Sunday, February 23, 2020

, , , ,

How do you solve a problem like Suella? The legal aspects of breach and termination of the withdrawal agreement






Professor Steve Peers, University of Essex

Today’s papers bring another storythat the UK might be planning to breach the Brexit withdrawal agreement – including the report that the new Attorney-General Suella Braverman is poised to attempt to justify this legally. So what happens if the UK (or the EU) breaches the withdrawal agreement is not a hypothetical issue. I’ve previously discussed the dispute settlement rules in the agreement, as well as the Withdrawal Agreement Act giving effect to the treaty in the UK, but it’s useful to bring the key issues together, and also to look at the issues from the point of view of international law.  This blog post summarises in turn the dispute settlement provisions of the withdrawal agreement and the rules on termination etc of treaties in international law, discussing the interaction between them.  (Update, Feb 24 2020: I've summarised the main points of this blog post in a Twitter thread)

A few basic parameters first. It’s likely that any (alleged) breach of the withdrawal agreement would have political consequences as well as face legal action. Political consequences might include not only reaction from the EU, but reluctance of the US Congress, concerned about the position of Ireland, to agree to the legislation implementing a US/UK trade deal. Which of the two would be most important is hard to guess in advance. It’s also harder to guess the route that the political aspects of the dispute would take, as compared to the legal route, which in principle follows a set of rules laid down in advance. Also the two routes will overlap: as discussed below, there’s political discretion as to whether and how to pursue legal proceedings.

Second, international law aims for treaties to remain in force if possible. A breach of a treaty by one side does not trigger its automatic collapse, or even an automatic reciprocal breach by the other party. The goal is to resolve disputes about treaties, not terminate them with every passing spaff.

Third, it’s not just the view of the parties to the agreement (the UK and EU) that will be relevant. Arguments about the breach of the treaty can be brought via the courts of both sides by individuals; and even the termination of the treaty could be challenged via the national courts and the CJEU, at least on the EU side. 

Fourth, the withdrawal agreement is distinct from the political declaration on the future relationship (discussed here), which is not binding as such. While the withdrawal agreement does require the parties to negotiate their future relationship in good faith, it might be difficult to establish a breach of such a vague obligation. The withdrawal agreement, including in particular the provisions on citizens’ rights (discussed here) and the Northern Ireland protocol, is also distinct from any failure to negotiate any treaties on the future relationship: in the event that negotiations on future treaties between the UK and EU are unsuccessful, this does not lead to automatic termination of the withdrawal agreement.

Finally, I am not assuming that the UK government has breached or will breach the withdrawal agreement, just discussing what might happen if it did from a (mainly) legal perspective. Recent statements by the UK government refusing to carry out checks in the Irish Sea have not been framed as an intention to breach the agreement, and the relevant part of the treaty doesn’t apply yet, so can’t yet be breached in practice. Although the title of this blog post focusses on the UK, because it’s the UK government’s recent statements that have caused some people to doubt its intentions to apply the withdrawal agreement, it’s also possible that the EU might breach the withdrawal agreement. The blog post therefore applies equally to any breach that might be committed on the EU side.

It’s obviously relevant that, as things stand, the UK has passed an Act of Parliament to give effect to the withdrawal agreement in domestic law. While a further Act of Parliament could amend the current Act in order to give effect to an intention to breach the agreement, there’s no proposal to do so at present. Any secondary legislation or government action in breach of the agreement could be struck down by the courts; although the government has indicated an intention to restrain judicial review of government action, it has not put its intentions into effect yet. (While the recent Withdrawal Agreement Act gives the executive powers to direct the judiciary how to interpret retained EU law, such powers do not apply to the withdrawal agreement itself). 
   
Dispute settlement and the withdrawal agreement: an overview

There are two distinct parts to the withdrawal agreement (and the UK legislation giving effect to it fully reflects this). During the transition period (discussed here), applying to the end of 2020 unless the two sides agree to an extension of one or two years (which the UK currently opposes in principle), substantive EU law as such, including all the usual rules of CJEU jurisdiction (references from national courts on the interpretation of EU law, Commission infringement actions for alleged breach of EU law) still apply to the UK. There’s no provision to reduce the transition period, either unilaterally or bilaterally, whether as a ‘punishment’ for not complying with the agreement or for any other reason.

After the end of the transition period, things change significantly. Most substantive EU law stops applying to the UK, along with most CJEU jurisdiction. There are special rules for cases pending at the end of the transition period, along with pending proceedings that might lead to litigation later. There is also special jurisdiction for the CJEU over citizens’ rights (for eight years after the end of the transition period), for disputes over EU budget legislation, and for cases concerning parts of the Irish border protocol and the whole protocol on UK army bases in Cyprus.

Furthermore, the main dispute settlement system then also kicks in. It provides for disputes on the interpretation of the agreement between the two sides to go to arbitration, if they cannot be settled by negotiation.  There’s nothing in the agreement to rule out parallel cases under national and EU courts on the one hand, and the dispute settlement system on the other.

There’s a general obligation for the parties to stick to the processes in the withdrawal agreement to settle disputes about that agreement (Article 168, which applies from Brexit day already):

For any dispute between the Union and the United Kingdom arising under this Agreement, the Union and the United Kingdom shall only have recourse to the procedures provided for in this Agreement.

If a dispute is not settled after three months of consultation, either side may request an arbitration panel to be set up. The two sides could jointly agree to start arbitration even before that point. The panel will consist of five arbitrators, from a list of 25 to be drawn up jointly (the EU and UK have not agreed this list yet). It must be set up within 15 days of the request to set it up. If the two sides cannot agree on a list of arbitrators, or the arbitrators who will serve on a particular case, there are tie-break rules. So it will not be possible to paralyse the dispute settlement system by refusing to agree on these names. The arbitration panel must give a ruling within 12 months. If it agrees to a request to give its ruling urgently, that deadline is reduced to six months.

If a dispute raises questions of EU law (which is likely as regards citizens’ rights in particular), the arbitrators must ask the CJEU to rule on it. The CJEU’s judgment will bind the arbitrators. Asking the CJEU to rule on the EU law stops the clock on the deadline for the arbitrators to give their ruling. 

An arbitration decision is binding on both sides. But when does it have to be complied by, and what happens if it isn’t? If the arbitrators have determined a breach of the agreement, the losing party has 30 days to notify the complainant of how long it thinks it needs to implement the ruling. If the complainant is unsatisfied by this, the issue of how long is necessary to comply with the ruling can go back to arbitration.

At the end of the deadline to comply with the ruling, if the winning party is unsatisfied with what the losing party has done to comply with the ruling, it can go back to the arbitrators to ask if the losing party has properly complied with it. The CJEU might again be asked to rule on an issue of EU law.

If the arbitrators rule that the losing party has not complied with the prior ruling, then at the request of the winning party, they may impose a ‘lump sum or penalty payment’ on the losing party. The arbitrators must consider ‘the seriousness of the non-compliance and underlying breach of obligation, the duration of the non-compliance and underlying breach of obligation’ when deciding how big the financial penalty should be.

If the losing party refuses to pay up after one month, or refuses to comply with the ruling confirming its non-compliance with the first ruling after six months, further sanctions might apply. The winning party can suspend any part of its obligations under the withdrawal agreement other than the citizens’ rights part, or under another treaty as agreed between the two sides (obviously, no such other treaties exist yet). This suspension has to be ‘proportionate’, and must take into account ‘the gravity of the breach and the rights in question’. If the losing party thinks that the suspension is disproportionate, it can ask the arbitrators to rule on the point within 10 days; the suspension is postponed from taking effect until the arbitrators rule on the point.

Suspensions are meant to be temporary until the losing party complies with its obligations, or the two sides agree to settle the overall dispute. If the losing party claims later on that it is now complying with its obligations and the winning party disagrees, the arbitrators can be asked to rule on the point; again the CJEU could be involved if there is an EU law issue. If it’s agreed, or the arbitrators rule, that the losing party is now complying with its obligations, the financial penalties or suspension of obligations by the winning party have to end.

As we can see, the dispute settlement system is not a means to terminate the withdrawal agreement. Even if part of the agreement is suspended by one side, the citizens’ rights provisions cannot be suspended. The intention is that any suspension or financial penalty is temporary, until the party breaching the agreement complies with its obligations. This suggests that termination is not a possible remedy – as does the provision stating that the parties must settle disputes in accordance with the agreement.

Also, as noted above, the dispute settlement system co-exists with some jurisdiction for the CJEU (as distinct from the CJEU jurisdiction within the dispute settlement system), as well as possible litigation in national courts. Traditionally, the CJEU has said that WTO dispute settlement rulings do not form part of EU domestic law, since the EU institutions preserve their political discretion whether to comply with WTO dispute settlement rulings or face proportionate suspension of trade from the winning party under the WTO rules – which are broadly similar to the dispute settlement system in the withdrawal agreement. On the other hand, the big difference between the WTO and withdrawal agreement dispute settlement systems is that the CJEU must be involved under the withdrawal agreement, where there is an issue of EU law. Whether that is enough to convince the CJEU that it should take a different approach remains to be seen. 

For the UK, the Withdrawal Agreement Act makes no specific reference to dispute settlement rulings forming part of UK law or not. The UK’s ‘dualist’ approach to international treaties (treaties do not form part of domestic law unless Parliament says otherwise) would suggest that they do not form part of UK law, unless it could be argued that Parliament’s implementation of the withdrawal agreement in domestic law was implicitly intended to apply to dispute settlement rulings too. Also, given that the CJEU still has jurisdiction to rule on citizens’ rights via the UK courts for awhile yet, the approach that the CJEU takes to the legal effect of the dispute settlement system is relevant to the UK as far as EU citizens are concerned.

Termination of treaties

As it relates to treaties, international law has two sources: customary international law and the Vienna Convention of the Law of Treaties (VCLT). The VCLT sought to codify the customary rules but differs in some respects. Some EU countries have not ratified the VCLT; nor has the EU, as the Convention is open to States only.

The starting point in the VCLT is that the validity of a treaty or a State’s consent to be bound by it can only be impeached under the VCLT. Termination, suspension, denunciation or withdrawal can only take place under the treaty or the VCLT. Other international law still applies between the parties even if the treaty no longer does.

Denouncing, withdrawing from or suspending a treaty (as provided for in that treaty, expressly or by implication) must apply to the whole treaty unless the treaty otherwise provides. Invalidating, terminating, withdrawing from or suspending a treaty as provided for in the VCLT must apply to the whole treaty, except (among other exceptions) if there is a ‘material breach’ of the treaty, or if the ground for termination etc relates to certain parts of the treaty and they are severable from the others and they were not an ‘essential basis’ of the consent to the treaty. A State loses its ‘right to invoke a ground for invalidating, terminating, withdrawing from or suspending the operation of a treaty’ if it agrees expressly that the treaty is valid or remains in force, or its conduct suggests the same.

The conclusion of a treaty can be invalid where a State’s consent to be bound was very obviously in violation of its internal law, if an error in a treaty was an ‘essential basis’ the consent to be bound, by fraud by another negotiating State, corruption of the State’s representative procured directly or indirectly by another negotiating State, or coercion (acts or threats against the State’s representative, or the threat or use of force against the State in violation of the UN Charter). A treaty is void if it conflicts with ‘a peremptory norm of general international law (“jus cogens”)’.  

States can terminate or withdraw from a treaty only in accordance with its provisions or consent of the parties. If a treaty does not provide for termination, denunciation or withdrawal, a State cannot denounce or withdraw from it unless either: ‘(a) it is established that the parties intended to admit the possibility of denunciation or withdrawal; or (b) a right of denunciation or withdrawal may be implied by the nature of the treaty.’ Twelve months’ notice to leave is required in that case.

As for suspension of a treaty, this is possible ‘in conformity with the provisions of the treaty’ or if all the parties consent. Either termination or suspension is possible in the event of a ‘material breach’ by one party, which ‘entitles’ the other party ‘to invoke the breach as a ground for terminating the treaty or suspending its operation in whole or in part’. The VCLT defines a ‘material breach’ of a treaty as ‘(a) a repudiation of the treaty not sanctioned by the present Convention; or (b) the violation of a provision essential to the accomplishment of the object or purpose of the treaty’. But all this is ‘without prejudice to any provision in the treaty applicable in the event of a breach’ and does ‘not apply to provisions relating to the protection of the human person contained in treaties of a humanitarian character, in particular to provisions prohibiting any form of reprisals against persons protected by such treaties’.

A party can also terminate or withdraw from a treaty due to the ‘impossibility’ of applying it if that ‘results from the permanent disappearance or destruction of an object indispensable for the execution of the treaty’, but this cannot be invoked if the impossibility was caused by the party which wants to escape its obligations. Termination, withdrawal or suspension could also follow a ‘fundamental change of circumstances’ since the treaty was concluded, if it was ‘not foreseen by the parties’, if the original circumstances were an ‘essential basis of the consent of the parties to be bound by the treaty’ and ‘the effect of the change is radically to transform the extent of obligations still to be performed under the treaty’. Again, this cannot be invoked if the change in circumstances was caused by the party wanting to end its obligations.

The VCLT also regulates the process of termination or other forms of ending a treaty. A party seeking to end a treaty must give the other party three months’ notice. It can then proceed as planned if there’s no objection. But if there is an objection, the parties must try to find a solution. This does not affect anything in force regarding dispute settlement between the parties, so it is obviously possible that either the EU or UK would invoke the dispute settlement provisions in the withdrawal agreement.

If there is no negotiated solution within 12 months from raising the objection, one party may ask the International Court of Justice to rule, or both can ‘agree to submit the dispute to arbitration’, or the conciliation process set up by the VCLT itself might apply. The process of termination etc must be in writing and communicated formally, and can be revoked at any time before it takes effect.

As for the consequences of termination etc, an invalid treaty is void and has not legal force. In case of termination, ‘[u]nless the treaty otherwise provides or the parties otherwise agree’, the parties have no further obligation to perform the treaty, but this ‘does not affect any right, obligation or legal situation of the parties created through the execution of the treaty prior to its termination’.

As for suspension, ‘[u]nless the treaty otherwise provides or the parties otherwise agree’, suspension means that the treaty does not apply ‘during the period of the suspension’ and ‘does not otherwise affect the legal relations between the parties established by the treaty’. The parties must ‘refrain from acts tending to obstruct the resumption of the operation of the treaty’. As noted already, however, the withdrawal agreement has specific provisions on suspension.

In general, the various references in the VCLT to suspension and termination of a treaty unless the treaty otherwise provides need to be read in light of the rules in the withdrawal agreement itself, as summarised in the dispute settlement section above. These rule out dealing with disputes otherwise than in accordance with the treaty, allow for suspension as a possible outcome of the dispute settlement process except for the citizens’ rights chapter, and make no mention of termination of the withdrawal agreement, referring instead to the ‘temporary’ nature of partial suspension or financial penalties. The only reference to termination of the withdrawal agreement is in the Irish protocol, which provides (in Article 18 of the protocol) for the unilateral termination of part of that protocol (not the whole withdrawal agreement) in the event that it lacks continued consent in Northern Ireland, following a particular detailed procedure in that protocol. One can therefore argue that termination of the entire withdrawal agreement in response to a breach of it is therefore ruled out by the agreement itself.

In the alternative, what happens if the EU or UK believe that it is still possible to terminate the entire withdrawal agreement on the grounds set out in the VCLT? Even if actions by one side are arguably a ‘material breach’ of the withdrawal agreement as defined by the VCLT, it’s important to remember that invoking that breach in order to terminate that treaty is an option: as pointed out at the outset, a breach of a treaty by one side does not automatically terminate or suspend that treaty.  So in the event, for instance, of a breach of the Irish protocol by the UK, the EU might judge that it would be unwise to invoke its right to terminate the withdrawal agreement, as that would (among other things) throw the rights of EU citizens in the UK under the proverbial bus.

What about the role of individuals in all this? More than you might think, at least on the EU side. The international law rules on termination of treaties have been applied by the CJEU in the context of cases brought by individuals. In Racke, the EEC (as it was then) terminated the EEC/Yugoslavia cooperation treaty immediately (rather than in accordance with the terms of that treaty) on grounds of a fundamental change in circumstances, when the Yugoslav war broke out. An importer of wine from Yugoslavia, being affected by the termination of the treaty, challenged the EEC’s decision in the German courts, which asked the CJEU if the EEC’s termination of the treaty was valid.

According to the CJEU, ‘even though the Vienna Convention does not bind either the Community or all its Member States, a series of its provisions, including Article 62 [on change in circumstances], reflect the rules of international law which lay down, subject to certain conditions, the principle that a change of circumstances may entail the lapse or suspension of a treaty.’ The International Court of Justice had ruled already that on this point, the VCLT ‘may in many respects be considered as a codification of existing customary law on the subject of the termination of a treaty relationship on account of change of circumstances’.

Next, the Court asserted that it had jurisdiction to rule on the validity of an EEC act in light of the rules of public international law.  Before it applied those rules, it insisted that the rules in the treaty which the importer sought to invoke had to confer rights on individuals. It found that they did, based on its usual test for ‘direct effect’ of an international treaty in EEC (now EU) law. (This test should easily be satisfied as regards the citizens’ rights provisions of the withdrawal agreement).

The Court then observed that international treaties concluded by the EEC (now EU) form an integral part of EU law, and that if the termination of the treaty were invalid, the company would still have rights regarding the import of Yugoslavian wines.  So the EU has to ‘respect international law in the exercise of its powers’ and ‘is therefore required to comply with the rules of customary international law when adopting a regulation suspending the trade concessions granted by, or by virtue of, an agreement which it has concluded with a non-member country’. On that basis ‘[i]t follows that the rules of customary international law concerning the termination and the suspension of treaty relations by reason of a fundamental change of circumstances are binding upon the Community institutions and form part of theCommunity legal order.’

Applying these rules, the Court noted that international law was based on compliance with treaties (the pacta sunt servanda principle), ‘which constitutes a fundamental principle of any legal order and, in particular, the international legal order. Applied to international law, that principle requires that every treaty be binding upon the parties to it and be performed by them in good faith (see Article 26 of the Vienna Convention).’ Its importance had been further underlined by case law of the International Court of Justice, ruling that 'the stability of treaty relations requires that the plea of fundamental change of circumstances be applied only in exceptional cases.

Individuals could therefore invoke ‘obligations deriving from rules of customary international law which govern the termination and suspension of treaty relations’ to challenge the EU’s termination of a treaty. (Note that the Court did not confine itself to the ‘fundamental change in circumstances’ rule here). However, since the rules in question were complex and imprecise, the Court limited its judicial review to whether ‘the Council made manifest errors of assessment concerning the conditions for applying those rules’. It decided that the civil war in Yugoslavia met the conditions to justify terminating the treaty.

As for application of the procedural rules for terminating treaties in the VCLT (ie sufficient notice and consultation), the CJEU thought that prior warnings of termination were sufficient, and that ‘[e]vven if such declarations do not satisfy the formal requirements laid down by’ the VCLT, ‘the specific procedural requirements there laid down do not form part of customary international law.’

The Court came back to these procedural requirements in the recent well-known judgment in Wightman(discussed here), on the issue of whether the UK could unilaterally revoke its notification to leave the EU. In the Court’s view, its various reasons for ruling that unilateral revocation was possible were ‘corroborated’ by the provisions of the VCLT, ‘which was taken into account in the preparatory work for the Treaty establishing a Constitution for Europe’ (which later became Article 50 TEU). So the ‘clear and unconditional’ possibility in the VCLT to revoke a notification of withdrawal from a treaty before it takes effect supported the view that the Article 50 notification could be revoked too.

How would this work in practice for the withdrawal agreement? If the EU terminated or suspended the agreement, anyone who argued that their rights conferred by the agreement (UK citizens in the EU, or traders with Northern Ireland, for instance) were affected as a result could bring a challenge to the EU’s decisions, arguing that they were in violation of public international law generally and/or the withdrawal agreement in particular. A direct challenge before the EU courts would likely lack standing, but a challenge via the national courts (as in Racke), which could request the CJEU to rule on the issue, probably would not. Challenges concerning the Brexit process and withdrawal agreement are hardly hypothetical, as we have seen over the last two years.

Even if suspension or termination is valid, there are specific issues relating to citizens’ rights. As we have seen, suspension of the agreement expressly cannot affect them. And while the VCLT provisions concerning retention of rights in the event of termination of a treaty, or the ban on reprisals in the event that a treaty of a ‘humanitarian character’ is terminated for a material breach, arguably do not literally cover those covered by the citizens’ rights rules, it could be argued that in conjunction with the EU law principle of legitimate expectations, such rights cannot be removed.

What about the UK side? Public international law rules do not form part of the domestic legal order. But there are domestic legal and political issues nonetheless. In the event that the government aims to breach or terminate the withdrawal agreement by primary legislation, the House of Lords can delay it by up to a year. It rarely uses such powers, but it could be argued that there’s a ‘reverse Salisbury convention’ justifying it doing so when a government explicitly aims to reverse a commitment in its election manifesto – throwing its own ‘oven-ready deal’ in the bin. If the government aims to breach or terminate the withdrawal agreement by secondary legislation or executive action, there could be legal challenges on the basis that the secondary legislation is ultra vires the Withdrawal Agreement Act, or that executive action cannot simply suspend rights which are guaranteed by an Act of Parliament – leading to litigation which we might dub ‘Miller III’. But this brings us back to the government’s reported intention to nobble the judges.  

Barnard & Peers: chapter 27
Photo credit: dw.com

0 comments:

Post a Comment