Estimating a Bargaining Model with Asymmetric Information: Evidence from Medical Malpractice Disputes

Games with asymmetric information play a prominent role in the theoretical literature of malpractice disputes. The common modeling framework in many papers is a game in extensive form which consists of two stages. In the first stage, one agent makes a settlement demand, and the other agent accepts or rejects the demand. If the demand is accepted, the case is settled out of court. Otherwise the case is taken to court and decided by a jury. This article develops a strategy for estimating such a model and focuses on reconciling the theoretical literature with observed regularities in malpractice data. Estimation of these types of models is complicated by the fact that key variables are (partially) unobserved and must therefore be treated as latent variables. The estimation strategy requires a complete specification of the bargaining model, including distributional assumptions of the latent variables. The parameters of the model are estimated using a simulated method of moments (SMM) estimator. The results of this study suggest that a simple bargaining model with private information can explain many of the qualitative and quantitative regularities observed in the data.