Robust estimation of the VAT pass-through in the Netherlands

This paper introduces the Common Correlated Effects Estimator into the study of Value-Added-Tax pass-through and compares this method to various other methodologies used in the literature. To this end, we study two Value-Added-Tax increases in the Netherlands, in January 2001 and October 2012. We show that the Common Correlated Effects Estimator produces robust estimates, especially when divergent macroeconomic trends make identification more difficult. Furthermore, we show that the choice of the control group is of lesser importance once sufficient control variables are included. Our results indicate, in accordance with most findings in the literature, that we cannot reject the null-hypothesis of a full pass-through for both Dutch tax-hikes.