Markups and Misallocation with Evidence from an Exchange Rate Appreciation

With non-homothetic preferences, a monopolistic competition equilibrium is inefficient. In a setting with heterogeneous firms that charge variable markups, this paper finds a sufficient statistic for changes in allocative efficiency that can be directly measured with data. The model also predicts differential effects of competition and cost shocks on the reallocation across domestic firms, which I test empirically with firm and industry-level data in Chile, a country that has experienced large exchange rate shocks. I find important changes in misallocation over time due to the way firms pass-through productivity gains into markups. From industry-year variation, there is evidence that industries that import a larger share of their inputs become more misallocated as a result of exchange rate appreciations compared to “open” sectors whose output competition becomes fiercer. The observations of markup responses at the firm level are consistent with the aggregate measure of allocative efficiency.