Using retail scanner data, we find two previously unknown facts about price setting. First, the probability of price adjustment increases with product revenue. Second, the absolute size of price adjustment decreases with revenue. We show that these facts are consistent with a menu cost model where the fixed cost of adjustment does not scale with product revenue. In addition, these facts suggest that the price stickiness of individual products decrease with revenue. Over the business cycle, both the mean and variance of the cross-sectional revenue distribution decrease with the unemployment rate. These empirical facts imply that monetary policy has a stronger effect on the economy in recessions than in expansions. We quantify this property using a calibrated menu cost model, and find that the effect of policy on output is 53.7% larger in recessions.
What inventory behavior tells us about markups (with Alberto Arredondo Chavez)
We use firm-level data on stocks of final goods and sales in order to estimate the change in firm-level markups through time. We extend a model of stock-out prevention to accommodate a variety of price-setting behavior. This allows us to be agnostic about the product market structure. Everything else equal, a higher markup implies a higher stock to sales ratio as the loss from a stock-out is larger. Three other components are key in order to identify markups with our model: the rate at which the firm discounts future flows, the expected growth in marginal costs and the strength of the need for inventories to prevent stock-outs. We use our information at the firm level to measure these components and recover markups.
Airline Competition, Oil Price Pass-Through, and Carbon Taxes (with Paul Brehm and Anirudh Jayanti)
Investment dealer collateral and leverage procyclicality (with Jason Allen) Empirical Economics (2018): 1-17.