A Two-Dimensional Approach to the Political Opportunity Structure of Extreme Right Parties in Western Europe

Abstract

Previous studies on the electoral fortunes of extreme right parties (ERPs) have pointed to the importance of variables of party competition for the success – or failure – of ERPs. These studies vary greatly when it comes to describing the political opportunity structure of the extreme right. Apart from their methodological differences, existing studies differ especially with regard to the assumed underlying dimension of party competition. This article tests the impact of three frequently discussed variables in the political opportunity structure of ERPs (mainstream party convergence, position of the established right and party system polarisation) on the vote share of ERPs in Western Europe. In addition to examining previous studies in this field, it focuses on the interplay between the economic and the cultural dimensions as part of the political opportunity structure. The authors show that a decrease in polarisation with regard to economic questions is accompanied by a growing salience of ERPs' core issues, leading in the end to an increase in ERPs' vote share.

Acknowledgements

We are grateful to André Kaiser, this journal's editors Peter Mair and Klaus Goetz, and two anonymous reviewers for their helpful comments and constructive suggestions. An earlier version of this article was presented at the ECPR General Conference in Potsdam 2009. We want to thank all participants for their valuable feedback.

Notes

1. Arzheimer and Carter (2006) refer to these variables as long-term, medium-term and short-term contextual factors.

2. Note that our approach differs in important aspects from Meguid's (2008), although both approaches see the strategic decisions of parties as decisive. While Meguid (2008: 22–40) clearly separates left–right positions, issue ownership and saliences from each other as strategic tools, we claim that emphasising issues always has an effect on the positioning of a party. Hence, we do not separate these two strategic tools. Furthermore, Meguid (2008: 12–13 and 23) is sceptical about the explanatory power of sociological approaches and assumes that parties create new policy dimensions. Although we share her view of the importance of actors' behaviour in analysing party competition as well as in the activation of cleavages by parties, we do not dismiss sociological findings to the extent that she does. We assume that the evolution of ‘new’ policy dimensions can be explained by value changes in the society (Inglehart 1984; Flanagan and Lee 2003), but for our analysis, this is an exogenous fact which parties can use strategically and is therefore part of the political opportunity structure.

3. It should be noted that it might be the case that the ERP is not the most extreme party of the right. In 75 cases for which we have data for both the position of the ERP as well as the mainstream right party, this is the case in 18 (left–right dimension), 17 (economic dimension) and 15 cases (non-economic dimension). Nearly exclusively, this scenario is given in Flanders (Vlaams Blok), Italy (Lega Nord) and Finland (True Finns). Interestingly, these ERPs are often perceived as somehow borderline-cases of ERPs.

4. Since 1978, no party has competed for votes in both regions of Belgium in the national elections.

5. The special mode of government formation in Switzerland (Zauberformel) ensures that the five largest parties become members of the federal government. This makes it problematic to identify two mainstream parties, since in the Swiss case, one can plausibly speak of five mainstream parties.

6. Using a Tobit model is theoretically appropriate only when one assumes that the variables that influence ERPs' vote share also influence the existence or non-existence of an extreme right party (Golder 2003). While we know of only one study that directly addresses this question (O'Malley 2008), we want to claim that the theoretical argumentation for a Tobit model is convincing. Alternatively, we have also estimated an OLS model without these 23 elections, as other studies have done (Abedi 2002; Carter 2005; Meguid 2005), and the results are to a large extent comparable.

7. The idea of this approach is to take all out of the raw data that are not positional. Using this approach, first confrontational and non-confrontational issues are identified. In a second step it is determined whether a confrontational issue is left or right. Then the raw CMP scores itself are transformed. A minimal value is substracted for each election and party system in order not to overestimate issue stances that each party in a particular country and in a particular election does. The scores of these differences will be summed up and transformed to an 11-point scale, leading us to determine party positions between 0 (leftmost) and 10 (rightmost). Using moving averages as smoothing procedure, the raw value of a manifesto before and after the particular election is used for error correction. This approach reveals not only sufficient reliability scores (Dinas and Gemenis 2010), but also a high construct validity compared to voters' left–right positions in mass surveys (Franzmann and Kaiser 2006; Franzmann 2009). Data are available online: http://www.uni-potsdam.de/db/ls_regierungssystem_brd/index.php?article_id = 498&clang = 0.

8. The formula reads:

where p_i is a party's vote share at a given election, _X is this party's position on a given dimension, and is the weighted mean of the positions of all parties on this dimension. is calculated by multiplying the vote share by the ideological position of a party. This step is repeated for each party in the party system and finally, these values are summed up.

9. We obtained our data on international migration flows from the OECD Database on International Migration (http://www.oecd.org).

10. While the impact of the disproportionality of the electoral system has been analysed extensively by different authors (Carter 2002; Golder 2003; Norris 2005), only Golder's study also analyses elections where no ERPs were present. While Golder finds evidence that ERPs actually benefit from more proportional electoral systems – the opposite of our finding – his period of analysis is from 1970 to 2002. Reanalysing his data, we found that the statistical significance of his (dis)proportionality variables is due to the inclusion of the elections before 1980 and turns out to be insignificant (but still positively correlated) for the period from 1980 to 2002.

11. This consideration suggests that in Germany, the rise of the Linkspartei and the dominance of economic issues after reunification have prevented the German party system from developing a successful ERP at the national level, despite having a considerable share of voters within the electorate who support ERP policy positions – especially in eastern Germany and in parts of the south. The comparatively high level of post-materialist attitudes among the German electorate reveals that there is strong potential for a rise in the importance of non-economic issues for the policy agenda. Given that the German mainstream parties – Christian Democrats and Social Democrats – have narrowed the gap between their policy positions on the non-economic dimension (Franzmann 2008), we expect that with a further reduction in polarisation on the economic dimension, the probability of a successful ERP will rise.