Are power indices a valid measure to quantify changes in influence in the EU's Council of Ministers, following the re-weighting of votes in the Treaty of Nice?

Table of Content

Introduction

1. The Council of Ministers – a power index analysis
1.1 QMV in the Council - past, present and future
1.2 Methodology of the analysis
1.3 QMV after the Treaty of Nice – a change for the better?

2. Applying power indices to the EU – a critique
2.1 Arguments against the applicability of power indices to the Council of Ministers
2.2 The concepts of power
2.3 Power indices and ‘true power’

3. Are there alternatives?
3.1 Power indices versus spatial analysis
3.2 Arguments against the applicability of spatial analysis and for the relevance of power indices
3.3 Power indices in the Council – imperfect, but without any real alternatives

Conclusion

Figures

Bibliography

Are power indices a valid measure to quantify changes in influence in the EU’s Council of Ministers, following the re-weighting of votes in the Treaty of Nice?

Introduction

At the beginning of the 21st century, the European Union is preparing to face the probably biggest challenge of its nearly fifty year-long existence; the binding-back into the West of many of the once-isolated countries of Central and Eastern Europe. But for enlargement to proceed successfully, the EU’s ways of working need to change. One of the top priorities with regards to this is that the relative weights of countries in the Council of Ministers (hereafter referred to as “the Council”) must be adapted in order to prevent collective decision-making from freezing and the EU from becoming immobile.

At the December 2000 Intergovernmental Conference (IGC) in Nice, EU heads of state attempted to do exactly that. Voting weights in the Council were modified and new rules for its Qualified Majority Voting (QMV) procedure were drawn up. However, the results have subsequently attracted considerable disapproval. Whereas the Union itself praises the achievements for providing the flexibility necessary to cope with enlargement, its critics claim that, far from this, the new rules will possibly make it even more difficult to adopt decisions, while preserving much of the general inequality of influence between member states that has previously characterized the Council throughout much of its existence.^[1]

In a recent study, Machover & Felsenthal (M&F) apply power indices to the new QMV-rules in the Council in an attempt to assess the degree of equitability (that is, comparative levels of influence on outcomes) among member states as well as the practical workability they provide. They conclude that, whilst a ‘contingency plan’ in case of non-enlargement by 2005 does offer genuine improvements in both areas, the most likely outcome will be continued inequality of power, in spite of the IGC’s explicit goal of giving to the bigger and more populous countries (especially Germany) more influence than to their smaller partners.^[2] Furthermore, adopting decisions seems set to become even more difficult than at present.

But is this a valid analysis? Given the EU's notoriously complicated inter-institutional arrangements and decision-making structures, can a priori power indices be a valid measure to assess such changes in influence in an body like the Council of Ministers? In this paper I will try to defend the concept of power indices against claims that they cannot. Also, I will attempt to qualify M&F ’s results in the light of my conclusion.

I shall firstly summarise M&F ’s study and results in order to allow for an easier assessment of the situation. Subsequently, I shall depict the arguments against the application of power indices to an organisation such as the European Union and put them into perspective. In the third part of this essay, I shall then attempt to show that methods proposed as alternatives to power indices do not, in fact, offer any workable alternatives when analysing the EU-Council after enlargement, and conclude that power indices can provide an important approximation of a priori power distribution and do carry relevance in assessing the Council and especially in understanding the behaviour of actors when deciding on issues like the re-weighting of votes in the Nice Treaty.

1. The Council of Ministers – a power index analysis

M&F analyse and evaluate two decision rules for the Council of Ministers.

The first one (hereafter referred to as ‘N15’) will apply from 1 January 2005 to a fifteen-member EU, in the case that enlargement does not proceed as planned and no new countries will have been included by then. The second rule (‘N27’) will apply to an enlarged Union of (prospectively) up to 27 member states.^[3]

1.1 QMV in the Council - past, present and future

Previously, QMV in the Council has taken the form of a pure weighted decision rule. Each member state is assigned a certain number of bloc-votes to be cast, and a proposed measure is adopted if the total weight of the votes in favour equals or exceeds a certain quota (currently about 72 per cent of all votes). Both new rules, N15 as well as N27, however, differ from this in that they are presented as the conjunction of three such weighted rules:^[4]

1) a bloc-vote of the nature described above, however with new numbers of votes for the member states (see Figure 1^[5] )
2) a numerical quota, specifying a minimum number of Council members that have to be in favour of a measure for it to be adopted
3) a population quota, specifying a minimum percentage of the Union’s total population which, as represented by their governments’ ministers, must be in favour of a measure for it to be adopted

All three of these conditions must be met in order for a proposed measure to pass. However, upon closer examination, it becomes clear that the numerical quota in N15 has basically no effect on the final outcomes of the rule, as a majority of member states is in any case necessary to reach the bloc quota of 170 votes. N15 can thus be approximated by a conjunction of the bloc vote and the population quota.^[6]

The same applies to N27. Moreover, both the numerical and the population quotas hardly affect the final outcomes of the rule. Due to some further characteristics, this means that N27 can very precisely not only be approximated by its bloc-vote component but reformulated as a single weighted voting rule and thus evaluated by means of a power index analysis.^[7]

1.2 Methodology of the analysis

M&F analyse the rules in terms of various criteria, the most interesting of which are:

1) The voting power they give to individual member states. For this they use absolute (illustration not visible in this excerpt) and relative (100 e) Banzahf index values.^[8]

The latter value here signifies the usual Banzahf measure of voting power, defined as the number of a player’s critical defections from distinct minimal winning coalitions (DMWCs) as a proportion of the total number of such critical defections of all players combined.^[9] A critical defection is critical if and only if it transforms a minimal winning coalition (with respect to at least one player) into a non-winning coalition. By Banzahf ’s definition, an actor’s voting power is then proportional to the number of DMWCs in which his/her presence is necessary for the coalition to remain winning. In order to allow for better comparability, the value is here expressed in percentage terms, all values consequently adding up to 100.

However, since the original Banzahf index expresses power in relative terms and can thus only be used for comparing the voting powers of several voters under he same decision rule, it is not a reliable yardstick for comparing the voting powers of different voters, or even of the same voter, under two different decision rules.^[10] For this, the absolute Banzahf score illustration not visible in this excerpt (originally devised by Penrose) has to be used, defined as

illustration not visible in this excerpt^[11]

where r denotes the proportion of all divisions in the voting body where the actor is on the successful side. illustration not visible in this excerpt is basically only a ‘de-normalised’ version of the relative Banzahf score (thus the denotation as absolute Banzahf score); the latter can be obtained from it by simply dividing the value of illustration not visible in this excerpt for that voter by the sum of all illustration not visible in this excerpt values of all voters in the assembly.^[12]

[...]

^[1] Grow, Europe, The Economist Magazine, London, 09 December 2000, pp. 23

^[2] The Many Tricks of Widening Europe, The Economist Magazine, London, 09 December 2000, pp. 53

^[3] M. Machover, D.S. Felsenthal, The Treaty of Nice and Qualified Majority Voting, London/Haifa, 2001, pp. 1

^[4] ibid, pp. 3

^[5] Sources: European Council (www.consilium.org) and So That’s All Agreed, Then, The Economist

Magazine, London, 14 December 2000, p. 5 (as accessed via www.economist.com/display/story_id=45129)

^[6] Machover & Felsenthal (1), pp. 7

^[7] ibid, pp. 9

^[8] ibid, pp. 12

^[9] S. Brahms, Game Theory and Politics, London, 1976, pp. 165

^[10] Machover & Felsenthal (1), pp. 6

^[11] ibid

^[12] ibid