Organized after an almost complete term of office, but the end of which was marked by the resurgence of the community-linked problems and by the departure of the Ministers of the Volksunie, the parliamentary elections of 24th November 1991 will remain characterized by the punishment inflicted by apart of the voters, not only on the majority's parties, but also on the traditional parties as a whole. The opposition of the dissatisfied voters did not show itself either in a reduced participation to the vote, a rather normal phenomenon in a country where voting is compulsory, or in a rise of the blank and spoilt ba/lot papers. The 1991 figures are, in these respects, very similar to the average figures of the last thirty years. The opposition was first characterized by high variations in the choice of the voters, which dit not, however, exceed the size of the movements noticed at previous elections. The rate of external mobility, as computed by us, is ranked fifth among the 22 levels that have been counted since universal suffrage has been introduced in Belgium. In 1991, this rate was the highest in Brussels, which is traditional, but it was also considerably higher in Flanders than in the Walloon Region, which quite seldom happens. Then, the opposition found expression in a setback for the Jour ruling parties. This set back amounted to about 8% and even 10% if account is taken of the Volksunie which was part of the Government until the very last weeks before the dissolution of the Houses of Parliament. Such a setback for an outgoing majority is not exceptional; a more unusual phenomenon lies in the f act that this decline was not profitable, on the whole, to the third traditional "family", i.e. the Liberals. As a result of this simultaneous setback for the three traditional families, these total only about 70% of the valid votes, which is the worst result of the whole Belgian electoral history. The political formations that are progressing are the Ecologists, on the one hand, and the far-right lists, on the other hand. Ecolo improves its results considerably in the Walloon Region and in Brussels, without however reaching the level obtained at the 1989 European elections, while Agalev only registers a slight progress.On the contrary, the winner of the elections in the Flemish Region, is unquestionablythe Vlaams Blok, as well as the "Rossem" lists that draw some 5% of the angry voters. If the far-left trend has almost disappeared from the Belgian political scene, the far-right parties, on the contrary, are making a breakthrough on it. The current made up by the Vlaams Blok and the openly far-right lists wins a bit less than 8% of the votes in the whole country. It is likely, however, that those who voted for the Vlaams Blok do not all adhere to the far-right ideas, but that some of them are attracted by the Flemish autonomist stands. The real electoral power of the far-right parties can then be assessed at 6 to 7%, which is much more than in 1987, hut does not make a record in comparison with other European countries. The votes of opponents without any clear political orientation, the bulk of which is won by the "Rossem" lists, amount to some 3%, which is new for Belgium. In a country where voting is optional, many of these voters would probably have stayed at home. This being so, and as f ar as these concepts still remain meaningful, the Belgian electoral pendulum shifted some 5% towards the right, at the expense of the left for more than 3% and the centre for a bit less than 2%. As far as the Parliament is concerned, the situation is clear in the Walloon Region and in the French-speaking Community where the Socialist Party, by far the most important party, is almost inevitable. It is however much more vague on the Flemish side, where the CVP's setback and the dispersion of the polical farces make several types of coalitions possible.

At the general elections in Belgium, the voter bas the possibility to bring out a vote fora party or a vote for a candidate of the party (a preferential vote). At the general elections of November 24, 1991, for the House of Representatives, the voters have voted for 48% by preference, whereas, for the Senate 41 % of them have done so. The evolution is stagnating. The use of preferential votes is varying from one electoral district to another: from 62% in the district of Tongeren-Maaseik to 34% in Mechelen. There is also a difference from one political party to another. The highest percentage is obtained by the christian-democratic parties: 58%; the lowest by the last remaining communists: 27%. The preferential v'ótes individual candidates obtain differ widely. This is not only a result of their popularity, but also of their political position, rendering of services and media-attraction. Preferential voting is characteristic to politically conscious voters.

Voters probably do not take a series of relevant issues into consideration but rather vote on the basis of the few issues that happen to be on their minds at the moment of the decision. Issue salience, i.e. the availability of issue-schema's, is to a certain extent determined by the political communication during the election campaign. A content analysis of the debates during the 1991 campaign shows that the socio-economic issues, the ethnic issue and the immigrant issue were on top of the agenda. A similar analysis of the party political broadcasts and the ads in the national newspapers indicates that the parties tried to focus the campaign on the socio-economie issues (christian-democrats, socialists), the functioning of the political system (socialists, liberals) and to a much lesser extent on the environmental issue (greens) and the communal issue (left-wing and right-wing Flemish nationalists). Only the latter nationalist party attempted to prime the immigrant issue. Survey data show that this issue was exceptionally salient in the electorate, as were the ethnic issue and the political system issue.

In spite op compulsory voting, the number of non-voters increased at the last general elections in Belgium to 7.3 per cent. This evolution can largely be explained by demographic factors. The number of blank or invalid voters reaches nearly the same level. Concerning this form of political non-participation, we noticed considerable differences occur between the types of elections (local, provincial, House, Senate, European) due to factors such as the importance and the proximity of the proper institution, the social distance between candidate and citizen and the main issue of the elections. The analysis of both phenomena over time at the level of the individual voters based on surveys shows that electoral absenteeism is rather accidental, white blanc and invalid voting is more permanent. From the analysis of absenteeism and blanc and invalid voting emerge on the one hand socio-economic factors (like age, gender, professional activity, income, marital status and family composition) which jeopardize electoral participation in a direct or indirect way, and on the other hand, attitudes and behaviour reflecting political indifference, alienation and hostility (like low levels of political interest, information, knowledge, satisfaction, party identification, and participation).

This study evaluates the impact of new voters and deceased voters in the statistical analysis of shifts in voting behaviour from one election to another. Conclusions about the shifts between the political parties from one election to another are ajfected by the fact that at the time of the analyzed election, the population of the former election bas substantial by changed. In order to fix the marginal distributions of the turnover table, estimations should be made about the distribution of the new voters and the deceased voters over the political parties. Information about the new voters can be extracted from the sample in use. Estimations of the last voting behaviour of the deceased are based on prior surveys. The conclusions about the mutual shifts between parties are affected to the degree that the parties differ according to the age distributions. The method has been applied to exit poll datafrom the 1991 general elections in Limburg.

On the 24th of november 1991 the Belgian voters elected the 716 members of the nine provincial councils. The socialists are the biggest losers of this election, with the Volksunie as a close second. Also the Christian Democrats suffered a serious decline, mainly caused by the loss of the CVP in Flanders. The electoral gain of the Flemish Liberals is neutralized by the decline of the Liberal party in Wallonia. The Greens gain 32 seats, the Far Right 35. These national aggregates hide striking regional differences. The national success of the Green is mainly due to the spectacular growth of Ecolo in the Walloons. The success of the Far Right is the sole result of a multiplication of votes for Vlaams Blok in Flanders. These results show that both the Flemish and the Walloon voters have sanctioned the traditional parties in a similar way. They opted, however, for totally different alternatives: the Flemish for the Far Right, the Walloons for the ecologists. The outcome of the provincial elections in the bilingual province of Brabant neatly mirrors these tendencies. In 1991 the outcome of the provincial elections showed a profile quite different from that of the national elections which were held on the same day. This is explained by the fact that the Flemish party Rossem, which won 3.2% of the votes, only ran for the national elections and not for the provincial elections. Hence, a considerable difference in voting behaviour on the national versus the local level. The comparison of the results of the national elections with those of the provincial confirms the claim that smaller parties generally score better at a lower level. In 1991 it can, however, not be said that the bigger parties did better on the national level.

The four-yearly social elections in Belgium took place for the tenth time in june 1991. The results show a renewed victory for the christian trade union ACV; all the other unions and lists lost. The socialist union ABW, which was once the most powerful in Belgium, felt the decline most strongly. It could maintain a narrow relative majority in votes in Wallonia but was outnumbered in seats by the ACV. For many years Wallonia was considered to be a "red bastion". Another important shift was that the three "general" unions together gained the majority of votes of the executive staff members, at the expense of their "specific" lists: those of organizations and groups representing only members of the executive staffs. These elections showed a further decline in attendance, in spite of all the propaganda made by the Ministry of Labour, urging the employees to take part in the elections.