Essays on Human Capital Investment

Given the large returns to education for the labor market and for other indicators of wellbeing, understanding the factors that determine educational attainment are an important concern of policymakers. In this thesis, I explore the contribution of two such factors: 1) the availability of career and technical education (CTE) in high school and 2) the strength of the local labor market in college. In Chapter 1 of the thesis, I test the hypothesis that offering CTE courses encourages students to complete high school. I use data on public high school students from the North Carolina Education Research Data Center that allow me exploit within-school variation in the availability of CTE courses to obtain the causal effect of CTE availability on graduation. My study finds that a 10 percentage point increase in the proportion of a school’s course offerings that are CTE (one standard deviation is 8 percentage points) raises the probability a student graduates by 0.63 percentage points. The average dropout rate in the data is about 10 percent, so such an increase reduces the dropout population by about 6 percent. I also find notable differences in the strength of the graduation response across student characteristics. The effect of increasing CTE supply is particularly strong for Hispanics, those with above average 8th grade math and reading test scores, and those with below average time spent on homework in 8th grade. The effect is also stronger when the local economy is stronger, suggesting that CTE helps keep people in school who would otherwise drop out and work. In Chapter 2 of the thesis, I construct a local labor demand index that provides new causal evidence that college enrollment goes up in response to negative local labor demand shocks and down in response to positive ones. Unlike other proxies of local labor demand, such as the unemployment rate or wages, the index is exogenous to local labor supply shocks that may be correlated with other factors that determine the decision to enroll. In the end, I find little difference between estimates using demand proxies vulnerable to labor supply shocks and the demand index I construct. I also look for gender heterogeneity in the enrollment response to demand shocks, and find that men are more sensitive than women are. This indicates that the positive gap between women and men in enrollment grows during booms and shrinks during busts. I explore the possibility that the gender heterogeneity is the result of gender differences in labor demand, with mixed results.