UC San Diego

This dissertation explores questions in financial aid, student loans, and homework time use. In chapter one, I use quarter of birth instruments to estimate the relationship between calculated financial need and actual financial aid received. A dollar increase in financial need is associated with 15 cents in federal grants and 47 cents in loans. However, the relationship between financial need and institutional aid is, on average, small and statistically insignificant. In chapter two, I use instrumental variables to estimate the effect of student loans on fertility later in life. First, I exploit a FAFSA loan-eligibility rule along with state-level variation in the school-entrance age to instrument for differences in loan availability. Next, I exploit state-by-SAT-score variation in tuition at public universities to instrument for differences in loan demand. I find that an increase in undergraduate loans, due to an increase in loan availability, does not significantly affect fertility. However, an increase in loans, due to higher tuition, decreases the number of children the student has 10 years after graduation. Back of the envelope calculations suggest that doubling tuition at in- state public colleges will lead to a 14 percent decrease in total fertility 10 years after graduation. In chapter three, we use the American Time Use Survey (ATUS) to document that Asian- American high-school students, on average, spend 2.3 times more time on homework compared to white students and 3.9 times more time on homework compared to black students. Our estimates suggest that the Asian-white homework gap is approximately 7.4 hours per week. Contrary to certain stereotypes, Asian-American students do not spend less time on leisure or sleep compared to white students. Rather, Asian-American students spending less time working for pay, doing household chores, and playing sports account for the entire Asian-white homework gap. We hypothesize that the differences in time use across ethnic groups may be related to the differences in educational outcomes across ethnic group