Electing Regional Representatives: Conceptualizing and Measuring Regional Electoral Systems

Paper proposal for consideration for panel 3 on ‘Political Parties and Party Competition in Multilevel States’. The main research objective of this paper is to introduce indicators that describe the most important characteristics of regional electoral systems. The need for such an exercise is based on three observations. First, the literature has largely ignored the impact of regional electoral systems. Second-order election and nationalization scholars use very crude and simple measures that tap whether proportional or majoritarian (or mixed) rules are applied in regional elections. Moreover, scholars have developed measures that are attuned to gauge differences between regional and national electoral systems rather than the characteristics of regional electoral rules. As a result, the impact of electoral rules on regional electoral outcomes have received scant attention in the literature. Second, regional electoral systems have unique features that are not present in national and European electoral systems. For example, in Greece voters cast one vote which both determines the seat allocation in the council as well as which candidate will become the regional president. In addition, classifications based on national electoral systems do not lend themselves to assess electoral rules in a multilevel context and thereby are not suitable to study how electoral rules applied at one level impact electoral outcomes at another level. For example, in this paper we develop the concept of ‘the absolute vote threshold’ which is the total number of votes a candidate needs to obtain to win one seat. We argue that the ratio between absolute vote thresholds between regional, national, and European electoral arenas –i.e., how many more/less votes a candidate needs to gain representation in another electoral arena—is a key variable driving the extent of spill-over between electoral arenas. Third, national and European electoral systems can play out very differently across the statewide territory. For example, party competition in national elections held in Valle d’Aosta and the Åland islands boil down to a one- to two-party system even though proportional rule applies. Existing classifications of national and European electoral systems are not attuned to measure this regional variation. We argue that the relative size of a region vis-à-vis other regions in the state should be taken into account in the measurement of electoral rules translating votes into seat in order to better understand how electoral rules impact party competition at the regional level. In this paper we start with a critical review of ‘methodological nationalism’ in electoral system research and we provide an in-depth discussion of the three caveats identified above. We then proceed by developing a set of indicators that are attuned to tap meaningful variation at the regional level and are suitable to probe multilevel electoral contextual variables that are thought to be conducive for vertical and horizontal electoral spill-over. The final section provides (descriptive) analyses to further expose and underline ‘methodological nationalism’ in electoral system research.