Joseph G. Altonji, Thomas Dewitt Cuyler Professor of Economics, Yale University.
We extend the control function approach introduced by Altonji and Mansfield (2016) for bounding the variance of group treatment effects by 1) introducing complementarities between individual and group-level inputs and 2) allowing for multiple group levels.
We implement the augmented control function approach using a mixed effects specification with data from two panel surveys, and use the results to characterize the importance of neighborhood, school, and commuting zone inputs during childhood for determining student long-run educational attainment and wages as well as the degree to which certain types of students are differentially sensitive to these inputs.
We find that experiencing a school/location combination at the 90th versus 10th percentile of the school/location treatment effect distribution increases the high school graduation probability and college enrollment probability for a randomly selected student by at least .05 and .18, respectively, with dropout rates for disadvantaged students exhibiting particular sensitivity to school/location inputs.
Co-authored with Mohit Agrawal (Yale University) and Richard K. Mansfield (University of Colorado-Boulder and NBER).
About the speaker
Joseph G. Altonji is the Thomas DeWitt Cuyler Professor of Economics at Yale University. He previously held faculty positions at Columbia and Northwestern and has been a visiting professor at Princeton and Harvard (2011-2012).
Altonji specializes in labor economics and applied econometrics. His interests include labor market fluctuations, labor supply, consumption behavior, the economics of education, economic links among family members, race and gender in the labor market, wage determination, and econometric methods.
His current research focuses on the role of families and schools in inequality, school choice programs, the link between siblings in substance use, dynamic models of earnings and family income, the effects of high school curriculum and college major on economic outcomes, the effects of labor market conditions on recent college graduates, and the use selection on observed variables to address selection on unobserved variables.
Altonji has served as a consultant to the Federal Reserve Bank of Chicago and on a number of advisory panels, including the NAS/NRC Committee on National Statistics, the NAS/NRC Panel on Measuring Discrimination, the Federal Economic Statistics Advisory Committee, and the President’s Council of Advisors on Science and Technology STEM Undergraduate Education Working Group.
He is an elected fellow of the Econometric Society, the Society of Labor Economists, and the American Academy of Arts and Sciences. Altonji received BA and MA degrees from Yale in 1975 and a PhD from Princeton in 1981, all in economics.