Representation by election is the widespread formal pattern of democratic representation. Our research under this heading is very much inspired by the increasing criticism of electoral representation and by the question whether the electoral and partisan representation processes are able to properly catch the varying and changing demands and expectations of divided, multileveled and multicultural societies. There are several projects that are or have been conducted under this general label.
We concluded and published in 2014 the results of a large comparative survey among MPs in 15 countries and in 73 statewide and substate parliaments, evaluating – among others – the degree to which individual MPs can make individual choices on how to represent the people, or are, to the contrary, very much affected by the institutional environment in which they have to function (like the electoral system or the level of government). Our research confirmed the latter reality was predominant. The data gathered with this international survey have also allowed us to shed light on the variation in representational styles and focus between levels of government, and to get a better understanding of representational roles, of constituency orientation and of the balance between the seeking of partisan or personal votes (see also below).
The representation of diversity is quite central in our research. We have worked extensively on the electoral mobilization of regional identities, with particular attention to linguistic identities in Belgium. Most work in this respect has been done with survey data, but we have also successfully used mental maps – asking people to draw their country – to evaluate the way in which they perceive the territorial organization of politics. A similar project is now in progress in Cyprus.
Against the background of increasing criticism of the traditional political parties, we have paid attention to the rise and success of new political parties that try – with varying success – to respond to representational demands left open by the traditional parties. We have been able to show that the mortality rate among new parties is rather high, and that survival chances depend very much on the degree to which new parties can rely on existing organizations.
We are also analyzing (in a project started in 2015) the way in which voters with Eurosceptic attitudes behave electorally in a country – Belgium – where the Eurosceptic position is not offered by the political parties. While radical left and radical right parties offer Eurosceptic programmes and can attract the votes of the radical electorate, centre voters with Eurosceptic attitudes have no party to vote for.
An often mentioned indicator of the weakening link between parties and voters is the increasing personalization of politics. It means that voters who search for good representation do so more by voting for candidates rather than for parties. We have several projects that analyze both for Belgium and comparatively the evolution of the alleged personalization of politics.
The work of Tom Verthé focuses on strategic voting i.e. assessing the degree to which voters choose for the party that truly represents their interests or rather make a strategic choice for a party that has better chances of winning. Strategic voting has in the past mostly been studied in environments that were conducive to (relatively) easy model construction and testing, namely elections with single member districts in plurality voting systems. This means that the large majority of studies on strategic voting have nothing to say on the most common democratic format: multiparty (i.e. more than three) parliaments and coalition governments. Because of the necessity for coalition formation the question of winning and losing has a different meaning for parties and voters and, furthermore, requires more complex strategies than the relatively simple classic wasted-vote logic (abandoning the hopeless preferred party for the second party with better chances). The last few years Tom Verthé has worked on several projects that are related to strategic voting and coalition formation in Belgium, a complex and highly fragmented proportional multiparty system. The overarching research question that ties the three projects together is: to what extent can coalition and threshold insurance voting explain strategic voting in Belgium?
A first exploration of the complex strategic motives of voters was an analysis of voting motive questions that were part of an Exit Poll that we organized for the 2012 Belgian municipal elections. From this data we learnt that a substantial part of the voting population in fact attributes strategic considerations to their vote choice. Based on this evidence we added a question to the 2014 surveys for the Belgian federal and regional elections in order to investigate the link between voters’ perceptions of parties’ coalition potential (i.e. government viability) and their vote choice. Our analyses showed that there is indeed a significant and substantive positive effect of parties’ perceived government viability on vote choice. A third project used an online experiment in the period leading up to the 2014 Flemish regional elections to investigate the effects of parties behaving as agents who try to (positively) influence their perceived coalition potential by sending coalition signals to voters. The results indicate that these signals have an appeasing effect on voters in their choice between parties they like and therefore increase the rate of sincere voting among their supporters.