Governments and researchers are frequently forced to predict the impact of perspective mergers on markets. This dissertation provides structural methods to empirically evaluate mergers. We first build a static model in which players are boundedly rational to evaluate the welfare consequence of mergers in that environment. Then we use that model studying bidding in the Texas electricity market, a market in which bidding by some firms departs significantly from what Bayesian Nash models predict, while bidding from other firms closely resembles these predictions. Our results show that exogenously increasing sophistication may significantly increase efficiency and additionally, mergers may increase efficiency even without cost synergies. The next chapter provides a structural method to empirically evaluate mergers in a dynamic setting. We build an infinite five-step repeated game. Then, we propose a three-step estimation method to estimate the game in which Markov perfect Nash equilibrium is played. Our three-step estimation method is flexible and can be easily modified to estimate various market structures. These dissertation studies mergers in more realistic settings. We first show that mergers that do not generate cost synergies may also increase efficiency when some of the firms in the market are boundedly rational. Then we build a dynamic endogenous merger game and provide a new method to estimate it. Our simulation result shows that our estimation method is computational feasible and can be applied to real data.
