Does Cohabitation Matter? French European Policy-Making in the Context of Divided Government

Abstract

This article analyses French European policy-making in the context of divided government, arguing that in the area of European policy-making cohabitation increases the number of veto players, as compared with unified government. The increase of veto players, in turn, reduces France's acceptance-sets at the European table and thereby introduces a status quo bias. In order to test this theory, use is made of a comparative case study design. Process-tracing and counterfactual analyses reveal that veto player theory can, indeed, be applied to the interactions of a split-executive government.

Acknowledgements

Research underlying this article has generously been supported by the Mannheim Center for European Social Research. The author particularly wishes to thank Beate Kohler, Wolfgang C. Müller, Walter Müller and Franz U. Pappi for their support.

Notes

1. Whether this is driven by a sincere policy orientation or by a vote-maximising strategy does not make a great difference analytically (cf. Laver and Shepsle 1996; Strom and Müller 1999: 10).

2. It is generally acknowledged that during cohabitation the PM takes the lead in domestic politics. Apart from nomination rights and the signing of ‘ordonnances’, the President here is basically reduced to ‘inaugurating the Chrysanthemums’, to cite Raymond Barre (cf. Chevalier and Conac 1991: 925).

3. I do not test a model of bureaucratic decision-making since I assume that the bureaucracy affects the preferences of the two camps. Also, I do not include a legislative model since the influence of the French parliament on European policy-making is generally not considered to be very important (cf. Kessler 1999: 224; Cohen 1986: 29; Guyomarch et al. 1998: 60). These two models have not proved very successful in past empirical tests (cf. Laver and Shepsle 1994, 1996: 13). Similarly, the cabinet model's relevance for the Fifth Republic has often been denied (cf. Elgie 1993: 32; Wright 1989: 88; Thiébault 1994: 141).

4. I interviewed 35 former ministers and members of the high bureaucracy. In addition, I analysed written primary sources and secondary sources such as newspaper reports, memoires and also public speeches. Finally, some interviewees allowed me access to internal documents. This was extremely helpful information especially for reconstructing decision-making processes of the 1980s. For further details on the case studies, compare Leuffen (2007).

5. Note that my cases, for example the budgetary Delors-I package but also the Amsterdam intergovernmental conference, contain a plurality of issues which is why the actual number of observations is higher than the number of case studies mentioned here (cf. Gerring 2004).

6. In addition, two cases were analysed that started during unified government and ended during divided government. One case is on the GATT negotiations that began during the socialist government and were concluded under the Balladur government; the second case is on the Amsterdam Treaty where the Juppé government was replaced by the gauche plurielle. Splitting up these cases into sub-cases, they take a quasi-experimental form. While almost all other factors remain unchanged there is a change from unified to divided government.

7. However, Chirac did not succeed in representing France at the European summit of The Hague together with Foreign Minister Raimond. Mitterrand vetoed any changes from the practice of occupying two seats per nation (cf. Raimond 1989: 49; Attali 1995: 79). As to the content, the two sides basically agreed on the French positions.

8. In the past, cohabitations resulted under off-year elections. However, it should also be noted that future cohabitations cannot be completely avoided since, for instance, a resignation or death of a President could lead to another de-synchronisation of the election calendars (cf. Jérôme et al. 2003).