Micro and Macro Cost-Price Dynamics in Normal Times and during Inflation Surges

We develop a unified approach to studying cost-price dynamics in the cross-section of firms in order to explain the time series of aggregate inflation, both during normal times and inflation surges. A key novelty is the use of microdata on firms’ prices and production costs to construct an empirical measure of price gaps — the deviation between a firm’s listed and optimal price. We characterize the mapping between price gaps and the size and frequency of price adjustments and take them to the data to test nonparametrically how firms’ pricing strategies align with the predictions of different pricing models, conditional on shocks of different magnitudes. The microdata provide strong evidence of state dependence: the passthrough of costs into inflation increases more than proportionately when the economy is hit by large aggregate shocks. In contrast, in normal times, the frequency of price adjustment is approximately constant, and the microdata conform to the predictions of time-dependent models (e.g., Calvo 1983). Conditional on the path of aggregate cost shocks extracted from the data, a generalized state-dependent pricing model accounts well for both the low and stable inflation of the pre-pandemic period and the nonlinear surge that followed.