Abstract How far should an industry be allowed to consolidate when competition and innovation are endogenous? We develop a stochastically alternating-move game of dynamic oligopoly and estimate it using data from the hard disk drive industry, in which a dozen global players consolidated into only three in the last 20 years. We find plateau-shaped equilibrium relationships between competition and innovation, with heterogeneity across time and productivity. Our counterfactual simulations suggest the current rule-of-thumb policy, which stops mergers when three or fewer firms exist, strikes approximately the right balance between pro-competitive effects and value-destruction side effects in this dynamic welfare trade-off.