Personality, IQ, and Lifetime Earnings
Talented individuals are seen as drivers of long-term growth, but what makes them realize their full potential? In this paper, I show that lifetime earnings of high-IQ men and women are substantially influenced by their personality traits, in addition to intelligence. Personality traits directly affect men's earnings, with effects only developing fully after age 30 and increasing in education. Personality and IQ also influence earnings indirectly through education, which has sizeable positive rates of return for men in this sample. Women's returns to education past a bachelor's degree are lowered through worse marriage prospects, which offset gains to education in terms of own earnings. The causal effect of education is identified through matching on detailed background information. This paper complements the well-established results regarding the role of education and personality traits in explaining life outcomes of disadvantaged children by demonstrating that they also account for considerable variation in lifetime productivity at the opposite end of the ability distribution.
The Web Appendix is hosted on this website as well.

Skills, Occupational Choice, and Educational Attainment - Work in Progress
I propose a model of educational choice in which the formation of wage expectations accounts for anticipated occupational choices as well as education levels. Occupations are characterized as bundles of tasks that are explicitly linked to multi-dimensional abilities and personality traits. These traits are measured in Project Talent data that I augment with information from the Dictionary of Occupational Titles. The importance of skills is then contrasted with the influence of parental background and credit constraints in this mutually dependent decision.

Reanalyzing the Effects of Chicago Child-Parent Centers with a Focus on Neighborhood Quality - Work in Progress with Maryclare Griffin and James J. Heckman
We supplement a well-known non-randomized cohort study of the Child Parent Center (CPC) programs, an intensive preschool, kindergarten, and follow-on program for students in poor neighborhoods of Chicago, with neighborhood quality data (Census and police records). We find that the generally positive associations between program participation and students' long-term outcomes can greatly depend on where these students were born and grew up. For example, CPC has a significantly positive effect on educational attainment, SES, and income of males but only in less disadvantaged neighborhoods. Contrary to expectations, CPC participation does not differentially decrease criminal behavior by neighborhood characteristics, including homicide rates.