The Brexit Bill and the Law of Treaties

Written by

As has been widely reported in the media (e.g. The Guardian, the BBC), the House of Lords reached two main legal conclusions in its March 2017 report on Brexit and the EU budget:

  1. Article 50 TEU allows the UK to leave the EU without being liable for outstanding financial obligations under the EU budget and related financial instruments, unless a withdrawal agreement is concluded which resolves this issue.(para. 135).
  2. The jurisdiction of the CJEU over the UK would also come to an end when the EU Treaties ceased to have effect. Outstanding payments could not, therefore, be enforced against the UK in the CJEU. (para. 133).

The UK government appears to have adopted a similar position on the Brexit bill as the House of Lords. The German newspaper Frankfurter Allgemeine Zeitung published an account of a ‘disastrous Brexit dinner’ at the end of April 2017 between UK Prime Minister Theresa May and Commission President Jean-Claude Juncker in which PM May reportedly argued that the UK does not owe anything to the EU upon its departure. The fact that this dinner conversation was leaked led to strong criticism, particularly in the UK as the campaign for the general election in June is currently underway (see for example here and here).

On 3 May 2017, the UK’s Brexit Secretary David Davis in a TV interview emphasized that he had not seen any official figure of the EU’s demands, and left open room for compromise:

[The UK] have said we will meet our international obligations,  but there will be our international obligations including assets and liabilities and there will be the ones that are correct in law, not just the ones the Commission want.

However, he indicated that the UK would not pay €100 billion upon leaving the EU.

The Commission’s draft negotiating directives for Article 50 negotiations with the UK, published later on the same day, emphasize the need for a ‘single financial settlement’ of the UK’s financial obligations as a member ‘in full’ – referring to it as a ‘settling of accounts’, rather than ‘punishment’. In February, the EU Commission claimed that the UK owes the EU around €60 billion as a result of its EU membership since 1973 (which the EU Commission reportedly revised to around €100 billion in early May).

The House of Lords rightly noted the considerable uncertainty about the UK’s financial obligations arising out of its departure from the EU. Like other aspects of their past and future relationship, the UK’s financial obligations are primarily a matter for negotiation between the EU27 and the UK. The EU27 and the UK will bargain about the UK’s financial obligations in the shadow of the UK’s legal obligations. And if the EU 27 and the UK were to fail to reach agreement on this point, the legal position is even more important.

This blog post shows that both the conclusions of the House of Lords, and the UK government’s apparent position, are likely erroneous. It argues first that the UK is liable, in principle, for a share of the EU’s budget commitments which all current EU member states (incl. the UK) have assumed, as well as for pensions of EU officials. Second, the European Court of Justice  may well have jurisdiction over the UK’s financial obligations arising out of its membership in the EU. It first considers the UK’s liability in principle, before turning to the jurisdiction of the European Court of Justice.

The UK’s Liability in Principle

The House of Lords relies on Article 50 TEU for its conclusion that the UK is not liable for financial obligations arising out of the UK’s EU membership. Article 50 provides:

Any Member State may decide to withdraw from the Union in accordance with its own constitutional requirements.

A Member State which decides to withdraw shall notify the European Council of its intention. In the light of the guidelines provided by the European Council, the Union shall negotiate and conclude an agreement with that State, setting out the arrangements for its withdrawal, taking account of the framework for its future relationship with the Union. That agreement shall be negotiated in accordance with Article 218(3) of the Treaty on the Functioning of the European Union. It shall be concluded on behalf of the Union by the Council, acting by a qualified majority, after obtaining the consent of the European Parliament.

The Treaties shall cease to apply to the State in question from the date of entry into force of the withdrawal agreement or, failing that, two years after the notification referred to in paragraph 2, unless the European Council, in agreement with the Member State concerned, unanimously decides to extend this period.

Article 50 TEU is an example of a withdrawal clause. Such clauses are found in many treaties. Withdrawal clauses allow states to evolve from a party to a non-party without breaching the treaty. They allow a unilateral, voluntary departure from the treaty. Upon withdrawal, the treaty is no longer binding on the withdrawing party. It brings treaty relationships to an end.

The House of Lords’ report takes the view that Article 50 contracts out of customary international law on withdrawal (the report refers specifically to Article 70 of the Vienna Convention on the Law of Treaties, ‘VCLT’). Articles 70 (1), which is customary international law, provides:

Unless the treaty otherwise provides or the parties otherwise agree, the termination of a treaty under its provisions or in accordance with the present Convention:

  • releases the parties from any obligation further to perform the treaty;
  • does not affect any right, obligation or legal situation of the parties created through the execution of the treaty prior to its termination.

A preliminary question – which the House of Lords’ report correctly answered in the affirmative – is whether international law, and specifically the VCLT, applies to the EU treaties. While EU law is a new legal order of international law, this order still exists against the background of general international law, particularly regarding basic questions of treaty law. International law serves as a fall-back for particular issues that the EU treaties do not regulate, or regulate only incompletely. This is the position with respect to Article 50 TEU, as the paragraphs below show.

Crucially, the House of Lords’ report concludes that Article 50 TEU is lex specialis, or an exception, expressly mentioned in the first part of Article 70 (1) VCLT. In other words, because the TEU contains a specific provision on withdrawal, Article 70 is irrelevant to the UK’s financial obligations arising out of its EU membership.

In reality, however, Article 50 TEU only partly contracts out of relevant customary international law. For the most part, Article 50 TEU lays down a specific procedure (the timeline and notification requirements) for a member state to withdraw from the complex edifice of the EU treaties. It also confirms the default rule in Article 70 VCLT that the EU treaties cease to apply from the critical date (Article 70(1)(a) speaks of “releases the parties from any obligation further to perform the treaty”).

That said, the contracting out is only partial because the VCLT has three provisions that deal, in part, with withdrawal, rather than just one. In addition to Article 70 VCLT, there are Articles 56(2) and 65-68 VCLT. Article 56 (1) is about the possibility of withdrawal in the absence of a withdrawal provision. Given the presence of Article 50 TEU, it is not relevant for present purposes. Article 56 (2) establishes a default notice period of 12 months. Article 65-68 provides for default procedural steps for, among others, withdrawal from a treaty (the same default steps also apply to invalidity, termination or suspension).

Contrary to the House of Lords’ report, Article 50 is no exception to Article 70 (1) (b) VCLT. All that Article 50 does is depart from the default procedural rules set out in Article 56 (2) and Articles 65-68. For instance, Article 56(2) TEU lengthens the notice period to 2 years (rather than the default period of one year). It also confirms the release from obligations under treaties going forward under Article 70 (1)(a).

The House of Lords’ report errs in taking the more specific procedural steps for withdrawal in Article 50 TEU as evidence that Article 50 contracted out of all other rules of international law on treaty withdrawal. Article 70(1) is the decisive provision for the UK’s financial obligations arising out of its EU membership because it expressly states that withdrawal does not affect these rights and obligations (“does not affect any right, obligation or legal situation of the parties created through the execution of the treaty prior to its termination”).

Importantly, the general rule is that silence of the treaty parties does not mean contracting out of customary international law. In other words, there is a presumption that there is no contracting out – but like other aspects of the law of treaties, this is ultimately a matter for interpretation. The UK may be able to rebut the presumption – for instance by adducing travaux préparatoires that show that member states intended to contract out of the VCLT’s provisions on withdrawal in their entirety. But this is likely to be a long shot – particularly given that the travaux expressly say that they are ‘partly inspired by the Vienna Convention on the Law of Treaties’.

In sum, Article 50 is lex specialis only with respect to the procedure for withdrawal. It preserves the customary international law rule in Article 70 VCLT (1) (b) that the rights and obligations of the treaty parties prior to withdrawal are untouched. And importantly, this includes the UK’s financial obligations arising out of its EU membership.

Whither the CJEU’s Jurisdiction?

The House of Lord’s second conclusion mentioned above would mean that the main compulsory dispute settlement mechanism is unavailable in the event that the EU and the UK fail to agree on the UK’s financial obligations on its departure from the EU.

It is uncontroversial that once the UK leaves the EU, the European Court of Justice will no longer have jurisdiction over the UK in respect of the EU treaties (unless the EU27 and the UK otherwise agree). Conversely, prior to the critical date, the EU treaties continue to apply to the UK and the UK remains subject to the European Court’s jurisdiction. Under Article 50 TEU, the critical date is either March 29 2019, unless the UK and the EU27 unanimously extend this period, or the date of entry into force of the withdrawal agreement.

However, there is an important temporal caveat. Neither Article 259 TFEU (infringement proceedings) nor Article 267 TFEU (preliminary rulings) contains any temporal limitation to the CJEU’s jurisdiction. Article 259 TFEU merely refers to a situation in which “another Member State has failed to fulfil an obligation under the Treaties”, and Article 267 TFEU refers to “the interpretation of Treaties”.

First, one, several or all EU member states other than the UK could request that the Commission examine whether the UK has failed to fulfil its obligations under the treaties. To the extent that the UK’s financial obligation at issue arose prior to the critical date, any dispute concerning such an obligation is likely to fall within the CJEU’s temporal jurisdiction because it concerns an allegation that the UK ‘has failed to fulfil an obligation under the treaties’ and this dispute arose before the critical date. As a result, the dispute over the UK’s financial obligations could reach the CJEU in the form of infringement proceedings against the UK.

Second, the CJEU’s temporal jurisdiction might also extend to give a preliminary ruling on the UK’s financial obligations arising out of its departure. The CJEU’s interpretation of EU law is likely binding on the UK with respect to obligations that arose prior to the UK’s withdrawal from the EU. However, a national court would need to refer such a dispute to the CJEU in the first place. Given that this is a dispute between the EU27 and the UK, it is not straightforward how this dispute would appear before a national court of an EU member state (given that the individuals and companies that are the ultimate beneficiaries will look to the EU, rather than the UK for payment, for example retired Commission officials).

Of course, it would be possible for the EU27 and the UK to agree on another method of dispute settlement, such as inter-state arbitration. However, to the extent that the CJEU is likely to have temporal jurisdiction over the UK’s financial obligations arising from its membership in the EU, the EU has few incentives to opt for inter-state arbitration instead. Finally, the EU has allegedly also considered the International Court of Justice as a possible forum for such a dispute – but this is unlikely to be the EU’s preferred option.

The ICJ would only have compulsory jurisdiction to the extent that the UK’s optional clause declaration and the optional clause declarations of other EU member states overlap (the minimum common denominator). A case against the UK would face several hurdles in the ICJ, though these may be surmountable. First, as is well known, only states could be claimants (rather than the EU in its own rights). Second, there is no existing example of multiple states bringing a case against a state in the ICJ. Third, six EU member states have not made an optional clause declaration in the first place, and two further EU member states are members of the Commonwealth (Malta and Cyprus). As a result, the carve out in the UK’s optional clause declaration for disputes with other Commonwealth member states would apply.

Print Friendly, PDF & Email


Leave a Comment

Comments for this post are closed


Elvira Dominguez-Redondo says

May 4, 2017

EU countries may not have any incentive to use another mechanisms when they have their own, but in a dispute between the EU and the UK as a third party, it seems that the European Court would be part in its own cause. This kind of disputes (if no deal is reached) are normally settled through arbitration or claims commissions. It is, in my view, the most likely outcome.

I don't see how the ICJ would deal with a case between an IO and a State, unless its advisory jurisdiction is misused to resolve a contentious case (procedural obstacles seem unsurmountable too). The only precedent (and it is not very useful because the Court avoided the issue) was the case brought by Serbia and Montenegro against NATO countries. However, the consideration of NATO as an IO at the time was questionable

Michael Waibel says

May 4, 2017

Dear Elvira,

Many thanks for your comments.

The CJEU, if it came to formal dispute settlement over this issue and the CJEU indeed had jurisdiction, seems the natural forum for this dispute to be resolved. Under the treaties, it is the compulsory and exclusive dispute settlement mechanism for questions regarding the interpretation of EU law, including the extent of the UK's financial obligations under the treaties on departure. In addition, it has unrivaled expertise on EU law.

I agree that it could be problematic for the UK not to be represented on the CJEU after the UK's withdrawal from the EU to the extent that the Court still were to resolve disputes that concern the UK. One way of addressing this concern would be for the withdrawal agreement to provide for the appointment of an hoc judge to the CJEU in such cases.

As my post noted, the ICJ would not have jurisdiction over any dispute involving any international organisation, including the EU. In principle, it may be possible for EU members acting jointly as states to initiate contentious proceedings against the UK, subject to the caveat explained above. Unlike the NATO case (which involved multiple respondents), it would involve multiple claimants.

Dapo Akande says

May 4, 2017

Hi Michael,

Thanks very much for this post. Although it is more common in the ICJ to have claims by a single claimant against multiple respondents than claims by multiple claimants against a single respondent, the latter is not be entirely without precedent. If multiple means more than one, there is the South West Africa cases where two states (Liberia & Ethiopia) brought claims against South Africa. In PCIJ, there was the Wimbledon case a group of states (the Great Britain, France, Italy and Japan) brought a case against a single state – Germany – and did so by means of a single application.
I mention these cases in an earlier post dealing with a trivia questions on provisional measures and joinder at the ICJ.

PS I knew those trivia questions and answers would be useful one day!
PPS Stay tuned for a post asking brexit-related trivia question. Perhaps later today!

Pierre d'Argent says

May 4, 2017

Dear Michael,

Very interesting and timely post -- thanks. Two quick points, in reverse order.

1. I agree that ECJ infringment proceedings brought prior to withdrawal date (not withdrawal notification) do not need to be concluded before that date and that the ECJ would retain temporal jurisdiction after such date, and that its judgment would still be binding on UK -- precisely because of Art. 70 (1) VCLT to which, as you rightly point out, Art. 50 does not derogate from.
However, I think the UK and the EU27 may not agree to put the "Brexit bill" dispute to international arbitration (instead of ECJ) prior to the withdrawal date; in my opinion, this would be in breach of Art. 344 TFEU.

2. I agree with you that the House of Lords errs in considering Art. 50 as a derogation from Art. 70 (1) VCLT. However, interestingly, the House of Lords' legal advice admits that Art. 70, 1 b) VCLT means that “existing rights and obligations under the treaty must be respected until they are fulfilled” (para. 11).

Therefore, the core issue is to know what are the [financial] “obligations… created through the execution of the treaty prior to its termination”, knowing that any such obligation is not affected by the withdrawal of the UK, from the date of such withdrawal? In my opinion, a reasonable argument can be made to the effect that budgetary commitments decided prior to the date of the effective withdrawal are indeed “obligations… created through the execution of the treaty”. If that is the case, then the UK is correct to consider that such obligations are left unaffected by the withdrawal, which means that they must still be fulfilled after it – which does not amount to performing the treaty further.

However, if (i) the UK is correct about fulfilling obligations “which arise as a result of participation in that treaty” (legal advice, para. 11) but (ii) that the UK it is wrong, as you rightly point out, about the setting aside of Art. 70, 1 b) VCLT by Art. 50 TUE, then it must be concluded that past British commitments to the EU budget stand under international law, together with past EU budgetary commitments in favour of the UK.



Jakob Cornides says

May 4, 2017

This is probably one of the best posts I have recently read on this blog, raising interesting legal issues rather than just stating political opinions. I agree with the author that Art 50 of the EUT probably cannot be read as a derogation of Art 70 VCLT.

While as an EU official I must be careful not to make any comments that might seem disloyal against my employer, I nevertheless feel tempted to make the following observations:

First, I think that real life does not always follow the script that lawyers think it should follow, and secondly the script is at times open for diverging interpretations. In this case, the main question that arises is: what precisely are the UK’s “outstanding financial obligations under the EU budget and related financial instruments” that have been “created through the execution of the treaty prior to its termination”?

There are certainly many obligations that the UK government cannot, and will not, seriously contest, such as the UK’s fair share in paying the salaries of EU staff up to the date on which the UK will leave the EU, or the pensions of staff who, like I, have worked for many years to defend and further the interests of (inter alia) the UK. But there are other obligations that are far more debatable. For example, while the EU budget is proposed annually by the European Commission, it is pre-determined by the Multiannual Financial Framework, laid down for a seven-year period by the Council (requiring the unanimous approval of every Member State). The current Multiannual Financial Framework extends from 2014 to 2020, i.e. beyond the anticipated day of Brexit, and the UK has agreed to it. It could thus be argued that the UK is still liable to fulfil its obligations under this Framework, even if its membership ends in March 2019. I gather that much of the divergences between the figures that are currently being tossed around by various sources stems from the question whether it is already the Framework, or only the annual budgets, that create financial obligations. One should not underestimate the difference that this makes: obviously, whatever approach is going to be taken the obligations will be to some extent counter-balanced by corresponding entitlements – but given that the UK is a net payer, the former are in any case likely to exceed the latter. Other Member States are have already planned their budgets on the basis of the payments they are expecting to receive from various EU funds...

I guess that rather than tossing around fancy figures, it would be far better to seek agreement on a principled approach, distinguishing between various categories of uncontested and (more or less) debatable claims. This could take some emotion out of this debate. The UK (not only the Government, but also the populace) has manoeuvred itself into a very weak bargaining position – but it has not lost a World War, and this is hopefully not going to be a new Versailles Treaty engineered by some latter-day Clemenceau. This separation, if it really has to take place, should be negotiated in a gentleman-like manner. The EU cannot afford giving the image of a club in which members stay members not because they view membership as beneficial, but only for fear of being punished. Besides this, the EU and the UK will in any case remain neighbours and trading partners - and it would be unwise to handle Brexit in a way that leaves the UK in the role of a disgruntled and hostile ex-member...

As regards the famous “Disaster Dinner”, I guess that what was leaked by and large corresponds to reality: the UK government seems to be grossly under-estimating the difficulties that need to be resolved, and/or over-estimating the strength of its negotiating position. Brexit cannot be turned into a “success”, it is a lose-lose situation that inflicts serious damage on the interests of both sides. These negotiations are about damage control and burden sharing, let us be realistic about this.
Finally, the interesting question – as in any negotiation – is: what happens if there is no agreement? Whatever the legal situation may be, I have serious doubts that the EU could then simply turn to the CJEU or the ICJ, get a judgment, and have it enforced in the UK. Things would play out differently, probably in a more chaotic and nasty way.

Lorand Bartels says

May 5, 2017

Nice work. Some questions about the ICJ option. First, do the MS have any legal interest? The bill concerns the EU, not them. Second, would this not be precluded (presumably via inadmissibility) by the exclusive jurisdiction of the CJEU on disputes directly involving EU law involving the MS (different from indirectly involving EU law, even if Opinion 2/13 did not see this distinction)? Third, one implication of the jurisdictional hurdles you identify is that for some of the EU Member States the indispensable third party rule applies. one question: can this be waived? I would say so, but whether it would be waived is another question.

Gary says

May 7, 2017

So if the EU makes any decisions in the next 22 months then the UK may be obliged to pay towards those after departure?
So the UK will not agree the MFF mid term review?
And will need to block all EU business?

Michael Waibel says

May 9, 2017

Many thanks Dapo, Pierre, Jakob and Lorand for your very helpful comments and questions and for engaging with my blog post.

- The Article 344 TFEU issue: I agree with both Pierre and Lorand that Article 344 TFEU would prevent the EU27 and the UK from submitting a dispute concerning the UK’s financial obligations to a dispute settlement mechanism other than the CJEU. This option only exists as from the withdrawal date. Prior to withdrawal, other fora, such as the ICJ, should conclude that there is a temporary impediment to the exercise of their jurisdiction, and declare the action inadmissible.

- Pierre, I believe we are also on the same page with respect to fulfilment of existing rights and obligations. The recognition by the Legal Advisor to the EU Committee of the House of Lords that ‘existing rights and obligations under the treaty must be respected until they are fulfilled’ (para. 11) means that to the extent that Article 50 TUE is only a partial carve out from Article 70 VCLT – the position the blog post adopted – the UK is liable in principle for financial obligations arising out of EU membership

- That brings us squarely to the central question that Jakob identified: what exactly are the UK’s financial obligations? Jacob’s suggestion to seek agreement on a principled approach seems to me the best way forward. Alex Barker of the Financial Times has provided the best break down of the various liabilities in play.

- Finally, while the ICJ option is very unlikely in practice, Lorand points to two further, potentially important obstacles in contentious proceedings brought by a group of EU members states against the UK. I agree that because the financial obligations are likely owed to the EU (and not the individual member states), member states may lack standing due to a lack of a legal interest. The indispensable third party rule could probably be waived, but in all likelihood only with the agreement of the respondent.