Voter characteristics and leader effects in a post-Communist context: the case of the 2012 legislative elections in Romania.

AuthorGheorghita, Andrei
PositionPOLSCI PAPERS - Case study

Introduction

In the contemporary political environment, the added value brought by leaders to the electoral performance of their parties appears to be significant and growing. Although the degree to which such a contribution is decisive for the results of elections is unclear, the increased public attention to leaders instead of issues is a finding hard to contest. This phenomenon is largely referred to in the literature as 'personalization of (electoral) politics' (Kaase, 1994; Karvonen, 2010; McAllister, 2007), 'electoral face of presidentialization' (Mughan, 2000; Poguntke and Webb, 2005), 'candidate-centred politics' (Wattenberg, 1991) or simply as 'leader effects on voting' (Aarts et al., 2011; Barisione, 2009). Most of the research on leader effects focuses on enduring Western democracies and regards the phenomenon as an aggregate effect of a sequence of transformations: changes in the patterns of mass communication in the age of electronic media (Bean and Mughan, 1989; King, 2002; McAllister, 2007, 1996; Poguntke and Webb, 2005), a significant erosion of traditional electoral alignments (McAllister, 2007, 1996; Schmitt and Ohr, 2000; Wattenberg, 1991), an increase in the complexity of political issues correlated with a shift of interest from local to national politics (McAllister, 1996), and the internationalization of politics (Poguntke and Webb, 2005).

Although extensively studied for the case of Western polities, the scenario of a similar shift from parties to leaders as main actors of the electoral scene occurring in the Eastern post-communist bloc is largely ignored in the literature. The exceptions are rather few (Colton, 2000, 2002; Flacco, 2014; Gheorghita, 2014; Grbesa, 2004; Rudi, 2014; Stefuriuc, 2003). Still, it is very likely that many CEE countries have reached the same outcome of electoral personalization following an alternative pathway, as a consequence of several local-specific evolutions: an accelerated development of media systems, with fast-growing audiences of privately-owned, commercially-targeted TV stations (Hallin and Mancini, 2013), a lack of solid party alignments, reflected in rather fluid structures of political cleavages and high volatility rates (Evans and Whitefield, 1993), inexperienced electorates, unprepared to deal with the complexity of political issues after the fall of communism (Lewis, 2002), and a rapid internationalization of politics due to NATO or EU accession issues. Added to the historical tradition of almighty leaders in the region, all these developments may have facilitated a 'fast track' towards leader-centred electoral politics in post-communist CEE countries, in contrast to the traditional evolutions in Western democracies.

The personalization literature is rather discordant in assessing the magnitude of leader effects on the vote (3): while some studies discuss big and growing effects across recent elections (Costa Lobo, 2006; Mughan, 2000; Stewart and Clarke, 1992), others identify small effects or inconsistent evolutions (Dinas, 2008; Holmberg and Oscarsson, 2011; Schoen, 2007). At least up to a certain point, such variations in results might rely on differences in the conceptual definitions and methodological strategies employed (Barisione, 2009). But an increasing number of voices in the literature suggest that, in order to understand better the variations in the magnitude of personalization, a closer look at the 'conditionality' of leader effects is needed (Barisione, 2009). In other words, the research on personalization should give more consideration to the conditions mediating the manifestation of leader effects, acting as stimuli or inhibitors. In a theoretical article, Barisione (2009) discusses four categories of such conditions: structural constraints (institutional, political, territorial and media contexts in which elections take place), political opportunities (opinion climate, economic situation, systemic crisis and campaign environment), individual moderators (availability of leader-oriented segments of voters) and image variables (features of the leader). Aardal and Binder (2011) discuss three large categories of conditions: individual factors (either linked to voters or candidates), system-specific factors (party systems and electoral laws) and party-related factors (type, functional characteristics, organizational structure and ideological platform).

In spite of such a diversity of possible pathways of investigation, rather few of them have been empirically explored. This article aims to bring a contribution to the understanding of how individual voter characteristics moderate the occurrence of leader effects in legislative elections. It investigates the influence of objective political knowledge, subjective political information, party identification, political engagement and time of vote decision on the mechanisms of personalization in the context of the 2012 legislative elections in Romania. The case of Romania is particularly challenging for investigating the conditionality of leader effects in post-communist polities for several categories of reasons. First, it fits perfectly the 'fast track' profile of electoral personalization for CEE post-communist countries (for an extensive discussion, see Gheorghija, 2014). Second, it is a semi-presidential system, with a directly elected President, hence more likely to experience stronger leader effects (Curtice and Hunjan, 2011; Curtice and Lisi, 2014). Moreover, given the unclear power settlements between the President and the Prime Minister, top political leaders are even more visible during cohabitation periods. Third, the 2012 Romanian legislative election provides an excellent opportunity to investigate simultaneously the positivity and negativity scenarios of leader effects on vote choice, given the particular one-leader-centred electoral context.

For the purpose of investigating the influence of voter characteristics on the magnitude of leader effects, this article employs data from the Romanian Election Studies (RES) three-wave panel survey. It starts with a review of the previous findings in the literature on the effects of voter characteristics on the personalization of electoral behaviour. Then it introduces the general context of the 2012 legislative elections in Romania. The following sections are dedicated to the methodological aspects of the research. The main findings and conclusions are then discussed extensively.

Voter characteristics and leader effects in the literature

The assumption of working with a heterogeneous electorate (Bartle, 2005) is frequently ignored in the study of personalization of electoral behaviour. However, it is quite reasonable to believe that some voters place more weight on the leaders than others in making electoral decisions: for some of them, leader evaluations might be the main thing to consider, while for the others they might be more or less irrelevant (Gidengil, 2011). Some recent studies have explored the degree to which several individual-level variables do moderate the magnitude of leader effects on vote choice: political sophistication, political engagement, political indecision, party identification, and exposure to television news. The findings were frequently inconsistent, leaving a lot of room for debate on the comparability of measures or on the importance of the context of elections. Nevertheless all these studies addressed the issue of leader effects in Western democracies. Let us have a look at the main findings.

The conventional wisdom used to place political engagement in a negative relation with the personalization of electoral choice (Campbell et al., 1960; Nie et al., 1976): leader evaluations should matter most to voters who are less politically aware, as it is a less sophisticated behaviour. However, subsequent research has cast serious doubt on this assumption. Gidengil et al. (2000) find that leader effects tend to be strongest among the Canadian voters who are the most interested in the election. Similarly, leader evaluations appear to weight more for the Swiss voters with higher levels of involvement (Lachat, 2014). In a comparative study, Gidengil (2011) comes out with mixed findings: Canadian and Dutch voters who are more interested in the elections are more likely to weigh the leaders in their vote choice, while it is quite the reverse for the American voters.

Things are rather similar in the case of political sophistication, traditionally regarded as an inhibitor of leader effects: sophisticated voters (more knowledgeable, more educated) will make choices based on political issues and policy concerns, while the less sophisticated ones will pay more attention to candidate images (Converse, 1964; Popkin and Dimock, 1999). However, many influential studies do contradict this classical assumption, arguing that highly educated/knowledgeable individuals are more concerned with candidate personal attributes than less educated/ knowledgeable ones (Bean, 1993; Clarke et al., 2013; Glass, 1985; Kroh, 2003; Miller et al., 1986; Rico, 2014). This trend appears to be a consequence of the fact that sophisticated voters use leader personal qualities as a guide to how he/she will perform in office and, consequently, are more focused on performance-related characteristics (Brown et al., 1988; Miller et al., 1986). There are also studies who formulate 'softer' conclusions, arguing in favour of little or no effects of political sophistication on the salience of leader evaluations (Gidengil, 2011; Pierce, 1993). An additional perspective comes from Catellani and Alberici (2012), who investigate the effect of political sophistication in interaction with the time of voting decision: in the Italian case, high-sophisticated early deciders appear to rely more on the challenger traits, while low-sophisticated late deciders rely even more on the incumbent.

Party identification is widely regarded as following a negative...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT