Panel Paper:
Surprise Medical Bills, Arbitration, and the Effect on Bargaining: A Structural Model of Insurance Network Formation
Friday, November 8, 2019
I.M Pei Tower: Majestic Level, Majestic Ballroom (Sheraton Denver Downtown)
*Names in bold indicate Presenter
New York's surprise medical bill law stipulates a final-offer arbitration system (“baseball-style arbitration”) to resolve payment disputes between insurers and providers when a patient receives a surprise out-of-network medical bill. The arbitration mechanism is one of a number of possible policy options for resolving compensation disputes between the insurer and provider after a surprise bill. By examining never-before-seen data from New York's arbitration records, we document several empirical facts about the arbitration mechanism in New York. First, we show an increase in the utilization of arbitration over time. Second, we show variation in who wins (whose bid is selected by the arbitrator) as a function of what kind of specialty the provider practices. Third, we document a reluctance of providers to budge from their initial charges, which may be a function of uncertainty about the arbitration process. We fit parties' offers in arbitration to a model first developed by Farber (1980) and Ashenfelter and Bloom (1984) to describe final-offer arbitration (where the arbitrator cannot split the difference between bids). We embed the arbitration model inside a settlement and negotiation model, describing how the presence of a surprise bill patient protection law (like New York's) can influence provider network formation at the beginning of the year. This structural network formation approach, which borrows from the design of Ho (2009), affords us two benefits. First, we can rationalize the findings of Cooper, Scott-Morton and Shekita (2017), who present reduced-form evidence that the New York surprise bill law lowered the incidence of surprise bills. Second, given the embedding of the final-offer arbitration game inside of a Nash-in-Nash bargaining process of network formation, we can examine the impact of counterfactual (alternative) policies that aim to reduce surprise out-of-network medical bills, such as rate-setting or alternative arbitration benchmarks. These counterfactual exercises, estimated with forthcoming All-Payer Claims Data from New York, are important to guiding discussions with policymakers about optimal policies to address surprise bills, such as the policies considered by the US Senate and in other states.
