Ratification of the Constitution of the EU: A Minefield

Ratification of the Constitution of the EU: A Minefield


On 18 June the Brussels summit gave the go-ahead to the draft Constitution which now moves into its final phase. The text has to be ratified, unanimously, by all Member States through a variety of procedures, so there is still a real possibility of the whole process floundering. This article is a brief survey of the ratification procedures to be employed and the pitfalls that might appear. It also looks at the possible effects on the legitimacy of the Union.


The agreement reached on the European Constitution by the European Council on 18 June ended the process of drafting the document but by no means ended what could be called the EU’s ‘constitutional policy’. In fact, the most hazardous period for the Constitution begins now, as it can only come into effect once ratified unanimously by the 25 Member States, including Rumania and Bulgaria (whose entry is scheduled for 2007, the same date as the Constitution is supposed to come into force). Ratification will be no pushover, given the variety of hazards that can crop up. The requirement for unanimity shields individual States from the political cost of being responsible for rejecting the document; instead, they can use their voting power as a means of leveraging national interests. More than a ‘social contract’ conferring legitimacy on the Union, the European Constitution risks being interpreted as a response to different, and often conflicting, national ambitions.


Parliamentary Ratification

The Constitution should be ratified according to national constitutional requirements (Article IV-8, CONV 86/04)[1], leading to three possible procedures, which might be used alone or concurrently: ratification by parliament, a referendum or constitutional reform. In each case various bodies are directly implicated: political parties, pressure groups, Constitutional courts, etc. In all cases parliament must take at least the first step. Even in those cases in which the EC is handled as an international treaty, with ratification delegated to the executive, such a controversial issue is unlikely to get through with just a government rubber stamp. Some sort of parliamentary debate will have to be held, even if no vote is taken.

Table 1. Parliamentary Ratification Procedures in the Member States of the EU

StateParliamentary Ratification
AustriaSimple majority of the Congress (and of the Senate if its competences are affected) of two thirds of the Congress (and of the Senate if as above), if the transfer of powers implies Constitutional reform (Articles 50, 42 and 44).
BelgiumTreaties affecting citizen rights must be approved by both Houses. If they affect the competences of the Regions, the Councils of both must also approve them (Article 163).
CyprusAdopted by the Cabinet and approved by the House of Representatives (Article 169).
Czech RepublicApproval by three fifths of the Congress and the Senate (Articles 10 and 39).
DenmarkApproval by a majority of five sixths; otherwise, a referendum (Articles 20 and 42).
EstoniaSimple majority and other procedures (Articles 120 and 121).
FinlandBy law. Simple majority or two thirds majority if it affects the Constitution (Article 94).
FranceBy law (Articles 52-55 and 88). Discretionary referendum at the initiative of the President (Article 11).
GermanyBy law. Majority of two thirds of the Bundestag and two thirds of the Bundesrat (Articles 23 and 79).
GreeceBy law, majority of three fifths (Article 28)
HungaryMajority of two thirds of both Houses (Article 2a).
IrelandNo specific rule. Each reform of the EU requires a parallel reform of the Constitution by means of referendum (Articles 29, 46 and 47).
ItalyRatification by both Houses; no referendum (Articles 80 and 75).
LatviaParliamentary ratification, but if half the parliamentarians so wish, a referendum must be held (Article 68).
LithuaniaParliamentary ratification; referendum required for treaties which affect major aspects of the lives of Lithuanians (Articles 135,1 and 5).
LuxemburgBy law approved by two thirds of members of parliament (Articles 37, 49 and 114).
MaltaThere are no constitutional regulations, unless ratification requires Constitutional amendment.
NetherlandsBy two-thirds parliamentary majority (Article 91)
PolandBy parliamentary procedure, the conditions of which are established in another Act of Parliament (Article 90).
PortugalParliamentary majority (Article 161)
SlovakiaMajority of three fifths (Articles 7 and 84).
SloveniaMajority of three fifths (Articles 3 and 8)
SpainAbsolute Majority of both Houses (Article 93).
SwedenApproval by three quarter of the members of the Riksdag (Article 10.5).
United KingdomParliamentary majority.

Sources: http://www.uc3m.es/uc3m/inst/MGP/NCR/portada.htm; see also: http://www.european-referendum.org/materials/di/refsum.pdf.

Parliamentary approval is the main procedure in those countries opposed to or hesitant about holding a referendum, although only three have explicitly refused: Germany, where a proposal to this effect by the liberals of the FDP was turned down even though the opinion polls show that most voters are in favour; Malta and Sweden. In Germany, the Bundesrat can veto any decision taken by the lower house, making it the dominant partner (besides which, no German government is likely to negotiate something that the Länder are known to oppose). Aside from these three countries, a simple parliamentary vote is an attractive option where there is a broad-based consensus; it avoids the unforeseeable consequences of a referendum. This is the case with Greece, where government and opposition are agreed on the issue and where, although the Constitution admits referendums, there is no enabling legislation establishing the procedure and the last one was held in 1974. It is also the case in Finland, where all parties represented in parliament (the Riksdagen) accept the European Constitution other than the Finnish Alliance of the Left and the Christian Democrats, which have not yet taken a stance, and the tiny ‘True Finns’, with one seat, which is adamantly opposed.

A parliamentary decision is even more desirable among those political parties which are keenly aware that previous referendums did not necessarily reflect the opinions of MPs. Both in Malta (where the EC was approved by a narrow majority) and Sweden (where a referendum recently rejected adoption of the euro as the national currency), the EC enjoys broad cross-party support in parliament, guaranteeing ratification of the Treaty. However, there are also cases in which a referendum may be the solution to the lack of support in parliament. For example, the two Czech opposition parties (the Civil Democrats and the Communists) are opposed to the Constitution, meaning that the government might lose a vote in parliament and opt for a referendum instead. Finally, the main parties in both Latvia and Poland opposed the idea of a referendum, but in each case, they are under pressure to hold one. In Latvia, the main opposition party (20 seats out of 100), the Popular Party, has called for a referendum to decide on a Constitution which would open the way to a Federal State, according the Atis Slakteris, party leader, and the possibility of a referendum cannot be ruled out. In Poland in September 2003 the Sjem rejected a motion in favour of a referendum and the two majority parties (the Democratic Alliance of the Left, or SLD, and the Civic Platform, or PO) were opposed. However, doubts have been raised as to the ability of a weakened prime minister to forge a parliamentary majority in favour of the EC, and the parties opposed to it, Justice and Law and the League of Polish Families, have begun to campaign in favour of a referendum.

Referendums on the Constitution and their implications

These cases suggest a subtle shift: whereas in the past the usual choice was for approval by parliament, now referendums have become the favoured means of approving the European Constitution at national level. The reasons are various. The primary one is where this is the legal norm for cases in which the national Constitution is to be reformed. Ireland, which has held a referendum each time a new European treaty has been passed since the Single European Act (SEA), is the leading exponent of this method of ceaseless ratification of changes in the primary legislation of the EU. In the opposite camp, a number of new members interpret that their ‘treaties of membership, approved in referendum, cover the Constitution also. This is the case in Cyprus, Estonia and Slovakia, where there is support for referendums. In Hungary, where no firm decision has been taken, it looks as though a referendum will not be held, while in Latvia, a direct vote will be sought only if the Constitution makes ‘significant changes’ to the government of the EU.

Much has been said on the huge symbolic importance of the European Constitution and on the need to involve citizens in this new social contract. Denmark, where all previous treaties other than Nice have been voted on, is the obvious example, but Spain also comes into this category, although from a diametrically opposed starting point (wide consensus in favour of the EC). The significance of this approach is underlined by the fact that referendums are novelties in two of the eight Member States that have undertaken to hold one: in Belgium (where it will not be legally binding) and in the Netherlands, where it will be the first referendum in Dutch history and where, although the result will not be binding on the government, the latter has undertaken to respect the majority view. In Italy, where a referendum was held to provide a constituent mandate for the European Parliament, there is a widespread feeling that one should be held on the European Constitution, also, but the fact that the Italian Constitution excludes its application to international treaties creates significant ‘technical difficulties’.

Figure 2. Referendums in the European Union

StateReferendum on the ConstitutionPrevious referendums on the EUResults
AustriaUndecidedMembership (1994)Yes 66%; No 33%
CyprusProbably notNo*
Czech RepublicUndecidedMembership (2003)Yes (77%); No (23%)
DenmarkYesMembership (1972)Single Act (1986)Maastricht I (1992)Maastricht II (1993)Amsterdam (1998)Adoption of the euro (2000)Yes (53%); No (33%)Yes (42%); No (33%)Yes (41%); No (42%)Yes (49%); No (37%)Yes (41%); No (34%)Yes (41%); No (46%)
EstoniaProbably notMembership (2003)Yes (67%); No (33%)
FinlandUndecidedMembership (1995)Yes (57%); No (43%)
FranceProbably yesEnlargement (1972)Maastricht (1992)Yes (68%); No (32%)Yes (51%); No (49%)
GreeceProbably notNo
HungaryProbably notMembership (2003)Yes (84%); No (16%)
IrelandYesMembership (1972)Single Act (1987)Maastricht (1992)Amsterdam (1998)Nice I (2001)Nice II (2002)   Yes (94%); No (6%)Yes (46%); No (54%)Yes (63%); No (37%)
ItalyPossibly yesConstituent mandate for the EP (1989)Yes (88%); No (14%)
LatviaProbably not Yes (67%); No (33%)
LithuaniaUndecidedMembership (2003)Yes (91%); No (9%)
MaltaNoMembership (2003)Yes (54%); No 46%
PolandProbably notMembership (2003)Yes (77%); No (23%)
SlovakiaProbably notMembership (2003)Yes (94%); No (3%)
SloveniaProbably yesMembership (2003)Yes (66%); No, (34%)
SwedenNoMembership (1994)Adoption of the euro (2003)Yes (52%); No (47%)Yes (42%); No (56%)
United KingdomYesMembership (1975)Yes (67%); No (33%)

(*) The referendum of 24 April 2004 in Cyprus was on reunification of the island.

Source: Various.

The third reason for recourse to a referendum is the most important one. It is where there are no clear parliamentary majorities and/or where there is a danger of undermining the structure of traditional political decisionmaking on which government and opposition operate within a parliament. The two most obvious instances of this are France (where calling a referendum is one of the powers of the President of the Republic and where politicians of such varying political persuasions as Giscard, Mosovici and even Villepin have repeatedly called on President Chirac to call one) and the United Kingdom (where, however, a referendum would not be legally binding). In both cases, the aim is to allow the voting public to have its say without creating a further split in the ranks of the two main parties, exposing the level of internal opposition to the official party line. While in some cases, such as the Czech Republic, euroscepticism is a new development providing one more option to an already long list of political parties, the situation in the United Kingdom and France is more worrying, as there new parties such as the United Kingdom Independence Party (UKIP) or the Movement for France, of Philippe de Villiers, have sprung up with withdrawal from Europe as their sole election platform. Both parties demonstrated their ability to attract broad swathes of voters in the recent elections for the European Parliament. In fact, together with the Swedish and Danish June Movements (Junilistan and JuniBevægelsen), plus the Dutch ChristenUnie and the League of Polish Families, they look set on establishing a powerful eurosceptic movement within the European Parliament, with opposition to the European Constitution as their primary political ambition.

But new parties are not the only problem. Europe has become a potential divisive issue for some of the traditional parties such as the UK’s Labour and Conservative parties. Now the French Socialist Party is also affected, following longstanding unrest within its ranks over the lack of ‘social’ provisions in the new Constitution.

In this context, a rash of referendums across Europe will provide the perfect opportunity to create a cross-party rift in a number of Member States between, on one hand, supporters of greater European integration and, on the other, eurosceptics or those who are content to see Europe as a purely trading platform. The resulting cleavage will leave traditional political categories such as left and right, centralist and regionalist, totally at sea.

With this prospect looming up, the use of referendums as a means of legitimising foregone conclusions may well backfire. To begin with, the turnouts will probably be small, as they were in the recent elections to the European Parliament. The opinion polls show a persistent ‘diffuse support’ for the single European project, but this will be difficult to mobilise in terms of yes votes for the European Constitution. The referendum format itself may prove more of a hindrance than a help. Referendums are notoriously unsubtle instruments, forcing voters to give a Yea or a Nay on issues of immense complexity, with an even chance that many will vote for reasons that have precious little to do with the matter in hand. Many will undoubtedly be voting to punish their national governments. Even in those countries in which there is a widespread consensus in favour of the EU and the Constitution, and the likelihood of a party political contest over the outcome is remote ­(Spain, for example), politicians may well be dismayed by the result.

Then again, the ability of the pro-European camp to mobilise its supporters is much less than that of the eurosceptics or the free-traders, as the recent parliamentary elections showed only too well. In a growing number of cases, the referendum will have to paper over wide differences of opinion on the Constitution, and appeals to ‘national interest’ and ‘free trade only’ will probably steal the show. The need to satisfy national aspirations and the need to obtain unanimous approval creates the following conundrum: the Constitution must band together sufficient majorities in each country formed by specific national circumstances. The daring proposal heard at the Intergovernmental Convention of holding a Europe-wide referendum died a quick death, with only the Austrian Chancellor coming out in favour. On the other hand, a negative vote in just one country will affect all the Member States of the European Union, preventing both the Treaty and the Constitution being implemented anywhere. Such an outcome does not appear to have been contemplated by any of the Member States. It is well known that the Danish referendum on Maastricht and the Irish vote on Nice (in both cases rejecting them) obliged the EU to find alternative formulas and hold further referendums. This has created the view among many that countries can be ‘made’ to accept whatever is put to them. In fact, the apparent exercise of the sovereignty of the people is a ruse to ‘outsource costs’, by making all countries foot the bill for one country’s default. The referendums held on the euro in Denmark (2000) and in Sweden (2003) show that when the decision is not to outsource costs onto other Member States, the EU remains totally aloof.

The chances of a single country, such as the United Kingdom or Denmark, deciding not to ratify the Treaty are real. The means of dealing with such a situation are sketchy. The Final Act of the Intergovernmental Conference says that if only four fifths of the Member States ratify the Constitution in the two years following its promulgation, the Council will take cognisance of this fact; a vague formula indeed, and one which gives no indication of how to advance in such circumstances, seeing that the legal problems arising if a dissenting country wants to remain within the EU have not been, nor (or so it would seem a priori) could be, solved.

Constitutional reform

If the two first procedures suggest future problems, the third is no less relevant, among other things because it will impinge not only on parliamentary approval but on referendums, too. This is constitutional reform, which in some cases, such as Ireland, is essential. Ireland has adapted its Constitution to all the reforms of the EU by means of Article 46.2 of the Irish Constitution which provides for ‘limited authorisation’. In other cases there are portmanteau clauses, such as Article 23 of the German Constitution (although the German Constitutional Court, in its well-known sentence by Brunner on the Maastricht Treaty, specifically allows limits on the process of European integration). In the context of the European Constitution, some people have already pointed out the need to modify national Constitutions, such as, for example, in Finland. In Spain, the proposed Article 10, which specifically establishes the supremacy of community law, has been criticised by several leading constitutionalist experts as being difficult to reconcile with the existing Article 93.

Without getting too involved in the variety of circumstances likely to emerge from this requirement, suffice it to say that ratification of the Constitution will allow the constitutional or supreme courts of the Member States to interpret how European integration advances. To judge by past results, there is scant room for optimism on a quick and efficient ratification once these august bodies get into gear.


The breadth and complexity of the procedures and circumstances outlined in this article suggest that there is a real possibility of encountering serious obstacles to ratification by all 25 Member States and, thus, a good chance of the whole process grinding to a halt. This highly likely outcome invites two questions: first, Is unanimous ratification by 25 Member States a realistic goal? and, second, Would it not be wiser to submit the qualitative changes implicit in the new Constitution to a common vote? These are questions to be discussed in the future.

Carlos Closa Montero

Political and Administrative Science, Public Law Department, Faculty of Law of the University of Zaragoza

[1] http://ue.eu.int/uedocs/cms_Data/docs/pressdata/en/misc/81243.pdf.