Unified voters in a divided society: Ideology and regionalism in Belgium

ABSTRACT

In societies divided along ethnocultural lines, intergroup cooperation can often be a challenging task. This process can be even more complex if political parties and voters are divided along those same social cleavages. This study focuses on the case of Belgium and explores whether divided societies with separate party systems necessarily lead to distinct partisan alignments. Using electoral survey data from the 2014 Belgian federal election, we investigate whether political ideology is stronger than ethnolinguistic group membership in shaping electoral behaviour. The results demonstrate that although Belgian voters are divided along linguistic lines when it comes to preferences about centralization, they remain aligned along party families on social and economic dimensions.

KEYWORDS:

• Social cleavages
• ideology
• regionalism
• elections
• Belgium

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

1 There are also German-speaking parties, who stand for elected office in the Parliament of the German-Speaking Community.

2 Designed and supervised by the PartiRep team, the survey was carried out in two (pre and post-election) waves in both Flanders and Wallonia using geographically stratified sampling (in terms of population size) according to the Belgian National Registry. The respondents were questioned on their political attitudes and voting choices in the federal, regional and European elections, which were administered simultaneously on May 25th, 2014.

3 There are also far-right parties, the Dutch-speaking Vlaams Belang (VB) and the French-speaking Front National, that are generally not very successful in elections. However, in the recent 2019 federal and regional elections, the VB had an important surge in votes; making them the second party in terms of vote share in both the Federal and Flemish legislatures. Though it would have been interesting to explore the determinants of vote choice for the VB in the current study, the dataset only includes 21 respondents who indicated having had voted for VB. As for the far-left, the Workers’ Party of Belgium (PVDA/PTB) functions as a single party and fielded in 2014 a common list in all 11 constituencies. While also traditionally not very popular in elections, the party was relatively successful, especially in Brussels and Wallonia, in the recent 2018 provincial and local elections.

4 There are examples of, active and defunct, Walloon regionalist parties; for instance: Rassemblement wallon, Front pour l'Indépendance de la Wallonie, Rassemblement populaire wallon, and the irredentist Rassemblement Wallonie-France. As for the Démocrate Fédéraliste Indépendant party (DéFI; formerly known as FDF), it has traditionally been more of an ethnonationalist party focused on the interests and rights of Francophones in in Brussels and its periphery (in Flemish territory).

5 This technique has been shown to be preferable to gauge ideological positioning than self-reported questions (Treier and Hillygus 2009).

6 We recognize that these questions form more of a social diversity acceptability scale rather than represent the complete dimension regarding social values. This follows the logic of Gauvin and colleagues (2016).

7 Although economy and centre-periphery do load on their own factor with an eigenvalue very close to 1, a reliability analysis shows a Cronbach’s α score of 0.06; this is well under any sufficient score and supports our expectations that these two items are not part of the same dimension as well.

8 Using Canadian Election Study data from 2015, a difference in means test using weighted data reveals that Quebecers hold an average of 0.48 on the left-right scale compared to 0.50 for the rest of Canada, with a p-value of 0.076 (n = 1,243).

9 The question for the ‘Most important political problem’ in the Comparative Candidate Survey for Belgium was open-ended. We therefore coded the candidates’ responses into the following categories: State reform, economy, social, environment, Europe, and other.