Estimating the cross-sectional distribution of price stickiness from aggregate data

We estimate a multi-sector sticky-price model for the U.S. economy in which the degree of price stickiness is allowed to vary across sectors. For this purpose, we use a specification that allows us to extract information about the underlying cross-sectional distribution from aggregate data. Identification is possible because sectors play different roles in determining the response of aggregate variables to shocks at different frequencies: sectors where prices are more sticky are relatively more important in determining the low-frequency response. Estimating the model using only aggregate data on nominal and real output, we find that the inferred distribution of price stickiness is strikingly similar to the empirical distribution constructed from the recent microeconomic evidence on price setting in the U.S. economy. We also provide macro-based estimates of the underlying distribution for ten other countries. Finally, we explore our Bayesian approach to combine the aggregate time-series data with the microeconomic information on the distribution of price rigidity. Our results show that allowing for this type of heterogeneity is critically important to understanding the joint dynamics of output and prices, and constitutes a step toward reconciling the extent of nominal price rigidity implied by aggregate data with the evidence from price micro data. -- Heterogeneity ; price stickiness ; micro data ; macro data ; Bayesian estimation