Heterogeneity in the tax pass-through to spirit retail prices: Evidence from Belgium

On 1st November 2015, the Belgian government increased the excise tax on alcoholic beverages. For spirits with 40% of alcohol and bottle size of 70 cl, this tax change is equivalent to an amount of 2,43 € per bottle of spirits. This paper studies the impact of this tax reform at the store level on the (posted) retail price of six major brands of spirits, using a difference-in-differences method. The estimation is based on a balanced panel of scanner data from a major supermarket chain (with a 33% market share) and uses the retail prices of the same brands sold in France by the same supermarket chain as a control group. Having information on each store location, we show spatial variations in the tax pass-through for homogeneous products. We find that these variations are strongly related to the intensity of local competition and to a lesser extent to the proximity to the borders (mainly with Luxembourg which is the low-price country). We find that the tax was quickly passed through during the first month of tax implementation and that it was mostly over-shifted. However, we also find that both the border and the competition effects are not instantaneous, but arise several months after the tax reform. These findings have important implications for alcohol control policies as they highlight that the incidence of alcohol taxation can vary greatly across space and affect differently households depending on where they live.