Research

This paper presents new evidence on the impact of a need-based financial aid program on first-year undergraduate students' major choice. Using rich administrative data, we exploit sharp discontinuities in grant eligibility in Canada to identify the causal effect of eligibility for low-income and middle-income students separately. Our findings show that grant eligibility leads to more grants and fewer loans, as well as a higher likelihood of majoring in the humanities rather than business among low-income students. Students from middle-income families are more likely to pursue health-related majors when they are eligible for more grants instead of loans. However, we do not observe significant differences in major-specific degree completion rates within six years.

Increasing borrowing for post-secondary education, along with high student loan default rates, increasing labor market risk, and the fact that some youth need parental assistance in repaying student loans, indicates that youth may be inadequately insured through the current student loan repayment scheme. Some suggested that student loan repayment schemes should provide more insurance against labor market risk through more accessible income-based repayment plans. However, there is concern that the provision of more insurance for student loan repayment could lead to large revenue losses for the student loan program. To make the program self-sustaining, interest rates may need to increase, raising the costs of post-secondary education for many who are currently paying their loans in full. This, in turn, may discourage some youth from attending college altogether. To study the effects of student loan repayment policy when considering labor market risks and the insurance provided by parents, this paper develops a dynamic life-cycle model with endogenous parents-to-children transfers, together with children's education, borrowing, repayment, and labor supply decisions. After estimating the model using the National Longitudinal Survey of Youth 1997, I quantify the impacts of introducing an income-based repayment (IBR) plan while keeping the government budget constant. IBR crowds out savings and parental transfers as it provides more insurance to borrowers. Labor supply response to IBR is very small so moral hazard is not a concern. The college enrollment rate increases, and it increases the most for low income and low ability families. Aggregate welfare increases with relatively low income families benefiting the most.

This paper shows that repeated cross-section data with multiple skill measures (one continuous and repeated) available each period are sufficient to nonparametrically identify the evolution of skill returns and cross-sectional skill distributions.  With panel data and the same available measurements, the dynamics of skills can also be identified.  Our identification strategy motivates a multi-step nonparametric estimation strategy. We further show that if any continuous repeated measurement is shown to be linear in skills, a much simpler GMM estimator can be used. 

Using Health and Retirement Study (HRS) data on men ages 52+ from 1996-2016, we show that one of the available (continuous and repeated) skill measures is linear in skills and implement our GMM estimation approach.  Our estimates suggest that the returns to skill were fairly stable from the mid-1990s to the Great Recession, rising thereafter.  We document considerable differences in skills and lifecyle skill profiles over ages 52--70 across cohorts, with more recent cohorts possessing lower skills in their mid-50s but experiencing much weaker skill declines with age.  We also document skill differences by education and race, which are stable across ages and explain roughly one-third and one-half, respectively, of the corresponding differences in wages.  We observe substantial differences in skills for men in their mid-50s choosing to retire at different ages, but no clear evidence of sharp declines in skills surrounding retirement ages.  Finally, we show that individual fixed effects account for more than a third of all skill variation at age 60, with considerable persistence in year-to-year skill innovations.