Simulating Hospital Merger Simulations

Abstract

We assess the performance of recently developed hospital merger simulation methods in a Monte Carlo setting. We specify a theoretical model of hospital markets and generate true price eects of mergers by solving a simultaneous Nash bargaining game between the hospitals and MCOs for pre- and post-merger cases. We then take the data generated by the theoretical model that would be available in a real-world prospective merger analysis to the merger simulation methods and compare their predictions to the true price eects. We perform this exercise for multiple parameterizations, including dierent levels of competition in the hospital and insurance markets. We examine two related merger simulation methods and nd that both are preform well, although each exhibits some tendency to under- or over-predict merger price eects. We also nd some sensitivity to variation in the model parameters that govern consumer preferences over hospital networks and the division of the joint surplus in the bargaining game. 1