Voter–Representative Congruence in Europe: A Loss of Institutional Influence?

Abstract

What is the level of voter–elite congruence in Europe and how is it affected by institutions? This article presents a different conceptualisation and new data to comparative research on congruence between voters and their representatives in 15 European countries. The originality of this work is mainly in its use of the most appropriate cross-national data for the conceptualisation of congruence as a ‘many-to-many’ relationship, using for the first time a survey of representatives to replace expert and public opinion on legislators’ attitudes. The study’s results show that congruence in European countries is relatively high in terms of left–right positions and, surprisingly, even higher regarding the question of EU integration. However, while we find enough evidence to link ideological congruence to mostly electoral institutions, it seems the same factors have no relation to the European unification dimension of congruence. This indicates the different nature of congruence in both the ideological and EU integration dimensions. Additionally, the present study found congruence to be higher for the group of voters rather than non-voters, and also higher for voters interested in politics as well as voters with a university degree.

Keywords:

• congruence
• representation
• European integration
• left–right
• electoral institutions
• parliaments

Acknowledgements

We need to express our gratitude to Darina Malová for her continuous support, encouragement and inspiration, and two anonymous reviewers for their many helpful comments and suggestions. We should also like to thank Comenius University, Faculty of Arts support scheme for providing assistance in editing the manuscript.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

1. Studies using the conceptualisation of congruence as a many-to-many relationship are relatively rare. There are a few examples of this conceptualisation’s application in comparative analyses (Dagefőrde 2013; Golder and Stramski 2010) using public opinion polls to estimate the preferences of representatives. The same conceptualisation and measurement of congruence as used in the present article can be found in a number of case studies (see Andeweg 2011; Bengtsson and Wass 2012; Ditmars and de Lange 2014).

2. There is also some conceptual ambiguity in congruence research connected with the term ‘responsiveness’, which is often mentioned as an important feature of democracy (e.g. Dahl 1971: 1) or democratic representation (e.g. Pitkin 1967), and sometimes merely as a synonym for congruence (Powell 2014). The present authors agree with Powell (2014: 20–21) and his clarification, where congruence refers to static comparisons and responsiveness refers to a more dynamic democratic relationship in which some kind of change in citizen preferences is followed by some kind of change in policy-maker positions or behaviour (see also Lax and Phillips 2012). However, there is no compelling reason to limit congruence to comparisons between the preferences of citizens and positions of government as he stated in his definition. It can also be used for comparisons with a wider range of representatives.

3. Empirical literature sometimes considers the citizenry as a whole, sometimes the electorate, and sometimes actual voters (Powell 2014), so there is no accepted general rule on whose preferences should be taken into account for congruence studies. Although it would be possible to find compelling arguments for inclusion of the entire population, the present authors follow the majority opinion on this matter and use the preferences of voters for measuring congruence. As democratic representation is provided by elections for its assessment and measurement, it is more appropriate to consider only the explicitly expressed preferences of voters. However, this issue is important, and therefore the present study also examines differences in congruence for voters and non-voters.

4. Despite the theoretical basis for the simultaneous influence of DM and the type of candidate lists (open v. closed), we limited our research to the influence of DM. The analysis of the mutual and interconnected influence of DM and candidate list type on congruence would require better data, at least in terms of an increased sample size.

5. Rohrschneider and Whitefield (2012a) offer a different explanation. The impact of institutions in the past was largely due to the small number of independent voters. There is no institutional effect nowadays because most party-systems are more or less equally divided between partisans and independents, and parties face difficulties in representing the diversity of voters in places where majoritarian institutions may hold a representational advantage.

6. Strictly speaking, the PDF is only defined for continuous data. What Andeweg actually works with are frequencies and histograms, and it would be more exact to speak of empirical CDF (ECDF) rather than CDF. Andeweg himself admits this issue; however, he argues that approximating the PDF and presenting the results in a line graph is much easier to understand. We use the ordinal scale in this paper as well and we therefore face the same problem. We follow Andeweg’s approach and visualise our results in a line graph later on.

7. All Serbian respondents had already voted in national elections and there were no non-voters. Therefore, we have omitted Serbia from testing those hypotheses which compare a group of voters with other groups of people.

8. Spearman’s Rho coefficients were compared to Pearson’s product moment correlation coefficient and the results were almost the same.

9. Spearman’s Rho correlation coefficients vary from –0.089 to 0.189. The correlation between the two congruence dimensions is r = 0.257.

Additional information

Funding

This work was supported by the Slovak Research and Development Agency [grant number APVV-0413-11].