Files
Abstract
Shortly after the first European elections were held in 1979, scholars noted that, relative to national elections, governing and large parties tend to lose votes, smaller (and ideologically “extreme”) parties gain votes, and turnout is significantly lower (Reif and Schmitt 1980; Reif 1984). They therefore classified European elections as “second order national elections”. Second order elections (SOE) theory has since become one of the most important theories of European comparative electoral behavior. The general aim of this thesis is to test and unpack the key arguments of this ‘grand’ theory of European elections, which has become a sort of received wisdom in the field of European comparative electoral behavior over time. In particular, I concentrate on (1) the determinants of governments’ performance in European elections; (2) the determinants of performance of other, particularly extreme and radical, parties in European elections; and (3) the determinants of variation in turnout in European elections. I put special emphasis on the effects of election timing, party size, as well as some novel variables, such as coalition composition.