Architects of regional regime complexity: states and regional organizations in Europe

ABSTRACT

Regional Organizations (ROs) have become a central pillar of governance beyond the nation-state. This paper investigates why European states turned into architects of regional regime complexity: they have created and joined numerous different ROs and equipped them with a broad range of different policy competencies. Thereby, European states – some more than others – have increasingly duplicated identical policy competencies in multiple ROs. The phenomenon is puzzling as it is potentially costly and might undermine the effectiveness of regional cooperation especially if incompatible regional rules trigger non-compliance. Therefore, we ask why states differ in the extent to which they cover identical competencies in different ROs. Drawing on a unique dataset and analyzing cross-sectional temporal variation, we show that both indirect factors, such as late accessions and the number of states in Europe, as well as direct factors, such as state power and democracy, drive regional regime complexity.

KEYWORDS:

• Comparative Regionalism
• regional Regime Complexity
• regional Organizations
• overlapping Regionalism
• europe

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. We would like to thank the participants of the KFG closing conference ‘The Transformative Power of Europe Revisited’ for the lively discussion and the helpful comments to an earlier version of this paper, in particular Tanja Börzel, Tobias Hofmann, Anja Jetschke, Arie Kakowicz and Fredrik Söderbaum, as well as two anonymous reviewers. In addition, we are grateful to our student research assistants Nikolay Aleksiev, Chiara Fury and Klara Leithäuser.

2. ROs are defined as institutionalized cooperation between three or more states in more than one narrowly defined policy area and membership criteria are based on the geographical location (Panke, Stapel, and Starkmann 2020, 1).

3. The ROCO database codes RO policy competencies on the basis of RO primary law. Only the substantive paragraphs of RO primary law were coded. Preambles and declaratory parts were excluded since they do not transfer policy competencies to ROs. To capture these competencies, the coding scheme was created by a multi-stage inductive strategy. The goals of the coding scheme were to capture the legal basis for RO action and to make more than 70 ROs comparable. The final composition and number of policy competencies per policy field is based on a theory-driven clustering of coding buzzwords. For the full list of competencies and buzzwords, see Table A2. By contrast, the ROCO dataset does not include any measure for the ‘depth’ of an RO’s policy competency in a given area (see Börzel 2005) nor does it capture the delegation and pooling of authority to an RO (Hooghe, Lenz, and Marks 2019).

4. Multiple competency coverage would also decrease if member states decided to scale back the policy mandate and reduce the competencies of individual ROs. However, these involve political costs, such as reputation, that can be evaded by other alternatives. Instead of formally curtailing RO mandates, actors could simply not act upon the competencies, which would more likely result in diminished vitality of the RO. Less frequently, actors could also decide to dissolve ROs (Gray 2018; Eilstrup-Sangiovanni 2020). Alternatively, actors might choose to pursue their objectives through other outlets and set up ROs that promise to achieve their objectives and goals. However, this would further increase the number of ROs and likely also multiple competency coverage.

5. We performed two additional robustness checks. First, the number of ROs increased over time. Since the number of ROs and their policy competencies form part of the DV (the instances of competency duplication per country and year), we run all models using an alternative measurement of the DV that controls for the number of RO memberships per state and year. We divided the instances of competency duplication by the number of RO memberships per state in a given year. The data of RO memberships also stems from the ROCO database (Panke, Stapel, and Starkmann 2020). All substantive findings and significance levels remain robust and do not change across specifications (see appendix, Table A4). Second, we also excluded Russia as the country with the highest numbers of identical policy competencies covered in ROs. Again, all of our findings remain robust (table available upon request).

6. Darker areas signify more instances of duplicated policy competences.

Additional information

Funding

The research was financially supported by the Deutsche Forschungsgemeinschaft (DFG) under grant number (PA 1257/7-1; 421167407).