Prices are a significant driver of high health care spending in the US, but how to reduce prices remains an open question. I examine one widely-touted solution – setting prices via competitive bidding – in the context of a Medicare payment reform. The reform gradually replaced administratively-set prices with prices from competitive bidding for durable medical equipment (DME) in 100 metropolitan statistical areas. Using detailed claim-level data, I estimate that the competitive bidding program reduced the prices of covered items by 45%. However, the program also generated an 11% reduction in quantity, which several pieces of evidence suggest is associated with inefficient supply shortages. One likely cause of the shortage is the auction design, which allows winning bidders to renege on supply commitment. Leveraging novel bid data, I estimate an equilibrium model of optimal bidding and find that the program generated prices that were on average 6% below the market clearing price, consistent with the observed supply shortages. I use the results to show that counterfactual auction designs could reach the desired market quantity while saving 43% in government spending relative to administratively-set prices. The analysis highlights the importance of auction design in achieving desirable outcomes, and suggests that a well-designed competitive bidding program could potentially generate large savings in health care.
Job Market Paper
- Media coverage: The Washington Post
- Award: 2022 American Society of Health Economists Student Paper Award