Voters in the predominantly Dutch-speaking Belgian region of Flanders are often assumed to be more conservative than those in predominantly French-speaking Wallonia. Yet as Daan Delespaul shows, there is reason to question whether the two regions constitute distinct political worlds.
The structure of western European party systems has long been shaped by sticky political divides – termed cleavages – that originate in longstanding group conflict. Historically, these cleavages were based on a small number of social divisions (mainly on class and religion) that caused stable and durable alignments between parties and voters.
Belgium has always been characterised as an archetypal case of such structuration along socio-structural conflicts. In this context, three mass parties – Christian democrats, social democrats and liberals – managed to dominate political life by integrating their constituents in several “pillar” organisations.
From the 1970s onwards, however, western European party systems entered a state of flux. Increased prosperity, rising education levels, secularisation and the erosion of old class structures undermined political bonds that were forged through strong collective identities of modernity.
At the same time, cultural and economic globalisation led to diffuse feelings of threat for low-skilled workers, who experienced fewer opportunities for social mobility. The result of this development is the emergence of a new cultural cleavage, revolving around ideological differences of “universalism” and “particularism”.
The formation of this new cleavage is linked to the recent success of green and far-right parties. Belgium has been a frontrunner in many of these developments. In Flanders, the Dutch-speaking part of Belgium, the green Agalev and the far-right Vlaams Blok managed early successes in the 1980s.
Although a green party also emerged in Francophone Belgium, the success of far-right parties proved far more limited: both the far-right Front National (FN), and the right-wing populist Parti Populaire (PP) dipped below the electoral threshold after an initial breakthrough. In line with this pattern, the Walloon far-right party Chez Nous managed only a disappointing 2.83% of the vote in the most recent regional election in June 2024. As a result, the Francophone cultural cleavage has remained only half mobilised.
The absence of the far right in Francophone Belgium means the country now forms an exception to the ubiquitous processes of political change observed elsewhere. Simultaneously, it raises the question whether the new cultural cleavage has materialised in both parts of the country. Indeed, previous research suggests cultural conflicts in Francophone Belgium have been overshadowed by a stronger class cleavage. Since Francophone Belgium is comparatively less wealthy than Flanders, has a higher class consciousness and has historically tended to vote for left-wing parties, there is much to suggest that this hypothesis holds water.
Belgium, an asymmetric political space?
In a recent co-authored study, I investigated this potential asymmetry in the Belgian cleavage structure by comparing the ideological structure of Flemish and Francophone electorates. Our starting point is the relational method of (class-specific) multiple correspondence analysis (CSA/MCA), which was used to extract coherent ideological dimensions from 17 salient political issues among voters (based on a post-electoral survey from 2020). This analysis is based on the demand side of politics and thus informs us of ideological conflicts in both linguistic groups.
Figure 1: The political space of Flanders
Note: The chart shows a cloud of active categories with above-average contributions for the Flemish subspace. For more information, see the author’s accompanying paper in the European Political Science Review.
Figure 2: The political space of Francophone Belgium
Note: The chart shows a cloud of active categories with above-average contributions for the Francophone subspace. For more information, see the author’s accompanying paper in the European Political Science Review.
The results of this analysis are projected in figures 1 and 2, which visualise the political spaces of Flanders and Francophone Belgium. Categories lie close to each other if they were selected by the same respondents. Two key ideological dimensions emerge in both parts of the country: a cultural and an economic one. The cultural dimension combines issues of anti-immigration, euroscepticism, welfare chauvinism and morality. On the economic dimension, these are issues related to social benefits, fiscal policy, international trade and the organisation of the welfare state.
In broad strokes, these analyses tell us two things. First, correlations between the dimensions in both figures are high, which means that cleavages are likely composed of similar issues in the two linguistic groups. Nonetheless, analyses of the relative contribution of the categories show some differences.
The Flemish economic dimension centres slightly more on questions of “who” deserves access to the welfare state (such as the country’s “own people” or “hard workers”) while Francophones are more polarised about “what” and “how” to distribute. It is also notable that questions of state reform (“communitarianism”) matter greatly to the Flemish cultural dimension, but not at all to any of the two dimensions in Francophone Belgium.
Next, we see that the polarisation encapsulated by these dimensions differs marginally between the two electorates. Flemish voters are more polarised around cultural attitudes while Francophones are less divided by economic issues. Because Francophones are also economically more left-wing, there appears to exist a welfare consensus that obstructs a breakthrough of the far right.
Party agency and social structure
We can learn from these observations that the differences between Flanders and Francophone Belgium are not solely due to contextual factors but are also anchored in public opinion. In a second step of the analysis, we looked at two additional factors of potential asymmetry: social structure and voting behaviour.
Figure 3: Region-specific cleavages and electorates in Flanders
Note: For more information, see the author’s accompanying paper in the European Political Science Review.
Figure 4: Region-specific cleavages and electorates in Francophone Belgium
Note: For more information, see the author’s accompanying paper in the European Political Science Review.
Again, two conclusions can be drawn from a projection of these factors. First, statistical tests indicate that only three structural characteristics are significant in the political space. These are education, income and sub-national identity (feeling more Flemish or Francophone than Belgian). The influence of these social divisions is roughly equal in both parts of the country, although the linguistic cleavage carries far greater weight in Flanders.
Second, we find that working-class voters are positioned notably closer to social-democratic electorates in Francophone Belgium (PS), than in Flanders (sp.a). This difference in structural alignment is substantial, but it should not be surprising given the historically different electoral strategies pursued by the two parties. While the sp.a has pushed a “third way” agenda aimed at middle-class socio-cultural professionals, the PS has remained closer to core principles of social insurance favoured by the working class. By consolidating this working-class vote, the PS appears to have been able to create an effective buffer against the far right.
The Francophone cultural cleavage
All this suggests that a cultural cleavage has materialised in Francophone Belgium – albeit in a different form to the one in Flanders. Yet, the politicisation of this new divide seems to have been obscured by more salient class issues that are fuelled by the social insurance agenda of the social-democratic PS.
The Belgian case thus indicates that national cleavage variations arise through the combined influence of ideology and party agency. This recursive model of cleavage formation ensures that political realignment only takes place when new issues and conflicts are articulated by parties or politicians.
To conclude, the main implication of these analyses is that Flemish and Francophone electorates do not necessarily form separate political worlds. Instead, the asymmetry in the Belgian political space is the product of different party strategies and the welfare consensus espoused by Francophone voters.
For more information, see the author’s accompanying paper (co-authored with Cecil Meeusen, Koen Abts and Marc Swyngedouw) in the European Political Science Review.
Note: This article gives the views of the author, not the position of EUROPP – European Politics and Policy or the London School of Economics. Featured image credit: European Union
Discussion about this post