Research
Research
JOB MARKET PAPER: Reducing Medicare Costs by Accounting for Selection
Abstract: The U.S. Medicare Advantage (MA) program was designed to harness competition among private insurers and the cost-containment potential of managed care to reduce spending and improve care for beneficiaries. However, imperfect risk adjustment—where payments are linked to beneficiaries’ health—fails to fully account for the selection of healthier individuals into Medicare Advantage. This generates a trade-off between higher costs from selection into MA and lower costs from managed-care efficiency. Using detailed administrative claims data combined with additional survey measures of health, I develop a structural model that jointly captures individuals’ plan choices and insurers’ equilibrium price setting. The model links these demand and supply behaviors to quantify how their interaction affects government expenditure. Counterfactual simulations show that eliminating MA would reduce government spending by 20.8 billion annually. Adjusting for selection beyond current risk adjustment into the subsidy design would realign the MA program with its cost-saving intent, reducing CMS expenditure by 7.56 billion while resulting in a combined loss of 0.585 billion in consumer surplus and producer profits.
The Impact of Hospital-Sponsored Health Plans on Readmission Rates (with Alberto Cappello)
Abstract: Hospital systems are increasingly vertically integrated (VI) with healthcare financing by sponsoring their own health plans. This paper studies whether admission to a (VI) hospital is associated with different 30-day unplanned readmission rate compared to standard non-VI hospitals. To address patient selection into hospitals, we develop an instrumental-variables design that leverages quasi-experimental variation in hospital assignment generated by (i) the ambulance transportation in emergency care episodes and (ii) the patient’s relative distance to VI versus non-VI hospitals. We address unobserved hospital quality that may influence the decision to become a vertically integrated entity by including hospital fixed effects and year fixed effects, which allow us to exploit variation in hospitals’ vertical integration status over time. We implement this identification strategy using administrative Medicare Advantage (MA) enrollment and claims data. We find that in less concentrated hospital markets the readmission probability at VI hospitals is about 3.4 percentage points lower than at other hospitals, and that the estimated treatment effect is smaller in markets with higher observed hospital concentration.
Spillover Effects between Medicare Programs (with Alberto Cappello, and Yunus Coskun)
Abstract: Medicare’s major coverage programs share providers, creating scope for spillovers across segments. We study whether incentives in the Medicare Shared Savings Program (MSSP) spill over from Traditional Medicare to Medicare Advantage (MA). The MSSP can affect MA through two channels. First, benchmark channel: because MA benchmarks are tied to local fee-for-service spending, MSSP-induced spending reductions mechanically lower MA plan payments. Second, provider channel: because providers treat both Traditional Medicare and MA beneficiaries, MSSP-induced changes in care delivery may spill over directly to MA enrollees. We focus on the provider channel. To study it, we condition on plan-level MA benchmarks and exploit quasi-experimental variation generated by the 2019 Pathways to Success reform, which changed the timing of non-rebasing years by MSSP entry cohort and thereby altered the strength of savings incentives. Combining this policy variation with pre-reform county ACO shares and baseline plan enrollment weights, we construct a Bartik-style instrument for plan-level exposure to MSSP activity. We find that greater MSSP exposure lowers MA plans’ expected cost of coverage and increases plan generosity. A 1% increase in exposure reduces expected cost per beneficiary by 0.17%, increases rebates by 2.1%, lowers out-of-pocket maxima by 0.37%, and raises the probability of offering a zero-premium plan by 0.57 percentage points. The results imply that MSSP affects MA not only mechanically through benchmarks, but also through provider behavior that spills over across Medicare programs.