This article explains the ebb and flow in Belgian trade union membership from 1946 to 1995 by replicating the econometric model by Bain and Elsheikhn in which changes in macro-economic variables are highly significant. Since the automatic indexation of wages and the extension of collective labour agreements invite free riding, the relevance of the change in inflation and real wage is quite striking. However, the free riding-effect is slowed down by the institutionalised presence of the trade unions on the work floor. The Ghent system explains the positive impact of the unemployment rate. The model is furthermore improved by the trade union density as a structural variable. The linear form reflects the enforcement effect, while the quadratic form mirrors the saturation effect on the trade union membership. Mainly due to the 'Allgemeinkoalitionsfähighkeit' of the Belgian government system, the impact of left parties on union growth and decline is not significant in a quantitative framework. With only four explanatory variables the model clarifies more than 75% of the fluctuations in Belgian trade union membership.

Despite the fact that various authors have expressed concern about a general decline of civic engagement in Western societies, other indicators portray a transition from traditional and formal participation formats to more informal participation forms. This replacement thesis, however, entails the question whether these new forms can still be regarded as a form of political participation. The Alternative Food Circles in Belgium can be considered as a typical grass-roots example of 'political consumerism', which is portrayed as a contemporary alternative for institutionalised politics. In a member survey, 163 members of the Circles were questioned about their motives to participate. They clearly paid little attention to influencing the political system, but notions of solidarity and social change were clearl y present. This form of political consumerism therefore cannot be considered a full form of political participation (using an institutionalist definition of 'politics'), but it clearly is a form of 'life style politics' (Giddens).

Since 1989 candidates in Belgian elections have to declare their campaign expenses, but there is a widespread doubt as to the validity of the declared expenses as an indicator of the real cost of the campaign. After the 2000 local elections, data were collected concerning the campaign expenses of 2762 candidates spread over 134 municipalities in the Flemish region of Belgium. An analysis relating the declared expenses to a number of predictor variables reveals a sensible and systematic pattern, which can be interpreted as an indirect indication of their validity. There is a limit on spending, but the research shows that, on average, the candidates only spend 25% of what they are allowed to. Controlling for a number of relevant background variables, the campaign expenditures were found to have a substantial effect on the preference votes for the candidates.

Gender quota do not always go hand in hand with a considerable rise in the number of women elected. Although the number of fe/male representatives elected depends on several factors, we argue that the stipulations of gender quota acts influence to a large extent their impact on the sex ratios in politics. This is not so much due to the share of fe/male candidates parties have to present than to the extent to which a gender quota act anticipates the particularities of the electoral system. A comparative analysis of three prominent cases, the Argentinean, Belgian and French gender quota acts, shows that the more a gender quota rule targets at the specificities of the electoral system in order to promote a gender balance in political representation, the more this rule 'guarantees' a result, even though gender quota acts concern but the input side of the electoral process.

Flemish community senators hold seats in both the Flemish Parliament and the Belgian Senate. A lot of expectations are related to their specific mandate. On the basis of our analysis, though, we come to the conclusion that the community senators do not meet these high expectations. While one would expect Flemish community senators to be very active in putting forward parliamentary questions concerning deliberation and cooperation between the different governments, this was clearly not the case in 1995-1999. 'Community senators' in other words, do not exist. They do not act like a community senator, and according to the interviews, nor do they feel like one. Prime Minister Verhofstadt, who wants to reform the Senate and revalue the role of the community senators by turning every senator into a community senator, is a warned man. In the future, the concept of community senator will only work if significant structural changes are made.