A social network-based approach to assess de facto independence of regulatory agencies

Abstract

This article uses a policy network perspective to assess the independence of regulatory agencies (RAs) in liberalized public utility sectors. We focus on the de facto independence of RAs from elected politicians, regulatees and other co-regulators. We go further than previous studies, which only undertook a general analysis of the de jure independence of RAs from political authorities. Specifically, we apply a social network analysis (SNA), which concentrates on the attributes and relational profiles of all actors involved in new regulatory arrangements. The concept of de facto independence is applied to the Swiss telecommunications sector in order to provide initial empirical insights. Results clearly show that SNA indicators are an appropriate tool to identify the de facto independence of RAs and can improve knowledge about the issues arising from the emergence of the ‘regulatory State’.

Keywords:

• Regulatory agencies
• regulatory independence
• regulatory state
• social network analysis
• telecommunications

ACKNOWLEDGEMENTS

The presented data was gathered within the Project ‘Multi-level regulation of the utilities sector in Belgium, the Netherlands, Ireland and Switzerland’ led by KU Leuven and UC Louvain. The authors are also very thankful for helpful comments from Mario Diani and Dimitris Christopoulos, as well as to Philip Leifeld for the computation of the index within the R package Version 0.0.1.https://r-forge.r-project.org/projects/polnet/.

Notes

1 To reproduce and calculate the index on R, you may follow these steps and link: Reference and R Package: Leifeld, Philip (2013). polnet: Analysis of Political Networks. University of Konstanz, Germany. R Package Version 0.0.1.https://r-forge.r-project.org/projects/polnet/. As an input, the functions takes into account (1) the adjacency matrix of the influence network; and (2) the reputation measure as (a) a vector/attribute, or (b) as an adjacency matrix where every actor seeing another actor as ‘important’ would send a tie. Both data can be entered as network objects.

2 Figure 2 is available at http://www.ipw.unibe.ch/content/professuren/policy_analyse/regulatory_policy/index_ger.html.

3 Insight gained from personal interview with the Secretary General of the Swiss telecoms regulator ComCom in December 2009.

4 Normalized measures show an actor's degree centrality as a fraction of the overall possible degree centrality in the network where all ties would exist (1).

Additional information

Biographical notes:

Karin Ingold is Assistant Professor at the Institute of Political Science and the Oeschger Centre for Climate Change Research at the University of Bern, and Head of the Policy Analysis and Environmental Governance Cluster at the EAWAG.

Frédéric Varone is Professor within the Department of Political Science at the University of Geneva.

Frans Stokman is Professor of Social Science Research Methodology at the University of Groningen.