Working Papers

"Human Capital Investment and College Sorting" - available upon request

Abstract: College is traditionally viewed as one of the most important ways to promote upward mobility. However, there is a substantial over-representation of high-income families at the top of the college distribution. Difference in pre-college parental investment in human capital contributes to uneven access to the selective colleges. In this paper, I study how this investment affects sorting of students into colleges. I then estimate the efficiency of the decentralized allocation and explore the implications of pre-college investment for intergenerational mobility. To quantify the size of output losses due to the mismatch between students and colleges, I embed a student-to-college assignment model into a two-period overlapping generations model with endogenous human capital investment. Households compete for a fixed number of seats at the top-ranked colleges, and parental investment in their child’s human capital promotes access to them. After controlling for human capital, I find that income and the parent’s willingness to pay for a college education are major determinants of enrollment in highly selective colleges. Average human capital of enrolled students affects the quality of a college which partially mitigates the effect of income in the college admission process. The peer effects give rise to the tension between sorting students on willingness to pay and on human capital. I calibrate the model to NLSY97 cohort and find that the race to the top induces overinvestment in pre-college human capital and associated output losses relative to the first best. The effect is more pronounced for high-income families which promotes income persistence at the top of the college distribution. 

"The Allocation of Teaching Talent and Human Capital Accumulation" (with Simeon Alder and Ananth Seshadri) - available upon request

Abstract: The educational landscape in the U.S. has gone through major changes since the end of World War II. Real expenditure per student has risen from approximately $2,100 to more than $10,000 by the turn of the century and more than $12,000 since the Great Recession. At the same time, the student-teacher ratio has fallen from a national average of almost 27 in 1955 to 16 by the 2010s. Yet despite the rise in expenditures and the reduction in class sizes, educational outcomes in the U.S. don’t compare very favorably with countries at similar levels of the income distribution. One aspect of U.S. education that has not garnered a lot of attention until fairly recently is occupational choice. We add an education sector to an otherwise standard Hsieh et al. (2018)-style model to explore the extent to which changes in career opportunities in other occupations affect the selection of workers into teaching careers. In our model, changes in the allocation of teaching talent have implications for the evolution of class size as well as quality of instruction and hence the accumulation of human capital during the workers’ formative years. This gives rise to a trade-off between static and dynamic efficiency, which we quantify by way of a structural model. In order to discipline the parameterization of the model, we compare model-generated moments to their empirical counterparts in three longitudinal surveys: Project TALENT, the NLSY79, and NLSY97.

"The Effects of State Pregnancy Accommodation Laws on Infant Health" (with Jessica Pac and Alejandra Ros Pilarz) - available upon request

Abstract: Prior to the 2022 passing of the Pregnant Workers Fairness Act implemented in June of 2023, the absence of a federal law requiring employers to provide reasonable workplace accommodations to pregnant women led 30 states and D.C. to pass state-level laws beginning in 1990. In this paper, we evaluate the rollout of state pregnancy accommodation laws on fertility, infant mortality, and infant health at birth between 1999 and 2018 using a two-way fixed effects triple differences framework. Our estimates suggest that pregnancy accommodation laws are associated with a 10.81 percent reduction in the incidence of small-for-gestational age births (at or below the 10th percentile weight-for-age), a 9.43 percent reduction in low birthweight births, and a modest 0.24 percent increase in the rate of infants born with a ‘healthy’ Apgar score greater than 6. We attribute these effects to improved intrauterine growth vis-à-vis maternal employment continuity, earlier attainment of prenatal care, reduced rates of maternal smoking and longer interpregnancy intervals. We detect substantially larger, positive improvements in all measures of infant health and mortality in states with paid family and medical leave or temporary disability insurance policies. These findings provide critical insight into the potential impact of the federal Pregnant Workers Fairness Act in states without a state pregnancy accommodation law.

Work in Progress

"The Effects of State Pregnancy Accommodation Laws on Maternal Health" (with Jessica Pac and Alejandra Ros Pilarz)

Abstract: Pregnancy accommodation laws enable women to work under safe conditions throughout their pregnancy by requiring employers to provide reasonable accommodations, such as temporarily assigning pregnant workers to less physically demanding positions. Until the recently passed federal Pregnant Workers Fairness Act, which requires employers to provide reasonable accommodations to pregnant workers, goes into effect in June 2023, pregnant workers’ access to accommodations varies across states. As of January 2023, 31 states have independently passed pregnancy accommodation laws. Using data from the 2004 – 2020 Pregnancy Risk Assessment Monitoring System (PRAMS), we evaluate the causal impact of state pregnancy accommodation laws on maternal health using a staggered difference-in-differences (two-way fixed effects) approach. Specifically, we examine pregnancy and postpartum outcomes—including type 2 diabetes, hypertension, preterm labor, and depression—as well as health investments and behaviors—such as healthcare visit quality and quantity, smoking, and drinking. As historically marginalized women are more likely to work low-paid jobs that necessitate accommodations, in addition to estimating the average effects of the laws, we consider heterogeneous effects by mothers' race and ethnicity, education, and family structure. 

"Active Labor Market Policies and Multidimensional Skill Mismatch" (with Nisha Chikhale)

Abstract: This project investigates how the allocation of firms and workers with multidimensional skills changes over the business cycle, and explores the role for active labor market policies (ALMP) in improving welfare through allocation (efficiency) or the distribution of wages (equity). To answer these questions,  we extend Baley et al. (2019) to allow for the possibility of learning about abilities during unemployment via participating in a training program. We parameterize the model using NLSY79 and NLSY97. Then, using the calibrated model, we develop the optimal ALMP over the business cycle.