Essays on Education, Policy, and Household FormationPublic Deposited
This thesis is organized into three essays, each with particular insights involving public policy and family formation: In the ﬁrst essay, I address the well-documented correlation between the prestige of the university and the labor market income of its graduates by investigating whether a possible medium for this eﬀect is the choice of a major ﬁeld of study. Using data from two Texas universities, I predict students' choice of major from their high school qualiﬁcations and demographic characteristics. Because of Texas' implementation of a Ten Percent Law-which guarantees admission to the more elite school for those students in the top decile of their High School graduating class-I am able to implement a regression discontinuity design. I ﬁnd that the more prestigious institution makes students 22% more likely to major in Business than in Education, Nursing, or Social work. Results are conﬁrmed through multinomial logit analysis. In the second essay, I examine how increases in women's education aﬀect marriage rates of men. I use State-Level Broad-Based Merit Aid Programs as an instrument to increase the education levels of cohorts of women. Because men often marry younger women, there are a number of men who are too old to have been exposed to the scholarship programs, but whose prospective spouses would have been. I ﬁnd that men out of this group with college degrees who receive the treatment see reductions in marriage rates by as much as 4% when compared with similar men in control states. This gives evidence that the gains-from-trade eﬀect on marital surplus that is caused by wage diﬀerential is of stronger magnitude than that of matching of traits. In the ﬁnal essay, my co-author and I estimate the impact of the introduction of no fault divorce laws on divorce rates. In order to deal with the problems of law endogeneity and unobserved state heterogeneity, we employ a new method in which we use only data from counties that border adjacent states. We ﬁnd no evidence of any long-term increase in divorce associated with the passage of these laws. This means that the Coase Theorem-predicting the ability of parties to compensate each other within contracts in response to law changes-holds in equilibrium.