Fuel taxes can be employed to correct externalities associated with automobile use and raise government revenue. The general understanding of the efficacy of existing taxes is largely based on empirical analyses of consumer responses to fuel price changes. In this paper, we directly examine how fuel taxes, as distinct from tax-exclusive fuel prices, affect fuel demand. To do so, we use a Markov-switching approach on monthly observations of French fuel prices from 1983 to 2013. Our analysis reveals that consumers respond significantly faster to increases in fuels taxes than to increases in tax-exclusive fuel prices. This result raises questions about our understanding of the efficacy of existing fuel taxes and of the optimal tax to achieve the various goals for which they are implemented.