Working Paper: NBER ID: w20155
Authors: Fernando Alvarez; Hervé Le Bihan; Francesco Lippi
Abstract: We document the presence of both small and large price changes in individual price records from the CPI in France and the US. After correcting for measurement error and cross-section heterogeneity, the size-distribution of price changes has a positive excess kurtosis. We propose an analytical menu cost model that encompasses several classic models, as Taylor (1980), Calvo (1983), Reis (2006), Golosov and Lucas (2007) and accounts for observed cross-sectional patterns. We show that the ratio of kurtosis to the frequency of price changes is a sufficient statistics for the real effects of monetary policy in a large class of models.
Keywords: price changes; monetary shocks; kurtosis; menu cost model
JEL Codes: E3
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| kurtosis of the size distribution of price changes (kur_pi) (D39) | cumulative output effect (m) (E23) |
| average number of price changes per year (n_pi) (C43) | cumulative output effect (m) (E23) |
| ratio of kurtosis of the size distribution of price changes to the average number of price changes per year (kur_pi/n_pi) (C46) | cumulative output effect (m) (E23) |
| measurement error (C20) | cumulative output effect (m) (E23) |
| cross-sectional heterogeneity in the data (C21) | cumulative output effect (m) (E23) |