We study the relationship between market structure and innovation in the global automobile industry for the 1982-2004 period. We use the dynamic industry framework of Ericson and Pakes [1995] and estimate the parameters of the model using a two-step procedure proposed by Bajari et al. [2007]. Since the industry has seen a lot of consolidation since 1982, mergers are an important ingredient of our model. After estimating the parameters of the model, we simulate the industry forward and study how changing market structure (mainly due to mergers) affects innovative activity at the firm as well as at the industry level. Our findings are the following: (1) The effect of market structure on innovation in the global auto industry depends on the initial state. If the industry is not very concentrated, as it was in 1982, some consolidation may increase the innovative activity. However, if the industry is already concentrated, as in 2004, further consolidation may reduce the incentives to innovate. (2) Mergers reduce the value of merging firms though they may increase the aggregate value of the industry.