Interdependency, competition, and industry dynamics. Management Science 53(4): 599–615

A systematic understanding of industry dynamics is critical to strategy research because individual firm performance dynamics both reflect and affect change at the industry level. Descriptive research on industry dynamics has identified a dominant pattern where prices fall, output rises, and the number of firms rises and then falls over time. A number of models have been advanced to explain these patterns with a particular focus on explaining why a shakeout in the number of firms occurs. In the most prominent models, shakeout is generated by rising realized heterogeneity among firms that is either assumed to be unrecognized but determined ex ante or is generated by stochastic innovation outcomes coupled with convex adjustment costs and scale advantages in innovation and learning. In this paper, we develop an alternative model where heterogeneity develops among firms over time (leading to a shakeout) because firms must make choices over highly interdependent productive activities where the ideal combinations cannot be easily deduced or imitated. By combining two established models (a Cournot model of competition with an NK model of interdependency in production activities) we are able to advance an alternative explanation for the observed patterns of industry behavior including shakeout. We show that variation in the potential for interdependency in activities among industries is able to explai