Research

I show equilibrium existence for a price-setting game among multi-product firms facing a nested logit or nested CES demand. I generalize previous results in two directions. First, the nesting structure allows for a two-level nesting hierarchy. And second, I allow for an arbitrary firm-nest overlap. Additionally, if the game satisfies an easy-to-verify condition, I show that there exist two extreme equilibria that are the most and least preferred by consumers. Furthermore, I provide an algorithm to find these extreme equilibria and argue that the algorithm allows researchers to numerically verify equilibrium uniqueness, i.e., if the extreme equilibria are equal to each other. Finally, as a by-product of the equilibrium existence result, I show that FOCs are sufficient for profit maximization, which implies that the standard approach to marginal cost estimation in the empirical literature, i.e., inverting FOCs, correctly identifies the costs that rationalize pricing decisions.

I use an Iterated Strict Dominance (ISD) argument to build bounds on the distribution of outcomes of games and use them to pin down an identified set for the parameters of interest. These bounds (ISD Bounds) are robust to equilibrium multiplicity, pure and mixed, and to any non-equilibrium play as long as it is consistent with ISD. Furthermore, ISD Bounds apply to games of complete or incomplete information, with discrete or continuous actions of any dimensionality, as well as games with unobserved heterogeneity. To maximize the "bite" of ISD Bounds, I introduce Strategically Monotonic Supermodular Games, i.e., games of strategic complements/substitutes where players’ payoffs are supermodular in their actions. I show that ISD rules out large swaths of the strategy set for this type of game via an easy-to-compute sequence of best-response iterations. Moreover, I show that for binary games the resulting identified set is sharp. Finally, I apply ISD Bounds to an entry game for the airline industry and use the estimates to evaluate the impact of the proposed merger between JetBlue and Spirit.

We consider the oligopsony competition among beef packers to purchase cattle from feedlots. We explore in particular the competitive implicates of an increasingly popular contract in which the price of a (future) transaction is pegged to future cash market prices. These contracts create an additional incentive for packers to depress cash market prices, and as the largest four packers account for more than 80% of purchases, they likely have an ability to do so. We provide descriptive regressions that show a negative correlation between contract quantities and cash market prices on a week-to-week basis over 2005-2019. We then construct an empirical model and estimate it using data from the same period. The model allows us to assess the extent to which the rise of contracts may have contributed to an increase in the observed spread between the prices that packers pay for cattle and the prices that they receive for beef.

When multi-product firms endogenously choose their product variety, mergers can have welfare effects that go well beyond their immediate impact through prices. The structural models needed to quantify these effects, however, quickly become computationally intractable as the number of products grows. In this paper, I propose a dynamic structural model of multi-product firms and endogenous product variety that bypasses this dimensionality problem. I show that, under a nested logit demand assumption, firms introduce/remove products in pre-determined order. As a result, the number of products each firm supplies is a sufficient statistic for the identity of the products themselves and, therefore, for all market outcomes and firms’ continuation strategies. I use this result to argue that the strategies and states that need to be considered to solve the model are a small fraction of the whole strategy set and state space, making it possible to estimate the model using standard techniques in structural econometrics. I apply the model to the RTE-Cereal Industry and use it to simulate a hypothetical merger between two large players. Results show that ignoring endogenous product variety leads to a 30%  overestimation of the post-merger number of products and a 19% underestimation of consumer welfare loss.

We study the effects of mergers on incumbent behavior and firm entry in the U.S. dialysis industry. Using an event-study approach, we find that facilities and dialysis stations decrease in markets exposed to mergers, relative to markets without any merger activity. This is driven by a sharp reduction in both variables by the merged entity, which is only partially offset by an increase corresponding to other incumbent firms. We do not find evidence for novel entry. Additionally, we find declines in inputs per station at merged firm facilities and an increase at other facilities.