Mortality gradients by education and income have been rising in the United States and elsewhere. However, their impact on Social Security progressivity has received relatively little attention, and the impact on Medicare has received effectively none. This paper uses the Future Elderly Model to estimate the effects of increased mortality gaps on the progressivity of Social Security and Medicare for those born between 1928 and 1990. It finds significant reductions in progressivity of both programs if current mortality trends persist and noticeable effects on total program costs. The effects are large enough to warrant more attention from both policy-makers and researchers.

Medicare Part D relies upon drug plan competition. Plans have enormous scope to design benefits and to set premiums, but they may not charge differential premiums based on risk. We use the formulary and benefit design of all Medicare prescription drug plans and pharmacy claims data to construct a simulation model of out-of-pocket drug spending. We use this simulation model to examine individual incentives in Medicare Part D for adverse selection. We find that high drug users have much stronger incentives to enroll in generous plans than do low users, thus there is significant scope for adverse selection.

Restricted government spending along with universal health insurance has led to longer queues for surgical procedures in Canada versus the United States. Yet it is unclear whether these treatment delays affect health outcomes. This paper tests this hypothesis by comparing the determinants of wait time for hip-fracture surgery and its impact on postsurgery length of stay and inpatient mortality in Canada and the United States. Hazards for surgery/no surgery and discharge alive versus dead are modeled using a competing-risks model. Day of the week of admission is used to help identify the surgery wait-time distribution. We control for unobserved (to the econometrician) health status which may affect wait times and outcomes by assuming a semiparametric distribution for unobserved heterogeneity. We find that predicted hazards for inpatient mortality are virtually identical in Canada and the United States. Yet wait times for surgery are longer in Canada, and surgery delay has a significant impact on postsurgery length of stay in both countries. However, the magnitude of this effect is small relative to other patient and hospital-specific factors. Focusing attention on treatment delays as a weakness in the Canadian health care system may be misleading policymakers from hospital-specific inefficiencies that may have more-important implications for health care costs and patient welfare.