How do Gasoline Prices Respond to a Cost Shock ?

Some Implications of Micro Price-Setting Evidence for Inflation Dynamics and Monetary Transmission

“…This price selection effect is maximal in a standard menu cost model, such as Golosov and Lucas (2007), and minimal in a standard Calvo model. In two distinct empirical contributions, Alvarez et al (2021) and Gautier, Marx and Vertier (2022) test this theoretical proposition using micro price data and find strong supporting evidence for it. Using both consumer and producer price data for France, Alvarez et al (2021) find that the sectoral price response to a monetary policy shock is proportional to the ratio of kurtosis over frequency of price changes and that both kurtosis and frequency contribute to this result.…”

“…Frequency and kurtosis of price changes are also closely correlated with the price reaction to cost shocks across gas stations. In Gautier, Marx and Vertier (2022), the empirical test focuses on gasoline retail prices, for which we can observe high-frequency price data and can precisely measure a marginal cost shock. In this case study, where more precise information is available on the kurtosis of price changes and on the reaction of prices to an observed shock, the authors find very robust correlations, not only between the cumulated price response of gas stations and the ratio of kurtosis over the frequency of price changes, but also with both the kurtosis and the frequency of price changes taken separately (Chart 10).…”

“…Source:Gautier, Marx and Vertier (2022). Note: Correlations are calculated across gas stations; statistics are measured for every gas station in France, based on a dataset of daily gasoline prices between 2007 and 2018, collected by the French Ministry of the Economy.…”

Empirical Investigation of a Sufficient Statistic for Monetary Shocks

“…The sufficient statistic proposition that we analyze supplements the predictions for the role of frequency with a prediction for the role of kurtosis, which indeed our data confirm to be relevant. Following the same empirical strategy as this paper, Gautier, Marx, and Vertier (2021) test the sufficient statistic prediction on gasoline price data for France. They calculate the frequency and kurtosis of price changes at the gas station level, and estimate the CIR P of each gas station to a shock of the gasoline input cost.…”

“…Section 6 concludes and discusses avenues for future research. 5 The analysis of Gautier, Marx, and Vertier (2021) has three main advantages over ours: first, the good is very homogenous and each gas station has a long time series (relative to the frequency of price changes and to the half-life of the impulse response). Second, prices have almost no measurement error.…”