Gasoline Demand in Brazil: An Empirical Analysis

I estimate the elasticity of demand for gasoline in Brazil taking into account that gasoline and ethanol prices are endogenous variables in the aggregate demand curve. Combining di erent identi cation strategies, I nd that consumers are more elastic to price than earlier studies suggest. An increase, ceteris paribus, of 1 percent in gasoline and ethanol prices reduces gasoline demand by 1.67 percent and raises gasoline demand by 0.73 percent respectively. To analyze the behavior of gasoline demand in the face of exogenous shocks, I also calculate impulse response analysis. The results suggest that a 1 percent temporary increase in gasoline and ethanol prices reduces gasoline demand by 0.4 percent and raises gasoline demand by 0.2 percent respectively. Both of these e ects are permanent. In this study I also look for evidences that the introduction of ex-fuel automobiles in Brazil led to consumers more elastic to price change. I divide my sample in three periods and analyze the rst and the last subsample. The elasticities estimated using the rst subsample are closer to the ones found for other countries and in meta-analysis studies, being consistent with the idea that both fuels are not close substitutes at the initial period of my sample. The results found using the last subsample suggest a completely di erent scenario. Although we cannot interpret such results as a causal relationship between the introduction of ex-fuel automobiles and more elastic consumers, we can interpret these results as evidence that this new technology may have had an impact over consumers' behavior.