Beyond the average elasticity - applying quantile panel regression to German household mobility data

摘要

This paper employs quantile panel regression to the study of fuelprice elasticities. Contrasting with standard panel approaches, thismethod reveals the impact of explanatory variables across all points inthe conditional distribution of the response variable while controllingfor unobserved heterogeneity. Applying quantile panel regression toGerman household data demonstrates that fuel price elasticities arevery high in magnitude - below -0.8 - for a small segment of householdswhose car mileage is low, but that this effect tapers off rapidly amonghouseholds with higher car mileage. These findings have implicationsfor policy instruments that rely on estimates of fuel price elasticities,for example fuel taxation.