Decomposing the Rise in Markups with Measurement Error

A rise in the aggregate markup can be decomposed into an effect coming from changes in firm-level markups and an effect coming from reallocation of economic activity toward high-markup firms. I show that such a decomposition is heavily influenced by measurement error of firm-level markups. When firm-level markups are mismeasured the decomposition can give opposite results depending on whether actual or measured markups are used. The data suggests that measurement error has increased over time within Compustat data. Correcting for measurement error lowers the reallocation effect. Moreover, correcting for measurement error also lowers the estimated sales-weighted average markup, and I find that the estimated increase in the sales-weighted markup over the last 60 years in Compustat is for a large part due to this rise in measurement error. The cost-weighted average markup is not affected by idiosynchratic measurement error and is therefore a more robust estimate of market power