Plan selection in Medicare Part D: Evidence from administrative data

We study the Medicare Part D prescription drug insurance program as a bellwether for designs of private, non-mandatory health insurance markets that control adverse selection and assure adequate access and coverage. The value that Part D or other prescription drug coverage has for an individual crucially depends on her current level and future development of health conditions, prescription drug needs, and life expectancy. We use administrative data on medical claims in Medicare Part D. These data provide us with detailed and reliable information on health conditions, prescription drug use, as well as information on the plan an individual has chosen and out-of-pocket costs. We can also construct the set of plans that were available to an individual when she made her decision, along with all their characteristics, including formularies and co-payment tiers. The data are available for a 20% random sample of the entire Medicare population, so that sample selection is not an issue and rare conditions, which often require expensive treat-ment, can be studied. The focus of this paper is on the ability of consumers to evaluate and opti-mize their choices of plans, an essential ingredient if adequately regulated private health insurance markets are to be successful in allocating health resources efficiently.