The standard Medicare Part D drug insurance contract is non-linear—with reduced subsidies in a coverage gap—resulting in a dynamic purchase problem. We consider enrolees who arrived near the gap early in the year and show that they should expect to enter the gap with high probability, implying that, under a benchmark model with neoclassical preferences, the gap should impact them very little. We find that these enrolees have flat spending in a period before the doughnut hole and a large spending drop in the gap, providing evidence against the benchmark model. We structurally estimate behavioural dynamic drug purchase models and find that a price salience model where enrolees do not incorporate future prices into their decision-making at all fits the data best. For a nationally representative sample, full price salience would decrease enrolee spending by 31%. Entirely eliminating the gap would increase insurer spending 27%, compared to 7% for generic-drug-only gap coverage.

Prior researchers have deployed the Vietnam-era draft lottery as an instrument to estimate causal effects of military service on health and earnings. However, household and residential outcomes may be more sensitive to the psychological effects of military service. Using 2SLS analyses of the 2000 Census and the 2005 American Community Survey, we find mixed results for residential stability, housing tenure, and extended family residence. While in the ACS white veterans are less mobile, veteran status has no effect on homeownership. Veteran status reduces extended family living for whites in the Census but increases it for ACS veterans of “other” races.