Power analysis of the Nice Treaty on the future of European integration

Abstract

Power analysis of changes in voting weights and rules in the Nice Treaty is done by applying methods that use Shapley–Shubik and Banzhaf indices. Significant decreases in voting power of small countries make widening of integration more acceptable to incumbent members due to small size of the applicants. Relative increases in the conciliatory power of smaller members, and relative increases in the independent power of bigger members make smaller members compromise more, and improve the position of large members for further deepening of the integration. Lastly, the fairness analysis reveals a more federalist EU in the way votes are distributed.

Acknowledgements

The author has greatly benefited from the comments and suggestions from participants of seminars at the Kellogg Institute for International Studies, and the Nanovic Institute for European Studies at the University of Notre Dame, the European Union Centre at the University of Michigan, and the International Conference on Social Sciences in Hawaii.

Notes

Election process to the European Parliament and groups formed within or after the elections are based on ideology.

For more on the division of power among EU institutions, refer to Laruelle and Widgren (1997).

There has been some criticism of using power indices to study the EU by Garrett and Tsebelis (1999). Discussion of this criticism can be found in Lane and Berg (1999), and Felsenthal and Machover (2001b).

Hosli (1993) gives the areas that require each voting rule prior the Nice reforms.

This is the voting rule for the usual procedure, where the proposal comes from the Commission. Under rare circumstances, decisions are taken outside this procedure. Therefore, the analysis is confined to this normal procedure.

Galloway (2001) observes that while 258 is observed in the main body of the treaty, a quota of 255 appears in the declaration annexed to the treaty. He concludes that the latter is purely political rather than a legally binding statement. This view is also shared by Felsenthal and Machover (2001a). However, Moberg (2002) claims that the order of texts show that 255 should be the binding one. In this analysis, calculations have been made using both: although the numbers change, the main conclusions drawn in this paper are still valid on both cases. The calculations presented use 258, which is ‘[what] appears to be prescribed in the Nice Treaty’ according to Felsenthal and Machover (2001a).

Official Journal of the European Communities, C 80, p. 82.

This threshold is the percentage that will be used when the accession of all 12 applicants is complete. The Nice Treaty leaves the threshold for the interim steps ambiguous, although it sets minimum and maximum thresholds, which are both above the traditional 71%.

Widgren (1994) observes a logarithmic relation between population size and number of votes assigned in the Council of Ministers. This formula deduced before the 1995 EU enlargement has been tested and found successful by Hosli and Wolffenbuttel (2001). Therefore, pre-Nice distribution of votes in Table 1 is determined according to Widgren.

Here and elsewhere, by ‘favour’, is meant a distribution of votes, where small member states get more votes than implied by their population.

The absolute Banzhaf Index produces numbers that do not sum to 1. In the rest of the paper, these numbers are normalized to obtain relative Banzhaf Indices that measure relative voting power.

The idea is simple: when the threshold is 100%, all members have equal relative power. As the threshold is lowered small member states become less crucial in the outcome; and they lose power. In turn, bigger member states gain power. Leech (2001) illustrates this relationship between the threshold and the relative voting power at high values of threshold.

The power indices given in the table assume simple majority voting rule in the distribution of power within a bloc, since countries in the blocs act conciliatorily. When compared to SSI or BI, although the power of a country according to partial homogeneity is affected by the particular coalition the country is in, one observes some patterns: partial homogeneity implies larger power for small countries than SSI and BI implies. Furthermore, the power of large countries according to partial homogeneity lies in between SSI and BI, larger than BI, smaller than SSI.

The model behind Laruelle and Widgren (1998) is only applicable for Banzhaf index.

In EU6, x = 0.363; in EU9, x = 0.357; in EU10, x = 0.392; in EU12, x = 0.374; in EU15, x = 0.387.