How does fuel demand respond to price changes? Quasi-experimental evidence based on high-frequency data

This article exploits quasi-natural experiments provided by both the 2022 crude oil price surge consecutive to the Russo-Ukrainian war and fuel excise tax cuts in France to infer the price sensitivity of fuel demand. The granularity of bank account data available at the transaction level permits to shed new insights on how to properly disentangle anticipation effects from price effects. After controlling for anticipatory behavior, we obtain a price-elasticity comprised between -0.4 and -0.21. The average elasticity exhibits sizeable dispersion with respect to fuel spending, but varies little with income and location. Counterfactual simulations enable us to assess both financial and distributive impacts of the tax policy at stake as well as its effect on CO2 emissions.