Informational Lobbying in the European Union: The Effect of Organisational Characteristics

Abstract

Information supply is an important instrument through which interest groups can exert influence on political decisions. However, information supply to decision-makers varies extensively across interest groups despite the common objective to influence policy-making. Drawing on resource mobilisation and organisational theory, a new theoretical framework is developed that identifies organisational characteristics, more specifically the resources, the functional differentiation, the professionalisation and the decentralisation of interest groups as determinants of information supply. These theoretical expectations are tested based on a large new dataset. Using multilevel modelling, this article examines information supply to the European Commission across a large number of policy issues and interest groups by combining an analysis of consultation submissions with a survey conducted among interest groups. The findings confirm the theoretical expectations suggesting that interest groups cannot equally exploit their access to decision-makers, but that resource endowment and organisational structures play a crucial role for effective informational lobbying.

Acknowledgements

Research for this article was financially supported by the Graduate School of Economic and Social Sciences at the University of Mannheim, the Landesstiftung Baden-Württemberg, the German Academic Exchange Program and the Volkswagen Foundation. I would like to thank the anonymous reviewers and numerous colleagues and friends who have repeatedly commented on earlier drafts of the article or the entire research project, most notably Gema García Albacete, Doreen Allerkamp, Christian Arnold, Patrick Bayer, Tanja Dannwolf, Oshrat Hochman, Thomas Meyer, Sven-Oliver Proksch, Bettina Trüb and Arndt Wonka. Special thanks go to Sabine Saurugger, Daniel Stegmüller, Thomas Gschwend and in particular Berthold Rittberger for continuous invaluable support throughout the research process.

Notes

1. Further details concerning the selection of interest groups and policy issues as well as regarding the measurement of information supply will be provided in the research design section.

2. This applies to the first pillar of the European Union, which is arguably the most important and therefore the focus of this study.

3. Concerning associations and companies, Beyers et al. (2008: 1108–9) suggest distinguishing between ‘interest groups’ which are membership organisations and ‘interest organisations’ denoting all actors that also fulfil the above-mentioned criteria, but do not have any members. According to this definition, associations correspond to ‘interest groups’ while companies correspond to ‘interest organisations’. To keep the terminology in this study as simple as possible, I use ‘interest groups’ to denote all actors fulfilling the three criteria while I refer to ‘associations’ and ‘companies’ when distinguishing membership organisations from corporate actors.

4. According to the Kaiser criterion, these three indicators measure the same underlying factor since the Eigenvalue is above 1.

5. According to the Kaiser criterion, the factor analysis retained only one factor as the Eigenvalue is higher than 1.

6. Poisson regression might suffer from overdispersion. However, to my knowledge there is so far no statistical test that checks for overdispersion in multilevel models. In order to test whether the results hold when a negative binomial multilevel regression is estimated that addresses potential overdispersion, I tried to estimate such a model using the recently developed R package glmmADMB, which is to my knowledge the only statistical package that can currently estimate multilevel negative binomial models. The algorithm failed to estimate the model however. I checked the robustness of the results estimating a linear multilevel model and a single level negative binomial model with clustered standard errors. The direction of the effects is the same for all variables, only the statistical significance of the effect of professionalisation is not confirmed.