Recent empirical work has shown that valence considerations are fundamental for theoretically and empirically understanding voting behavior (Bélanger and Meguid, 2008; Serra, 2010; Johns, 2011). Initially valence theory was considered as an alternative to spatial theory in explaining vote choice (Stokes 1992; Davis et al., 1970; Enelow and Hinich, 1984). However, scholars have now admitted that spatial considerations often interact with valence judgments and can both simultaneously inform vote choice (Abney et al. 2011; Sanders et.al. 2011). The detachment of valence criteria from ideological consensus (Pardos-Prado, 2012) opened the way to the study of valence voting as a feature of competitive politics still connected to individual spatial considerations.
This paper takes this debate a step further by answering the questions: What is the role of systemic factors on the impact of valence considerations on vote choice for the incumbent party taking into account spatial constraints?
Using a multilevel approach for the 2009 European election we test cross-level interactions between the impact of spatial and valence and the conditions posed by systemic factors such as party polarization, effective number of parties and electoral system at the national elections.
This paper makes a twofold contribution. On the one hand it tests valence theory in a comparative perspective and advances our knowledge both on how valence is used across countries and how its impact is conditioned by party system and electoral system differences across Europe. On the other hand it allows a methodological evaluation of the current tools of measuring valence across Europe, as it facilitates judgments about the meaning of valence in the different countries.
