru Child’s ranking for vocabulary (percentile) 65th 57 57 Child’s ranking for mathematics (percentile) 65th 58 56 50th 35th 32 50th 35th 33 35 37 4 5 6 7 8 9 10 11 12 13 14 15 16 4 5 6 7 8 9 10 11 12 13 14 15 16 Child’s age (years) Child’s age (years) d. Viet Nam Child’s ranking for vocabulary (percentile) 65th 57 55 Child’s ranking for mathematics (percentile) 65th 58 57 50th 35th 39 41 50th 35th 39 37 4 5 6 7 8 9 10 11 12 13 14 15 16 4 5 6 7 8 9 10 11 12 13 14 15 16 Child’s age (years) Child’s age (years) Mother’s educational attainment: Primary or less Some secondary or more Source: Original figure for this publication, using data of Young Lives Study (dashboard), Oxford Department of International Development, University of Oxford, https://www.younglives.org.uk/. Note: Percentile rankings were calculated separately by round, marked by points. Only the youngest cohort (individuals born in 2001) are included. The older cohort (children born in 1994) were not observed at age 5. Vocabulary refers to receptive vocabulary as measured by the Peabody Picture Vocabulary Test (Dunn and Dunn 1997). Mathematics skills are measured through a subset of questions from TIMSS (Trends in International Mathematics and Science Study) (data repository), International Association for the Evaluation of Educational Achievement, https://www.iea.nl/data-tools/repository/timss. For data by region and country, refer to the interactive figures online at https://humancapital.worldbank.org/en/building-human-capital-where-it-matters. xxvi Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces children’s human capital that can persist into adulthood. These interventions have frequently proven difficult to scale up, however. Alternatively, human capital accumulation can be achieved through programs that increase the coverage of preschool. These programs often allow women to join the labor force and, if the quality of preschool is high, can foster the development of cognitive and social- emotional skills that are rewarded in the labor market. Education can confer greater skills among parents as they build the human capital of their children. This is particularly true for women, who tend to bear most of the responsibility for providing care in the home. Therefore, policies that increase the education of girls will also increase the human capital of the next generation. Human capital accumulation in the neighborhood Although the children of parents who have more resources and more education generally have substantially better human capital outcomes, the neighborhood (or village) where children grow up can also have substantial effects on human capital trajectories (refer to chapter 3). This has been shown in the United States and, more recently, in Brazil, where the children of low-income parents will complete two more years of schooling, learn more while in school, and earn twice as much in adulthood, if they grow up in a rich neighborhood rather than a poor neighborhood (refer to figure ES.4). Neighborhoods matter because families generally use their local school and local primary health center. In practice, the quality of these local services varies widely. In rural Punjab, Pakistan, for example, a child growing up in a village in the highest decile of school quality learns 44 percent more per year than a child in the lowest decile. Service availability is not the only neighborhood attribute that matters. Air quality, clean water, and sanitation are largely shaped by where people live, and these conditions can vary considerably across neighborhoods or villages. In Indonesia, for example, children living in open defecation–free communities during their first two years of life are more than 10 percentage points less likely to be stunted and have higher cognitive test scores than children living in communities where all other households defecate in the open. In Mexico, exposure to lead from battery recycling plants reduced cognitive development and school performance among children living close to the factories emitting the toxins. There is also substantial variation across neighborhoods in the exposure to violence that residents experience. In San Salvador, people living in gang-controlled neighborhoods had fewer assets, less income, and lower educational attainment even relative to people living only 50 meters away. Local role models matter, too. In India, exposure to the leadership of women on village councils has been shown to influence the career aspirations and educational attainment of adolescent girls. Executive Summary xxvii FIGURE ES.4 Neighborhood characteristics shape human capital in Brazil a. Years of schooling b. Probability of formal employment Number of years Probability (%) 11.7 12.0 11.5 11.0 10.5 10.0 9.5 9.0 9.4 0 25 50 75 100 Share of low-income parents in childhood neighborhood (%) 80.5 90 80 70 60 50 0 55.1 25 50 75 100 Share of low-income parents in childhood neighborhood (%) c. Probability of earning more than parent d. Income at ages 25–29 Probability (%) Income (Brazilian reais) 60.6 65 60 55 50 45 40 35 30 25 R$26,500 R$30,000 R$25,000 R$20,000 R$15,000 R$10,000 29.9 R$13,500 0 25 50 75 100 0 25 50 75 100 Share of low-income parents in childhood neighborhood (%) Share of low-income parents in childhood neighborhood (%) Source: Original figure for this publication, based on Britto et al. 2025. Note: The figure shows the relationship between the share of low-income parents in the neighborhood during childhood and the following measures of average adulthood outcomes of children from low-income families growing up in these neighborhoods: years of schooling, probability of formal employment, probability of earning more than parent, and income at ages 25–29. The scatterplots use an aggregation of neighborhoods as the observation unit. Neighborhoods are divided into 10 equal groups based on the percentage of low-income parents in each. Low income is defined as income at or below the 33rd percentile of the national income distribution. The share of low-income parents in the childhood neighborhood is used as a proxy for neighborhood characteristics growing up. For data by region and country, refer to the interactive figures online at https://humancapital.worldbank .org/en/building-human-capital-where-it-matters. xxviii Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces In terms of policy, this means that it is important to target struggling neighborhoods and identify the main constraints that individuals face to accumulate human capital in these neighborhoods. Policies should provide resources and incentives to improve service quality, environmental quality, and social capital in struggling neighborhoods. Human capital accumulation at work Traditionally, workplaces have been considered as settings where human capital is used. More recently, however, a consensus has emerged that human capital is also built at work. For example, a nurse will be more effective as she learns to work in a team of health professionals in a hospital and, critically, as she builds tacit knowledge on how to interact most effectively with patients. Although there is potential to accumulate significant human capital at work, relatively few people in low- and middle-income countries have an opportunity to do so (refer to chapter 4). Some people are not accumulating human capital at work because they are not even part of the labor force. In low- and middle-income countries, around 50 percent of women are out of the labor force, while around 20 percent of youth are neither studying nor working. Those who are employed, meanwhile, are concentrated in jobs where little learning occurs. Many workers are in small firms that operate with low technology and minimal organizational capital, often only front-line workers, without managers or engineers or other technical personnel. In fact, about 70 percent of workers in low- and middle-income countries, but only 20 percent in high-income countries, are working in small-scale agriculture, low-quality self-employment, or microfirms. These are generally jobs with limited formal training and few on-the-job learning opportunities. Even with the same gain in experience, earnings in low- and middle- income countries rise only half as much among the self-employed as among wage employees (refer to figure ES.5). These challenges call for policies that expand learning on the job, ease transitions into work, and create more jobs with strong learning potential. Formal apprenticeships, for instance, have had positive effects on skills and earnings, even when implemented at scale, in numerous Sub-Saharan African countries. These policies should be supported by broader reforms that reduce market failures and misallocation. Policies can increase learning on the job in all types of employment. Farmers can benefit from extension programs to learn new techniques and adopt new technologies. The self-employed can benefit from soft-skills and business training. Formal job training among wage workers can be effective, but it is undersupplied even in large firms. This is because workers can take their newly acquired skills and move to a different employer unless the training is firm-specific. Incentivizing firms Executive Summary xxix FIGURE ES.5 Returns to experience are lower among the self-employed than among wage workers in low- and middle-income countries 75 50 25 0 10 20 30 Sources: Original figure for this publication, based on data of GLD (Global Labor Database Repository), World Bank, https://worldbank.github.io/gld/README.html; I2D2 (International Income Distribution Database) (internal database, discontinued in 2020), World Bank; SEDLAC (Socio-Economic Database for Latin America and the Caribbean), https://www.cedlas.econo.unlp.edu.ar/wp/estadisticas/sedlac/. Note: The figure shows estimated experience–wage profiles for working-age men grouped by potential experience. Hourly wages are total labor earnings, divided by hours worked. Returns are calculated in five-year experience bins, following Jedwab et al. (2023), using population weights. The results exclude high-income countries. For data by region and country, refer to the interactive figures online at https://humancapital.worldbank.org/en/building-human-capital-where-it-matters. to invest in on-the-job training, particularly in the formation of general skills, can therefore help. Creating more jobs with stronger potential for learning requires incentives for firm growth and expanded education to develop needed talent. Governments and stakeholders can promote access to technology, finance, markets, and research and development (R&D), particularly among young, innovative firms that drive radical innovations and create jobs demanding skilled labor. Well-targeted R&D credits can have lasting effects on human capital and productivity. Facilitating firm growth and structural transformation—from subsistence agriculture and low-productivity services to modern firms—is therefore critical for human capital policy. Wage workersSelf-employedWage increase (%)Experience (years) xxx Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces Integrating settings into policy to tackle long-standing challenges Human capital enables people to contribute productively to society. Within countries, investments in human capital spur economic growth and reduce inequality. Despite these well-recognized benefits, trends in human capital accumulation in low- and middle-income countries over the last two decades paint a bleak picture. In many countries, the situation has gotten worse, rather than better. This report argues that consideration of key settings in which human capital is built—the home, the neighborhood, and the workplace—allows governments to design and implement interventions more effectively to improve health status, raise educational attainment and achievement, and increase on-the-job learning. Figure ES.6 summarizes key policy priorities. Policies to strengthen human capital would benefit from a settings lens. Strategies to address malnutrition, for example, must address constraints in the home. Families need resources to purchase and prepare nutritious food, and they should engage in early stimulation activities with their children. Other interventions will need to target the neighborhood, including policies to ensure garbage collection, provide clean water and sanitation, and offer access to health centers. FIGURE ES.6 Policy priorities for the home, the neighborhood, and the workplace Key challenge Priorities HOME Improvements in care in the home Resources and jobs for the poor, parenting programs, and girls’ education NEIGHBORHOOD Key challenge Opportunities to build human capital in struggling neighborhoods Priorities Resources and incentives to improve service quality, environmental quality, and social capital Key challenge More jobs with stronger potential for skill development at work Priorities Apprenticeships, childcare, extension services, soft and business skills training, and incentives for firms to expand and to invest in on-the-job learning WORKPLACE Source: Original figure for this publication. Executive Summary xxxi Workplaces are also important because parents and other caregivers need to have good jobs to be able to purchase the inputs required to provide a healthy diet. A policy to tackle malnutrition therefore requires coordinated action across the three settings: the home, the neighborhood, and the workplace. It also requires coordinated action across multiple sectors, including health care, social protection, agriculture, transport, water, sanitation, and labor, alongside regulation of the private sector and support for local food markets. The same logic can be used to design effective policies to increase learning or the acquisition of skills at work. A policy that acknowledges the role of multiple settings requires tools that can help coordinate investments within and across settings. Social registries are information systems that aggregate socioeconomic information on individuals or households that can then be used in multiple programs to support households across settings. Social assistance centers can serve as single-window entry points that connect individuals and households to the full set of benefits and services for which they are eligible. Case management involves a trained professional who works closely with a household to identify the specific constraints it faces, codevelops a tailored plan, and coordinates access to the range of services needed. The common theme in all these approaches is that they attempt to coordinate multiple programs to reach the same household. Tools such as these could also be used to identify and design policy packages for struggling neighborhoods. Tracking progress in human capital investment across homes, neighborhoods, and workplaces also requires clear metrics of success. Given current data availability, however, a more ambitious global and national data agenda is needed. Human capital is essential to enable people to obtain good jobs and earn higher incomes. Despite tremendous progress in expanding access to education, health care, and social services, improvements in key outcomes have stagnated or declined in many low- and middle-income countries. This volume proposes an approach to tackle this stagnation. It argues that a careful consideration of the constraints and opportunities in three key settings—the home, the neighborhood, and the workplace—is needed to prepare people for jobs, to help them thrive, and to unlock productivity. References Britto, Diogo G. C., Alexandre de Andrade Fonseca, Paolo Pinotti, Breno Sampaio, and Lucas Warwar. 2025. “Do CCTs Create Conditions to Thrive? Bolsa Família and Social Mobility in Brazil.” WIDER Working Paper 96/25 (December), United Nations University–World Institute for Development Economics Research. Dunn, Lloyd M., and Leola M. Dunn. 1997. Peabody Picture Vocabulary Test, 3rd ed. American Guidance Service. Jedwab, Remi Camille, Paul Romer, Asif M. Islam, and Roberto Samaniego. 2023. “Human Capital Accumulation at Work: Estimates for the World and Implications for Development.” American Economic Journal: Macroeconomics 15 (3): 191–223. Chapter 1 Introduction Norbert Schady Summary This chapter serves as an introduction, motivation, and framing for the subsequent chapters in this report. It argues that human capital is multidimensional and encompasses all characteristics that make an individual more productive. Human capital is critical for development. It raises earnings, boosts economic growth, increases the likelihood that women join the labor force, and, because human capital is frequently the only capital that poor people have, has a disproportionate effect on poverty reduction. No country has gone through sustained periods of economic growth and concomitant reductions in poverty without first investing in human capital. The chapter demonstrates the considerable urgency in the need to address human capital shortfalls. In low- and lower-middle-income countries, many human capital outcomes have stagnated, including health, education, and the skills acquired at work, and, in some dimensions, outcomes are worse today than they were two decades ago. The chapter argues that a focus on the settings where human capital is built is important for policy design. Human capital policies have traditionally focused on specific sectors (for example, education, health, or social protection) or age-groups (for instance, children ages 5 or younger). The chapter posits, however, that focusing on how human capital accumulates in the home, in the neighborhood, and in the workplace is a useful complement to sectoral or age-specific approaches. A reproducibility package is available for this book in the Reproducible Research Repository at https://reproducibility.worldbank.org/catalog/461. 2 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces Why is human capital so important? The acquisition of talents during education, study, or apprenticeship . . . is capital in a person. Those talents are part of his fortune and likewise that of society. —Adam Smith ([1776] 1937, chapter 1) This report is about human capital: what it is, how it accumulates in distinct settings, such as the home, the neighborhood, and the workplace, and what this implies for policy design. Human capital is the health, knowledge, skills, and experience that people accumulate over their lifetimes. It encompasses all characteristics that make an individual more productive. It is combined with other factors (such as physical capital) in the production of goods and services. No country has ever achieved sustained periods of economic growth or significant reductions in poverty without investing in human capital. This chapter provides an introduction to the report and establishes the motivation and framework for the chapters that follow. It shows why investments in human capital are critical for people and countries. The chapter argues that there is considerable urgency in the need to address human capital shortfalls. In low- and lower-middle-income countries, many human capital outcomes have stagnated, and, in a number of dimensions, outcomes are worse today than they were two decades ago. The chapter argues that the standard organizing framework for human capital policy, the life cycle, is insufficient on its own to support the design of effective policies. An approach that focuses more closely on the settings where human capital accumulation takes place can fruitfully widen the scope of human capital policy. Human capital raises earnings and boosts growth Globally, every additional year of schooling attained by individuals raises their earnings by about 10.0 percent, while every year of experience raises earnings by about 2.5 percent.1 This means that, over a lifetime, a person with 10 years of education would earn about 50 percent more than a person with 5 years of education, while a person with 10 years of labor market experience should earn about 10 percent more than a person with 6 years of experience. School quality also yields large returns in the labor market.2 Individuals who are in better health are more likely to work and earn higher wages, and some of these effects can be traced as far back as health in utero or in early childhood.3 Human capital is also critical for economic growth. In neoclassical growth models, human capital is an input into the production function of gross domestic product (GDP). So, increases in human capital directly raise aggregate income. Moreover, because human capital and physical capital are complements in Introduction 3 production, an expansion in human capital spurs additional investments in physical capital.4 In endogenous growth models, human capital is even more important. It fuels innovation, accelerates the adoption of new technologies, and expands a country’s productive potential. Although human capital is thought to be critical to growth, quantifying the magnitude of the effect is a challenge. One approach, known as development accounting, estimates by how much the differences in per capita GDP between rich and poor countries would decline if their levels of human capital were equalized. Estimates indicate that human capital accounts for a substantial share of cross-country income differences: roughly two-thirds of the difference in per capita GDP between rich and poor countries is accounted for by differences in human capital.5 Human capital increases women’s labor force participation In rich and poor countries, most men enter the labor force as soon as they have concluded formal schooling, but a substantial share of women do not do so. Female labor force participation rates are particularly low in the Middle East and North Africa, where only one in five women are employed, and in South Asia, where the proportion is one in three. This represents an immense loss of talent, household income, and per capita GDP. Education changes this picture. A woman with more schooling has higher potential earnings and more agency, that is, the ability to make choices about work and family. As a result, education increases the likelihood that women work. Globally, the gap in labor force participation between women with university education and women with no education is 24 percentage points. Human capital reduces poverty Labor income represents by far the largest share of income among the poor, and, so, investments that raise the productivity of labor—principally, investments in human capital—will have a larger effect on poverty than what one might expect simply based on the effects of human capital on growth. A recent paper tests this intuition by estimating the effect of increases in school attainment rates on global poverty reduction since 1980. It concludes that increases in school attainment rates account for 70 percent of income gains among the poor and for 40 percent of the global reduction in extreme poverty.6 Investments in human capital may reduce inequality Human capital shapes income inequality primarily through its effects on wage inequality. Investments in human capital may therefore reduce inequality if they compress the wage distribution and weaken the link between family background and labor market outcomes. The total effect of investments in human capital on inequality is ambiguous, however, because such investments change both the 4 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces distribution of skills and the returns to skills.7 The effect ultimately depends on who acquires the skills, how technology interacts with skills, and whether policies ensure that the returns to human capital are broadly shared. Empirically, in Latin America during the 2000s, rapid growth in the supply of skilled labor outpaced demand, contributing to lower skill premiums and declining wage inequality.8 By contrast, wage inequality increased in China and India, alongside the rise in the returns to education and skills.9 In Indonesia, rising wage inequality has been linked more directly to technological change that boosts the demand for skilled labor.10 In sum, human capital is clearly central to economic development. It raises individual earnings and makes women’s entry into the labor market more likely. It is a principal determinant of aggregate economic growth rates, and it is key to reducing poverty and inequality. Human capital: Stagnating in low- and lower-middle- income countries This section shows that human capital accumulation has stagnated in most low- and lower-middle-income countries. Indeed, in some dimensions, human capital outcomes in the poorest countries are lower today than 15 years ago. This has major implications for the ability of countries to grow, reduce inequality, and increase welfare. Health Health is an important dimension of human capital. The height of people in any country varies because of genetics, but the average height of adults is generally taken to be a measure of the latent health status of a population.11 The average height in Western Europe rose by roughly 1 centimeter per decade in the twentieth century. This is also the rate at which height has increased among Chinese adults in recent decades. The picture is much less encouraging in many other countries. Figure 1.1 plots the height of adults born in 1966 and 1996. Countries in which people have grown taller are shown above the 45-degree line, while countries in which people have become shorter are shown below the line. The figure indicates that the average adult in upper-middle- and high-income countries has become taller. Meanwhile, the average adult in low- and lower-middle-income countries is shorter today that the average adult was 30 years ago. In Sub-Saharan Africa, the average adult born in 1996 is 3 centimeters shorter than the average adult born in 1966. This indicates that latent health is worse today relative to past decades. Introduction 5 FIGURE 1.1 Changes in adult height, by country income group Average adult height, 1996 birth cohort (centimeters) 180 175 170 165 160 155 Algeria 160 165 170 175 180 Average adult height, 1966 birth cohort (centimeters) Low and lower middle income Upper middle income High income Sources: 2017 data, Height (dashboard), NCD RisC (Non-Communicable Disease Risk Factor Collaboration), World Health Organization Collaborating Centre on NCD Surveillance, Epidemiology, and Modelling, Imperial College London, https://www.ncdrisc.org/data-downloads-height.html; World Bank Income Groups, 2024. Note: Adult height data correspond to adults ages 18 or more by birth cohort. Schooling School enrollment and the average years of education completed—educational attainment—have risen sharply in most low- and middle-income countries (refer to figure 1.2). It is also important to consider progress in learning (educational achievement). Has the increase in enrollment translated into increases in learning? This report uses data on Harmonized Learning Outcomes (HLOs), which render scores on various regional and global tests comparable for an examination of trends across countries. Figure 1.3 shows that little progress has been achieved overall since 2010 and that there is no evidence that poorer countries are gradually closing the gap relative to wealthier countries. In fact, on average, test scores have fallen in low-income and lower-middle-income countries (by 32 and 28 points, respectively), and the absolute declines in achievement are largest in the countries in Sub-Saharan Africa, the poorest region in the world. ArgentinaBrazilColombiaIndonesiaNetherlandsRussian FederationChadChinaGuatemalaIndiaMadagascarMexicoNigeriaSwedenUnited States 6 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces FIGURE 1.2 School enrollment and the tertiary completion rate a. Primary school enrollment rate b. Secondary school enrollment rate c. Tertiary completion rate Rate (%) 100 98 95 69 75 50 25 0 2000 98 96 89 2006 2012 2018 2024 Rate (%) 100 93 75 75 50 51 25 0 2000 96 89 71 Rate (%) 100 75 50 25 2006 2012 2018 2024 0 2000 43 22 7 47 23 18 2006 2012 2018 2024 Sources: Data of Population with Tertiary Education (dashboard), OECD Data Explorer, Organisation for Economic Co-operation and Development, https://www.oecd.org/en/data/indicators/population-with-tertiary-education.html; UIS Data Browser (dashboard), Institute for Statistics, United Nations Educational, Scientific, and Cultural Organization, https://databrowser.uis.unesco.org/; WIDE (World Inequality Database on Education), Institute for Statistics, United Nations Educational, Scientific, and Cultural Organization, http://www.education-inequalities.org/; World Bank Income Groups, 2024. Note: Primary and secondary enrollment rates are based on modelled estimates of the Institute for Statistics, United Nations Educational, Scientific, and Cultural Organization. Rates of four-year tertiary completion among individuals ages 25–34 are compiled based on household surveys. World Inequality Database on Education data are supplemented with tertiary completion data of the Organisation for Economic Co-operation and Development. Country-income group aggregates are population weighted and based on income classifications in 2024. FIGURE 1.3 Student learning across countries HLO test scores, 2025 650 600 550 500 450 400 350 300 250 HLO test scores, 2010 Sources: HLO (Harmonized Learning Outcomes) Database, World Bank, https://datacatalog.worldbank.org/search/dataset /0038001; World Bank Income Groups, 2024. Note: The data displayed refer to countries with HLO data for both 2010 and 2025. Country groupings are based on the 2024 income classification. High incomeUpper middle incomeLow and lower middle incomePeruRussian FederationSaudi ArabiaSouth AfricaTürkiyeCameroonCôte d’IvoireCubaMadagascarMalaysia200250300350400450500550600650Low and lower middle incomeUpper middle incomeHigh income Introduction 7 On-the-job learning Human capital accumulation does not end when people leave school. Individuals can continue acquiring human capital through work, but only if they work. Most men are in the labor force in all country income groups, and there has been little change over time (refer to figure 1.4, panel a). Meanwhile, female labor force participation rates vary widely across countries (refer to figure 1.4, panel b). There have been substantial increases in female labor force participation in high-income countries, but little change in low- and middle-income countries. This implies that the differences in the opportunities for human capital accumulation through work between richer and poorer countries are larger now than they were two decades ago. How much human capital is acquired at work depends on both the share of the working-age population that is working and the amount of on-the-job learning that occurs. Quantifying this learning is not straightforward, but, if workers are paid their marginal product of labor, as would occur in a competitive market, the increase in labor income as people acquire work experience should be driven by an increase in skills, not merely seniority.12 FIGURE 1.4 Trends in labor force participation rates, prime-age adults, ages 25–54, 2004–24 a. Male labor force participation rate b. Female labor force participation rate Rate (%) 100 95 92 94 90 80 70 60 50 93 90 92 Rate (%) 100 90 80 70 60 50 73 70 51 79 70 51 2004 2008 2012 2016 2020 2024 2004 2008 2012 2016 2020 2024 Sources: ILOEST Database (ILO Modelled Estimates Database), International Labour Organization, https://ilostat.ilo .org/methods /concepts-and-definitions /ilo -modelled-estimates/; ILOSTAT Indicators and Data Tools (dashboard), International Labour Organization, https://ilostat.ilo.org/data/; World Bank Income Groups, 2024. Note: Labor force participation rates for population ages 25–54 are based on a combination of survey data and International Labour Organization modeled estimates. Statistics are produced using a five-year trailing average of labor force participation rates. All income group aggregates are population weighted and based on World Bank income classifications for 2024. High incomeUpper middle incomeLow and lower middle income 8 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces The wage-experience profile is steepest in high-income countries (refer to figure 1.5). On average, an individual acquires only about half as much human capital through work in India relative to Brazil, and an individual in Brazil only half as much as an individual in the United States.13 This reflects the fact that the opportunities for learning offered by available jobs are greater in richer countries than in poorer countries.14 In this dimension, too, there are large differences in human capital between richer and poorer countries. In sum, critical human capital gaps between richer and poorer countries have widened substantially. In some measures, including height and learning, low- and lower-middle-income countries have worse outcomes than they had 15 years ago. FIGURE 1.5 Labor income as a function of on-the-job experience, men ages 18–67, by country income status 100 75 50 25 0 10 20 30 Sources: Data of GLD (Global Labor Database Repository), World Bank, https://worldbank.github.io/gld /README.html; I2D2 (International Income Distribution Database) (internal database, discontinued in 2020), World Bank; World Bank Income Groups, 2024. Note: The figure illustrates data on log wage increases by experience calculated from assorted labor force surveys. The data are limited to men because there is no straightforward way of accounting for selection into work by women, many of whom are not employed (refer to figure 1.4, panel b). Individuals with fewer than five years of potential experience are the omitted category. Difference in monthly wage relative to 0–4 years of experience (%)Potential experience (years)High incomeUpper middle incomeLow and lower middle income Introduction 9 The importance of settings to human capital policy Human capital accumulates over the life cycle Human capital accumulation is a dynamic process. The skills acquired in one stage of the life cycle affect both the initial conditions and the technology of learning at the next stage. —Heckman and Carneiro (2003, 5) The standard organizing framework to describe the process of human capital accumulation is the life cycle.15 Human capital accumulation starts as early as conception. During early childhood (from gestation through age 5), children physically grow, acquire immunity to debilitating and fatal diseases, and develop the cognitive, language, social-emotional, and motor skills that not only prepare them for formal schooling, but also directly translate into health and labor market success in adulthood.16 School-age children receive explicit instruction in school and continue to acquire skills and gain knowledge in subjects that are essential for contributing productively to society. In adulthood, skill acquisition continues through work, provided that people work and that the job they hold lends itself to on-the-job learning. Starting around age 50, human capital plateaus, on average, although there is a great deal of variation across individuals.17 The depreciation of human capital starts in old age, as an individual’s health declines and as the performance of productive tasks becomes more challenging than it was earlier in the life cycle. Human capital formation is inherently cumulative. Each stage builds on previous foundations. In this sense, skills beget skills. The ability of school-age children to grasp mathematical concepts and to focus and process new information in a classroom is strongly affected by their experiences during early childhood. In some dimensions of skill, investments may also exhibit dynamic complementarities. This means, for instance, that greater investment during one period may raise the returns to future investment. Thus, high-quality schools may not only improve the skills with which young people enter the labor market, but may also increase the returns to further investment through on-the- job learning.18 10 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces Human capital accumulates in particular settings The simple human capital investment model . . . does not address several important factors, such as where the . . . training takes place (school, on-the-job, at home), who provides . . . and pays for it, and what role the “state” or collective plays in these matters. —Goldin (2016, 70) Investments in human capital occur in different settings. The life-cycle model is an appealing framework because it stresses how investments made at different times may interact. This report complements the life-cycle approach by examining the settings in which human capital is built. The report considers three settings: the home, the neighborhood, and the workplace. The life-cycle approach reveals what investments are needed and when, while the consideration of settings reveals where and how these and potentially other human capital investments should occur. The household is critical to human capital accumulation, particularly among children and adolescents. Decisions related to nutrition, play, and schooling occur within the home, as do decisions on employment. Households invest in work-related experience because household members typically decide as a unit who works, how much they work, and where they work. Many policies are already designed with a view to the importance of the home. For example, early nutrition counseling tends to target mothers, and outreach related to essential childhood vaccinations or social services may include home visits. Nonetheless, the home is rarely recognized as a location in which learning occurs, that is, where individuals first pick up the cognitive and social-emotional skills that will position them for success in school and in the labor force. Missing from most human capital policy frameworks are strategies to ensure that this type of human capital is built in the home. Similarly, neighborhoods or villages in rural areas are not merely agglomerations of people. They are spaces where families live, children grow up, and individuals interact with one another. Neighborhoods affect human capital accumulation through numerous channels in the life cycle. From early childhood to adulthood, from child mortality to lifetime income, places matter because the locations where people live determine their access to services and local markets and their exposure to environmental hazards and social dynamics. Neighborhoods condition the schools and health facilities that people use. They also affect human capital through the health of the local economy, environmental factors, and social dynamics, ranging from cohesion to exposure to crime. Many policies recognize the importance of neighborhoods, such as housing policies and water, sanitation, and hygiene programs. Neighborhoods are, Introduction 11 however, rarely acknowledged as settings crucial to the broad range of human capital outcomes they influence. Most human capital policies fail to include strategies to internalize neighborhood externalities and engage additional sectors, such as infrastructure, sanitation, and urban planning. The workplace is another key setting for human capital accumulation. Learning in the workplace occurs through explicit training and on-the-job learning. Farmers refine techniques through trial and error, peer exchanges, and extension services. The self-employed build skills through hands-on experience, business networks, and structured training. Wage workers learn by practicing advanced techniques, adopting technology, solving problems, engaging with peers, and undertaking formal or informal training. Beyond these direct investments, the workplace has a pivotal role in shaping human capital accumulation. The returns to past human capital investments are reflected in wages and create incentives for additional investment. Many policies recognize the workplace as a setting in which previously acquired skills are used. Yet, the workplace is rarely acknowledged as a crucial setting for skill development. As a result, most human capital policies fail to include effective strategies to expand public and private investment in on-the-job learning or to create incentives for businesses to invest in workplace attributes that foster skill development. The settings that have received the primary attention in relation to human capital accumulation are government-provided services in formal facilities, such as schools or health centers. This is a narrow view of the way human capital is formed. The rest of this report argues that the role of government in human capital policy need not be limited to direct service provision, but may also encompass support for effective human capital formation by households, communities, and firms. The report (1) presents evidence on the importance of settings for human capital accumulation, (2) identifies the main reasons for underinvestment in each setting, and (3) articulates relevant policies for each setting that are supported by the extant evidence base in low- and middle-income countries. The rest of the report is organized as follows. Chapter 2 focuses on the home and highlights the importance of resources and care to human capital accumulation. Chapter 3 concentrates on neighborhoods and the importance of the quality of local schools and primary health centers as well as the role of environmental factors, local economic conditions, and social interactions in education and health. Chapter 4 highlights the importance of the workplace and the characteristics that make the workplace more conducive to on-the-job learning. Chapter 5 describes broader reforms that could enable governments to design policies that activate these settings to promote human capital accumulation more effectively. 12 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces Notes 1. Labor force surveys do not generally collect data on the actual number of years worked. For this reason, the reference here is to potential experience, defined as age minus years of schooling, minus six. It is assumed that any experience before age 18 produces zero returns. 2. Card and Krueger (1992) use state and time variations in the pupil-teacher ratio, average term length, and relative teacher pay as measures of school quality. Numerous papers by Hanushek and coauthors also argue that higher school quality results in higher earnings and higher economic growth rates. Refer to Hanushek and Kimko (2000); Hanushek and Woessmann (2008). Chetty and coauthors use data on teacher quality to show that children who, for random reasons, were assigned to better teachers show higher earnings in adulthood. Refer to Chetty et al. (2011); Chetty et al. (2014). 3. Low birthweight babies have worse health in adulthood and earn lower wages, as shown in numerous country settings. In Norway, for example, a 10 percent increase in birthweight raises earnings in adulthood by 1 percent (Black et al. 2007). Other studies also show that birthweight affects school attainment, as well as longer-run outcomes, such as height, IQ at age 18, and earnings (Almond et al. 2005; Bharadwaj et al. 2018). 4. A simple neoclassical model is often expressed as Yc = A*Hc α*Kc 1−α, where Yc is GDP, A is total factor productivity, Hc is the human capital stock, Kc is the physical capital stock, and the subscript c refers to countries. Hc may be rewritten as Lch, where Lc is the number of workers, and h is the average human capital per worker. Estimates of α, the output elasticity of labor, are generally around 0.65, and the output elasticity of capital, 1–α, is around 0.35. In this formulation, increasing human capital by 1 percent would raise GDP by ~0.65 percent. But, because human capital and physical capital are complements, this increase in human capital, Hc, raises the marginal product of physical capital and thus prompts additional investments in physical capital, Kc. As a result, once the economy has fully adjusted, increasing Hc by 1 percent boosts GDP by 1 percent. The fact that the ratio between physical and human capital is stable in most countries over long horizons suggests that, if one input (physical or human capital) increases or decreases, the other input adjusts proportionately. 5. Hendricks and Schoellman (2018; 2023); Jedwab et al. (2023). 6. Gethin (2025). 7. The intuition is as follows. Consider an economy with three kinds of labor: unskilled (workers with primary educational attainment or less), medium skilled (workers with secondary educational attainment or less), and skilled (workers with tertiary educational attainment). If a country increases the share of medium-skilled workers, for example, by expanding access to secondary school, the increase in labor supply will tend to depress the wages of these workers relative to the wages of unskilled and skilled workers. The magnitude of the effect will depend on the elasticity of substitution between workers with different skill levels. In the long run, increases in the supply of these medium-skilled workers will lead to changes in demand, which will also affect wages. Refer to Acemoglu and Restrepo (2022); Autor and Dorn (2013); Manacorda et al. (2010). 8. Messina and Silva (2018). 9. Ge and Yang (2014); Lee and Wie (2017). 10. Lee and Wie (2015). 11. Deaton (2007, 13232) concludes: “Although there is a large genetic component to heights within populations, the contribution of genetics to variation in mean heights across populations is much smaller” (italics added). 12. Workers may also find a better match between their abilities and the tasks required in different jobs over time, as emphasized in search models of the labor market. This, too, could lead to an increase in wages as the workers acquire more experience. Refer to Lagakos et al. (2018) for a discussion. 13. Similar results are reported by Jedwab et al. (2023) and Lagakos et al. (2018). Introduction 13 14. These results are based on regressions of the natural logarithm of wages on years of schooling and indicator variables for five-year experience intervals, with the lowest experience category (0–4 years) as the omitted category. These results thus assume that schooling and experience are additively separable, as in the standard Mincerian formulation. In practice, the returns to experience are higher among more highly educated workers, and, because the average number of years of schooling is higher in richer countries than in poorer countries, a share of the difference in the returns to experience between richer and poorer countries is accounted for, mechanically, by the differences in schooling. Lagakos et al. (2018) estimate that this share is 25 percent–40 percent of the overall difference between poor and rich countries. 15. This follows Heckman’s seminal work. Refer, for example, to Heckman (2007). 16. Almond and Currie (2011); Attanasio et al. (2020); Cunha et al. (2010); Grantham-McGregor et al. (2007); Heckman (2007). 17. Jedwab et al. (2023); Lagakos et al. (2018); Mincer (1974); Skirbekk (2004). 18. Caucutt and Lochner (2020); Cunha and Heckman (2007); Cunha et al. (2010). References Acemoglu, Daron, and Pascual Restrepo. 2022. “Tasks, Automation, and the Rise in U.S. Wage Inequality.” Econometrica 90 (5): 1973–2016. Almond, Douglas, Kenneth Y. Chay, and David S. Lee. 2005. “The Costs of Low Birth Weight.” Quarterly Journal of Economics 120 (3): 1031–83. Almond, Douglas, and Janet Currie. 2011. “Human Capital Development Before Age Five.” In Handbook of Labor Economics, vol. 4, part B, edited by Orley C. Ashenfelter and David E. Card. North-Holland. Attanasio, Orazio Pietro, Sarah Julie Cattan, Emla Fitzsimons, Costas Meghir, and Marta Rubio-Codina. 2020. “Estimating the Production Function for Human Capital: Results from a Randomized Control Trial in Colombia.” American Economic Review 110 (1): 48–85. Autor, David H., and David Dorn. 2013. “The Growth of Low-Skill Service Jobs and the Polarization of the US Labor Market.” American Economic Review 103 (5): 1553–97. Bharadwaj, Prashant, Petter Lundborg, and Dan-Olof Rooth. 2018. “Birth Weight in the Long Run.” Journal of Human Resources 53 (1): 189–231. Black, Sandra E., Paul J. Devereux, and Kjell G. Salvanes. 2007. “From the Cradle to the Labor Market? The Effect of Birth Weight on Adult Outcomes.” Quarterly Journal of Economics 122 (1): 409–39. Card, David E., and Alan B. Krueger. 1992. “Does School Quality Matter? Returns to Education and the Characteristics of Public Schools in the United States.” Journal of Political Economy 100 (1): 1–40. Caucutt, Elizabeth M., and Lance John Lochner. 2020. “Early and Late Human Capital Investments, Borrowing Constraints, and the Family.” Journal of Political Economy 128 (3): 1065–1147. Chetty, Raj, John N. Friedman, Nathaniel Hilger, Emmanuel Saez, Diane Whitmore Schanzenbach, and Danny Yagan. 2011. “How Does Your Kindergarten Classroom Affect Your Earnings? Evidence from Project STAR.” Quarterly Journal of Economics 126 (4): 1593–1660. Chetty, Raj, John N. Friedman, and Jonah E. Rockoff. 2014. “Measuring the Impacts of Teachers II: Teacher Value-Added and Student Outcomes in Adulthood.” American Economic Review 104 (9): 2633–79. Cunha, Flavio, and James J. Heckman. 2007. “The Technology of Skill Formation.” American Economic Review 97 (2): 31–47. Cunha, Flavio, James J. Heckman, and Susanne M. Schennach. 2010. “Estimating the Technology of Cognitive and Noncognitive Skill Formation.” Econometrica 78 (3): 883–931. Deaton, Angus S. 2007. “Height, Health, and Development.” PNAS, Proceedings of the National Academy of Sciences 104 (33): 13232–37. 14 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces Ge, Suqin, and Dennis Tao Yang. 2014. “Changes in China’s Wage Structure.” Journal of the European Economic Association 12 (2): 300–36. Gethin, Amory. 2025. “Distributional Growth Accounting: Education and the Reduction of Global Poverty, 1980–2019.” Quarterly Journal of Economics 140 (4): 2571–2618. Goldin, Claudia Dale. 2016. “Human Capital.” In Handbook of Cliometrics, edited by Claude Diebolt and Michael John Haupert. Springer. Grantham-McGregor, Sally M., Yin Bun Cheung, Santiago Cueto, et al. 2007. “Child Development in Developing Countries 1: Developmental Potential in the First 5 Years for Children in Developing Countries.” Lancet 369 (9555): 60–70. Hanushek, Eric Alan, and Dennis D. Kimko. 2000. “Schooling, Labor-Force Quality, and the Growth of Nations.” American Economic Review 90 (5): 1184–1208. Hanushek, Eric Alan, and Ludger Woessmann. 2008. “The Role of Cognitive Skills in Economic Development.” Journal of Economic Literature 46 (3): 607–68. Heckman, James J. 2006. “Skill Formation and the Economics of Investing in Disadvantaged Children.” Science 312 (5782): 1900–02. Heckman, James J. 2007. “The Economics, Technology, and Neuroscience of Human Capital Formation.” PNAS, Proceedings of the National Academy of Sciences 104 (33): 13250–55. Heckman, James J., and Pedro Manuel Carneiro. 2003. “Human Capital Policy.” NBER Working Paper 9495 (February), National Bureau of Economic Research. Hendricks, Lutz, and Todd Schoellman. 2018. “Human Capital and Development Accounting: New Evidence from Wage Gains at Migration.” Quarterly Journal of Economics 133 (2): 665–700. Hendricks, Lutz, and Todd Schoellman. 2023. “Skilled Labor Productivity and Cross-Country Income Differences.” American Economic Journal: Macroeconomics 15 (1): 240–68. Jedwab, Remi Camille, Paul Romer, Asif M. Islam, and Roberto Samaniego. 2023. “Human Capital Accumulation at Work: Estimates for the World and Implications for Development.” American Economic Journal: Macroeconomics 15 (3): 191–223. Lagakos, David, Benjamin Moll, Tommaso Porzio, Nancy Qian, and Todd Schoellman. 2018. “Life Cycle Wage Growth Across Countries.” Journal of Political Economy 126 (2): 797–849. Lee, Jong-Wha, and Dainn Wie. 2015. “Technological Change, Skill Demand, and Wage Inequality: Evidence from Indonesia.” World Development 67 (March): 238–50. Lee, Jong-Wha, and Dainn Wie. 2017. “Wage Structure and Gender Earnings Differentials in China and India.” World Development 97 (September): 313–29. Manacorda, Marco, Carolina Sánchez-Páramo, and Norbert R. Schady. 2010. “Changes in Returns to Education in Latin America: The Role of Demand and Supply of Skills.” Industrial and Labor Relations Review 63 (2): 307–26. Messina, Julián, and Joana C. G. Silva. 2018. Wage Inequality in Latin America: Understanding the Past to Prepare for the Future. Latin American Development Forum Series. World Bank. Mincer, Jacob A. 1974. “Age and Experience Profiles of Earnings.” In Schooling, Experience, and Earnings, edited by Jacob A. Mincer. Vol. 2 of Human Behavior and Social Institutions. National Bureau of Economic Research; Columbia University Press. Skirbekk, Vegard. 2004. “Age and Individual Productivity: A Literature Survey.” Vienna Yearbook of Population Research 2: 133–53. Smith, Adam. (1776) 1937. An Inquiry into the Nature and Causes of the Wealth of Nations, Book 2. Modern Library Series Reprint. Random House. Chapter 2 Human Capital Accumulation in the Home Alaka Holla Summary Human capital accumulation starts early and at home. Families spend resources and make choices that shape the health, skills, and work experience of all members of the household. Around the world, there are large gaps in nutritional status and cognitive skills based on family circumstance that are apparent in early childhood, before children start school, and that remain constant through adolescence. Deficits can emerge even if economic circumstances are similar, suggesting the strong role of care in the home. Policy can support human capital accumulation in the household by ensuring that households have more resources, but this will not be sufficient. It will also be critical for governments to deploy programs that target the care environments of children and adolescents. Because human capital investments during childhood not only yield lifelong benefits, but also extend to the next generation, policies that improve education today will also bolster human capital accumulation at home in the future. A reproducibility package is available for this book in the Reproducible Research Repository at https://reproducibility.worldbank.org/catalog/461. 18 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces The home matters In the early 1990s, families in high-income countries began to adopt large numbers of Romanian orphans after learning about the living conditions in orphanages. In the orphanages, these children were often deprived of necessities, such as food and clothing. They did not live with adults who provided cognitive stimulation or emotional support. They sometimes even lacked human contact. A study that tracked what happened to orphans adopted by families in the United Kingdom found stark differences between orphans adopted after they had spent less than six months in orphanages and orphans who had spent more time in orphanages. Throughout childhood and adolescence, the orphans adopted before the six-month threshold could not be distinguished from other adoptees who had been born in the United Kingdom. The Romanian children who had spent more than six months in orphanages experienced a markedly different trajectory. They were much more likely to show symptoms of autism spectrum disorder, inattention and overactivity, and cognitive impairment throughout childhood, deficits that were still present at age 25.1 Their earliest environments had been characterized by severe deprivation, and this seemed to have stuck with them for life. The home environment is critical for human capital accumulation, particularly among children and adolescents (refer to figure 2.1). Young children spend most of their time with their families in the home and depend on family members almost entirely for nutrition, cognitive stimulation, social-emotional learning, and FIGURE 2.1 Human capital investment at home across the life cycle Nutrition, early stimulation and learning, and hygiene Stimulation and learning, whether to go to school, and for how long Who works and where, time use, and migration Early childhood (under age 6) School (ages 6–14) Youth (ages 15–24) Prime working age (ages 25–54) Diet, whether to seek medical care, risky behaviors, and mental health Source: Original figure for this publication. Human Capital Accumulation in the Home 19 protection from harm. During adolescence, decisions about staying in school, working, and time-use tend to be made in the home, often jointly among family members. Because human capital accumulates unevenly across the life cycle, these early investments can determine an individual’s entire trajectory of human capital accumulation.2 Many policies are already designed around the importance of the home. For example, early nutrition counseling tends to target mothers, and outreach related to essential childhood vaccination or social services may include home visits. The home, however, is rarely recognized as a place of learning, that is, as a place where individuals first acquire the cognitive and social-emotional skills that will smooth their path to success at school and in the labor force. Missing from most human capital policies are strategies to ensure that this type of human capital is built at home. Data produced through nationally representative surveys show strong associations between early human capital accumulation and family background. Parental education proxies for a range of attributes that may be beneficial to the health and skill development of children and adolescents, such as resources and sanitation conditions within the home, knowledge, and access to services. Figure 2.2 examines two outcomes—the prevalence of stunting among children under age 5 and mathematics proficiency among children ages 7–14—in a low-income (Afghanistan), lower-middle-income (Bangladesh), and upper- middle-income country (Thailand). Children whose mothers have higher educational attainment accumulate more human capital. In Afghanistan, for instance, children whose mothers have less than primary educational attainment are 47 percent more likely to be stunted than children whose mothers have at least some secondary education. (To examine these relationships in more countries, refer to the accompanying interactive graphs online, with the link provided in the figure note.) The relationship between educational attainment and children’s human capital outcomes holds for paternal education as well. While these results may not be surprising in the case of the younger children, who spend most of their time at home, the strong association between human capital and maternal educational attainment also holds for older children, who can attend school and acquire foundational skills, such as reading and mathematics, at school even if their families may not be equipped to encourage learning at home. In Bangladesh, children are around two times more likely to have attained minimal proficiency in mathematics if their mothers have completed some tertiary 20 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces FIGURE 2.2 Child nutrition and skills, by maternal educational attainment a. Stunting prevalence, ages 0–4 years b. Mathematics proficiency, ages 7–14 years Percent Percent 47.6 50 40 30 20 10 0 36.1 30.4 13.1 18.7 10.3 72.2 45.5 41.3 80 70 60 50 40 30 20 10 0 62.5 23.0 14.5 Less than primary Primary Secondary Some tertiary Less than primary Primary Secondary Some tertiary Maternal educational attainment Maternal educational attainment Afghanistan Bangladesh Thailand Source: Original calculations based on data on Afghanistan (2022–23), Bangladesh (2019), and Thailand (2019) from MICS (Multiple Indicator Cluster Surveys) (dashboard), United Nations Children’s Fund, https://mics.unicef.org/. Note: The “Secondary” school category for maternal educational attainment in Afghanistan includes “Some tertiary” as well. The sample size for “Some tertiary” for women in Afghanistan is too small to estimate stunting prevalence or math proficiency. For data on countries other than the three above, refer to the interactive graphs online, at https://humancapital.worldbank.org/en /building-human-capital-where-it-matters. education than if their mothers have completed only primary school or less. Even in Thailand, where fewer children are stunted and more children have attained proficiency in mathematics, there is a gradient in children’s outcomes by maternal educational attainment. These cross-sectional data demonstrate that there are stark differences in health and skills among children with different family backgrounds. Figure 2.3 presents findings from an analysis of longitudinal data from four countries that participated in the Young Lives Study, which has followed the same 12,000 children since 2002. The gaps in early vocabulary and mathematics among children whose mothers are at different levels of educational attainment are initially large (18–29 percentiles). They remain virtually constant throughout childhood and adolescence at a time when most children have attended some school. Human Capital Accumulation in the Home 21 FIGURE 2.3 Skill deficits among children, by maternal educational attainment Child’s ranking for vocabulary (percentile) Child’s ranking for mathematics (percentile) a. Ethiopia 68 44 80th 65th 50th 35th 71 43 80th 65th 50th 35th 72 43 65 45 4 5 6 7 8 9 10 11 12 13 14 15 16 4 5 6 7 8 9 10 11 12 13 14 15 16 Child’s age (years) Child’s age (years) b. India Child’s ranking for vocabulary (percentile) Child’s ranking for mathematics (percentile) 64 44 65th 50th 35th 59 46 65th 50th 35th 62 44 65 43 4 5 6 7 8 9 10 11 12 13 14 15 16 4 5 6 7 8 9 10 11 12 13 14 15 16 Child’s age (years) Child’s age (years) c. Peru Child’s ranking for vocabulary (percentile) 65th 57 57 Child’s ranking for mathematics (percentile) 65th 58 56 50th 35th 32 50th 35th 33 35 37 4 5 6 7 8 9 10 11 12 13 14 15 16 4 5 6 7 8 9 10 11 12 13 14 15 16 Child’s age (years) Child’s age (years) d. Viet Nam Child’s ranking for vocabulary (percentile) 65th 57 55 Child’s ranking for mathematics (percentile) 65th 58 57 50th 35th 39 41 50th 35th 39 37 4 5 6 7 8 9 10 11 12 13 14 15 16 4 5 6 7 8 9 10 11 12 13 14 15 16 Child’s age (years) Child’s age (years) Mother’s educational attainment: Primary or less Some secondary or more Source: Original figure for this publication, using data of Young Lives Study (dashboard), Oxford Department of International Development, University of Oxford, https://www.younglives.org.uk/. Note: Percentile rankings were calculated separately by round, marked by points. Only the youngest cohort (individuals born in 2001) are included. The older cohort (children born in 1994) were not observed at age 5. Vocabulary refers to receptive vocabulary as measured by the Peabody Picture Vocabulary Test (Dunn and Dunn 1997). Mathematics skills are measured through a subset of questions from TIMSS (Trends in International Mathematics and Science Study) (data repository), International Association for the Evaluation of Educational Achievement, https://www.iea.nl/data-tools/repository/timss. 22 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces Why does the home matter? What causes these differences among children with different family backgrounds? What attributes of the home environment are important in human capital accumulation? Resources matter A first source of variation is the fact that households differ in the resources available to them. To thrive, children require nutritious food, safe and sanitary living conditions, and opportunities to learn. Many of these needs can be met only if families spend money. Healthier foods and diets tend to cost more than less healthy options.3 Even if schools and health care are provided free, households must still purchase books and medicines and cover transportation costs. The presence of children’s books in a home is particularly important for early language acquisition and cognitive development.4 If parents engage in early reading activities with their children, they are helping build early skills in literacy, such as the skills associated with vocabulary and comprehension. These skills also facilitate the development of other cognitive skills, such as proficiency in mathematics. In nationally representative data from Madagascar, children living in homes with at least three children’s books are around four times more likely to be proficient in mathematics than children without books at home (refer to figure 2.4, panel a). The data also demonstrate how infrequent the ownership of children’s books is around the world. In Nigeria, for example, the average child has access to only one book at home. FIGURE 2.4 Skills in childhood, by resources and care at home a. Mathematics proficiency and children's books at home Madagascar, 2018 Various countries Proficiency, ages 7–14 (%) 40 Average number of children's books at home 8 30 20 10 5.9 0 0 23.9 6 4 2 0 Thailand (2.7) Nigeria (1.0) Madagascar (0.4) 1 or 2 Number of children’s books at home 3+ Low income Lower middle income Upper middle income High income (Figure continues on next page) Human Capital Accumulation in the Home 23 FIGURE 2.4 Skills in childhood, by resources and care at home (continued) b. Child development and care or stimulation activities Nigeria, 2021 Development on track, ages 3–4 (%) 60 54.1 50 40 30 20 10 22.3 Various countries Average number of different care or stimulation activities 6 5 4 3 2 Thailand (5.2) Nigeria (3.6) Madagascar (2.2) 1 or 2 0 Number of different care or stimulation activities at home with an adult 3 or 4 5+ c. Mathematics proficiency and violent punishment at home Thailand, 2019 Proficiency, ages 7–14 (%) 75 70 65 60 55 50 45 53.4 Yes Various countries Average prevalence of violent punishment at home (%) 100 80 60 40 20 Nigeria (78.5) Madagascar (52.4) Thailand (31.1) 70.1 No Violent punishment at home Low income Lower middle income Upper middle income High income Source: Original calculations based on data of MICS (Multiple Indicator Cluster Surveys) (dashboard), United Nations Children’s Fund, https://mics.unicef.org/. Note: The mathematics proficiency threshold is met if a child can respond correctly to six questions on reading numbers, five on number discrimination, five on addition, and five on number pattern recognition. Children face violent punishment if their caregiver reports they have been shaken, spanked, hit with a hard object, or beaten with an implement. The number of children’s books are capped at 10 in the survey. For care or stimulation activities, the number of types of activities is graphed, and the maximum is capped at six in the survey. Care matters A second source of variation across households stems from the care environment in the home, that is, the amount of time parents invest in helping children learn and in playing with their children, their methods of maintaining discipline, and the amount of social-emotional support they offer to children and adolescents. 24 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces Children’s early development, for example, increases with the number of care or stimulation activities at home, such as an adult singing or playing with a child. In Nigeria, for instance, only around one in five children who had experienced no care activities in the previous three days is developmentally on track (that is, exhibiting at least 80 percent of the skills deemed necessary to rule out developmental delays), whereas more than half of the children in homes reporting five or more activities were classified as developmentally on track (refer to figure 2.4, panel b). By contrast, there is a negative association between children’s skills and the use of violent punishment as a form of discipline. In Thailand, for example, nearly 70 percent of children exhibit proficiency in mathematics if their parents refrain from physically violent forms of punishment, while only 53 percent of the children who face such harsh discipline at home achieve mathematics proficiency (refer to figure 2.4, panel c). Global data suggest that the prevalence of this form of discipline in the home is substantial. In Nigeria, for instance, nearly 80 percent of parents report that they rely on violent punishment to discipline their children. Box 2.1 defines violent punishment and presents data on its prevalence, alongside evidence on the belief among parents in the usefulness of physical violence in raising children. Though many parents report that they use physical violence, far fewer parents report that they believe that “to bring up, raise, or educate a child properly, the child needs to be physically punished.” There thus seems to be a belief-behavior gap in disciplining children, suggesting that parents might need support in adopting alternative strategies to manage the behavior of their children. BOX 2.1 Violent punishment at home: Belief versus behavior In their survey responses, caregivers—typically mothers—report significant reliance on physical punishment to discipline their children (refer to figure B2.1.1). On average, in low- and middle-income countries on which data are available, 60 percent of children ages 1–14 experience violent physical punishment at home, including being shaken, spanked, hit with a hard object, or beaten with an implement. (Box continues on next page) BOX 2.1 Violent punishment at home: Belief versus behavior (continued) FIGURE B2.1.1 Use and endorsement of violent punishment to discipline children ages 1–14 Kiribati Central African Republic Afghanistan Congo, Dem. Rep. Samoa Nigeria Tonga Ghana Chad Gambia, The Togo São Tomé and Príncipe Vanuatu Benin West Bank and Gaza Guinea-Bissau Madagascar Algeria Malawi Bangladesh Fiji Nepal Eswatini Lesotho Iraq Tunisia Suriname Yemen, Rep. Jamaica Guyana Turkmenistan Honduras Comoros Dominican Republic Zimbabwe North Macedonia Viet Nam Trinidad and Tobago Lao PDR Kyrgyz Republic Thailand Costa Rica Argentina Azerbaijan Cuba Georgia Mongolia Uzbekistan Montenegro Kosovo Belarus Serbia 0 10 20 30 40 50 Share of caregivers (%) 60 70 80 90 100 Endorse physical punishment to discipline children Use violent punishment to discipline children Source: Original calculations based on data of MICS (Multiple Indicator Cluster Surveys) (dashboard), United Nations Children’s Fund, https://mics.unicef.org/. Note: Countries were omitted if the entire sample of children numbered fewer than 1,000. One country (Sierra Leone) was not included because no households reported endorsement, a pattern not found in any other data set. (Box continues on next page) 26 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces BOX 2.1 Violent punishment at home: Belief versus behavior (continued) This pattern contrasts with the evidence on gender-based violence. There is a limited set of countries that report women’s average reported experience with intimate partner violence, but according to data on the World Bank Gender Data Portal, the countries in which women report a high prevalence of intimate partner violence are also the countries in which women themselves report that such violence is justified.a a. Gender Data Portal (dashboard), World Bank, https://genderdata.worldbank.org/. Causal evidence on resources and care From these simple correlations in cross-sectional data, it is not possible to ascertain, however, what would happen if households had more resources or if they started to engage more regularly in care that is conducive to children’s development. More well-educated mothers may possibly engage in other, unobserved behaviors that are correlated with resources and care and that are beneficial for their children, or the correlations between maternal education and children’s outcomes may reflect the genetic transmission of human capital. Thus, a policy-induced increase in resources or care may not generate the higher levels observed in the cross-sectional comparisons of children. Moreover, it is possible that resources and care may be partial substitutes, that is, it may be possible to make up for the lack of one with higher levels of the other. To assess whether policies that provide poor households with more resources or that seek to improve children’s care environments will lead to improvements in human capital accumulation, it is useful to examine causal evidence or situations that help isolate the resource channel from the channel involving care. In particular, what happens to human capital if (1) households experience a change beyond their control (an exogenous change) in either resources or care and (2) some children receive less care despite the availability of equal or more resources in the household? Fluctuations in household income often follow recessions or changes in the price of a cash crop. These shocks may alter investments in children’s human capital and subsequent human capital accumulation. Children are especially Human Capital Accumulation in the Home 27 sensitive around the time of birth. Recessions in low- and middle-income countries—specifically, a 1 percent decline in GDP per capita—appear to raise the infant mortality rate by 0.25 to 0.40 for every 1,000 births.5 Conversely, in Ghana, an increase in the price of cocoa, a cash crop, during the first year of life of a child leads to improvement in mental health observed in adulthood. A one standard deviation price increase at birth leads to a 3 percentage point (or 50 percent) reduction in severe mental distress decades later.6 What happens if a household suddenly begins receiving additional income? This occurred in the United States when a casino opened on an indigenous Native American reservation in the state of North Carolina in 1997, and a portion of the profits were distributed to all adult tribe members on a per capita basis every six months. Over the next four years, this raised average annual income by about US$4,000 (US$8,031 in 2025 purchasing power parity), approximately 20 percent of average annual income at the time. Increases in some years amounted to 60 percent of annual income.7 Numerous studies have found lasting effects on children and their parents stemming from this large injection of resources (refer to figure 2.5). Children in poor households who began benefiting from the transfers before age 10 had, by age 21, completed one more year of education relative to children who had not benefited from the transfers and were 22 percent less likely to have committed a misdemeanor when they were age 16 or 17.8 Children in households that had received cash transfers for four years also exhibited fewer symptoms of behavioral and emotional disorders as adolescents and fewer symptoms of depression and anxiety at age 30.9 These improvements in human capital may partly reflect changes in parent-child relationships. Parents receiving the casino payments reported increases in the amount of time they had spent supervising their children. Children reported that they had spent more time on enjoyable activities with their mothers and had engaged in fewer arguments with their parents.10 Other studies demonstrate the importance of resources that poorer households may often forgo because of budget constraints. For instance, a nutrient-rich diet may be expensive, particularly in remote areas with limited access to markets. Experiments have shown that small-quantity lipid-based nutrient supplements—concentrated doses of micronutrients that can be sprinkled on food—can help prevent stunting.11 Likewise, a cement floor may be out of reach for many poor households, but replacing a dirt floor with a floor of concrete could decrease children’s exposure to the fecal matter and parasites that cause nutritional deficiencies. In Mexico, for example, a program that replaced dirt floors with floors made of concrete improved children’s cognitive development.12 28 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces FIGURE 2.5 Resources build human capital across the life cycle Average annual income goes up by an average of US$4,000 (US$8,031, 2025 PPP). After a casino opened on their reservation, Native American households in North Carolina started to receive transfers from casino revenues. Children who were at least 9 years old were followed over time. Symptoms of anxiety and depression are 90 percent lower at age 30. Children are 22 percent less likely to have been arrested at ages 16–17. Adolescents report having six fewer arguments with their parents in the previous three months, a decline of more than 110%. Among adolescents, behavioral disorder symptoms decrease by 0.23 standard deviations, and emotional disorder symptoms decrease by 0.37 standard deviations. Children who started to receive transfers before age 10 completed one additional year of education by age 21, compared with children not receiving transfers. 9 years 13–17 years 21 years Source: Original figure for this publication, based on Akee et al. 2010, Akee et al. 2018, and Akee et al. 2024. Note: PPP = purchasing power parity. Some unfortunate situations provide evidence of the importance of nurturing care among children and adolescents. If parents suffer from an emotional or behavioral disorder, for example, they may face difficulty finding and keeping employment and emotionally connecting with their children.13 Children with parents who are suffering from a mental health disorder are much more likely to become malnourished and experience injuries and bouts of diarrhea.14 Indeed, declines in parental mental health, rather than declines in income, appear to underlie the negative impacts of parental job loss on children’s school performance.15 According to one estimate, many children may find themselves in this situation; close to one in five adults globally had a mental health disorder in 2021.16 Without adequate care at home, resources are insufficient While receiving less care despite the availability of more resources may seem unusual, millions of children are in this situation if their parents migrate for work but leave them at home. In China, for instance, some parents in rural areas migrate to urban areas but leave behind their children, who would not be eligible for health or education services in the new work locations. In many other Human Capital Accumulation in the Home 29 countries, the number of children who are not residing with their parents is high (refer to figure 2.6). According to nationally representative household surveys in Eswatini, Sierra Leone, and Thailand, for instance, more than 20 percent of children ages 5–14 are not living with either parent although the parents are still alive. FIGURE 2.6 Share of children ages 5–14 living with neither parent Thailand Eswatini Sierra Leone Zimbabwe Guinea-Bissau Lesotho Malawi Ghana Gambia, The Jamaica Togo Benin Dominican Republic Central African Republic Madagascar Honduras Congo, Dem. Rep. Mongolia Kyrgyz Republic Chad Viet Nam Nigeria Cuba Lao PDR Trinidad and Tobago Nepal Costa Rica Bangladesh Georgia Turkmenistan Argentina Uzbekistan Belarus Serbia Yemen, Rep. Azerbaijan North Macedonia Algeria Kosovo Tunisia Afghanistan Iraq West Bank and Gaza 0 5 10 15 20 25 Percent Low income Lower middle income Upper middle income High income Source: Original calculations based on data of MICS (Multiple Indicator Cluster Surveys) (dashboard), United Nations Children’s Fund, https://mics.unicef.org/. Note: The coresidence status of children ages 5–14 is defined for children whose parents are not deceased, divorced, or separated. 30 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces Analysis of data of the China Family Panel Studies (CFPS), a survey conducted among households in China since 2010, suggests that at least 25 percent of rural children ages 5–17 have been left behind at some point during their childhood, typically with grandparents or other relatives, while their parents are away working as migrant labor.17 These children report that they see their parents an average of zero times in a typical week. What happens to these children? Does the additional income earned by their parents in migration help build human capital? Does the care they receive match what they would receive from their parents if their parents remained with them? Identifying children as left behind if the survey reports that they do not see their parents at all in a typical week, figure 2.7 presents selected information on the welfare of such children, including household income, the scores of the children on tests of vocabulary and mathematics, and the children’s own reports of how depressed they feel. The results of the analysis suggest that the additional income earned by parents does not compensate for the parental care missed by the children who are left behind. Compared with children who have never been left behind, children ages 10–17 who reported that they were left behind during any period were living in households that enjoyed higher incomes. Children who are now left behind, but were never left behind in the past are members of households with an additional US$794 (2010 US dollars) per year, 14 percent more than the national average. Children who were left behind at some point in the past, but who now see their parents at least once a week are living in households with an additional US$485 a year, 9 percent more than the national average. Children who have been left behind in all periods— the chronically left behind—show a smaller, US$120 advantage, 2 percent more than the national average.18 Despite the resource advantage, children who have ever been left behind exhibit lower test scores and higher rates of reported depression. The effects are large in magnitude. For instance, children who are now being left behind are a half year behind in mathematics scores relative to children who have never been left behind. These children are also 20 percent more likely to report being depressed. In other contexts, children living in the same home—with, in principle, the same access to infrastructure and resources—benefit from varying levels of care and human capital investment. In rural India, for example, parents spend much more time caring for infants who are boys than infants who are girls, devoting 60 fewer minutes a day if their youngest child is a girl. Male infants are also breastfed longer and are more likely to receive vitamin supplements.19 Similarly, the prevalence of stunting in India is less widespread among eldest sons, who typically care for parents when they are elderly, than among daughters and younger sons, who tend to leave the home after marriage.20 Human Capital Accumulation in the Home 31 FIGURE 2.7 Selected indicators of well-being, children ages 10–17, by left-behind status, China Average for children never left behind Annual household income (2010 US$) Chronically Present Past +120 +794 +485 Mathematics scores (% of a year of learning) –92 –88 –50 Chronically Present Past Vocabulary scores (% of a year of learning) –126 –23 –21 Chronically Present Past Reporting feelings of depression (%) Chronically Present Past +7 +10 +20 Children chronically left behind Children left behind now, but not in the past Children left behind in the past, but not now Source: Original calculations based on data of CFPS (China Family Panel Studies) (dashboard), Institute of Social Science Survey, Peking University, https://www.isss.pku.edu.cn/cfps/en/. Note: Children are considered left behind if they see both of their parents zero times in a typical week although the parents are still alive and the children are not in boarding school. The data are longitudinal. A child may therefore be left behind in some survey rounds, but not in others. Children are considered chronically left behind if they are left behind in all data rounds. Children are, on average, observed in three rounds. Evidence from multiple countries indicates that older children, particularly older sisters, must provide childcare to younger siblings, leaving these caregivers with less time to study and invest in their own human capital. Among a sample in rural Mozambique in 2017, children ages 10–14 spent approximately two hours a week taking care of their siblings, the same amount of time they were able to devote to homework.21 In a sample in rural Kenya, girls more often engaged in stimulating activities with their younger siblings than their mothers did.22 32 Building Human Capital Where It Matters: Homes, Neighborhoods, and Workplaces Overall, these findings suggest that resource availability in the home environment matters for children. Not only do children in a home with more resources exhibit greater human capital accumulation, but the responses of household members to exogenous changes in income show that human capital is sensitive to fluctuations in resources. The evidence also suggests, however, that resources alone are insufficient to build human capital effectively. The care environment of the home is a key determinant of human capital accumulation, specifically, the time parents spend supporting their children’s health and c