ABSTRACT
We examine the long-run relationship between European far right, radical left and unemployment. A unique dataset is compiled for 31 European countries that span from 1900 to 2013. We reveal the long-run relationship between extreme right and unemployment. An increase in unemployment and radical left increases the far-right vote share.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Matakos and Xefteris (Citation2017) argue that when the economy is not in turmoil, there is a positive relationship between (relative small changes in) unemployment and support for systemic parties. They could not exclude the possibility of a negative relationship from a theoretical point of view.
2 Exploiting both dimensions (time series and cross section) was problematic as the panel is unbalanced, the observations are infrequent and there are numerous zeroes for individual countries. As a result, we could only aggregate the data, given that the panel is unbalanced and panel cointegration not possible.
3 Our results remain qualitatively similar when we consider mean European unemployment instead of weighted unemployment.
4 The literature on unit roots is voluminous and one can argue one way or the other. In our case, we did not make any a priori assumption about the stationarity of the series and followed the ‘let the data talk’ approach since the emphasis of this note is on the long-run coefficient. We have also considered both the logistic transformation (see Wallis Citation1987 and Farmer Citation2015) and the literature for unit roots for bounded variables (see Carrion-i-Silvestre and Gadea Citation2013) (see ).
5 Hansen and Seo (Citation2002) test for two-regime threshold cointegration in vector error-correction models. We focus on the long run and account for breaks rather than a model with two regimes.