Unilateral withdrawal from the EU or EMU

A. A legal right of withdrawal? Doctrinal considerations

1. Withdrawal under the treaties

Unlike the conditions for accession to the EU, which are addressed, even if not exhaustively, in Article 49 TEU, neither the founding treaties (which, with the exception of the European Coal and Steel Community (ECSC) Treaty, have no specified term), nor the successive amending treaties made until the ratification of the Lisbon Treaty, made any provision for a Member State's withdrawal (negotiated or unilateral) from the EU or EMU. As one author has written, there are three main reasons why the treaties were silent on withdrawal: first, it was in order to avoid putting question marks to the Member States' commitment to the achievement of their shared objectives; second, it was because providing for the possibility of withdrawal might have increased its likelihood; and third, because to provide for this possibility would entail the daunting task of spelling out the procedure and consequences of withdrawal. Whatever the force of the first two reasons, there can be little doubt that disentangling the legal and other implications of withdrawal would be a formidable challenge. For instance, withdrawal from EMU would entail: (i) creating a new currency or re-establishing the old currency of the withdrawing Member State; (ii) refunding the departing national central bank's (NCB) contribution to the European Central Bank's (ECB) capital, and reimbursing its foreign reserve assets transferred to the Eurosystem; and (iii) transferring full monetary sovereignty back to the seceding NCB, with all the practical difficulties and legal uncertainties that this would involve for outstanding monetary policy operations, especially in the case of unilateral withdrawal. As for a Member State's withdrawal from the EU, the complexities surrounding it are legion, affecting the rights and obligations of every natural or legal person inside or outside the territory of the withdrawing Member State who is or who may be affected by it. Whatever the explanation for the treaties' silence on the possibility of withdrawal, its legal consequence is that, until now, no Member State has had the clear option of withdrawing from the EU or EMU, especially not unilaterally. This situation is about to change now that the Lisbon Treaty has been ratified by all 27 Member States.

The silence of Community primary law on the existence or otherwise of a legal right of withdrawal was, in any event, inconclusive, lending itself to two fundamentally opposed interpretations. One is that a right of unilateral withdrawal existed even in the absence of any explicit reference to it in the treaties, since sovereign States were, in any case, free to exercise their sovereign right to withdraw from their international commitments. The other is that the lack, until recently, of a formal exit clause in Community primary law must have been intentional, testifying to the Member States' lasting commitment to the EU's objectives and to the irreversibility of the European unification process, which is irreconcilable with a unilateral right of withdrawal. There were, pre-Lisbon Treaty, several indications in the treaties that the latter interpretation was no less plausible than the former. One was that both the EC and the EU Treaties were concluded for unlimited periods; another was that the original Member States' commitment 'to lay the foundations of an ever closer union among the peoples of Europe' was renewed and strengthened with each reform of the treaties, pointing to the continuity and irreversibility of the European unification project and to the apparent impossibility of secession from the EU if the EU's objectives are to be fulfilled. The explicit acceptance, by new entrants, of the acquis, the restrictions on the Member States' power to renegotiate their accession agreements (except in connection with derogations or other transitional measures), the solidarity principle expressed in Article 10 EC, and the fact that the treaties can only be amended in accordance with the procedure in Article 48 TEU also supported the view that Member States were not free to withdraw from the EU or, by extension, EMU, whether by agreement or, a fortiori, unilaterally.

2. The relevance of the Vienna Convention and of public international law

In the absence of any clear guidance in the text of the treaties on the existence or otherwise of a right of unilateral withdrawal from the EU, and pending the entry into force of the Lisbon Treaty, the Vienna Convention on the Law of Treaties (the 'Vienna Convention') was the best source for a public international law-based answer to this question. The text of the Vienna Convention suggests that although 'unilateral denunciation from treaties not providing for withdrawal is contrary to the principle of pacta sunt servanda', in some cases customary international law (of which the Vienna Convention is deemed to be a codification) recognises a limited right of withdrawal. The Vienna Convention recognises two different situations where a signatory can unilaterally withdraw from a treaty that is silent on the possibility of its denunciation. One is where it is possible to establish that the parties to the treaty intended to recognise a right of denunciation or withdrawal (or where such a right can be inferred from the terms of the treaty); the other (the clausula rebus sic stantibus or 'fundamental change of circumstances clause') is where there has been a fundamental and unforeseen change of circumstances, and: (i) the circumstances in question were an essential basis of the parties' consent to be bound by the treaty; and (ii) the change radically transforms the extent of the parties' obligations under the treaty.

The consensus among commentators is that the first of these situations was not relevant to the EC and EU Treaties. Given the terms and spirit of the treaties, and the long-term integrative nature of the EU, reflected in the permanence of its institutions and in the transfer of significant decision-making powers to the supranational Community institutions by the Member States, there is nothing to suggest that they contained a right of denunciation or withdrawal. In the case of EMU, Articles 4(2), 118 and 123(4) EC and its Protocol 24 on the transition to the third stage of monetary union expressly refer to the 'irrevocable' fixing of the conversion rates at which national currencies are to be exchanged for the euro, and to the irreversibility of the process leading to the adoption of the euro. The fact that EU membership is voluntary is not in itself conclusive since '[S]overeignty is ... given full expression in the right of any State to join a particular organisation, or not; but once a State decides to enter an organisation it is no longer free, and its own wishes are no longer decisive'. The EC Treaty's recognition of the existence of Member States with a derogation from EMU is similarly inconclusive, since EMU participation is, in the long run, obligatory for all Member States except for those that have negotiated opt- outs. As for the fundamental change of circumstances doctrine, which is accepted however reluctantly by most jurists, the consensus is that, whether in the context of the treaties or in connection with any other international agreement, its practical utility as a ground for withdrawal is limited to exceptional circumstances and that it should only be invoked sparingly, given the obvious 'threat it poses to the stability of treaties' and to the sanctity of contracts. At least one commentator has convincingly challenged the conceptual compatibility of the fundamental change of circumstances doctrine with the Community legal order, concluding that it 'might have difficulty passing legal muster in the EEC'. Another commentator has argued that its application should be limited to changes of circumstances that the signatories to a treaty have expressly identified, or to situations where the European institutions, including the Court of Justice (ECJ), have acted ultra vires and in breach of the treaties, leaving a Member State with no option but to withdraw.

To argue that, because public international law sometimes recognises a limited right of unilateral withdrawal, the same right must have existed by analogy in the context of the treaties prior to the Lisbon Treaty would be to err threefold. First, it would be to disregard the sui generis constitutional nature of the Community legal order and the ECJ's well-established interpretation of the treaties as being permanently binding on the Member States. Second, it would be to subscribe to an extreme and largely obsolete concept of sovereignty, at least as regards the relations of the EU Member States with each other. Third, it would be to overemphasise the affinities between public international law and Community law. On the first and second of these points, see the van Gend en Loos case where the ECJ famously ruled that '[T]he European conomic Community constitutes a new legal order of international law, for the benefit of which the States have limited their sovereign rights, albeit within limited fields'; and the Costa v ENEL case, where the ECJ stated that:

'By creating a Community of unlimited duration, having its own institutions, its own personality, its own legal capacity and capacity of representation on the international plane and, more particularly, real powers stemming from a limitation of sovereignty or a transfer of powers from the States to Community, the Member States have limited their sovereign rights and have thus created a body of law which binds both their nationals and themselves ... The transfer by the States from their domestic legal system to the Community legal system of the rights and obligations arising under the Treaty carries with it a permanent limitation of their sovereign rights'.

On the impact of sovereignty on the fulfilment by any State of its international commitments, the Vienna Convention's narrowly circumscribed right of denunciation or withdrawal clearly suggests that sovereignty does not automatically absolve States from their treaty obligations. In other words, a claim of sovereignty is not a fool-proof legal defence to a State's failure to comply with its voluntarily assumed obligations, nor can it entitle a State to withdraw unilaterally from the performance of its obligations as it sees fit. As for the EU dimension of sovereignty, it can no longer be argued that, in their intra-Community relations, the EU Member States retain the full measure of sovereignty which they can exercise by withdrawing unilaterally from the treaties. The Member States have ceded some of their sovereignty, not to the EU as such (which is not a 'State'), but to its supranational institutions, including the ECJ, in which they are represented. Already decades ago the ECJ expressly recognised the obligation of national courts to set aside their domestic law, including constitutional law, if it conflicts with Community law. The foregoing is less an argument in support of the bold and unrealistic proposition that Member States have altogether surrendered their sovereignty, and more a reminder that the concept of sovereignty subscribed to by the advocates of a free right of withdrawal is outdated. This is at least so in the context of the EU, given the challenge to a Hobbesian 'all or nothing' understanding of sovereignty represented by directly applicable Community law for the best part of 50 years. The position has been aptly summarised as follows: 'The sovereignty of member states has not been lost but rather subjected to a process of division and combination internally, and in a way enhanced externally. But the process of division and combination has taken us beyond the sovereign state'.

On the link between Community law and public international law and on the utility of public international law as a source of guidance on the existence or otherwise of a Member State's right of unilateral withdrawal, Community law differs markedly from public international law, despite the similarities which have led many commentators to treat Community law as a successful emanation of public international law. The ECJ has clearly established that one of the key differences between Community law and public international law is that 'with the latter, the effect of a norm in the national legal order is determined by national law, not international law; inEC law however, such effect is a matter of EC law, not national law'. This fundamental difference between the application of international treaties in the domestic law of the Member States on the one hand, and the application of the EC/EU Treaties in domestic law on the other hand, not only makes a nonsense of the traditional distinction in international law between monist and dualist systems of law, but ultimately 'justifies the conclusion that the Community constitutes a separate legal system that may be contrasted with traditional public international law in terms of institutional structures and outcomes'. It follows that whatever limited rights of unilateral withdrawal States may enjoy under public international law and the Vienna Convention, these rights are not necessarily immediately relevant in an EU context, where a right of unilateral withdrawal would be at odds with some of the basic assumptions underlying the Community legal order and the idea of a lasting union among the peoples of Europe, as a 'response to the specific internal logic of post-war Europe'.

3. A right of unilateral withdrawal as a remedy?

On a more practical level, the extent to which unilateral withdrawal is legally possible must ultimately depend on whether, viewed as a remedy or a relief measure from the withdrawing Member State's standpoint, secession would be compatible with the scheme of the treaties. There are three hypothetical circumstances where a Member State could, in extreme circumstances, assert a right of unilateral withdrawal, whether as a remedy or by way of relief. These are where: (i) another Member State(s) has fundamentally infringed and continues to infringe the treaties; or (ii) the European institutions have acted ultra vires (in both of the foregoing cases, without the treaties appearing to offer any remedy guaranteeing an early return to legality); or (iii) a Member State faces extraordinary difficulties that prevent it complying with its treaty obligations. While such circumstances are within the realm of the possible, the remedies provided in the treaties for infringements of Community law do not include a right for an aggrieved Member State to withdraw 'in protest', and difficulties that temporarily prevent compliance with treaty obligations are already catered for in Community primary law in ways that are inconsistent with the assertion of a unilateral right of withdrawal. Specifically, in order to ensure compliance with the EC Treaty, the ECJ has jurisdiction to hear cases brought against a Member State by the Commission or by another Member State. The EC Treaty also gives Member States an unlimited right to challenge, before the ECJ, the legality of acts of the European institutions or the ECB that are 'intended to produce legal effects vis-à-vis third parties' or where their failure to act constitutes an infringement of the Treaty. In both of these cases there is a treaty obligation for the European institution whose acts have been declared void or whose failure to act has been established by the ECJ 'to take the necessary measures to comply with the judgment of the Court of Justice'. Similarly, where a Member State's grievance against any of its partners relates to a 'serious and persistent breach' of the fundamental principles laid down in Article 6(1) of the EU Treaty, the Council may suspend 'certain of the rights deriving from the application of this Treaty to the Member State in question', including its voting rights in the Council. Where extraordinary domestic or international situations affect a Member State's ability to fulfil its treaty obligations, the EC Treaty provides for the possibility of Member States taking temporary measures, in derogation from the Treaty, in order to resolve 'serious internal disturbances affecting the maintenance of law and order, in the event of war, serious international tension constituting a threat of war, or in order to carry out obligations it has accepted for the purpose of maintaining peace and international security'. These exclusive treaty measures are applicable to a Member State's infringements of Community law and/or to extraordinary situations affecting its ability to meet its treaty obligations, and what they clearly show is that the unilateral withdrawal of an aggrieved or distressed Member State was never contemplated as a legitimate remedy or relief measure within the scheme of the treaties. Only if these remedies and/or relief mechanisms were to fail altogether, or if a novel situation were to arise which the treaties do not cater for, could a right of unilateral withdrawal perhaps be asserted, and even then only as a last resort, if the consequences of the continuing infringement of the treaties or of the crisis affecting a Member State were of a nature such as to leave it no option but to withdraw. Even then, a question would remain about the compatibility of a right of unilateral withdrawal with Article 48 TEU and with the procedure provided in it for amending the treaties.

Without prejudice to the possible application of the doctrine of fundamental change of circumstances, the conclusion is that the assertion of an implied right of unilateral withdrawal from the treaties, even in exceptional circumstances, would be highly controversial (especially in the case of EMU, where in the text of the EC Treaty it is clear that no such right was intended) except, perhaps, as a last resort in the event of an extremely serious and lasting infringement of the treaties or extraordinary circumstances affecting a Member State's ability to fulfil its treaty obligations. While public international law or the Vienna Convention may be useful sources of guidance on the existence or otherwise of a right of unilateral withdrawal, their utility is limited to contradicting the argument that, as the EU Member States are sovereign, they can unilaterally renounce their obligations under the treaties. As the law stood pre-Lisbon Treaty, 'the only acceptable way to withdraw from the Union is an agreement, ratified by all Member States'.

Needless to say, these conclusions do not touch upon the all-important political aspects of national sovereignty. It is, no doubt, political considerations that explain why, despite the founding treaties' silence on the possibility of secession, no Member State contested the UK's threatened withdrawal in 1975 and why Greenland was allowed to leave the European Communities in 1982, following domestic opposition to the common fisheries policy and growing demands for home rule. As one commentator has observed, '[A]s a practical matter if a Member State were determined to withdraw, the EEC has no sanctions that can be applied to compel lawful compliance with the Treaty. Thus, from this point of view, it really is of no consequence whether a legal right of withdrawal exists'. Nevertheless, the conclusions drawn in the preceding discussion provide useful insights into the rationale for the introduction, in the text of the new Treaty, of a clause expressly providing for the voluntary withdrawal of a Member State from the EU as well as a basis for assessing its implications in a Community law context. The recently enacted exit clause is considered is some detail in the following section.