Experimentation by firms, distortions, and aggregate productivity

Recent empirical research has documented that distortions of allocative efficiency among heterogeneous firms can have large aggregate consequences. This paper evaluates the size of these effects when distortions affect not only resource allocation but also the evolution of firm-level productivity itself. To this end, we partially endogenize the evolution of firm-level productivity in a standard heterogeneous firm model by allowing firms to engage in costly, purposeful experimentation: Firms can engage in risky experiments, which take the form of productivity shocks. Results from failed experiments can be discarded. We then show that endogenous productivity implies up to twice as large effects of productivity-dependent distortions on aggregate consumption.