Abstract: We build a real business cycle model in which monopolistic competition departs from the Dixit-Stiglitz assumption of consumers’ taste for variety, in order to better characterize the cyclicality of the labor share. When market structures allow for “winner-takes-all” behaviors, firms’ markups can either increase or decrease in the number of market competitors, depending on the shape of the probability distribution of consumers’ taste shocks, especially their tail properties (Gabaix et al. (2016)). In our macro-dynamic setup with endogenous firm entry, a positive productivity shock unambiguously increases market competition, but can in turn generate either a countercyclical or procyclical labor share. We estimate the model on US data over the period 1997-2022 at the three-digit NAICS level and show that it quantitatively matches both the slightly procyclical labor share at the aggregate level and the countercyclical labor share of more than half of US industries.

Marlène's page: https://sites.google.com/view/marleneisore/