European Elections and Domestic Politics: Lessons from the Past and Scenarios for the Future

4. Comparing the views of parties and voters in the 1999 election to the European Parliament

F O U R Comparing the views of parties and voters in the 1999 election to the European Parliament andreas m. wüst and hermann schmitt The increasing transfer of policy-making powers from the member states to the European Union during the past decades has generated relatively little political controversy in most of the member states. The exceptions to this general pattern—particularly Denmark and Britain, to some extent also Sweden—indicate that this state of affairs cannot be regarded as a ‘natural’ development. Both the general trend as well as the exceptions raise important questions with respect to the role of the electoral process in European integration and its democratic character. Addressing these questions requires, at the very least, a focus on political parties, on voters, and on the electoral connection between them. This chapter deals mainly with political parties, which are among the most important actors in the process of politicization of issues. Where do parties position themselves on the issue of European unification, and how are these positions perceived by voters? To what extent do parties in the various member states offer voters a choice between different views on the future of the EU? The question of political representation—how well voters’ attitudes toward European unification are represented by parties—will be discussed briefly in this chapter, and will be assessed in more detail in chapter 10. 73 chapter The analysis of parties’ positions in this chapter will be based on data from a systematic analysis of the contents of their manifestos for the European elections (appendix C). The question of how adequately voters perceive parties’ positions will be addressed by juxtaposing parties’ positions as derived from their manifestos with voters’ perceptions as derived from the voter survey of the European Election Study (appendix A). We use party manifestos as a source of information to gauge parties’ positions on various issues. Manifestos are authoritative statements of party policy. In spite of the facts that they are usually drafted by small groups of specialists within a party and that they are issued prior to an election so that they do not address issues that will become salient at later times, it has been established that what parties advocate in parliament and what parties do when in government is indeed to a large extent in line with the contents of their manifestos for national elections (Klingemann, Hofferbert, and Budge 1994). Thus, election manifestos constitute an important source of information on parties’ input into the electoral connection, and it is therefore that they take a prominent place in any ‘extended design’ of comparative electoral research (Thomassen 2000). Their relevance as sources of information on party positions is not diminished by the fact that they are usually hardly read by voters, although this observation prompts the question of how manifesto content and voter perceptions of party positions are related.1 Empirical analyses of party manifestos must be based on a systematic coding of the contents of these documents. This can be done in different ways. Probably the most prominent strategy is to code the issues raised, or referred to, in the manifesto. Statistical analyses can then determine how often (in absolute and relative terms) certain issues are mentioned, and the analyst can use this as a proxy for the party-specific saliency of these issues. This is in a nutshell what the multilateral Manifesto Research Group (MRG) does. The beginning of this research group dates back to the mid-1970s (cf. Robertson 1976). By that time, a coding frame was conceived that later was further developed and utilized to generate empirical information on the content of election programs of parties in twenty-five countries from 1945 onward (Budge et al. 2001). The MRG scheme focuses on the emphasis that parties put on various issues as manifested by the number of arguments devoted to them (see also the next section).2 All in all, MRG coding provides frequencies for arguments in fifty-six coding categories which are grouped into the following seven policy domains: external relations, freedom and democracy, the political system, the economy, welfare and quality of life, the fabric of society, and social groups. 74 Andreas M. Wüst and Hermann Schmitt In this chapter we report first findings of a research project which closely follows the footsteps of the MRG. In this project the experience and the particular approach of the MRG research group has been...