| In this dissertation I study economic decision-making in three contexts---retirement, prescription drug insurance take-up, and college enrollment.;In the first chapter, I examine the relationship between early retiree health benefits and early retirement. Although there is a positive association between the availability of retiree health benefits and early retirement, this association could be driven by other factors. I show that individuals in poor health and with poor outside insurance options value retiree health benefits more and then use variation along these dimensions to examine whether the estimated correlation between retiree health benefits and early retirement reflects demand for health insurance. The effect of retiree health benefits is not statistically significantly larger for those in poor health, but it is larger for those who lack insurance from other sources, particularly from their spouses. I conclude that retiree health benefits do increase the early retirement hazard and that health insurance demand among the near-elderly is not closely tied to health.;Second, I study the market for prescription drug insurance supplements to Medicare (Medigap). Regulations prohibit insurers from using information on customer health to set prices or deny coverage, creating ideal conditions for adverse selection. I test for the presence of adverse selection in this market using the Health and Retirement Study (FIRS). Controlling for a range of demographic and economic characteristics, those in worse health at age 64, before the Medigap purchase decision, are more likely to purchase insurance at age 66. The estimates are large, with the probability of coverage 40% greater for those reporting fair to poor health and 28% greater for those with an additional ;In the final chapter, Harvey Rosen, Cecilia Rouse and I study the effects of a change in financial aid policy introduced by a Northeastern university in 1998. Previously, the university's financial aid packages consisted of grants, loans, and jobs. After the change, all loans for low-income students were replaced with grants. We find the program had a statistically insignificant positive effect on the likelihood of matriculation by low-income students, with a larger, marginally significant effect among minorities. |
