Presenter: Andreu Arenas (EUI)

When: Tuesday, 24 November 2015, 17h00-18h30

Where: Seminar Room 2, Badia Fiesolana

Abstract: Many governments and constitutions ban the existence of parties or organisations which support or justify violence. Are political bans effective in reducing political support for the targeted political movements? Do they have side effects? In this paper, I try to answer these questions by exploiting the differential length of the ban of Batasuna across municipalities. Batasuna was a political party which was banned due to its links to the Basque terrorist organization ETA. In some municipalities it was banned for one election; in some others, for two elections. I exploit the rule used by the public prosecutor -based on observables- which gave rise to the differential length of the ban across municipalities to identify the effect of the differential length of the ban. Differences-in-differences (DiD), and DiD-Instrumental Variables estimates show that in municipalities where the ban did last for two electoral terms rather than one, Batasuna obtains significantly less support after the ban. The effect is large, persistent for at least two elections, and it has spillovers to support for Batasuna in regional elections. The effect is driven both by the extensive and the intensive margins, i.e. Batasuna running for office in a lower number of municipalities and obtaining less votes wherever they run after the ban. Moreover, I find that a longer ban increases street violence in the very short run (one month after), but it decreases it afterwards.