Equilibrium Effects of Tiered Cost Sharing in the Market for Medical Care (Job Market Paper)
Abstract: In the market for medical care, prices paid by consumers and those received by providers are mediated by insurance plans, dampening the market forces that typically facilitate price competition. Consumers often face little to no variation in out-of-pocket prices across providers, giving little motivation to shop around. Providers, facing price-inelastic consumers, have little incentive to compete in price. In this paper, I study the effects of tiered cost sharing, an innovative plan feature designed to tackle these issues. Tiered plans apply a steep gradient in out-of-pocket costs to steer consumers toward low-cost providers. Using detailed administrative data from the State of New Hampshire, I estimate the impact of these incentives in both the short and long run. I first examine the rollout of tiered programs by the state's two largest insurers in an event-study framework to quantify the average effects on provider choices and spending. I next examine underlying mechanisms on the demand-side by estimating consumer responsiveness to out-of-pocket prices in a discrete-choice context. Finally, I develop a structural model of supply and demand to study the general equilibrium implications of this plan feature and to compare it with other consumer-directed mechanisms. Reduced form estimates imply savings of 35%, 7%, and 5% on lab, arthroscopic, and endoscopic services, respectively. I show that consumers respond modestly to spot prices when choosing a provider, implying that incentives must be sizable to shift demand meaningfully. The equilibrium model suggests that tiered cost sharing affects expenditures by generating substitution toward low-cost providers and driving equilibrium price reductions. Estimates indicate that the tiered design as currently constructed compares favorably to alternatives, although some potential improvements are brought to light. Counterfactual simulations project that moving all consumers to a tiered plan can save as much as 12% on shoppable services.
Work in Progress
Health Care Demand Elasticities on the Intensive Margin (draft coming soon)
Abstract: The majority of prior work estimating the demand for health care focuses on the response of total spending to cost sharing. With the growth in transparency of medical service prices, policy makers and insurers have increasingly embraced strategies to encourage price-shopping among consumers. I study the effectiveness of these incentives by estimating the response of provider choice to price variation within an individual’s choice set. To accommodate endogenous out-of-pocket prices, I develop an instrumental variable strategy to estimate demand which using transaction prices and average inertial plan cost sharing to instrument for individuals’ experienced spot prices. This framework can be employed in both reduced-form and structural models of utilization. Using detailed administrative claims data, I apply this method to provide new estimates of the intensive margin elasticity of demand for a variety of shoppable services such as MRIs and arthroscopic knee surgery.
The Welfare Effects of Surprise Medical Bills
The perils of surprise hospital bills have been well documented in both the popular press and as academic articles. I estimate two sources of potentially significant welfare loss due to these bills using plausibly unanticipated treatment by an out-of-network provider. First, I estimate the static welfare loss using a demand model in which realized prices differ from expectations. Second, I study the effects of bill shock on dynamic consumption behavior using an event-study framework. I find that families hit with bill shock reduce medical care expenditures relative to families who experience a similar admission but were not subject to an out-of-network physician.
Bringing Down the Costs of Health Care: Evidence from New Hampshire (with Cori Andriola)
Spillover Effects of Parental Insurance Coverage: Evidence from Medicaid Dental Benefits (with Xiaoxi Zhao)