Decomposing the Rise in Markups

An increase in the aggregate markup can be due to both firm-level markups increasing and economic activity reallocating toward high-markup firms. I show that standard decomposition methods estimate a non-zero reallocation effect even when the allocation does not change and that this bias is potentially large. In opposite to the existing literature, I find that reallocation does not seem to be the major driver of the rise in markups among Compustat firms.