Market Structure and Multiple Equilibria in Airline Markets

We provide a framework for inference in discrete games that involve multiple decision makers and use it to study airline market structure in the US. We make inferences on a ``class of models'' rather that looking for point identifying assumptions that pin down a unique model. We extend the empirical literature on entry and market structure started in \citeasnoun{bres_reiss_restud}. We allow for a more flexible model of entry, heterogeneity and player identities without making assumptions on equilibrium selection. Our estimation strategy is directed at a class of models that obey the fundamental assumption that if a firm enters a market it expects nonnegative profits. This fundamental condition provides a set of inequality restrictions on regressions that we exploit to learn about the profits of various firms. We then examine airline market structure focusing on the strategic behavior between a set of airlines where we allow for and find heterogeneity in the effects airlines have on each other, and for correlation among the unobservables. We find that there are multiple equilibria in the number and identity of firms. Finally, we also test for particular selection rules and find that a rule that picks the equilibrium with the largest total profits is consistent with the data and the model