Labor Earnings and Community College Enrollment Since COVID: Evidence from California

(with Brian Johnson, Christopher Avery, and Elise Swanson)


Community college enrollments have fallen nearly 20 percent since 2019. In this paper, we examine the extent to which earnings growth has depressed enrollments in the California Community Colleges system, the largest network of community colleges in the U.S. We instrument for county labor earnings growth using a Bartik instrument approach that allows us to estimate labor earnings elasticities of enrollment by county for individuals over age 25 with a high school education. We estimate a pre-COVID elasticity of -0.99, and we find modest evidence that enrollment has become less sensitive to labor earnings since 2019. Enrollment declines are far greater than we would predict using observed earnings growth and estimated elasticities: Of the 18 percent decline in enrollments from Fall 2019 to Fall 2022, we attribute just 8 to 10 percent to earnings growth.

Economics of Foster Care

(with Joseph Doyle, Max Gross, and Brian Jacob)

Journal of Economic Perspectives, Vol. 36, No. 2, pp. 223-246. 2022. Link.


Foster care provides substitute living arrangements to protect maltreated children. The practice is remarkably common: it is estimated that 5 percent of children in the United States are placed in foster care at some point during childhood. These children exhibit poor outcomes as children and adults, and economists have begun to estimate the causal relationship between foster care and life outcomes. This paper provides background on the latest trends in foster care policy and practice to highlight areas most in need of rigorous evidence. These trends include efforts to prevent foster care on the demand side and to improve foster home recruitment on the supply side. With increasing data availability and a growing interest in evidence-based practices, there are a range of opportunities for economic research to inform policies that protect vulnerable children.

The Causal Impact of Removing Children from Abusive and Neglectful Homes

(with Eric Chyn, Justine Hastings, and Margarita Machelett)

Journal of Political Economy, Vol. 130, No. 7, pp. 1919-1962. 2022. Link.


This paper measures impacts of removing children from families investigated for abuse or neglect. We use removal tendencies of child protection investigators as an instrument. We focus on young children investigated before age 6 and find that removal significantly increases test scores and reduces grade repetition for girls. There are no detectable impacts for boys. This pattern of results does not appear to be driven by heterogeneity in pre-removal characteristics, foster placements, or the type of schools attended after removal. The results are consistent with the hypothesis that development of abused and neglected girls is more responsive to home removal.