Power and the Constitutional Treaty

The study argues that majority-voting rules are the central element in the draft EU constitution because most EU laws are adopted by majority voting and these are binding in all Member States, including those who opposed them.

The EU’s draft Constitutional Treaty proposes a radical reform of EU institutions. Many changes have been debated, but little notice has been paid to what is perhaps the most critical reform – the change in majority voting rules in the EU’s prime law-making body – the Council of Ministers.

It is easy to see why majority-voting rules are the central element in a constitution. Most EU laws are adopted by majority voting and these are binding in all member states – including those who opposed them. Plainly, nations should care about such rules.

Our research shows that the draft Constitution voting rules would:

• Shift a great deal of power to large members, modestly increase the power of tiny members, and reduce everyone else’s power; and
• Make it much easier to find a majority and so improve the EU’s ability to act.

Under current rules, qualified-majority voting (QMV) involves each member casting a certain number of votes with more populous members having more votes (e.g. France has 10, Denmark has 3). In total, there are 87 votes in the EU15 and at least 62 are required to pass a proposal.

These rules, however, are scheduled to be replaced in November 2004 according to the Accession Treaty. The rules that replace them – those decided in the Treaty of Nice – will strongly increase the vote shares of big members, slightly increase the majority threshold, and make the procedure more complex by adding two new majority thresholds, 50% of the membership and 62% of the population. These Nice/Accession Treaty rules, however, will be replaced in 2009, according to the draft Constitutional Treaty. (By 2009, the EU is likely to have 27 members, so we look at the impact on the EU27.)

The draft Constitutional Treaty (CT) proposes a much simpler and more transparent rule – a winning coalition must contain at least half the EU member states representing at least 60% of the EU population. In essence, each member’s vote is assigned two different weights – its population share and its membership share (e.g. 1/27 in the EU27) – and the weighted votes are checked against two different thresholds – 60% on population and 50% on membership.

How do the draft Constitution’s rules affect the power distribution?

There are many ways to define power, but to be concrete we take power to mean influence, or more precisely the ability to influence EU decisions by being in a position to make or break a winning coalition in the Council of Ministers. Of course, no nation has absolute power in the EU, so we focus on the likelihood that a member will be influential.

To determine this likelihood, we ask a computer to identify whether a given member’s vote is critical to a given winning coalition. Adding over all possible winning coalitions, the member’s power share is the number of times that its vote is critical as a fraction of the number of times that any member’s vote is critical. The result tells us the likelihood that the member will critical on a random issue. In the mathematics of voting, it is called the normalised Banzhaf index.

Performing these calculations for the Treaty of Nice rules and the draft Constitution rules gives a rough indication of how the proposed reform redistributes power. The figure shows the results of our calculations for all EU27 nations.

Plainly the four big members – Italy, France, the UK and Germany – gain a great deal, with Germany’s share of power rising 65%; from roughly 8% to 13%. The sum of big five’s power share rises from 31% under Nice rules to 40% under draft Constitution rules. Our calculations show that 17 of the EU27 would lose power – all the nations with populations between 3 and 40 million. Somewhat unexpectedly, the tiny members also gain from the draft Constitution’s rule changes.

Why do small nations also gain?

To understanding this odd result we have to step back and ask what is really going on. While the Treaty of Nice rules are complex on paper, they are in fact simple since only the weighted votes matter. In EU27, there are about 2.7 million winning coalitions according to Nice’s weighted-vote majority criterion and only 23 of these fail on the other two criteria. In short, the Nice system is, de facto, a weighted voting scheme with big nations having more votes than small nations, but far fewer votes than would be suggested by population proportionality. For example, Germany has 17% of the EU27 population but only 8.4% of the votes, while Luxembourg has 0.1% of the population and 1.1% of the votes.

The draft Constitution’s scheme of dual weighting (population share and membership share) shifts voting weights in two opposing directions. Under the population weighting, Germany’s vote share rises to 17%, but under the new scheme’s membership weighting, Germany’s vote share falls to 1/27th. Plainly the effective power share of each nation will be some sort of average of its two weights, but how important is population share versus the membership share?

The precise answer is given by the Banzhaf index, but intuitively we can understand the result as follows. For the biggest nations, the membership share is almost irrelevant to their power, so the big increase in population weight matters a lot. For the smallest nations, the population share is almost irrelevant, so the big increase in their membership share is what matters. Because power shares must sum to 100%, the in-between nations lose. They see only mild differences between their weighted-vote share under Nice and their population and membership shares under the draft Constitution.

Why the ability to act improves?

The last point concerns the draft Constitution’s impact on decision-making efficiency. As was widely pointed out at the time, the Treaty of Nice rules made it very hard to form a winning coalition. Indeed, of the 134 million possible coalitions in the EU27, only about 2% met the Nice Treaty’s QMV criteria. It should be much easier to find winning coalitions under the draft Constitution’s reforms, since 21.9% of all coalitions meet the new QMV rules. There are two main reasons for this improved ability-to-act. The highest majority threshold is lowered to 60% from 72%, and 40% of the power is concentrated in the hands of just four nations.

Because it will be so much easier to put together a qualified majority under the draft Constitution’s rules, it will be – statistically speaking – as easy to pass laws in the EU27 as it was in the EU6. This, in turn, will shift power from the Council to the Commission and European Parliament.

Richard Baldwin, HEI, Geneva, Switzerland, and CEPR
Mika Widgren, Turku University, Finland, and CEPR

For more on this, see the essay

Decision Making and the Constitutional Treaty: Will the IGC discard Giscard?.

For more analyses from the Centre for European Policy Studies visit the

CEPS website.