EU Constitution: an economist’s perspective, Part 4

Richard Baldwin

13 June 2007



The EU already has a constitution. Every organisation does. A group without a constitution is just a social gathering. The EU’s constitution is called “the Treaties” – basically the Treaty of Rome and the Maastricht Treaty as modified by subsequent treaties. But this is a “small c” constitution – rules on how the organisation is governed, not a “big c” Constitution that sets out a republic’s basic laws and principles. For a while it looked like Europe would get a “big c” Constitution, but French and Dutch voters vetoed the idea, and Britain and France will veto its revival. In all likelihood, EU national leaders will lay the big-c Constitution to rest at their 21-22 June 2007 Summit.

Europe’s small-c constitution, however, is in for big changes. If things go according to the German Presidency’s plans, the Summit will announce the outline of a new treaty. Most likely, the new treaty will implement its changes by amending the Treaties in order to avoid the big-c label. The details would be negotiated during an Intergovernmental Conference in 2008.

This column, the last in a series of four, discusses the changes that must be in the new treaty to consider it a success. As I argue in the second instalment, the key is what happens to the four major legal changes in the rejected Constitution – Council of Ministers’ voting rules, the composition of the European Commission, removal of the pillars, and the passerelle and flexibility clauses.

Back to basics

Over the past 15 years, EU leaders have explicitly and repeatedly identified two must-do reforms – Council voting rules and Commission composition. The new treaty must, at minimum, tackle these.

Starting with Council voting reform, there are two options. Option #1 is to stay with the Constitution’s double majority rules where passing a law requires at least X% of the member states that represent at least Y% of the EU population (in the Constitution, X=55, Y=65); negotiators may tinker with X and Y. Option #2 is to stay with the status quo Nice Treaty rules but to repair them. While Option #1 is strongly backed by Germany, Poland threatens to veto the whole show unless negotiators are allowed to consider Option #2.

Evaluating these options involves the mathematics of voting theory – reasoning too knotty for a Vox column to deal with satisfactorily. Mika Widgrén and I have been studying Council voting for 15 years and will soon post a CEPR Policy Insight on the matter.1 Our key findings are used as assertions in this column.

The first thing to note is that both options allow the EU to function effectively and legitimately with a greatly expanded membership.

The main differences between Options #1 and #2 emerge when thinking about Turkish membership. A good way to think about the Constitution’s double majority is as two separate weightings of each nation’s vote in the Council. For the membership threshold, nations have equal weight. For the population threshold, a nation’s weight is its population share, so national voting weights are automatically awarded according to population. Turkey’s population is about 70m, so it would be the second most powerful member if it joined today. Using standard population projections, Turkey would be the most powerful member in about 14 years. Giscard d’Estaing (an arch foe of Turkish membership) knew this when he put the dual majority scheme into the draft Constitution in 2003. Turkey’s top-ranked power may be unacceptable to some, so the Constitution’s voting scheme may be a barrier to Turkish membership.

Poland’s version of Option #2 would allow votes according to the square-root of a nation’s population. By automatically linking votes and population, this too would make Turkey the most powerful member. The Nice Treaty repairs that Mika Widgrén and I propose avoid this feature by sticking with an equal number of votes for all large nations (UK, France, Italy, German and Turkey)2. More generally, the Nice Treaty’s approach of divorcing population and voting weights allows political flexibility on vote allocation that the Constitution’s voting rules do not.

The Constitution’s fudge on the Commission is probably good enough. Now, there is one Commissioner per member, but as the EU expands, the Commission could become ineffective. The Constitution’s compromise was to stick with one-per-member up to 2014, after which the number is capped at two-thirds the number of EU members, with posts rotating equally among members.

Pillars and the Charter

Things get tricky when it comes to the three other major changes in the Constitution. Conditional statements are necessary since the various elements interact in complex ways. The pillars are the linchpin. If the new treaty stays with the basic structure of the the Treaties, the pillars should remain. Let’s assume they do.

With the pillars in place, the Charter would create a lot less uncertainty about the EU’s role in Justice and Home Affairs. The Court and Commission would have no automatic authority over such matters. Still, as I argued in an earlier instalment, putting the Charter into a treaty is a big step. The very power of the words in capital letters "Fundament Rights, Charter and Treaty" suggests an acknowledgement of their primacy over national labour market legislation.

The Charter is now a ‘solemn proclamation’ by the European Council (heads of state), Commission and Parliament. This is not a bad solution given its nature. It is not – despite its name – a succinct statement of principles. The US Bill of Rights has ten articles. The UN’s 1948 Universal Declaration of Human Rights has 30. The EU’s Charter has 54 articles and it contains elements that seem too specific for a high-minded declaration. Article 1 is fine: “All human beings are born free and equal in dignity and rights. They are endowed with reason and conscience and should act towards one another in a spirit of brotherhood.” But Article 29 is: “Everyone has the right of access to a free placement service". Does that sort of thing really belong in the Treaties?

Moreover, some articles are very subjective. Bulgarians, for example, are likely to interpret them very differently than Italians. Article 31: “Every worker has the right to working conditions which respect his or her health, safety and dignity". And “Every worker has the right to limitation of maximum working hours, to daily and weekly rest periods and to an annual period of paid leave". Given the economic, political and social diversity of the EU, such language invites perpetual conflict.

Or, to put it differently, if one wanted to put more ‘social Europe’ into the EU by the back door, putting the Charter into the Treaties would be a good way to start. I am quite sure that EU governments will increasingly pool sovereignty in these areas; cooperation at the European level is often the best –  and sometimes the only – way of dealing with the challenges of globalisation. But EU national politicians should make the case directly to their voters rather than arranging for it to happen on auto-pilot.

Since the Treaties are the highest level of EU law, any amendment requires a new treaty. This would be a normal state of affairs if the Treaties only contained really fundamental things – like German’s Basic Law, or Italy’s Republican Constitution. The Treaties, however, run to hundreds of pages and cover many issues. The EU will want to change some of these without being dragged through the full treaty-writing-and-ratification process. The passerelle clauses are therefore necessary. The Treaties already have passerelle clauses attached to specific items; the rejected Constitution generalised their application to everything but foreign and defence matters. It would seem sensible for the new treaty to stay with current practice rather than the Constitution’s shotgun approach. Transferring sovereignty to the EU is something that should be well thought out and publicly defended on an item-by-item basis. For the same reasons, it would seem sensible to be very specific about which items can be subject to an accelerated procedure for sovereignty transfers. Some thought should also be given to making passerelle clauses two-way – allowing particular items to be shifted from unanimity to majority voting as well as vice versa.

If the pillars are kept, flexibility will continue to be limited to economic integration where its logic is strongest. Innovation means that the nature of trade changes continuously and it makes no sense to write a new treaty every time something like internet services arise. This is exactly why the flexibility clause was included in the Treaty of Rome and why it continues to make sense in today’s world.

That – in my view and in less than 1500 words – is the minimum reform the new treaty needs to be a success.



1 “Pandora voting: Can the Constitution voting rules be improved?” CEPR Policy Insight No. 4, posted 15 June 2007. We analyse various voting rules on three criteria: impact on the EU’s ability to act, impact on the power distribution, and robustness to future enlargements, especially Turkey’s accession. We also look at the implications for the Council-Commission-Parliament balance of power.
2 See the proposal in “Nice Try: Should the Treaty of Nice be Ratified?” Richard Baldwin , Erik Berglöf , Francesco Giavazzi , and Mika Widgrén, 8 June 2001, CEPR.



Topics:  EU institutions

Tags:  Constitutional Treaty, EU Constitution, EU Council, EU Charter

Professor of International Economics, Graduate Institute, Geneva; Director of CEPR; Vox Editor-in-Chief