Jere R. Behrman on the Impacts of Community Infrastructure on Children’s Learning

Jere R. Behrman is the William R. Kenan, Jr. Professor of Economics in the Department of Economics and Professor of Sociology in the Department of Sociology at the University of Pennsylvania, a Penn IUR Faculty Fellow, and a Research Associate at Penn’s Population Studies Center (PSC), Population Aging Research Center (PARC), and the Penn Development Research Initiative (PDRI).

Jere R. Behrman is a globally renowned economist and prolific scholar with over 530 articles published in leading journals, spanning a multitude of disciplines including empirical microeconomics, labor economics, and economic demography. A unifying theme in his work is improving empirical knowledge about human capital formation and outcomes, especially considering unobserved factors such as innate ability and health and institutional imperfections including incomplete markets and information asymmetries. Behrman’s research sheds light on relevant insights for policy outcomes in developing nations worldwide. The benefactors of his research contributions include the World Bank, Asian Development Bank, Inter-American Development Bank, United Nations Development Program, and UNICEF, among other leading international development organizations and government bodies.

In a recent publication in Economics of Education Review, “Floods, community infrastructure, and children’s heterogeneous learning losses in rural India,” Behrman, together with co-authors Nazar Khalid (lead author and Ph.D student in Demography and Sociology), Emily Hannum (Stanley I. Sheerr Term Professor in the Social Sciences and Professor in Sociology), and Amrit Thapa (Senior Lecturer in GSE), examines the role of community infrastructure in alleviating children’s learning losses as a result of flood events in rural India.

Behrman and his colleague’s novel research parses the effects of both physical infrastructure—proximity to towns, presence of flood-resistant roads, availability of schools and medical facilities, and access to electricity, potable water, and sanitary facilities—and social infrastructure—village size, communal living arrangements, social organizations, prevalence of crime and conflict, education quality, and teacher characteristics. The paper also considers the potential for disparate impacts to children of marginalized caste/religious groups, gender, and socioeconomic backgrounds.

Reflective of Behrman’s contributions to the human capital literature, the paper is impactful in its nuanced approach to analyzing the numerous variables that interact to influence the heterogeneous effects of flood events on educational outcomes. The lessons learned from the case of rural India have practical and scalable applications for developing nations worldwide, reinforcing the importance of geographic, social, and economic context to inform critical human development outcomes in education and beyond.

Excerpts from “Floods, community infrastructure, and children’s heterogeneous learning losses in rural India”

In rural India, flood exposures are associated with poorer educational outcomes for children, and children of certain marginalized caste/religious, gender, and socioeconomic backgrounds are more likely than others to experience floods (Khalid et al., 2024)... [We] go beyond the previous literature by considering whether facets of physical and social infrastructures are associated with reduced learning losses for children in flood-exposed communities, and whether children from marginalized backgrounds have more, equal, or less learning associated with infrastructure in the absence of and in the presence of floods.

(1) Do children in villages with better infrastructure have smaller learning losses when exposed to floods?

We find that floods are associated with less learning, but villages closer to towns experience significantly smaller learning losses during floods compared to more remote villages. We find no evidence that other components of the physical infrastructure are significantly associated, on average, with different learning losses during floods.

We also find that smaller learning losses are associated with a number of community social infrastructure measures including higher safety from crime, better average baseline student performance on standardized tests …, higher proportions of teachers perceived as fair and unbiased by households, and higher rates of parent-teacher interactions about children’s educational progress. These findings speak to the emerging body of literature that documents the positive roles played by strong community social capital, especially during challenging times posed by natural disasters.

(2) Are components of infrastructure associated with different learning outcomes for children from more-marginalized versus other groups in the absence of floods?

[I]n the absence of floods, some infrastructure components are associated with lower learning disparities between marginalized and other children and some components are associated with higher learning disparities between marginalized and other children…

Among these, village electrification is associated with greater learning disparities for marginalized children along socioeconomic lines, but lower disparities for girls versus boys… On the other hand, most social infrastructure components with significant associations, including the presence of more social organizations, conflict-free communities, and communal problem-solving, are associated with lower learning disparities between marginalized and more-privileged children, particularly those whose parents have no schooling. However, there are exceptions where certain social infrastructure components are associated with greater disparities, such as safety against crime for girls versus boys… Impacts of some infrastructure components vary depending on the dimension of marginalization.

(3) Are components of infrastructure associated with smaller learning loss gaps due to floods between children from more- and less-marginalized groups?

[S]ome physical and social infrastructure components have different significant associations for children depending on their background… The majority of the infrastructure components are not associated with significant learning divergences between marginalized and other children in the presence of floods… Nevertheless, a number of the infrastructure components are associated with significant learning divergences and convergences between marginalized and other children in the presence of floods… Among these, piped drinking water stands out as the only physical infrastructure component that significantly contributes to the convergence of learning disparities between children of poorer households and those of richer households during floods… In contrast, among the significant social infrastructure-marginalization interactions, the majority are associated with higher learning losses for marginalized children versus other children in the presence of floods.

[In] stratified societies like India, the relationships between infrastructure and flood-related learning losses are complex. [Our] varied findings underscore the need for a nuanced approach to policy interventions that are rooted in local contexts. Such policies must consider the unique challenges faced by different marginalized groups and how various aspects of community infrastructure—both physical and social—interact with these challenges to affect educational resilience.