Political effects of low turnout in national and European elections

This article estimates for member states of the EU the effect of low levels of turnout on parties' share of the vote in national elections. It does so by comparing the distribution of party choices in national elections for all those who participate in those elections on the one hand, and for the much more restricted group of those who participate in European Parliament elections on the other. As European elections register lower turnout than other nation-wide elections, this comparison provides an extreme, but empirically observed case of low turnout. Turnout effects prove comparatively small, and are non-negligible in only few cases. Turnout effects are slightly different for different kinds of parties: right-wing parties benefit slightly from them (on average) and left-wing parties are (on average) somewhat hurt. Although significant, these differences are exceedingly small, and explain no more than a few percent of variation in turnout effects. No significant effects are found from other party characteristics (such as their size, government status, position on European integration, or interactions of these with government approval or time since the last national election). The analyses are based on data from European election studies in 1989, 1994, 1999 and 2004. (c) 2006 Elsevier Ltd. All rights reserved.