Abstract
Medicare currently allows beneficiaries to choose between traditional Medicare and privately run plans in the Medicare Advantage (MA) program. Because enrollment in MA is optional, conventional observational estimates of the program’s impact are potentially subject to selection bias. To address this issue, we use a discontinuity in the rules governing MA payments to health plans that gives greater payments to plans operating in counties in metropolitan statistical areas with populations of 250,000 or more. The sharp difference in payment rates at this population cutoff creates a greater incentive for plans to increase the generosity of benefits and therefore enroll more beneficiaries in MA in counties just above versus just below the cutoff. We find that the expansion of MA on this margin reduces beneficiaries’ hospital use and mortality.