Electoral independence, dependence, or co -dependence?: A reassessment of second -order electoral expectations in subnational elections

Abstract

In the electoral hierarchy, subnational elections have frequently been relegated to second-order status, as are supranational elections (i.e., European Parliament elections). According to the predictions of the second-order electoral framework, subnational elections should be dependent on national trends, manifest substantially lower participation levels, exhibit anti-government swing and higher support for small and regional parties. However, as subnational governments gain increasing powers and jurisdictional competencies, the potential for greater electoral participation and/or divergent and complex electoral behaviors between the state and substate levels becomes more likely. This dissertation explores what political factors and attitudes affect the likelihood of second-order behavior in a specific set of subnational elections, namely second-tier subnational elections.The first part of the dissertation employs a large cross-national aggregate analysis of second-tier elections in twelve countries from 1980-2002. The principal hypothesis is that a variety of institutional, economic, election-specific, and contextual features affect differential participation and vote choice in second-tier contests. Initial descriptive statistics provide evidence that there are significant variations between and among second-tier units regarding levels of differential turnout, electoral incongruence or deviation, and anti-government swing. Multivariate models indicate that factors such as the level of subnational revenue-raising capacity, subnational electoral timing, and bicameralism (among others) exert significant power over differences in electoral participation and electoral choice in multilevel electoral environments.Individual case studies of subnational elections in Spain, Germany, and Finland further help to elucidate inter-country differences in second-order expectations. The case studies further provide evidence of individual-level factors that impact the incidence of national political dominance, interlevel vote switching and participation in these contests. In particular, the surveys indicate that individual political attitudes and preferences predispose individuals in different manners to participate in second-tier elections, to focus more on national or subnational issues in their subnational electoral decisions, and to vote differently based on the arena in question. The individual and aggregate findings both point to the fact that the stakes of the election (both perceived and actual) directly affect the nature of electoral behavior. The dissertation's findings have distinct implications and consequences for broader political issues of decentralization, democracy, accountability, and representation.-- Chris Palazzolo, PhDHead of Collection ManagementSocial Sciences LibrarianAdjunct Professor, Department of Political ScienceEmory UniversityAtlanta, GA 30322Cpalazz@emory.edu<mailto:Cpalazz@emory.edu>404-727-0143________________________________This e-mail message (including any attachments) is for the sole use ofthe intended recipient(s) and may contain confidential and privilegedinformation. If the reader of this message is not the intendedrecipient, you are hereby notified that any dissemination, distributionor copying of this message (including any attachments) is strictlyprohibited.If you have received this message in error, please contactthe sender by reply e-mail message and destroy all copies of theoriginal message (including attachments).

Nationalisation of regional elections in Central and Eastern Europe

“…With less at stake, regional elections gain a status of "barometer" (Anderson and Ward 1996) or "referendum" elections (Carsey and Wright 1998), which disproportionally attract opposition party supporters. As a result, the "anti-government swing" should be higher (Palazzolo 2007), because opposition party voters are more likely to turn out "to make a point" than are government supporters to express their satisfaction (Jeffery and Hough 2009). Some authors argue that the extent of government party loss can be predicted from the placement of the second-order election in the first-order election cycle.…”