The access of business interests to EU institutions: towards élite pluralism?

Abstract

It has often been suggested that bias in the access of interest organizations to policy-makers leads to biased policies. The paper analyses the access of interest organizations to the EU institutions drawing on data from 800 business interest associations and 34 large firms. Exploring the proposition that some form of élite pluralism may emerge in the EU, it argues that the contact patterns derive from resource dependencies among the political actors and the interest groups, institutional opportunities in the EU, and the characteristics of the interest organizations. The study identifies imbalances in the access of EU associations, large firms, and national associations to the EU institutions, with large firms being in the forefront. Nonetheless, the evidence does not hint at the emergence of élite pluralism in the EU but points to important variations across the EU institutions and among the working level and their political leadership in each institution.

Keywords:

• EU institutions
• EU policy-making
• interest organizations
• large firms
• pluralism in the EU

Acknowledgements

The paper draws on survey research conducted at the Mannheim Centre for European Social Research with Christine Quittkat and directed by Beate Kohler-Koch. I am grateful for the financial support provided by the European Centre for Public Affairs and the University of Mannheim as well as for the chance to conceive the initial parts of this study at the European University Institute. Earlier versions of the article were helpfully discussed at the European Union Studies Association Conference in Austin, at the European Consortium for Political Research General Conference in Budapest, at the FernUniversität in Hagen, and at the preparatory workshop of this special issue. In particular, I would like to thank Nathalie Behnke, Arthur Benz, Jan Beyers, Bernhard Boll, David Coen, Maria Green Cowles, Andrea Lenschow, Susanne Lütz, Gary Marks and the two anonymous referees for their insightful comments and suggestions.

Notes

1 To some extent, Bouwen also considers professional consultancies, but less systematically than the other three forms of organization so I do not consider them here.

2 The sample is based on the following sources: Oeckl 1997; Conseil National du Patronat Francais (1997); Henderson and Henderson (1995); European Commission (1996).

3 While T-tests and analyses of variance are designed to analyse interval data, as quite robust procedures, they are frequently considered to be applicable to the study of ordinal data such as is used in this analysis.

4 ANOVA for access to EU institutions by firms and associations: European Commission working level: F 56.109 df between 4, df within 794, p 0.000; European Commission leadership: F 52.172 dfb 4, dfw 756, p 0.000; Members F 29. 229 dfb 4, dfw 777, p 0.000; European Parliament Committees F 36.357 dfb 4, dfw 754, p 0.000; Council of Ministers F 17.829 dfb 4, dfw 718, p 0.000; COREPER F 36.969 dfb 4, dfw 737, p 0.000. Post-hoc comparisons indicate significant differences for all institutions between EU associations and firms, on the one hand, and national associations, on the other. There is just one significant difference between the national associations: German groups are more frequently in touch with the working level of the Commission than British groups. Finally, large firms have better access than EU associations to the Commission leadership, the Council of the EU, and the MEPs.

5 T-test for access of firms to the EP and the Council: T 1.034 (df 32), p 0.309. The other T-tests that compare access to the EP with that to the Council, access to the Commission with that to the EP, and access to the Commission with that to the Council for each type of organization are significant at the 1 per cent level (not reported).

6 ANOVA for access to information from EU institutions: European Commission: F 2.18 df between 4, df within 625, p 0.069; EP: F 3.64 dfb 4, dfw 574, p 0.006; Council of the EU: F 4.11 dfb 4, dfw 531, p 0.003.

7 Associational measures for the timing of interest representation by type of organization: agenda setting: CHI2 95.005 (df 8), p 0.000; Commission proposal: CHI2 69.152 (df 8) p 0.000; EP debate: CHI2 47.148 (df 8) p 0.000; EU Council debate: CHI2 32.217 (df 8) p 0.000; transposition: CHI2 66.938 (df 8) p 0.000; implementation: CHI2 62.565 p 0.000. The Cramer-V measures that indicate the strength of these associations range between 0.147 (Council debate) and 0.250 (agenda setting).