Policy Research Working Paper 10891 The Accumulation and Utilization of Human Capital over the Development Spectrum Federico Rossi Michael Weber People Practice Group Human Capital Project September 2024 Policy Research Working Paper 10891 Abstract This paper reviews how human capital is accumulated and capital on development. It is also a driver of human capital used over the process of development. It highlights that accumulation itself. In light of this interplay between accu- differences in worker productivity are the byproducts of a mulation and utilization, the paper discusses the potential wide variety of investments happening at different stages role of human capital in facilitating the adaptation to global of the life cycle, all subject to barriers in developing coun- challenges such as climate change, demographic transitions, tries. Moreover, the effective use of human talent in the and rapid urbanization. production process is a key mediator of the effect of human This paper is a product of the Human Capital Project, People Practice Group. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at mweber1@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team The Accumulation and Utilization of Human Capital over the Development Spectrum Federico Rossi* Michael Weber† Keywords: Human Capital Accumulation, Human Capital and Growth JEL Codes: J24, O15, I25 * Universityof Warwick, Department of Economics, Coventry CV4 7AL, United Kingdom. Email: federico.rossi@warwick.ac.uk. † World Bank and IZA, Bonn. Email: mweber1@worldbank.org. We are grateful to Aart Kraay and Omar Arias for useful comments and discussions. 1 Introduction Over the past few decades, human capital has been a central ingredient of the process of economic development. The XXth century has seen unprecedented increases in educational attainment across the world, as well as important improvements in the health and nutrition of the global population. Still, large cross-country gaps persist in several forms of human capital, and a growing literature in macro-development argues that these gaps contribute importantly to differences in the standards of livings between rich and poor countries (Rossi, 2020). In which ways is human capital accumulated across different countries? Through which channels does human capital contribute to development? How is this contribution likely to change in the future? This paper provides a review of the academic literature rel- evant to these questions. Our aim is to offer a comprehensive overview of the multifaceted processes of the accumulation and utilization of human capital, highlighting the key challenges that countries at different stages of development face in promoting these processes and the roles of policy in alleviating these challenges. Our work complements and adds to several reviews on human capital and economic development, such as Klenow and Rodr´ ıguez-Clare (1997), Caselli (2010), Flabbi and Gatti (2018), and Rossi (2020), which are mostly focused on the measurement of the contribution of skills for cross- and within-country income gaps. To set the stage for the rest of the paper, it is useful to start from a definition of human capital. Acemoglu and Autor (2011) define human capital as “any stock of knowledge or char- acteristics the worker has (either innate or acquired) that contributes to his or her productivity.” Two aspects of this definition are worth noticing from the onset. First, human capital is a very broad concept, which include cognitive skills but also many other characteristics potentially af- fecting productivity, such as health and physical traits, personality, socio-emotional skills, tacit and explicit knowledge, culture, and so on. Second, the definition puts no restrictions on how and when these relevant characteristics are acquired. In other words, formal education is by no means the only way to accumulate human capital; on the contrary, human capital accumulation can start in utero and then take place throughout the whole life cycle, from early childhood to late adulthood. Indeed, Section 2 reviews evidence supporting the idea that human capital gaps, both within and across countries, are due to differences in a variety of forms of investments. In- dividuals in low-income countries face stronger barriers to human capital accumulation through- out the life cycle, from before birth until old age, and these barriers reinforce each other due to dynamic complementarities between human capital investments. The resulting human cap- ital gaps go well beyond cognitive skills, and include physical health as well as a variety of non-cognitive traits. From a policy perspective, a narrow focus on educational attainment is likely to be ineffective, given the variety of environments and stages of the life cycle (beyond schooling) where human capital is accumulated. Section 3 argues that the effects of human capital accumulation on economic devel- 1 opment are crucially mediated by the way human capital is utilized in the production process. We discuss the key forces that shape the allocation of human talent across sectors, jobs, and firms, highlighting how the varying availability of high-skill labor affects the organi- zation of production in rich and poor countries. Several frictions prevent an efficient allocation of talent in poor countries, partially jeopardizing the effectiveness of the accumulation of in- vestments that have been taking place over the past few decades. Moreover, human capital ac- cumulation is itself a function of opportunities for utilization, as new generations have stronger incentives to invest in skills that can be utilized properly in the production process. In light of this interplay between accumulation and utilization, Section 4 concludes the paper by providing an outlook on the future importance of human capital for several mega-trends that are transforming the way production is organized. We argue that human capital will likely play an important role in shaping the effects of climate change, demographic change, and urbanization on the economy. More research is needed to identify the specific skills and traits that are more relevant for each of these trends, as well as the best ways to encourage their accumulation and effective utilization. Overall, these trends are all likely to increase the demand for highly specialized skills, which leads us to speculate that the XXI century will continue the human capital era. 2 How and when is human capital accumulated? 2.1 Conceptual Framework Human capital accumulates and multiplies across different stages over the life cycle (Fig- ure I). The productive capacity of an individual is shaped by the skills and traits acquired in early childhood, during adolescence (through formal schooling and other channels), during the working age period (through tertiary education, vocational training, and learning on the job) and at older ages. This accumulation process applies both to cognitive skills and other forms of human capital, such as non-cognitive skills, health, and culture. A large empirical literature documents that the human capital accumulation process features dynamic complementarity, such that the skills accumulated at one stage affect the productivity of further investments un- dertaken at later stages. This implies that barriers to human capital accumulation at early stages of the life cycle are particularly costly. In addition, investments in human capital, such as im- proving health, nutrition, and education, have a multiplicative effect over time, as they not only enhance an individual’s productivity and earnings potential, but also benefit future generations by creating a virtuous cycle of development. At each stage, the accumulation of human capital is driven by expected returns and perceived costs. Both returns and costs depend on the economic environment, available op- portunities, as well as a host of external shocks, as represented in Figure I. For example, in- dividuals might choose to acquire less schooling when most employment opportunities are in 2 sectors with low skill requirements (low return), when the opportunity cost of child labor is high (high perceived cost), and when a war disrupts the economy and educational system (low return and high cost). In addition, the incentives to accumulate human capital today depend on future accumulation opportunities, given the dynamic complementarity of these investments. All these factors are likely to vary dramatically over the development spectrum, contributing to the correlation between various forms of human capital and economic development. The next subsections review the existing evidence on some of the specific factors af- fecting human capital accumulation at each stage represented in Figure I, emphasizing whether and how they vary across countries. Given the broad scope in terms of the types of human capital investments being considered and the goal of keeping a cross-country per- spective, this review abstracts from important issues on the practical implementation of each of these investments, on which the evidence is more scattered and country-specific.1 Later in the paper (Section 3), the discussion will turn to how the utilization of the skills acquired at different stages shapes growth and development. Figure I: Human Capital Accumulation over the Life Cycle 2.2 Early Childhood and Parental Inputs Human capital accumulation starts before birth. An extensive literature suggests the course of fetal development and the quality of the in-utero environment are important determinants of health conditions and overall productivity later in life (Barker, 1990; Almond and Currie, 2011b). To establish causality, several papers study random shocks affecting pregnant mothers, 1 This includes important issues on the delivery of formal education, such as the advantages and disadvantages of private and public provision. 3 such as pandemics (Almond, 2006), conflict (Lee, 2014), or family bereavements (Persson and Rossin-Slater, 2016), typically concluding that children of mothers affected by negative shocks are characterized by higher incidence of various diseases, lower educational attainment and poorer labor market outcomes during adulthood. Human capital differences due to pre- birth conditions plausibly contribute to exacerbate inequalities within and across countries, as disadvantaged households are generally more exposed to negative shocks and poor nutrition. Early exposure to poor nutrition and diseases impairs human capital accumulation over the rest of the life cycle (Currie and Vogl, 2013). Policy interventions that provide food supplements to children in developing countries are often associated with large gains in terms of educational attainment and cognitive development, as shown for example in Field et al. (2009) for an iodine supplementation program in Tanzania and Maluccio et al. (2003) in the context of a randomized experiment providing high-protein drinks in Guatemala. In addition, eradication efforts for diseases such as malaria (Lucas, 2010) and pneumonia (Bhalotra and Venkataramani, 2011) are associated with life-time benefits in terms of schooling and earnings for cohorts covered by these programs during early childhood. These estimates might even understate the effects of improving nutrition and reducing exposure to diseases because of mortality selection, i.e., the idea that reductions in mortality might be associated with lower average health outcomes if relatively unhealthy individuals are more likely to die (Currie and Vogl, 2013). In addition, there is growing evidence that pollution experienced during childhood is another key determinant of human capital formation. Several studies show that environ- ments with strong exposure to carbon monoxide (Currie et al., 2009), lead (Troesken, 2008), and other airborne particulates (Chay and Greenstone, 2003) are associated with high infant mortality and low birth weight. The evidence (mostly from rich countries where long panels are more readily available) suggests that this early exposure also has long-term consequences. For example, Almond et al. (2009) find that Swedish children more exposed to radioactive fallout following the Chernobyl disaster perform worse in secondary school, while Ferrie et al. (2012) document adverse cognitive effects of early lead exposure for US Army enlistees. These sources of disruption to human capital formation are likely to be particularly important in developing countries, where levels of pollution are much higher than what was observed historically in the developed world (Currie and Vogl, 2013). A large body of work establishes that skills acquired during early childhood have an important impact on adult outcomes (Almond and Currie, 2011a). A key reason for this is that investments in skills feature important dynamic complementarities: acquiring skills and preserving health early on facilitates the further accumulation of skills later on in the life cycle (Heckman and Mosso, 2014). The investments identified as important include the quantity and quality of parental time with children (Del Boca et al., 2013), the parental involvement in children’s education (Heckman and Mosso, 2014) as well as the quality of other forms of childcare (Fort et al., 2020). These results highlight the importance of acting early to close 4 human capital inequalities, as investments in young adults lacking suitable foundations of basic skills are relatively less effective (Heckman and Mosso, 2014). The literature includes several evaluations of highly cost effective interventions; for example, Garc´ ıa et al. (2020) show that two early childhood programs implemented in the United States in 1970s had sizeable positive impacts on the participants’ education, health, and labor market outcomes over the life cycle. The benefits of these programs are even larger when their intergenerational effects are taken into account (Daruich, 2018; Garcia et al., 2021). This literature highlights the important role played by noncognitive skills, such as self-confidence and discipline, which are found to foster subsequent cognitive develop- ment (Cunha et al., 2010). An implication of this is that mentoring and guidance directed at fostering character and personality skills are often important ingredients of successful early childhood interventions; for example, Heckman et al. (2013) show that most of the positive impact of a US pre-school program on adult outcomes is due to improvements in noncogni- tive skills. Importantly, noncognitive skills also appear to be relatively more malleable during adolescence and early adulthood, suggesting that remedial programs targeting these skills can be somewhat effective in closing existing gaps, even though earlier interventions lead to larger returns (Heckman and Mosso, 2014). Moreover, accounting for the effects on noncognitive skills is crucial to properly evaluate the effects of policies involving monetary transfers or tax credits for parents (Morchio, 2022). While it is broadly accepted that early childhood inputs contribute to within country inequalities, the evidence is scarcer on whether they are also important to explain cross- country gaps. Schoellman (2016) shows that labor market outcomes of refugees arrived in the US during early childhood are independent of the age of arrival; in other words, spend- ing an extra year in a rich country environment does not significantly boost the formation of early childhood human capital. On the other hand, parents in poor countries plausibly face stronger informational, monetary and more generally poverty-driven constraints that lead to under-investment in young children (Attanasio et al., 2021). Relatedly, De Philippis and Rossi (2021) document that parental influence contributes to cross-country differences in standard- ized test performance. More work is needed to evaluate the quantitative importance of parental inputs for overall human capital accumulation and economic development. 2.3 Adolescence 2.3.1 Formal Schooling A large body of evidence shows that formal schooling is a key determinant of workers’ productivity. Estimates of Mincerian regressions from several countries show that an extra year of schooling is associated with about 10% higher wages, with limited variation between rich and poor countries (Caselli et al., 2016). While in principle these correlations could reflect differences in unobserved ability, estimates based on instrumental variable strategies suggest 5 that the causal effect of schooling is of a similar magnitude (Card, 1999). Moreover, schooling is associated with a broad variety of desirable outcomes, such as lower crime (Lochner and Moretti, 2004), higher civic participation (Milligan et al., 2004), improvements in child health (Currie and Moretti, 2003), and higher wellbeing (Easterbrook et al., 2015). Formal educational attainment varies substantially across rich and poor countries. Figure IIa displays the average years of schooling of the adult population against GDP per worker. In 2015, the average worker in Norway was a high-school graduate (about 12 years of schooling), the corresponding average worker in Mozambique had not completed primary ed- ucation (about 3 years of schooling). While most countries are relatively close to universal en- rollment in primary schools, large gaps remain for what concerns secondary school enrollment (Figure IIb). A large development accounting literature quantifies the contribution of these gaps in educational attainment in explaining cross-country income differences. The emerging consensus is that this contribution is limited: even low-educated workers in rich countries are much richer than high-educated workers in poor countries, leaving little space for the educa- tional composition to account for income gaps (see Rossi, 2020, for a review). Figure II: Educational Attainment and Development (a) Average Years of Schooling (b) Enrollment Rate 15 USA CZEISR CAN DNK NOR 100 NPLTJK GEO LKAIRN FJI BRB DZA ARG URY CRI BLR KNAMYS KAZ LTU GRCSVN LTU PRT CYP ESP ISR MLTKOR NZL ISR GBR FIN FRA GBR CAN BEL SWE ISL DNK SAU NOR IRL GBR JPN ITA AUS EST SVK KOR DEU SGP CHE SLE KHM BLZ PHL LAO MNG GRD UZBALB LCA SRB BRA AZEATG MUS LVA MDV MEX MNE RUS SYC TUR BLR CHL SVN ITA CYP NZL OMN FIN BHRNLD IRL NORLUX MAC IRL BRB SRB EST PRT OMN BEL FRA HKGUSA CHE ARE Average Years of Schooling (2015) HUN RUS SVN SWE LUX CPV PSE VCT PER ARM DOM LVAPOL GRC ESP AUS NLD ARM HRV CYP POL LTU AUS TWN HKG NLD MOZ TGO CMR VEN ECU COL GEO ZAFMKD BGR BGR HRVHUN CAN BHR USABRN KWT QAT SRB LVA MLT FIN HUN CZE ISL Net Enrollment Rate (%, 2015) UKR BEL ARE LSO STP KGZ GHA UZB BTN CHL HRV ARG MLT CZE DNKHKG DEU AUT KGZ BLZMDA BGR ROU KAZ BOL MDA ALB SUR ROUPAN TUR KAZ AUT DEUKWT TJK AUT SLV GTM ATG MUS CHE BRN LKA ALB MYSGRC ITA TTO ESP NZL ETH PSE JAM SWZ VCT ECU ZAF ROU SVK SVK LUX JAM CHL FRA DMA LCA BRA URY VEN ZAF BWA ISL 80 TCD KGZ HND PER CRIMEX SYC BHSVGB MAC 10 BOL JOR MNG BRB PER DOM ARGPAN BHRSAU SEN CIV VEN BOL MDA IDN COL THA FJI PRT QAT ECU MUS MAC GMB CPV JAM PHL GUY COL GAB URY BFA BLZ PAN PRY IRNBRA MEX DOM ZWE IDN CHN THA CRI STP PHL SLV TUN BTN GHA VNM SLV NER PAK DJI COD KEN ZMB NICSYR EGY IRQ DZA TUR 60 BGD SUR BHS KWT SDN CMR NAM MLI GHA LAO HNDIND LAO MDV LSO COG BGD TGO TZA BEN MWI HTI UGA PAK CMR GTM MAR GTM 5 RWA MMR CIV SWZ LSO HND SWZ GNQ CAF LBR GMB NPL MRT 40 DJI BEN KHM SDN SLE YEM MWI PAK MOZ SLE MLI SEN ETH MDG BDI RWA BFA MRT BDI MLI NER 20 MOZ TCD Primary NER LBR Secondary 0 7 8 9 10 11 12 7 8 9 10 11 12 Log GDP per worker in PPP (2015) Log GDP per worker in PPP (2015) Notes: Panel (a) plots average years of schooling in the population above 25 against log GDP per worker (in PPP terms) in 2015. Panel (b) plots the net enrollment rate in primary and secondary school against log GDP per worker (in PPP terms) in 2015. Average years of schooling is taken from Barro and Lee (2013), the enrollment rates are from the World Development Indicators, while GDP per worker is from version 9.0 of the Penn World Tables (Feenstra et al., 2015). More recent work, however, points to large gaps in learning achievements and educa- tional quality across countries, which compound the effect of gaps in schooling quantity. Figure III displays the average learning scores of primary and secondary school students from the Harmonized Learning Outcomes Database (Angrist et al., 2021a), a collection of average scores in standardized tests administered across 164 countries, against log GDP per worker. Students in rich countries vastly outperform their peers in poor countries, conditional on a given level of educational attainment; as a result, the vast majority of the young generations in the developing world do not achieve the skills needed to participate effectively in modern economies (Gust et al., 2024). In addition, Singh (2019) exploits panel data to show that an 6 extra year of primary schooling is associated with larger learning gains in Viet Nam com- pared to other developing countries.2 More broadly, large increases in educational enrollment in many developing countries have been associated with stagnant or even declining learning achievements by level of education (Pritchett, 2013; Le Nestour et al., 2023), highlighting the challenges brought by rapid educational expansions. Several studies argue that when quality differences are incorporated in the measurement of human capital, its contribution in account- ing for cross-country income gaps can be substantially larger (see for example Schoellman, 2012; Angrist et al., 2021b). Figure III: Average Learning Score and Development (a) Primary (b) Secondary 600 600 SGP TWN HKG SGP FIN LUX RUS LVA MAC BGR NLD DNK USA IRL KOR HUN KAZLTU CAN SWE DEU CHN TWN HKG PRTCZE AUT POL HRVGRC SVK SVN ITA ISR AUS 550 JPN FIN ESP NZLFRAISL 500 MDA ROU CYP EST CAN NLD Average Learning Score Average Learning Score NOR MAC TJK NZL GBR BEL AUS CHE IRL VNM TUR MLT SVN DEU MUS CHL TTO POL CZE FRA DNK AUT GEOAZE URYSYC HUN SWE USA ISL CRIMEX BHR KHM MMR KGZ GAB ARE 500 LTU RUS LVA PRT ESP NOR LUX MKD HRV SVK ITA SWZ IRNBRA ARG TZA KEN HND PER COL GRC ISR NIC IDNECU BWA SAU OMN 400 GTM PRY PAN MLT ZWE SLV QAT TUR KAZ CMR SRB ARE MOZ JAM BGR MYS ROU PHL JOR ZAF DOM 450 GIN SDN KWT THA CRI URYCHL NAM MDA BFAUGA NPL LSO VEN MUS TTO BDI TGO CIV COG LAO MEX MNE BEN SEN BLZMAR IRQ EGY JOR ETH GMB COM MDG TCD GEO AZE LBN ARG RWA BGD ALB COL TUN MKD BRA 300 MWI ZMB GUY IDN COD LBR HTI PAK 400 NER MRT DZA PAN PER QAT MLI NGA AGO YEM SLE IND DOM GHA 350 200 KGZ 7 8 9 10 11 12 8 9 10 11 12 Log GDP per worker in PPP (2015) Log GDP per worker in PPP (2015) Notes: The figure plots the average learning score for primary (left panel) and secondary school (right panel) students from the Harmonized Learning Outcomes Database (Angrist et al., 2021a) against log GDP per worker (in PPP terms) in 2015. The learning score is averaged across all available subjects and years. GDP per worker is taken from version 9.0 of the Penn World Tables (Feenstra et al., 2015). These results underline the need for effective investments making schools in poor countries well-equipped to deliver learning against the backdrop of a rapidly growing student population. The existing literature highlights a number of promising margins. Based on a comprehensive review of the evidence, the Global Education Evidence Advisory Panel (2023) concludes that improving classroom instruction through structured pedagogy and tar- geted support for students falling behind, and the provision of information on education quality are particularly cost-effective interventions. In many developing countries, it has been doc- umented that teachers perform poorly in terms of both subject knowledge and exerted effort (Bold et al., 2017); in light of this, the returns to policies that train teachers and provide them with appropriate incentives are likely to be high (Kremer et al., 2013; Evans and Popova, 2016). New technologies could help by both complementing teachers’ efforts and accommodating a growing and increasingly diverse student population; for example, Muralidharan et al. (2019) reports large performance gains associated with an instruction program that dynamically adapts the proposed content to the learning level of each student. More broadly, systemic reforms are 2 Indirect evidence on the crucial role of school quality for human capital formation comes from the learning losses experienced by several countries during the pandemic, which were large and highly heterogeneous across countries (Hanushek and Strauss, 2024). 7 needed to ensure that all parts of the educational system are aligned with the goal of improving learning, overcoming political challenges and technical constraints that are often particularly severe in developing countries (Pritchett, 2013; World Bank, 2018). Within countries, access to formal education varies substantially across gender, loca- tion and socio-economic classes, contributing to inequalities across these groups. Figure IVa highlights the gender dimension, displaying the difference between men’s and women’s average years of schooling against GDP per worker. In most of the poorer countries, women are less likely to be educated than men, while this gap disappears or even reverses with de- velopment.3 Figure IVb visualizes the extent of educational disparities across administrative units for countries at different deciles of the income distribution. Poorer countries feature very large gaps in average education between regions, while richer countries tend to be more ho- mogeneous. In addition, the importance of the family’s socio-economic status for children’s educational achievements has been widely documented in several rich countries (Bj¨ orklund and Salvanes, 2011), and its role might be even larger in poorer countries (though intergener- ational data are less readily available for developing economies). These different sources of within-country heterogeneity highlight the importance of understanding the barriers faced by different groups in human capital accumulation and utilization, an issue discussed in greater detail in Section 3. Figure IV: Schooling Gaps within Countries (a) Gender Gap (b) Across Regions Years of Schooling Difference: Men - Women (2015) 4 15 EGY TGO YEM LBR IND ARM USA COD BEN CAF HTI SEN Average Years of Schooling GMB PAK 2 UGA MRT 10 FRA SLE NPL KHM CIV GHA LAOMAR POL MOZ PER IRQ MWI ZWE KEN COG SYR TUN TUR PAN CMR BOL ZMBBGD BDI NER RWA TZA SDN GTM URY PER MLI TJK VNM SLV IDN DZA CHN KOR AUT SAU JOR IRN MDV HKG CHE ARE IRQ MUS THA MEX ZAF BWA BHR DEU SGP ALB NLD ROU CHL TWN LUX MAC 0 KGZ BLZ ECU PRY LKA SRB HRV MYS SVK GRC HUNCYP CZE GBR MLT ISR FRA AUS USA 5 ARM COL CRI KAZ TTO ITA NIC HND MDASWZ BGR RUS POL BEL NOR FJI UKR BRB SVN ESPJPN MMR JAM BRA ARGPANLTU NZL CAN IRL SWE PHL NAM DOM URY LVA PRT FIN DNK NPL MNG EST GUY ISL KWT SEN VEN LSO ETH GAB QAT -2 0 7 8 9 10 11 12 6 7 8 9 10 11 Log GDP per worker in PPP (2015) Log GDP per worker in PPP Notes: The left panel plots the difference in the average years of schooling in the population above 25 between men and women against log GDP per worker (in PPP terms) in 2015. The right panel plots (for countries belonging to the first to tenth deciles of the income distribution, plus the United States) average years of schooling in the population above 25, across regions (grey dots) and for the whole country (black dots), against the country’s log GDP per worker (in PPP terms). Data on years of schooling by gender is taken from Barro and Lee (2013), years of schooling by region is from IPUMS International (, n.d.), while GDP per worker is taken from version 10.0 of the Penn World Tables (Feenstra et al., 2015). 3 Evans et al. (2021) document that in most developing countries gender gaps in education have typically widened first (as boys started going to school) and then shrunk in a second phase (as educational expansions started involving girls as well). 8 2.3.2 Other Sources of Human Capital Accumulation Diseases contracted during adolescence represent a significant obstacle to skill accumula- tion. Part of the reason for this is that sick children can hardly attend or learn at school. This idea is supported by the analysis of the experimental introduction of deworming treatments in Kenyan primary schools (Miguel and Kremer, 2004; Hamory et al., 2020a), which resulted in higher school attendance and cognitive performance for the treated children. Similarly, the eradication of malaria is typically associated with higher educational attainment and labor mar- ket outcomes for the exposed school-age children (Lucas, 2010; Barofsky et al., 2015); in a structural framework calibrated to replicate the effects of a large-scale anti-malaria campaign in Tanzania, Kim (2022) shows that taking into account human capital formation substantially amplifies the economic gains from a universal malaria vaccination. These results highlight the important complementarities between different sources of human capital accumulation. In developing countries, child labor represents a significant barrier to human capital accumulation during adolescence. According to the International Labor Organization about 10 percent of the global child population is involved in child labor (International Labor Or- ganization, 2020), which dramatically reduces learning opportunities by being associated to reductions in the time and energy dedicated to schooling as well as various health hazards (Basu, 1999; Edmonds and Pavcnik, 2005). Child labor is mostly a consequence of poverty, and is way more common among low-income households in developing countries with little access to credit. The potential returns of policies targeted at reducing child labor are very large, though simple bans, fines or compulsory schooling laws are difficult to enforce in developing countries and might actually backfire by lowering the equilibrium wage and inducing more children to work to meet their family’s subsistence needs (Basu and Van, 1998; Edmonds and Pavcnik, 2005). Children and young adults also learn directly from their peers. A large literature in economics documents positive associations between students’ test scores and the ones of their peers, in particular when focusing on groups with similar socio-economic backgrounds and close friendships (Sacerdote, 2011; Hahn et al., 2019). These peer effects go beyond educa- tional performance and extend to health outcomes (Giulietti et al., 2022), migration decisions (Munshi, 2020) and risky behavior (Lindquist and Zenou, 2019), therefore shaping human cap- ital accumulation in multiple forms. In light of this evidence, the effects of all the barriers to human capital accumulation typical of developing countries are likely amplified by the fact that they not only harm individual children, but also their peers. 9 2.4 Post Basic Schooling and Working Age 2.4.1 Tertiary Education The prevalence of tertiary education varies dramatically across countries. As shown in Figure V, tertiary educated workers are virtually absent in poor countries, while they repre- sent half or more of the working age population in the richest countries. The rise of upper tail knowledge is thought to have played an important role in the historical development pro- cess of currently rich countries, by facilitating innovation and the adoption of new technologies (Mokyr, 2002; Squicciarini and Voigtl¨ ander, 2015). Today, in many countries inventors, en- trepreneurs and executives in leading firms are educated disproportionately in a small set of universities (Bell et al., 2018; Levine and Rubinstein, 2016; Martellini et al., 2022). Figure V: Tertiary Education and Development 80 Share with Tertiary Education (%, 2015) RUS USA 60 CAN ARM ARE SGP JPN ISR UKR KOR TWN DNK IRL FIN AUS NOR CYP GBR 40 MDA LTU EST SWE CHE BEL LUX ESPFRA HKGNLD VEN LVA NZL AUT PHL GRC ISL MAC IRN BGR POL BOL PAN SVN SAU TJK MNG DOM KAZ HUN BHR DEU CRI CHL PRT HRV CZE PER COL SRB 20 MYS SVK MLT ARG KGZ THA QAT JAM EGY ROU ITA JORALB TUN MEX TUR PRY ECU GAB DZA BRA KWT NIC LAO URY ZWE IND MAR IRQ TTO COD BGD PAKHND SLV IDN UGA GMB KEN MMR ZAF BWA YEM KHM SDN VNM BRB LBR NPL BEN CMR CIV SYRBLZ CHN FJI MUS TGO RWA MLI SEN GHA GTM LKA MDV COG CAF BDI NERMOZ SLE HTI TZA LSO MRT ZMB SWZ NAM GUY 0 MWI 7 8 9 10 11 12 Log GDP per worker in PPP (2015) Notes: The figure plots the share of the 25-64 population with some tertiary education against log GDP per worker (in PPP terms) in 2010. Data on educational attainment is taken from Barro and Lee (2013), while GDP per worker is taken from version 10.0 of the Penn World Tables (Feenstra et al., 2015). Recent work shows that the quality of tertiary educated workers also varies substan- tially across countries. Martellini et al. (2022) use data on the earnings of migrants across many host and origin countries to show that those educated in universities in developing coun- tries earn less than their peers from rich countries, everything else equal. Moreover, migration reinforces cross-country differences in the quality of the tertiary educated labor force, as mi- grants from poor to rich countries are positively selected in human capital. These results rein- force the conclusion (already discussed for primary and secondary schooling in Section 2.3.1) that closing educational gaps between countries requires more than increasing enrollment in the developing world. 10 The low prevalence of tertiary education in the developing world also reflects a lack of demand for the types of skills acquired in universities. As discussed in greater detail in Sec- tion 3, the demand for advanced skills is higher in environments with large and modern firms, complex technologies, and a low prevalence of agriculture, which are all features typical of rich countries. Standard growth models predict that countries closer to the productivity frontier rely more on innovation, an activity performed by highly skilled workers, while countries at early stage of development can grow out of imitation and investment-based strategies (Acemoglu et al., 2006). However, the accumulation of advanced skills can also initiate a virtuous cycle of firm creation and structural transformation towards modern sectors. This view is supported by the findings in Cox (2023) and Coelli et al. (2023), showing that policy-induced expansions in the provision of higher education result in an increase in firm size and a reallocation of la- bor away from agriculture and towards manufacturing or skill-intensive services in the most affected regions. 2.4.2 Vocational Training Vocational training refers to a variety of educational programs aiming to prepare students for a particular craft or trade. The prevalence of vocational education varies substantially across countries. Figure VI displays the share of upper secondary students enrolled in voca- tional (as opposed to general) training, against GDP per worker. In some countries vocational programs represent the norm at the secondary level, while in others they are virtually nonexis- tent. There does not appear to be a strong relationship with the level of development, though the overall enrollment rate in secondary education is higher in rich countries. This implies that the overall enrollment share in vocational education with respect to the schooling-age population is higher in those countries. Vocational education appears to facilitate the school-to-work transition. Zimmermann et al. (2013) reviews various studies documenting how the participation to occupation-specific training is associated with lower youth unemployment and higher labor market attachment at the beginning of workers’ career, with the effects on wages being more mixed (Ryan, 2001; Brunner et al., 2021). Importantly, vocational education can also act as a leveler of the playing field, by allowing a relatively larger improvement of the labor market outcomes of disadvan- taged students (Bertschy et al., 2009). However, vocational education might involve a trade-off between short-run gains and long-run losses. At times of rapid transformation of the labor market, the specific skills ac- quired through vocational programs might become obsolete relatively quickly, while more gen- eral education insures against this possibility by providing a stronger ability to adapt. In line with this hypothesis, Hanushek et al. (2011) shows that across 11 rich countries individuals with general education obtain worse labor market outcomes at entry, but gradually catch up and overtake their peers with a vocational education background. This result underscores the 11 Figure VI: Enrollment in Vocational Training across Countries Enrollment Share in Vocational Programs 80 CZE AUT SVK SVN 60 CHE POL LUX BEL NLD ITA TUR FIN HUN NOR 40 ISR LVA PRT MEX GBR FRA CRI DEU SWE COL EST CHL RUS LTU GRC ESP JPN 20 AUS DNK KOR NZL ISL IRL BRA ZAF CAN 0 SAU 9.5 10 10.5 11 11.5 Log GDP per worker in PPP Notes: The figure plots the share of upper secondary students aged 15-19 enrollment in vocational programs in 2019, against the country’s log GDP per worker (in PPP terms) in 2019. Enrollment by program orientation is taken from OECD (2022), while GDP per worker is taken from version 10.0 of the Penn World Tables (Feenstra et al., 2015). more general theme that the appropriateness of general and vocational education is context- specific, with faster technological process and higher labor market uncertainty favoring the accumulation of general skills, and increases in the division of labor favoring specialization (Alon, 2018; Alon and Fershtman, 2023). 2.4.3 On the Job Accumulation The accumulation of human capital on the job is an important determinant of wage growth over the life cycle. Such accumulation is facilitated by both formal training pro- grams provided by firms (Acemoglu and Pischke, 1998), as well as more informal learning opportunities associated with work experience (Bagger et al., 2014). Within countries, firms are heterogeneous in the quality of the learning environment they offer; for example, firms with more knowledgeable co-workers, (Jarosch et al., 2021) involved in exporting activities (Ma et al., 2020), of larger size (Arellano-Bover, 2020) and located in big cities (De La Roca and Puga, 2017) allow their workers to achieve faster human capital growth. The rate of wage growth over the life cycle is significantly higher in rich countries. Lagakos et al. (2017) document this fact for 18 countries; while wages roughly double over the life cycle in countries like Germany and the United States, they only increase by about 50% in poorer countries like Jamaica and Mexico. Jedwab et al. (2021) extend this result to a broader and more diverse sample of 145 countries. While higher returns to experience in rich coun- tries might also reflect lower labor market frictions and a more experience-biased organization 12 of production, several pieces of evidence point to human capital as an important driver. For example, Lagakos et al. (2018) show that, among US immigrants employed in the same labor market, the returns to experience are lower for those coming from poor countries, suggesting that the cross-country variation in life cycle growth is not entirely driven by technological and institutional factors. Moreover, Jedwab et al. (2021) show that experience accumulated under a subsequently overturned Communist regime has lower returns, consistently with the value (or obsolescence) of acquired skills being a determinant of returns to experience. Lower life cycle growth might be the endogenous response to lower returns to human capital investments in poor countries. Seshadri and Manuelli (2014) develop a model where the accumulation of human capital during the early childhood, schooling and post-schooling phases is the outcome of an income maximization problem. Their analysis show that relatively small increases in total factor productivity, reflected into higher wage rates and lower relative prices of goods used as inputs for the production of human capital, can lead to large differ- ences in human capital per worker, with pre- and post-schooling investments both playing a quantitatively important role. Ma et al. (2021) focus on the role of firm-provided training pro- grams: first, they document empirically that such programs are relatively more common in rich countries; second, their quantitative model suggests that a key reason for this is the higher job turnover in poor countries, which depresses firms’ incentives to invest resources in the training of their current workers. 2.5 Older Ages Population aging calls for continued investments in human capital at older ages. As pop- ulations around the world age, understanding the dynamics of human capital accumulation in old age becomes crucial. In contrast to traditional notions that suggest a decline in productiv- ity and skill acquisition in later life, recent studies reveal an encouraging trend of continuous human capital development among more mature workers. By debunking stereotypes and rec- ognizing the value of mature workers, (aging) societies can leverage their skills, experience, and expertise for continued economic development and growth. Potentials of learning later in life challenge stereotypes of diminished learning abilities while opening new opportunities. Contrary to popular belief, studies show that older adults can acquire new skills and knowledge effectively, challenging the stereotype of diminished learning abilities in old age. Lifelong learning programs and targeted training initiatives have been successful in enhancing the human capital of older individuals, enabling them to adapt to evolving job requirements (Jenkins, 2003; Fleming, 2011; UNESCO, 2022). Older adults who engage in learning activities have higher cognitive function. For example, Park and Bischof (2022) finds that older adults who participated in intellectual activities, such as reading, puzzles, and learning new skills, experienced slower cognitive decline compared to those who were less engaged. This suggests that continued learning in later life contributes to maintaining cognitive 13 abilities. Tapping into the experience and expertise of older workers can enhance productivity and economic performance while also benefiting younger cohorts. Older workers further- more possess a wealth of experience and expertise accumulated over their careers, making them valuable assets to organizations. Research indicates that age-related changes, such as improved problem-solving abilities and enhanced decision-making skills, can compensate for potential declines in cognitive functioning (Ng and Feldman, 2013). These characteristics contribute to the productivity and overall performance of mature workers. Furthermore, the transfer of knowledge from older to younger workers enhances organizational performance. The expertise and experience of older workers, when effectively shared, contribute to the overall success and competitiveness of businesses (Harvey, 2012; Dietz et al., 2022). 3 Human Capital Utilization and the Organization of Pro- duction 3.1 Conceptual Framework Figure VII displays a conceptual framework on the links between the accumulation and utilization of human capital and growth and development. Traditionally, the macro-development literature postulates that human capital increases workers’ productivity, and therefore directly shapes growth and development. For example, the development accounting approach relies on the cross-sectional dispersion in wages among workers with different levels of education and experience, or alternatively among migrants from different countries, to measure the underly- ing variation in average skills, and its contribution to cross-country income gaps. Moreover, quantitative structural models are typically used to study the extent to which human capital accumulation itself is driven by economic development (see Rossi, 2020, for a review of both accounting and structural approaches). More recent work emphasizes the importance of human capital utilization as a me- diating factor of the link between human capital and economic development. The het- erogeneity in the utilization of skills across firms, occupations, sectors and tasks demonstrates that human capital confers comparative advantages for some economic activities over others, suggesting that changes in the set of skills supplied by the labor force might drive economic development through changes in the organization of production. Moreover, various features of the organization of production are likely to affect in turn the incentives to accumulate hu- man capital in the first place. The next subsections review the existing evidence on the forces shaping the utilization of human capital across various types of economic activities and orga- nizations, illustrating the evidence supporting the channels represented by the arrows in Figure VII. 14 Figure VII: Human Capital Utilization and Development Human Capital Utilization Structural Transformation (i) (a) (b) Technology (j) (c) Adoption Human Capital Growth and Accumulation (d) (e) Development (k) Firms (f) (g) (l) Talent (h) Misallocation 3.2 Structural transformation Human capital is not equally distributed across sectors and space. Across multiple coun- tries, highly educated workers are less likely to work in agriculture and live in rural areas (Gollin et al., 2014; Young, 2013), and more likely to work in skill-intensive services such as education, finance, public administration and health (Herrendorf and Schoellman, 2017).4 Figure VIII visualizes this pattern by plotting average years of schooling across four broadly defined sectors against GDP per capita. High-skill services is the most schooling intensive sectors, followed by low-skill services and manufacturing, and then agriculture. Interestingly, these cross-sectors differences in schooling are largest in poor countries and tend to diminish with development. Returns to education are generally larger in sectors with many highly educated work- ers and urban areas. Herrendorf and Schoellman (2017) document across 13 countries that an additional year of schooling is associated with a significantly larger increase in wages in out- side of agriculture compared to agriculture. Figure IX makes use of the World Bank I2D2/GLD database to generalize this finding to a larger sample of country - on average, the return to schooling is 9% in non-agriculture and 7% in agriculture. One caveat for these comparisons is that many agricultural workers in poor countries are self-employed, and as such are not in- 4 Moreover, when focusing on workers moving across sectors, the more educated agricultural workers are more likely to leave agriculture, while the less educated non-agricultural workers are more likely to move to agriculture (Young, 2013; Hamory et al., 2020b). 15 Figure VIII: Years of Schooling by Sector Average Years of Schooling 15 10 5 High-Skill Services Low-Skill Services Manufacturing Agriculture 0 7 8 9 10 11 Log GDP p.w. Notes: The figure plots average years of schooling by sector for all country ˆ year pairs in IPUMS International, against GDP per worker from version 9.0 of the Penn World Tables (Feenstra et al., 2015). cluded for the computation of agricultural wages. However, the estimated returns to schooling are not dramatically affected when computed using self-employed income as well (Herrendorf and Schoellman, 2017; Rossi, 2022). New cohorts are disproportionately more likely to work outside agriculture, contribut- ing significantly to structural transformation out of agriculture (Schoellman and Hobijn, 2017; Porzio et al., 2022). Porzio et al. (2022) show that a key reason for this is that more recent cohorts are typically more educated, which gives them a comparative advantage in non-agriculture. Human capital growth across cohorts is both a cause and a consequence of structural transformation; on one hand, the entry of more educated cohorts allows the non- agriculture sector to expand (Figure VII.a), while on the other the increase in the demand for agricultural labor driven by technological change and income effects incentivizes the accu- mulation of human capital (Figure VII.b). These conclusions are consistent with a broader literature arguing that the expansion of formal education is a key ingredient of the transition from stagnation to growth (Galor, 2005). Overall, these findings describe a complex relationship between human capital, struc- tural transformation, and economic development. While human capital growth is a potential tool to accelerate the economy wide transition out of agriculture, a crucial factor is given by the extent to which the labor market is able to absorb an increase in the supply of highly edu- cated workers. Barriers to the growth of advanced skill-intensive sectors are likely to limit the productivity effects of educational expansions, leading to an inefficient utilization of human 16 Figure IX: Returns to Schooling by Sector .2 Returns to Schooling .1 0 Non Agriculture -.1 Agriculture 6 7 8 9 10 Log GDP p.w. Notes: The figure plots the estimated sector-specific coefficient on years of schooling from a regression with log wage income as dependent variable against log GDP per worker (in PPP terms). Each dot corresponds to a different country ˆ year sample. The estimation is performed on the World Bank I2D2/GLD database, while GDP per worker is taken from version 9.0 of the Penn World Tables (Feenstra et al., 2015). capital. This would result in skills mismatch, workers’ frustration, and lower returns to hu- man capital investments. In addition, given that agriculture is the sector displaying the largest productivity differences between poor and rich countries (Caselli, 2005), facilitating the real- location of workers away from this sector through human capital accumulation might support economic growth in the developing world (Figure VII.i). 3.3 Technology Adoption Human capital is a key driver of both the invention and the adoption of new technolo- gies (Figure VII.c). One reason for this is that understanding new technology developments and adapting to changes in the production process require advanced skills, which are typi- cally acquired through education and experience (Nelson and Phelps, 1965). The literature has offered several pieces of evidence in support of this idea. For example, Ciccone and Papaioan- nou (2009) show that countries with higher levels of education in 1980 saw faster growth in human-capital-intensive industries (relative to other industries) in the following decades, a pe- riod characterized by the ICT revolution and the rapid introduction of skilled labor augmenting technologies. Consistently with this idea, Maloney and Caicedo (2022) show that countries with a larger share of engineers in 1900 are characterized by faster technology adoption his- torically and higher levels of income per capita today. At a more micro level, Foster and 17 Rosenzweig (1995) document that Indian farmers with more experience or more experienced neighbors were more likely to adopt the high-yielding varieties associated with the Green Rev- olution. Given the important role played by technological change for economic growth (Figure VII.j), this makes human capital a key proximate driver of economic development. Recent work argues that younger generations play a particularly important role for technological diffusion. A fundamental reason for this is that new cohorts, in particular those at the formal education stage, have far more flexibility and malleability in the acquisition of new skills compared to older workers. When new technologies require a fundamentally differ- ent set of skills compared to the existing ones, their diffusion and consequent adjustment of the labor market can take several decades, as new cohorts acquire the necessary skills (Chari and Hopenhayn, 1991). In recent work, Adao et al. (2022) show that the ICT revolution is an exam- ple of a technological innovation which required very specific skills, and hence unfolded more gradually compared to the improvements in manufacturing productivity of the early 1900s. More generally, most recent forms of technological change have plausibly increased returns to skills (Acemoglu, 2002), providing a further incentive to acquire further human capital (Figure VII.d). 3.4 Firms Economic development is associated with a dramatic transformation of the organization of production. Most workers in developing countries are self-employed, while firms gener- ally employ few workers, operate in a single establishment, and are characterised by limited growth over time (Gollin, 2008; Hsieh and Klenow, 2014) On the contrary, a large part of the labor force in rich countries is employed in large multi-establishment corporations, with complex organizational structures (Poschke, 2018; Garicano and Rossi-Hansberg, 2015).5 The macro-development literature traditionally focuses on the relaxation of financial frictions and technological change as potential causes of this transformation of the production landscape (Gollin, 2008; Buera et al., 2015; Poschke, 2018), which in turn is associated with productivity growth and economic development (Figure VII.k). More recent work highlights the role of human capital in facilitating the rise of large and modern firms (Figure VII.e). Gomes and Kuehn (2017) point out that at the country level the share of secondary educated workers is positively correlated with firm size, even when controlling for proxies of financial development and institutional quality, and proposes a theory where more educated managers are able to grow larger firms. Seshadri and Roys (2014) and Guner et al. (2018) consider models with endogenous human capital accumulation, and show that higher productivity and lower barriers to firm growth incentivize the accumulation of managerial skills over the life cycle (Figure VII.f), which in turn is an important driver of 5 Firms are also highly heterogeneous within countries, and this heterogeneity contributes to worker earnings’ inequality (Song et al., 2019; Tomaskovic-Devey et al., 2020). More work is needed to characterize the evolution of firm heterogeneity and its contribution to inequality over the process of development. 18 cross-country differences in the firm size distribution and productivity. The hypothesis that the availability of managerial know-how grows with economic development is supported by the evidence in Hjort et al. (2022), which exploits data from a global consulting company to show that the relative remuneration of middle managers is much higher in poor countries. More generally, a growing literature documents the key role played by managers for firm-level performance. The World Management Survey has collected detailed firm-level in- formation on managerial practices over the past two decades, showing a wide variation across firms and countries closely associated to differences in productivity (Bloom and Reenen, 2007; Bloom et al., 2014). Moreover, the randomized controlled trial in Bloom et al. (2013) supports the existence of a causal relationship between managerial practices and performance, as plants that were randomly provided with free managerial consulting saw faster growth in total factor productivity and a persistent increase in the probability of adopting modern management prac- tices. Consistently with the literature on human capital and firms discussed above, managers’ education and know-how are commonly found to be positively associated with more effective managerial practices and firm-level upgrading (Bloom et al., 2014; Verhoogen, 2021). 3.5 Misallocation of Talent A key factor mediating the impact of human capital on economic performance is the al- location of talent. It is self-evident that workers vary widely in their comparative advantage for different occupations and tasks, as a function of their education, experience and idiosyn- cratic ability (Figure VII.e). Models of occupational choice such as Roy (1951) make it clear that frictions that prevent workers from pursing their comparative advantage are detrimental to aggregate productivity, partially undermining the positive effects of investments in the ac- cumulation of skills (Figure VII.l). A growing literature attempts to quantify the extent of the misallocation of talent and associated productivity losses in various contexts. Several studies emphasize the misallocation resulting from discrimination and social norms against specific demographic groups. Hsieh et al. (2019) documents that the occupa- tional distribution of women and black men has changed dramatically in the United States since the 1960s, and concludes that the relaxation of the labor market barriers faced by these groups can explain between 20% and 40% of the observed economic growth. Ashraf et al. (2022) ex- ploit personnel data from a multinational firm active in many countries to show that women’s wages are higher for countries with lower female labor force participation, consistently with the idea that these labor markets are characterized by larger barriers faced by women (and hence more positively selected female workers). Hurst et al. (2021) provides more evidence on the barriers faced by black workers, showing that in the United States the discrimination against this group is stronger for occupations involving interaction with others, suggesting that the growing demand for these tasks might underlie the stagnation of the wage race gap since the 1980s. It is plausible that the relaxation of these barriers might itself encourage human capital 19 accumulation (Figure VII.h), though it is challenging to establish this link empirically. The inequality of opportunities induced by differences in economic backgrounds is also an important source of talent misallocation. Bell et al. (2018) document that in the United States the probability of becoming an inventor is strongly related to parental socio- economic characteristics, suggesting that there might be many “lost Einsteins” in the poorer segments of society, i.e. individuals with high innovative potential who are prevented from creating innovations by their socio-economic background. A similar theme emerges from the development literature on poverty traps, which highlights how in the presence of capital market imperfections the distribution of wealth can shape the allocation of talent across different eco- nomic activities and economic growth (Banerjee and Newman, 1993; Galor and Zeira, 1993). Balboni et al. (2021) provide evidence for this mechanism through a randomized transfer of cows in Bangladesh, by showing that households treated with a large enough transfer are more likely to start their own business and grow out of poverty. 4 Global challenges and the future of human capital 4.1 Climate Change The anthropogenic emissions of carbon into the atmosphere are causing dramatic changes in the climate of our planet. A growing literature is debating the economic consequences of this process. Global warming is likely to reduce labor productivity - particularly so in agri- culture - and to induce migration flows and shifts in the geographic distribution of economic activities (Dell et al., 2014; Cruz and Rossi-Hansberg, 2021; Feriga et al., 2024). The rise in sea levels might lead to coastal flooding and the associated destruction of physical capital and infrastructure, as well as growing migration towards inland areas (Balboni, 2019; Desmet et al., 2021). More frequent natural disasters might slow down growth by destructing productive capital and reducing the incentive to accumulate it (Hsiang and Jina, 2014). Given their geo- graphical characteristics and strong specialization in agriculture, it is likely that poor countries will be hit hardest by these trends, which is particularly alarming as these countries are the least prepared (in terms of fiscal space and institutional quality) to deal with such shocks. Human capital might play an important role in facilitating the reallocation of labor made necessary by climate change. The evidence from Porzio et al. (2022) discussed in Section 3.2 shows that more educated workers are better positioned to leave the agriculture sector, which is the hardest hit by global warming, and join less climate-dependent sectors.6 Moreover, highly educated individuals are generally more geographically mobile, which makes them potentially more resilient to climate shocks hitting specific regions (Burzy´ nski et al., 2021). More work is needed to establish whether the labor reallocation facilitated by human 6 Though it is important to recognize that climate-driven productivity losses are not limited to agriculture (Dell et al., 2014). 20 capital growth can be considered a quantitatively important adaptation strategy. Human capital might also contribute to the reduction of carbon emissions through technology adoption, behavioral change and the political process. As discussed in Section 3.3, high-skill individuals are crucial for both the invention and adoption of new technologies, suggesting that a more educated labor force might contribute directly to the diffusion of green innovation (Acemoglu et al., 2016). Moreover, survey evidence shows that more educated cit- izens are both more aware of climate change issues and more supportive of policies aimed to mitigate its impact (Dechezleprˆ etre et al., 2022). Using compulsory schooling laws as a source of exogenous variation, Angrist et al. (2023) show that an extra year of education is causally associated with more pro-climate beliefs, behaviors, and policy preferences. The ongoing pro- cess of educational expansion might therefore contribute to creating a stronger constituency for more ambitious climate policies. 4.2 Demographic Change The demographic composition of the labor force is changing rapidly throughout the world. Due to a combination of lower fertility and mortality rates, the average worker is becoming older in both poor and rich countries. The implementation of policies aimed at widening con- traception provision and changing cultural norms on family size have led to a particularly fast decline in fertility in developing countries (de Silva and Tenreyro, 2017), whose demographic structure is projected to converge to the one in rich countries within this century. Existing studies argue that the demographic composition of the labor force has impor- tant macroeconomic consequences. Feyrer (2007) shows that a larger proportion of workers in the 40-49 age bracket is associated with greater productivity, which Feyrer (2011) argues to be due to the higher managerial talent of middle-aged workers. Population aging might hurt economic growth by reducing startup entry and labor market dynamism (Karahan et al., 2019; Peters and Walsh, 2021), while it might increase it by promoting the adoption of labor sav- ing technologies such as robots (Acemoglu and Restrepo, 2022). Moreover, the endogenous growth literature highlights that the rate of population growth is an important determinant of the rate of discovery of growth-enhancing ideas, through a scale effect in the innovation process (Jones, 2022). Increasing utilization of human capital of older individuals can foster economic growth and productivity levels while alleviating labor shortages. Increasing labor force participa- tion among older individuals has been observed in many countries, with individuals choosing to work longer due to financial necessity, personal fulfillment, or both (Coe et al., 2012). This trend presents an opportunity to tap into the human capital of older workers, fostering eco- nomic growth and alleviating potential labor shortages in aging societies (Isele and Rogoff, 2014). Research by Ng and Feldman (2013) reveal that older workers tend to have higher job satisfaction levels compared to their younger counterparts. Additionally, mature workers 21 demonstrate higher productivity and commitment to their work, attributed to their experience, work ethic, and sense of purpose. Older adults are also engaging in entrepreneurship at higher rates, utilizing their accumulated knowledge and experience to start businesses. Studies suggest that senior entrepreneurs have a higher likelihood of entrepreneurial success compared to their younger counterparts (Zhao et al., 2021). Among others, age has a positive effect on firm size and financial success. Encouraging entrepreneurship among older individuals can contribute to job creation. Still, research suggests that the development of entrepreneurial human capital should ideally start early in life, beginning with teaching adolescents a wide variety of skills and engaging youth in multiple activities (Krieger et al., 2021). The effects of population aging are likely to be shaped by differences in the human capital accumulated by younger and older workers. For example, the evidence reviewed in Sections 3.2 and 3.3 suggesting that young cohorts are key drivers of structural change and technology adoption implies that an older labor force may be less adaptable to changes in the productive landscape. Moreover, a longer lifespan for many workers suggests a growing importance of skill accumulation over the life cycle, the pace of which (as reviewed in Section 2.4.3) is highly heterogeneous across countries. More work is needed to shed light on the optimal design of training programs and other policy initiatives aimed at boosting and updating the human capital of older workers. 4.3 Urbanization The world is urbanizing at a rapid pace. Since 2007, the number of people globally living in cities is larger than the one living in rural areas; by 2050, the urban share is projected to reach two-thirds of the global population (United Nations, 2018). Historically, urbanization has been associated to economic development and structural transformation (Michaels et al., 2012; Henderson, 2010). However, developing countries are currently urbanizing at an unprecedented pace, so that most of the world’s largest cities today are located in relatively poor countries (Jedwab and Vollrath, 2019). A large body of work documents that urban areas in present times are typically characterized by higher consumption and income (Young, 2013; Gollin et al., 2014). In light of the evidence reviewed in the previous sections, it is plausible to postulate that human capital growth has been a driver of urbanization. As discussed in Section 3.2, more urban sectors such as manufacturing and services are relatively skill-intensive compared to agriculture, and the growing availability of educated labor has likely contributed to structural transformation and the associated reallocation from rural to urban areas. Moreover, given the role of skills in promoting technology adoption discussed in section 3.3, human capital growth might have contributed indirectly to urbanization by spurring economic growth and hence in- creasing the demand for mostly urban-produced goods and services. In addition, the existing evidence suggests that cities represent favorable environments 22 for human capital accumulation. Using data from Spain, De La Roca and Puga (2017) show that workers living in bigger cities see faster growth in earnings over the life cycle, consistently with the view that they acquire more valuable experience; this conclusion is supported by the evidence in Eckert et al. (2022), which leverages the quasi-random allocation of refugees across local labor markets in Denmark. In the context of developing countries, Monge-Naranjo et al. (2018) argue that the formation of slums (a key driver of urban growth in many countries) has accelerated human capital accumulation and provided a stepping stone to low-income house- holds. At the same time, both developing and developed countries face several challenges in ensuring that cities remain environments where human capital can easily form and thrive. The living conditions and standards of urban services are still very low in the developing world, which might constitute a barrier to skill accumulation and participation in urban labor markets. Another key issue is represented by the growth in residential segregation by income, which is amplified by and contributes to differences in educational opportunities across neighborhoods (Fogli and Guerrieri, 2019). Tackling these issues is crucial for urbanization and human capital growth to continue to be as intertwined as they have been in the recent past. 5 Conclusions The research presented in this paper provides some important lessons for the role of hu- man capital for economic development. First, the accumulation of human capital is a dynamic and multifaceted process, which involves different types of investments across all stages of the life cycle, as well as the accumulation of a wide variety of skills and traits. Individuals in developing countries face significant obstacles in all these dimensions, calling for a holistic approach when designing human capital policy for economic development. Increas- ing educational attainment is only one of the many possible margins for policy action in this context, and likely not very effective if not associated by improvements in educational qual- ity and the relaxation of barriers to the accumulation of other complementary forms of human capital. Second, there is growing evidence that the way human capital is utilized in practice shapes the extent to which its accumulation is reflected in a stronger economic perfor- mance. The accumulation of different forms of human capital represents both a cause and a consequence of structural changes associated with economic development, such as the realloca- tion of labor across sectors, occupations, and firms, as well as the adoption of new technologies. Barriers to these processes, such as those often encountered in developing countries, partially undermine the potential gains associated with human capital growth by preventing an efficient utilization of talent. This underscores important complementarities between policies targeting human capital accumulation and those aiming at removing obstacles to firm growth and labor mobility. 23 Third, several mega-trends will shape the global economic environment over the next century, likely leading to important transformations in the organization of production and the allocation of resources across different economic activities. The existing evidence suggests that human capital might be an important factor mediating the effects of these trans- formations, by facilitating adaptation and allowing to seize the opportunities brought about by new productive environments. However, more research is needed to understand which specific forms of human capital are going to be more important in this context, and how policy can facilitate their accumulation and efficient utilization. 24 References Acemoglu, Daron, “Directed Technical Change,” Review of Economic Studies, October 2002, 69 (4), 781–809. and David Autor, “Skills, Tasks and Technologies: Implications for Employment and Earn- ings,” in O. Ashenfelter and D. Card, eds., Handbook of Labor Economics, Vol. 4 of Hand- book of Labor Economics, Elsevier, 2011, chapter 12, pp. 1043–1171. orn-Steffen Pischke, “Why Do Firms Train? Theory and Evidence,” The Quarterly and J¨ Journal of Economics, 1998, 113 (1), 79–119. and Pascual Restrepo, “Demographics and Automation [Automation and Demographic Change],” Review of Economic Studies, 2022, 89 (1), 1–44. , Philippe Aghion, and Fabrizio Zilibotti, “Distance to Frontier, Selection, and Economic Growth,” Journal of the European Economic Association, 03 2006, 4 (1), 37–74. , Ufuk Akcigit, Douglas Hanley, and William Kerr, “Transition to Clean Technology,” Journal of Political Economy, 2016, 124 (1), 52–104. Adao, Rodrigo, Martin Beraja, and Nitya Pandalai-Nayar, “Fast and Slow Technological Transitions,” Working Paper 2022. Almond, Douglas and Janet Currie, Human Capital Development before Age Five, Vol. 4 of Handbook of Labor Economics, Elsevier, and , “Killing Me Softly: The Fetal Origins Hypothesis,” Journal of Economic Perspec- tives, Summer 2011, 25 (3), 153–172. , Lena Edlund, and M˚ arten Palme, “Chernobyl’s Subclinical Legacy: Prenatal Exposure to Radioactive Fallout and School Outcomes in Sweden,” The Quarterly Journal of Economics, 2009, 124 (4), 1729–1772. Alon, Titan, “Earning More by Doing Less: Human Capital Specialization and the College Wage Premium,” Technical Report 2018. and Daniel Fershtman, “What to Study and When: A Dynamic Roy Model of Specializa- tion,” Technical Report 2023. Angrist, Noam, Kevin Winseck, Harry A Patrinos, and Joshua S Graff Zivin, “Human Capital and Climate Change,” Working Paper 31000, National Bureau of Economic Research March 2023. 25 , Simeon Djankov, Pinelopi K. Goldberg, and Harry A. Patrinos, “Measuring human capital using global learning data,” Nature, 2021, 592 (7854), 403–408. , , , and , “Measuring human capital using global learning data,” Nature, April 2021, 592 (7854), 403–408. Arellano-Bover, Jaime, “Career Consequences of Firm Heterogeneity for Young Workers: First Job and Firm Size,” IZA Discussion Papers 12969, Institute of Labor Economics (IZA) February 2020. Ashraf, Nava, Oriana Bandiera, Virginia Minni, and Victor Quintas-Martinez, “Gender roles and the misallocation of labour across countries,” Working Paper 2022. Attanasio, Orazio, Sarah Cattan, and Costas Meghir, “Early Childhood Development, Hu- man Capital and Poverty,” NBER Working Papers 29362, National Bureau of Economic Research, Inc October 2021. ¸ ois Fontaine, Fabien Postel-Vinay, and Jean-Marc Robin, “Tenure, Bagger, Jesper, Franc Experience, Human Capital, and Wages: A Tractable Equilibrium Search Model of Wage Dynamics,” American Economic Review, June 2014, 104 (6), 1551–96. Balboni, Clare, “In harm’s way? Infrastructure investments and the persistence of coastal cities.” PhD dissertation 05 2019. , Oriana Bandiera, Robin Burgess, Maitreesh Ghatak, and Anton Heil, “Why Do People Stay Poor?*,” The Quarterly Journal of Economics, 12 2021, 137 (2), 785–844. Banerjee, Abhijit V. and Andrew F. Newman, “Occupational Choice and the Process of Development,” Journal of Political Economy, 1993, 101 (2), 274–298. Barker, David, “The fetal and infant origins of adult disease,” British Medical Journal, 1990, 301, 111. Barofsky, Jeremy, Tobenna D. Anekwe, and Claire Chase, “Malaria eradication and eco- nomic outcomes in sub-Saharan Africa: Evidence from Uganda,” Journal of Health Eco- nomics, 2015, 44 (C), 118–136. Barro, Robert J. and Jong Wha Lee, “A New Data Set of Educational Attainment in the World, 1950-2010,” Journal of Development Economics, 2013, 104 (C), 184–198. Basu, Kaushik, “Child Labor: Cause, Consequence, and Cure, with Remarks on International Labor Standards,” Journal of Economic Literature, September 1999, 37 (3), 1083–1119. and Pham Hoang Van, “The Economics of Child Labor,” The American Economic Review, 1998, 88 (3), 412–427. 26 Bell, Alex, Raj Chetty, Xavier Jaravel, Neviana Petkova, and John Van Reenen, “Who Be- comes an Inventor in America? The Importance of Exposure to Innovation*,” The Quarterly Journal of Economics, 11 2018, 134 (2), 647–713. Bertschy, Kathrin, M. Alejandra Cattaneo, and Stefan C. Wolter, “PISA and the Transition into the Labour Market,” LABOUR, March 2009, 23 (s1), 111–137. Bhalotra, Sonia and Atheendar Venkataramani, “Is The Captain of the Men of Death Still At Play? Long-Run Impacts of Early Life Pneumonia Exposure during Sulfa Drug Revolu- tion in America,” Technical Report 2011. orklund, Anders and Kjell G. Salvanes, “Education and Family Background: Mechanisms Bj¨ and Policies,” in Erik Hanushek, Stephen Machin, and Ludger Woessmann, eds., Handbook of the Economics of Education, Vol. 3 of Handbook of the Economics of Education, Elsevier, June 2011, chapter 3, pp. 201–247. Bloom, Nicholas and John Van Reenen, “Measuring and Explaining Management Practices Across Firms and Countries,” The Quarterly Journal of Economics, 2007, 122 (4), 1351– 1408. , Benn Eifert, Aprajit Mahajan, David McKenzie, and John Roberts, “Does Manage- ment Matter? Evidence from India,” The Quarterly Journal of Economics, 2013, 128 (1), 1–51. , Renata Lemos, Raffaella Sadun, Daniela Scur, and John Reenen, “Jeea-Fbbva Lecture 2013: The New Empirical Economics Of Management,” Journal of the European Economic Association, August 2014, 12 (4), 835–876. Boca, Daniela Del, Christopher Flinn, and Matthew Wiswall, “Household Choices and Child Development,” The Review of Economic Studies, 10 2013, 81 (1), 137–185. Bold, Tessa, Deon Filmer, Gayle Martin, Ezequiel Molina, Brian Stacy, Christophe Rock- more, Jakob Svensson, and Waly Wane, “Enrollment without Learning: Teacher Effort, Knowledge, and Skill in Primary Schools in Africa,” Journal of Economic Perspectives, Fall 2017, 31 (4), 185–204. Brunner, Eric J., Shaun M. Dougherty, and Stephen L. Ross, “The Effects of Career and Technical Education: Evidence from the Connecticut Technical High School System,” The Review of Economics and Statistics, 08 2021, pp. 1–46. Buera, Francisco J., Joseph P. Kaboski, and Yongseok Shin, “Entrepreneurship and Finan- cial Frictions: A Macrodevelopment Perspective,” Annual Review of Economics, August 2015, 7 (1), 409–436. 27 ´ Burzynski, Michał, Christoph Deuster, Fr´ ed´eric Docquier, and Jaime de Melo, “Climate Change, Inequality, and Human Migration,” Journal of the European Economic Association, 12 2021, 20 (3), 1145–1197. Card, David, “The causal effect of education on earnings,” in O. Ashenfelter and D. Card, eds., Handbook of Labor Economics, Vol. 3 of Handbook of Labor Economics, Elsevier, 1999, chapter 30, pp. 1801–1863. Caselli, Francesco, “Accounting for Cross-Country Income Differences,” in Philippe Aghion and Steven Durlauf, eds., Handbook of Economic Growth, Vol. 1 of Handbook of Economic Growth, Elsevier, 2005, chapter 9, pp. 679–741. , “Technology Differences over Space and Time,” Crei Lectures 2010. , Jacopo Ponticelli, and Federico Rossi, A New Data Set on Mincerian Returns Technology Differences across Space and Time, Princeton University Press, 2016. Chari, V. V. and Hugo Hopenhayn, “Vintage Human Capital, Growth, and the Diffusion of New Technology,” Journal of Political Economy, 1991, 99 (6), 1142–1165. Chay, Kenneth Y. and Michael Greenstone, “The Impact of Air Pollution on Infant Mortality: Evidence from Geographic Variation in Pollution Shocks Induced by a Recession*,” The Quarterly Journal of Economics, 08 2003, 118 (3), 1121–1167. Ciccone, Antonio and Elias Papaioannou, “Human Capital, the Structure of Production, and Growth,” The Review of Economics and Statistics, February 2009, 91 (1), 66–82. Coe, Norma B, Hans-Martin von Gaudecker, Maarten Lindeboom, and Jurgen ¨ Maurer, “The effect of retirement on cognitive functioning,” Health economics, 2012, 21 (8), 913– 927. Coelli, Federica, Difei Oouyang, Weidi Yuan, and Yuan Zi, “Educating Like China,” 2023. mimeo, University of Zurich. Cox, Alvaro, “From Classroom to Prosperity: Fostering Development Through Higher Educa- tion,” 2023. mimeo, Yale University. e-Luis and Esteban Rossi-Hansberg, “The Economic Geography of Global Warm- Cruz, Jos´ ing,” Working Paper 28466, National Bureau of Economic Research February 2021. Cunha, Flavio, James J. Heckman, and Susanne M. Schennach, “Estimating the Tech- nology of Cognitive and Noncognitive Skill Formation,” Econometrica, 05 2010, 78 (3), 883–931. 28 Currie, Janet and Enrico Moretti, “Mother’s Education And The Intergenerational Trans- mission Of Human Capital: Evidence From College Openings,” The Quarterly Journal of Economics, November 2003, 118 (4), 1495–1532. and Tom Vogl, “Early-Life Health and Adult Circumstance in Developing Countries,” An- nual Review of Economics, May 2013, 5 (1), 1–36. , Matthew Neidell, and Johannes F. Schmieder, “Air pollution and infant health: Lessons from New Jersey,” Journal of Health Economics, 2009, 28 (3), 688–703. Daruich, Diego, “The Macroeconomic Consequences of Early Childhood Development Poli- cies,” Working Papers 2018-29, Federal Reserve Bank of St. Louis October 2018. De Philippis, Marta and Federico Rossi, “Parents, Schools and Human Capital Differences Across Countries,” Journal of the European Economic Association, 2021, 19 (2), 1364–1406. de Silva, Tiloka and Silvana Tenreyro, “Population Control Policies and Fertility Conver- gence,” Journal of Economic Perspectives, November 2017, 31 (4), 205–28. Dechezleprˆetre, Antoine, Adrien Fabre, Tobias Kruse, Bluebery Planterose, Ana Sanchez Chico, and Stefanie Stantcheva, “Fighting Climate Change: International Attitudes Toward Climate Policies,” Working Paper 30265, National Bureau of Economic Research July 2022. Dell, Melissa, Benjamin F. Jones, and Benjamin A. Olken, “What Do We Learn from the Weather? The New Climate-Economy Literature,” Journal of Economic Literature, Septem- ber 2014, 52 (3), 740–798. Desmet, Klaus, Robert E. Kopp, Scott A. Kulp, D´ an Nagy, Michael Oppen- avid Kriszti´ heimer, Esteban Rossi-Hansberg, and Benjamin H. Strauss, “Evaluating the Economic Cost of Coastal Flooding,” American Economic Journal: Macroeconomics, April 2021, 13 (2), 444–86. Dietz, Laura, Anne Burmeister, and Ulrike Fasbender, “Age and knowledge exchange: Ability, motivation, and opportunities,” in “Age and Work,” Routledge, 2022, pp. 259–276. Easterbrook, Matthew, Toon Kuppens, and Antony Manstead, “The education effect: higher educational qualifications are robustly associated with beneficial personal and socio- political outcomes,” Social Indicators Research, 2015. Eckert, Fabian, Mads Hejlesen, and Conor Walsh, “The return to big-city experience: Evi- dence from refugees in Denmark,” Journal of Urban Economics, 2022, 130, 103454. Edmonds, Eric V. and Nina Pavcnik, “Child Labor in the Global Economy,” Journal of Eco- nomic Perspectives, March 2005, 19 (1), 199–220. 29 Evans, David, Akmal Maryam, and Jakiela Pamela, “Gender gaps in education: The long view,” IZA Journal of Development and Migration, January 2021, 12 (1), 1–27. Evans, David K. and Anna Popova, “What Really Works to Improve Learning in Developing Countries? An Analysis of Divergent Findings in Systematic Reviews,” The World Bank Research Observer, 10 2016, 31 (2), 242–270. Feenstra, Robert C., Robert Inklaar, and Marcel P. Timmer, “The Next Generation of the Penn World Table,” American Economic Review, October 2015, 105 (10), 3150–3182. Feriga, Moustafa Amgad Moustafa Ahmed Moustafa, Nancy Lozano Gracia, and Pieter Maria Serneels, “The Impact of Climate Change on Work : Lessons for Developing Countries,” Policy Research Working Paper Series 10682, The World Bank January 2024. Ferrie, Joseph P., Karen Rolf, and Werner Troesken, “Cognitive disparities, lead plumbing, and water chemistry: Prior exposure to water-borne lead and intelligence test scores among World War Two U.S. Army enlistees,” Economics Human Biology, 2012, 10 (1), 98–111. Feyrer, James, “Demographics and Productivity,” The Review of Economics and Statistics, 2007, 89 (1), 100–109. , “The US productivity slowdown, the baby boom, and management quality,” Journal of Population Economics, January 2011, 24 (1), 267–284. Field, Erica, Omar Robles, and Maximo Torero, “Iodine Deficiency and Schooling Attain- ment in Tanzania,” American Economic Journal: Applied Economics, October 2009, 1 (4), 140–69. Flabbi, Luca and Roberta Gatti, “A Primer on Human Capital,” World Bank Policy Research Working Paper 8309, World Bank January 2018. Fleming, Ted, “Models of lifelong learning: An overview,” Oxford handbook of lifelong learn- ing, 2011, pp. 29–39. Fogli, Alessandra and Veronica Guerrieri, “The End of the American Dream? Inequality and Segregation in US Cities,” Working Paper 26143, National Bureau of Economic Research August 2019. Fort, Margherita, Andrea Ichino, and Giulio Zanella, “Cognitive and Noncognitive Costs of Day Care at Age 0–2 for Children in Advantaged Families,” Journal of Political Economy, 2020, 128 (1), 158–205. Foster, Andrew D. and Mark R. Rosenzweig, “Learning by Doing and Learning from Others: Human Capital and Technical Change in Agriculture,” Journal of Political Economy, 1995, 103 (6), 1176–1209. 30 Galor, Oded, “From Stagnation to Growth: Unified Growth Theory,” in Philippe Aghion and Steven Durlauf, eds., Handbook of Economic Growth, Vol. 1 of Handbook of Economic Growth, Elsevier, 2005, chapter 4, pp. 171–293. and Joseph Zeira, “Income Distribution and Macroeconomics,” The Review of Economic Studies, 1993, 60 (1), 35–52. Garcia, Jorge Luis, Frederik Bennhoff, Duncan Ermini Leaf, and James J. Heckman, “The Dynastic Benefits of Early Childhood Education,” Working Papers 2021-033, Human Capital and Economic Opportunity Working Group June 2021. ıa, Jorge Luis, James J. Heckman, Duncan Ermini Leaf, and Mar´ Garc´ e Prados, ıa Jos´ “Quantifying the Life-Cycle Benefits of an Influential Early-Childhood Program,” Journal of Political Economy, 2020, 128 (7). Garicano, Luis and Esteban Rossi-Hansberg, “Knowledge-Based Hierarchies: Using Orga- nizations to Understand the Economy,” Annual Review of Economics, August 2015, 7 (1), 1–30. Giulietti, Corrado, Michael Vlassopoulos, and Yves Zenou, “Peers, gender, and long-term depression,” European Economic Review, 2022, 144, 104084. Global Education Evidence Advisory Panel, “2023 Cost-effective Approaches to Improve Global Learning - What does Recent Evidence Tell Us are “Smart Buys” for Improving Learning in Low- and Middle-income Countries?,” Technical Report, World Bank Group 2023. Gollin, Douglas, “Nobody’s business but my own: Self-employment and small enterprise in economic development,” Journal of Monetary Economics, 2008, 55 (2), 219–233. , David Lagakos, and Michael E. Waugh, “Agricultural Productivity Differences across Countries,” American Economic Review, May 2014, 104 (5), 165–70. e Kuehn, “Human capital and the size distribution of firms,” Review of Gomes, Pedro and Zo¨ Economic Dynamics, 2017, 26, 164–179. Guner, Nezih, Andrii Parkhomenko, and Gustavo Ventura, “Managers and productivity differences,” Review of Economic Dynamics, 2018, 29, 256–282. Gust, Sarah, Eric A. Hanushek, and Ludger Woessmann, “Global universal basic skills: Current deficits and implications for world development,” Journal of Development Eco- nomics, 2024, 166, 103205. 31 Hahn, Youjin, Asadul Islam, Eleonora Patacchini, and Yves Zenou, “Friendship and Fe- male Education: Evidence from a Field Experiment in Bangladeshi Primary Schools,” The Economic Journal, 11 2019, 130 (627), 740–764. Hamory, Joan, Edward Miguel, Michael W. Walker, Michael Kremer, and Sarah J. Baird, “Twenty Year Economic Impacts of Deworming,” NBER Working Papers 27611, National Bureau of Economic Research, Inc July 2020. , Marieke Kleemans, Nicholas Y Li, and Edward Miguel, “Reevaluating Agricultural Productivity Gaps with Longitudinal Microdata,” Journal of the European Economic Asso- ciation, 11 2020, 19 (3), 1522–1555. Hanushek, Eric A., Ludger Woessmann, and Lei Zhang, “General Education, Vocational Education, and Labor-Market Outcomes over the Life-Cycle,” Technical Report 2011. Hanushek, Eric and Bradley Strauss, “A Global Perspective on US Learning Losses,” Tech- nical Report, Hoover Education Success Initiative 2024. Harvey, Jean-Franc ¸ ois, “Managing organizational memory with intergenerational knowledge transfer,” Journal of Knowledge Management, 2012. Heckman, James J. and Stefano Mosso, “The Economics of Human Development and So- cial Mobility,” NBER Working Papers 19925, National Bureau of Economic Research, Inc February 2014. Heckman, James, Rodrigo Pinto, and Peter Savelyev, “Understanding the Mechanisms through Which an Influential Early Childhood Program Boosted Adult Outcomes,” Amer- ican Economic Review, October 2013, 103 (6), 2052–2086. Henderson, J. Vernon, “Cities And Development,” Journal of Regional Science, February 2010, 50 (1), 515–540. Herrendorf, Berthold and Todd Schoellman, “Wages, Human Capital, and Structural Trans- formation,” Technical Report 2017. Hjort, Jonas, Hannes Malmberg, and Todd Schoellman, “The Missing Middle Managers: Labor Costs, Firm Structure, and Development,” NBER Working Papers 30592, National Bureau of Economic Research, Inc October 2022. Hsiang, Solomon M and Amir S Jina, “The Causal Effect of Environmental Catastrophe on Long-Run Economic Growth: Evidence From 6,700 Cyclones,” Working Paper 20352, National Bureau of Economic Research July 2014. Hsieh, Chang-Tai and Peter J. Klenow, “ The Life Cycle of Plants in India and Mexico *,” The Quarterly Journal of Economics, 05 2014, 129 (3), 1035–1084. 32 Hsieh, Chang-Tai, Erik Hurst, Charles I. Jones, and Peter J. Klenow, “The Allocation of Talent and U.S. Economic Growth,” Econometrica, September 2019, 87 (5), 1439–1474. Hurst, Erik, Yona Rubinstein, and Kazuatsu Shimizu, “Task-Based Discrimination,” Work- ing Paper 29022, National Bureau of Economic Research July 2021. Isele, Elizabeth and Edward G Rogoff, “Senior Entrepreneurship: The New Normal,” Public Policy Aging Report, 10 2014, 24 (4), 141–147. Jarosch, Gregor, Ezra Oberfield, and Esteban Rossi-Hansberg, “Learning From Cowork- ers,” Econometrica, March 2021, 89 (2), 647–676. Jedwab, Remi and Dietrich Vollrath, “The Urban Mortality Transition and Poor-Country Urbanization,” American Economic Journal: Macroeconomics, January 2019, 11 (1), 223– 75. Jedwab, Remi Camille, Paul Romer, Asif Mohammed Islam, and Roberto Samaniego, “Human Capital Accumulation at Work : Estimates for the World and Implications for De- velopment,” Policy Research Working Paper Series 9786, The World Bank September 2021. Jenkins, A., “The determinants and labour market effects of lifelong learning,” Applied Eco- nomics, 2003. Jones, Charles I., “The End of Economic Growth? Unintended Consequences of a Declining Population,” American Economic Review, November 2022, 112 (11), 3489–3527. Karahan, Fatih, Benjamin Pugsley, and Ays ¨ S ¸egul ¸ ahin, “Demographic Origins of the Startup Deficit,” NBER Working Papers 25874, National Bureau of Economic Research, Inc May 2019. Kim, Minki, “How Will a New Malaria Vaccine Shape Africa’s Economic Future? A Macroe- conomic Analysis,” Working Paper November 2022. Klenow, Peter and Andr´es Rodr´ ıguez-Clare, “The Neoclassical Revival in Growth Eco- nomics: Has It Gone Too Far?,” in “NBER Macroeconomics Annual 1997, Volume 12” NBER Chapters, National Bureau of Economic Research, May 1997, pp. 73–114. Kremer, Michael, Conner Brannen, and Rachel Glennerster, “The Challenge of Education and Learning in the Developing World,” Science, 2013, 340 (6130), 297–300. Krieger, Alexander, Michael Stuetzer, Martin Obschonka, and Katariina Salmela-Aro, “The growth of entrepreneurial human capital: Origins and development of skill variety,” Small Business Economics, 2021, pp. 1–20. 33 Lagakos, David, Benjamin Moll, Tommaso Porzio, Nancy Qian, and Todd Schoellman, “Life-Cycle Wage Growth across Countries,” Journal of Political Economy, 2017. , , , , and , “Life-Cycle Human Capital Accumulation across Countries: Lessons from US Immigrants,” Journal of Human Capital, 2018, 12 (2), 305–342. Levine, Ross and Yona Rubinstein, “Smart and Illicit: Who Becomes an Entrepreneur and Do They Earn More?*,” The Quarterly Journal of Economics, 11 2016, 132 (2), 963–1018. Lindquist, Matthew J and Yves Zenou, “Crime and networks: ten policy lessons,” Oxford Review of Economic Policy, 2019, 35 (4), 746–771. Lochner, Lance and Enrico Moretti, “The Effect of Education on Crime: Evidence from Prison Inmates, Arrests, and Self-Reports,” American Economic Review, March 2004, 94 (1), 155–189. Lucas, Adrienne M., “Malaria Eradication and Educational Attainment: Evidence from Paraguay and Sri Lanka,” American Economic Journal: Applied Economics, April 2010, 2 (2), 46–71. Ma, Xiao, Alejandro Nakab, and Daniela Vidart, “Human Capital Investment and Develop- ment: The Role of On-the-job Training,” Working papers 2021-10, University of Connecti- cut, Department of Economics May 2021. , Marc-Andreas Muendler, and Alejandro Nakab, “Learning by Exporting and Wage Pro- files: New Evidence from Brazil,” MPRA Paper 109497, University Library of Munich, Germany August 2020. Maloney, William F and Felipe Valencia Caicedo, “Engineering Growth,” Journal of the European Economic Association, 2022, 20 (4), 1554–1594. Maluccio, John A., John Hoddinott, Jere R. Behrman, Reynaldo Martorell, Agnes R. Quisumbing, and Aryeh D. Stein, “The Impact of Nutrition during Early Childhood on Education among Guatemalan Adults,” PIER Working Paper Archive 06-026, Penn Institute for Economic Research, Department of Economics, University of Pennsylvania September 2003. Martellini, Paolo, Todd Schoellman, and Jason A. Sockin, “The Global Distribution of Col- lege Graduate Quality,” Working Papers 791, Federal Reserve Bank of Minneapolis March 2022. Michaels, Guy, Ferdinand Rauch, and Stephen J. Redding, “ Urbanization and Structural Transformation *,” The Quarterly Journal of Economics, 03 2012, 127 (2), 535–586. 34 Miguel, Edward and Michael Kremer, “Worms: Identifying Impacts on Education and Health in the Presence of Treatment Externalities,” Econometrica, January 2004, 72 (1), 159–217. Milligan, Kevin, Enrico Moretti, and Philip Oreopoulos, “Does education improve citi- zenship? Evidence from the United States and the United Kingdom,” Journal of Public Economics, August 2004, 88 (9-10), 1667–1695. Mokyr, Joel, The Gifts of Athena: Historical Origins of the Knowledge Economy, Princeton University Press, 2002. Monge-Naranjo, Alexander, Pedro Cavalcanti Ferreira, and Luciene Torres de Mello Pereira, “Of Cities and Slums,” Technical Report 2018. Morchio, Iacopo, “Policies for Early Childhood Skills Formation: Accounting for Parental Choices and Noncognitive Skills,” Bristol Economics Discussion Papers 22/755, School of Economics, University of Bristol, UK January 2022. Munshi, Kaivan, “Social Networks and Migration,” Annual Review of Economics, August 2020, 12 (1), 503–524. Muralidharan, Karthik, Abhijeet Singh, and Alejandro J. Ganimian, “Disrupting Educa- tion? Experimental Evidence on Technology-Aided Instruction in India,” American Eco- nomic Review, April 2019, 109 (4), 1426–60. Nelson, Richard R. and Edmond S. Phelps, “Investment in Humans, Technological Diffusion and Economic Growth,” Cowles Foundation Discussion Papers 189, Cowles Foundation for Research in Economics, Yale University 1965. Nestour, Alexis Le, Laura Moscoviz, and Justin Sandefur, “The Long-Run Decline of Ed- ucation Quality in the Developing World,” Working Papers 608, Center for Global Develop- ment 2023. Ng, Thomas WH and Daniel C Feldman, “Age and innovation-related behavior: The joint moderating effects of supervisor undermining and proactive personality,” Journal of organi- zational behavior, 2013, 34 (5), 583–606. OECD, Education at a Glance 2022 2022. Organization, United Nations Children’s Fund International Labor, “Child Labour: Global estimates 2020, trends and the road forward,” 2020. Park, Denise C and G´ erard N Bischof, “The aging mind: neuroplasticity in response to cognitive training,” Dialogues in clinical neuroscience, 2022. 35 Peters, Michael and Conor Walsh, “Population Growth and Firm Dynamics,” Working Paper 29424, National Bureau of Economic Research October 2021. Porzio, Tommaso, Federico Rossi, and Gabriella Santangelo, “The Human Side of Struc- tural Transformation,” American Economic Review, August 2022, 112 (8), 2774–2814. Poschke, Markus, “The Firm Size Distribution across Countries and Skill-Biased Change in Entrepreneurial Technology,” American Economic Journal: Macroeconomics, July 2018, 10 (3), 1–41. Pritchett, Lant, The rebirth of education: Schooling ain’t learning, CGD Books, 2013. Roca, Jorge De La and Diego Puga, “Learning by Working in Big Cities,” The Review of Economic Studies, 2017, 84 (1 (298)), 106–142. Rossi, Federico, “Human Capital and Macroeconomic Development: A Review of the Evi- dence,” The World Bank Research Observer, 04 2020, 35 (2), 227–262. , “The Relative Efficiency of Skilled Labor across Countries: Measurement and Interpreta- tion,” American Economic Review, January 2022, 112 (1), 235–66. Roy, A. D., “Some Thoughts on the Distribution of Earnings,” Oxford Economic Papers, 1951, 3 (2), 135–146. Ryan, Paul, “The School-to-Work Transition: A Cross-National Perspective,” Journal of Eco- nomic Literature, March 2001, 39 (1), 34–92. Sacerdote, Bruce, “Peer Effects in Education: How Might They Work, How Big Are They and How Much Do We Know Thus Far?,” in Erik Hanushek, Stephen Machin, and Ludger Woessmann, eds., Handbook of the Economics of Education, Vol. 3 of Handbook of the Economics of Education, Elsevier, January 2011, chapter 4, pp. 249–277. Schoellman, Todd, “Education Quality and Development Accounting,” Review of Economic Studies, 2012, 79 (1), 388–417. , “Early Childhood Human Capital and Development,” American Economic Journal: Macroeconomics, July 2016, 8 (3), 145–174. and Bart Hobijn, “Structural Transformation by Cohort,” Technical Report 2017. Seshadri, Ananth and Nicolas Roys, “Economic Development and the Organization of Pro- duction,” Technical Report 2014. and Rodolfo Manuelli, “Human Capital and the Wealth of Nations,” American Economic Review, 2014. 36 Singh, Abhijeet, “Learning More with Every Year: School Year Productivity and International Learning Divergence,” Journal of the European Economic Association, 06 2019, 18 (4), 1770–1813. Song, Jae, David J Price, Fatih Guvenen, Nicholas Bloom, and Till von Wachter, “Firming Up Inequality,” The Quarterly Journal of Economics, 2019, 134 (1), 1–50. ander, “ Human Capital and Industrialization: Evi- Squicciarini, Mara P. and Nico Voigtl¨ dence from the Age of Enlightenment *,” The Quarterly Journal of Economics, 07 2015, 130 (4), 1825–1883. Tomaskovic-Devey, Donald, Anthony Rainey, Dustin Avent-Holt, Nina Bandelj, Istv´ an Boza, David Cort, Olivier Godechot, Gergely Hajdu, Martin H¨ allsten, Lasse Folke Hen- riksen, and Are Skeie Hermanse, “Rising between-workplace inequalities in high-income countries,” Proceedings of the National Academy of Sciences, April 2020, 117 (17), 9277– 9283. Troesken, Werner, “Lead Water Pipes and Infant Mortality at the Turn of the Twentieth Cen- tury,” The Journal of Human Resources, 2008, 43 (3), 553–575. UNESCO, Reimagining our futures together: A new social contract for education, UN, 2022. United Nations, “World Population Prospects,” Technical Report, Department of Economic and Social Affairs 2018. Verhoogen, Eric, “Firm-Level Upgrading in Developing Countries,” NBER Working Papers 29461, National Bureau of Economic Research, Inc November 2021. World Bank, LEARNING to Realize Education’s Promise, The World Bank, 2018. Young, Alwyn, “Inequality, the Urban-Rural Gap, and Migration,” The Quarterly Journal of Economics, 2013, 128 (4), 1727–1785. Zhao, Hao, Gina O’Connor, Jihong Wu, and GT Lumpkin, “Age and entrepreneurial career success: A review and a meta-analysis,” Journal of Business Venturing, 2021, 36 (1), 106007. Zimmermann, Klaus F., Costanza Biavaschi, Werner Eichhorst, Corrado Giulietti, ˜ Michael J. Kendzia, Alexander Muravyev, Janneke Pieters, NAºria ˜ RodrAguez-Planas, and Ricarda Schmidl, “Youth Unemployment and Vocational Training,” Foundations and Trends(R) in Microeconomics, December 2013, 9 (1–2), 1–157. 37