LAUSR

Abstract This study tries to explain regional level variation in the far right-wing vote across more than 160 regions in 17 Western European countries from 1990 to 2013. With the help of a panel Tobit model, I first examine the impact of nine regional-level structural indicators on the dependent variable, the percentage of the far right-wing vote. I find that the far right performs better in territorial units with a high percentage of university-educated individuals, in rural regions and in areas that have a high percentage of foreigners. Second, I use a dynamic specification in first differences to evaluate how changes in the independent variables trigger changes in the dependent variable. The results of this second specification highlight that increases in unemployment rates and in the number of college-educated citizens trigger a better performance of the far right.