Constitutional Law Group September 3, 2019 Europe, European Union, International law, UK government, UK Parliament

Philip Allott: The Legality of a No-Deal Brexit Could Be Challenged

It may be that there is no such thing as a date of 31 October 2019 for a no-deal UK withdrawal from the EU. On 9 April 2019, according to Le Monde, Michel Barnier, chief negotiator for the European Council in the withdrawal negotiations with the UK, said: ‘The EU will never take a decision on a ‘no deal’. That will be a choice for the British.’

On 10 April, the European Council adopted a very obscure decision. ‘In response [to a British request], the European Council agrees to an extension to allow for the ratification of the Withdrawal Agreement. Such an extension should last only as long as necessary to allow for the ratification of the Withdrawal Agreement and, in any event, no longer than 31 October 2019. If the Withdrawal Agreement is ratified by both parties before this date, the withdrawal will take place on the first day of the following month.’ The European Council ‘also notes that…the United Kingdom has a right to revoke its notification at any time’.

‘Ratification of the withdrawal agreement’ was presumably supposed to mean acceptance within the UK of the then Prime Minister’s draft agreement; and ‘ratified by both parties’ was presumably intended to mean formal acceptance, rather than ratification in the technical sense of international treaty law.

The decision of the European Council does not say what would happen if the Withdrawal Agreement is not ratified by both parties before 31 October 2019. It does not use the formula of Article 50(3) to the effect that ‘the Treaties shall cease to apply to the State in question’ at the expiration of the time-limit.

Nevertheless, it seems to be taken for granted that the UK will cease to be an EU member state on October 31, 2019, if its notification of intention to withdraw is not revoked and the period of negotiation of a withdrawal agreement is not extended.

That view is correct only if the European Council’s decision of 10 April 2019 created a second time-limit for conclusion of an agreement and if such a decision would be legally valid, and if the decisions taken by the EU institutions to give legal effect to UK withdrawal without an agreement would be legally valid. Both of those assumptions are open to legal challenge.

Creating a second time-limit

Article 50 of the Treaty on European Union creates a two-year time-limit for negotiation of an agreement which, on this occasion, expired on March 29, 2019. Article 50 allows for the extension of the period of negotiation beyond that date. It does not expressly confer a power to create a second time-limit leading to the automatic withdrawal of a member state. It has apparently been assumed that Article 50 confers such a power by necessary implication.

The rules of International Law on treaty interpretation say that if the plain meaning of a treaty text, let alone an implied meaning, leads to a result that is manifestly absurd or unreasonable, another interpretation is possible, giving effect to the object and purpose of the treaty text as a whole.

Withdrawal from the EU under Article 50 is withdrawal by agreement. It does not confer a power of unilateral withdrawal permitted by many other international treaties. So-called no-deal Brexit on October 31, 2019 would amount to allowing the UK, one of the largest EU member states, to withdraw unilaterally by means of its own failure to act, completely regardless of the interests, policies and politics of all the other member states which would be profoundly affected by such unilateral withdrawal, and regardless of the enormous worldwide effects of a unilateral UK withdrawal, including legal chaos in the UK and the EU and in countless other legal systems and situations across the world. To give such a meaning to Article 50 is manifestly absurd and unreasonable.

On this view, the International Court of Justice, the ECJ or the UK Supreme Court (after obtaining a preliminary ruling from the ECJ) could decide that, if the agreement of 10 April 2019 extending the period of negotiation had created a second time-limit leading to an automatic UK withdrawal, then that would have been an unlawful act (lacking a legal basis) and hence invalid (without legal effect).

EU implementation of UK withdrawal

The EU institutions would give legal effect within the EU to the withdrawal of the UK without a withdrawal agreement. Unless the EU’s constitutional treaties were amended between now and October 31, which is not possible, the existing system of decision-making in the EU institutions would apply. The EU has seven primary institutions with decision-making powers. The European Parliament, the European Council, the Council, the European Commission, the ECJ, the European Central Bank, the Court of Auditors.

Leaving the last two institutions on one side for present purposes, each institution has its own decision-making powers, and also powers exercised in conjunction with one or more other institutions. In the case of legislation, such as regulations and directives, the typical process involves joint action by the Commission, the Council and the European Parliament. Typically also, decisions in the Council are made by unanimity or by a qualified majority.

Article 50(4) excludes representatives of the withdrawing member state from participating in the discussions of the European Council or the Council on the negotiation and conclusion of the withdrawal agreement ‘or in decisions concerning it’ (a painfully imprecise phrase). A special form of ‘qualified majority’ is created for this purpose. The text does not preclude the participation of representatives of the withdrawing state in the decision-making of the European Parliament, including legislation implementing UK withdrawal, in which the European Parliament must be involved as a co-legislator.

If Article 50(4) does not exclude the participation of UK members of the European Parliament in decisions implementing UK withdrawal, the UK members of the European Parliament would be entitled to take part in those decisions and, if they were prevented from participating, the decisions would presumably be invalid.

If there is no valid agreement or decision creating a second time-limit, then all decisions of the EU institutions giving effect to UK withdrawal without an agreement, whether made before or after October 31, would be invalid.

The ECJ is frequently called upon to consider the legality and validity of acts of the EU institutions. This classic constitutional and public law jurisdiction can be exercised before or after a decision is made.

Philip Allott is Professor Emeritus of International Public Law at Cambridge University.

(Suggested citation: P. Allott, ‘The Legality of a No-Deal Brexit Could Be Challenged’, U.K. Const. L. Blog (3rd Sept. 2019) (available at https://ukconstitutionallaw.org/))

15Comments

Add yours

1

Andy Jones on September 3, 2019 at 9:16 am
Reply

Art 50 to cut a long story short, allows a member state “to withdraw from the Union in accordance with its own constitutional requirements”

If the executive is found to have acted or is acting unconstitutionally in order to exit the EU, would that not mean that in accordance with both U.K. and EU law we had not left the EU?
2

whitenightf3 on September 3, 2019 at 9:34 am
Reply

People gave the British Government a clear instruction LEAVE THE EU. No mention of any deals, this is a ploy to keep us governed by EU Tyrants. If it was up to me all those who held public office who committed High Treason would be arrested for that heinous crime.
3

Paul Cavin QC on September 3, 2019 at 11:10 am
Reply

Attractive though that argument sounds…if it were the EU 27 failing to ‘agree’ that would mean the departing member state would not be able to leave..?
4

Russell Davies on September 3, 2019 at 11:38 am
Reply

How does one practically get the ECJ to make a consideration of the legality and validity of a no deal Brexit?
5

JohnAllman.UK on September 3, 2019 at 1:56 pm
Reply

Very clever.

Concerning the theoretical possibilities that you have documented, for (a) litigation nobody else has suggested as far as I know (making us birds of a feather – see JohnAllman.UK blog), (b) that litigation’s potentially surprising possible outcome, and (c) the withholding on the part of the EU 27 of its express consent to a no-agreement agreement (which is what a no-deal Brexit would need, you seem to be arguing), this very much reminds me of the question we considered at university, as to what would happen if a future Parliament were to repeal The Canada Act.

If we try to argue that the EU has raised no objection in the face of the near-universal assumption that we can leave without a deal, and therefore infer that she silently consents to a no-deal Brexit, we will get into trouble with the rape culture activists, whose doctrine is that “qui tacit consentire videtur” is no longer good law. (Why that argument doesn’t also apply to 29th March and 12th April, you haven’t explained.)

Perhaps the best-of-both-worlds solution is a new and lasting status quo in which the ECJ ruled that (for reasons you have argued so well) that the UK hadn’t left the EU after all, whilst the UK Supreme Court ruled that the we jolly-well had. Each court rejecting the authority of the other. This would allow the EU not to care about the porous Puckoon border in Ireland about which neither the UK nor the Irish Republic cares. Perfect!
6

Roger on September 3, 2019 at 2:50 pm
Reply

Or are you saying because the extension to 31/10/19 was to ratify the proposed WA but it wasn’t, then that extension is no longer applicable, so we left in April 2019?
7

peterb on September 3, 2019 at 6:22 pm
Reply

There is no definition of a withdrawal agreement or what it should, or should not, contain.
8

Richard Lang on September 3, 2019 at 7:51 pm
Reply

Thank you Philip.

On the procedure, it is past the two month deadline to challenge the act of the Council under Art 263 TFEU, and Art 267 TFEU (preliminary reference ie challenge to a related domestic act which is then referred) cannot usually be used to get around the expiry of the Art 263 deadline where admissibility was not in doubt. Thus it would very much depend on the identity of the challenger. However, even where admissibility was in doubt (say, if the challenger were a private individual with no individual concern), what exactly would they be challenging at domestic level? If, for example, they are challenging Mrs May’s asking for or agreeing to the April decision, then surely that is also out of time, as the domestic time limit for a challenge to an act of the executive is three months?

Turning to the substance, I would first like to check if the quotations given are from the Council’s Decision itself? I cannot see them in the Decision, which so far as I can make out is European Council Decision (EU) 2019/584 of 11 April 2019.

Secondly, and supposing that the challenger is challenging an act of Mr Johnson’s, this month or next, then it is his interpretation of the EU Decision that is at issue, and upon which the Court of Justice’s view would be sought. The Council’s interpretation of the Treaty does not seem to be at issue, or, put another way, the EU Decision is not incompatible with the proposed ‘non absurd’ interpretation of the Treaty. The Treaty talks about extending the period of negotiation, and the EU Decision extends it. The proposed absurdity would have come in only at the stage that Mr Johnson chose to interpret it (the EU Decision) as meaning that, on expiry of the new period but in the absence of ratification, notification or a new extension, the UK leaves the bloc without an agreement. It is hard to see what else it could mean, as, otherwise, the period of negotiation is ipso facto extended further, and that would seem to fly in the face of the words, “[s]uch an extension should last […] in any event, no longer than 31 October 2019”, which are only in the Recitals, so not binding, but are useful for interpretation.
Besides, other language in the Recitals, particularly Recitals 10 and 11 taken together, heavily indicates that the end of the extension is the date of withdrawal, and that the date of withdrawal is the date on which the Treaties cease to apply to the UK.

Finally, if the domestic court was convinced that this was not acte clair, and made a reference, and the Court somehow found the EU Decision invalid, even though only its interpretation and not its validity was at stake, then surely the result would be a return to the Council’s previous decision of 22 March 2019, with a resulting crash-out on, or back dated to, 12 April 2019.
9

Iain MELVLLE on September 4, 2019 at 3:24 am
Reply

A50(3) expressly contemplates a failure to make an agreement, in which case the Treaties cease to apply. If the period as extended is of the same character as the primary two-year period, surely the same consequences of a failure to make an agreement would follow?
10

A Matter of Confidence | Verfassungsblog on September 6, 2019 at 12:00 pm
Reply

[…] PHILIP ALLOT argues that a no-deal Brexit would also be highly questionable under European law. PAUL CRAIG elegantly shows why the British government’s justification of the prorogation of Parliament by the need for a Queen’s speech is a non-sequitur. ALEXANDRA SINCLAIR and JOE TOMLINSON point to an additional prorogation problem, namely the lack of control over secondary legislation already insufficiently controlled by Parliament. DEREK O’BRIEN examines from a post-colonial perspective how the former British colonies in the Caribbean deal with the issue of Prorogation of Parliament. […]
11

Stephen Allott (@AllottStephen) on September 8, 2019 at 4:43 pm
Reply

Let me address 2 questions raised by commentators on Philip’s article: one of substance and one of procedure.

The substance question is what happens if the Article 50(3) “period” is extended. Is the default now on 31 October 2019 a No Exit?

Although conventional wisdom rarely likes being challenged, the literal meaning of 50(3) needs more careful attention:

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.
Let’s start with “this period”. Which period?
2 years after the notification?
Or the period “from the date of entry into force of the withdrawal agreement” to the moment the “Treaties shall cease to apply”
Or could “extend this period” mean either of these?

Agreements can commonly enter into force on one date and have effects (such as ceasing to apply something or providing for a payment to be made) on a subsequent date. So there can be a “period” between entry into force and Treaties ceasing to apply which the Council and the Member State decide to extend.

Now look at the 2 words “failing that”. Well failing what? What are the possibilities?
The Treaties not ceasing to apply from the date of entry into force of the withdrawal agreement but ceasing to apply at some later date than its entry into force.
There being no concluded withdrawal agreement? If there were intended to be provisions for exit in default of agreement, it could easily have said as much.

Article 50(2) requires an agreement to be concluded.
the Union shall negotiate and conclude an agreement with that State, … that agreement shall be negotiated in accordance with Article 218(3)[12] of the Treaty

Generally agreements are binding unless there is an express provision for termination for convenience or unilateral departure. Article 50 provides for withdrawal by agreement but not explicitly for unilateral withdrawal. The former (agreement route) must override the latter (implied route). Even more so when there is no regime for liabilities under this implied right of unilateral withdrawal in the teeth of a mandatory requirement for withdrawal by agreement.

“Operation of the unless proviso” and its consequences
The unless has been triggered. No question about that. Everyone assumes that the effect is to extend simply the 2 year automatic no deal exit by the same amount, but it could also mean that the “Treaties cease to apply” automaticity is cancelled. In fact that would be very sensible. As has come to pass, things are not concluded. It would be very practical if the choice now would be to continue negotiations by agreement or let the current Article 50 notice lapse. If fact, this would be so sensible it would avoid the absurd chaos of no deal. Quite easy to imagine that once extensions are being agreed, then the parties would have wanted to stop the clock ticking.

So which should be preferred? A chaotic no deal exit under an implied unilateral right to exit on undefined terms at the end of a deadline extended by implication or a simple lapse of the time period for discussion on withdrawal by agreement as provided for which can be easily restarted with no harm done.

Procedure: surely the point could be brought up on a declaration but also every official and every citizen will want to know on 1 November whether borders are legal or not. As the UK will by default not exit the EU on 31 October, a border on 1 November will be a breach of the Treaty and immediately actionable. Boris Johnson’s state of mind is neither here nor there
12

georgiouleon91 on September 9, 2019 at 12:14 am
Reply

“Withdrawal from the EU under Article 50 is withdrawal by agreement. It does not confer a power of unilateral withdrawal permitted by many other international treaties. So-called no-deal Brexit on October 31, 2019 would amount to allowing the UK, one of the largest EU member states, to withdraw unilaterally by means of its own failure to act, completely regardless of the interests, policies and politics of all the other member states which would be profoundly affected by such unilateral withdrawal, and regardless of the enormous worldwide effects of a unilateral UK withdrawal, including legal chaos in the UK and the EU and in countless other legal systems and situations across the world. To give such a meaning to Article 50 is manifestly absurd and unreasonable.”

This is patently ridiculous. There is no ambiguity in Article 50 such that interpretation leads to a “manifestly absurd and unreasonable”, nor is the notion that unilateral withdrawal itself absurd given that membership of the EU is based on the consent of member states and unilateral withdrawal a normal feature of treaties generally.

If your argument were right, it would be ample evidence that Brexiters’ claims about the effect of EU membership on national sovereignty and self-determination is spot on. It would be akin to a prison.
13

Michael Detmold on September 9, 2019 at 6:33 am
Reply

With respect, Professor Allott has, I think, misread Article 50; in particular, his claim that it does not confer a power of unilateral withdrawal is mistaken. Withdrawal from the EU under the Article is not withdrawal by agreement (as he says); it is an automatic (effectively unilateral) withdrawal, which may or may not be accompanied by an agreement – and not an agreement to withdraw, but an agreement that assumes withdrawal by ‘setting out the arrangements for … withdrawal, taking account of the framework for [the] future relationship with the Union’. Whether there is or is not agreement on these matters, withdrawal occurs (unless the state in question withdraws its notice).

The time of withdrawal is decided either by the date of entry into force of a withdrawal agreement (if there is one) or in the case of no agreement by the stipulated two years. And then the Article gives a power of extension. As to this, Allott asserts that it ‘does not expressly confer a power to create a second time-limit leading to the automatic withdrawal of a member state’. This also is a misreading. The Article expressly empowers ‘the European Council, in agreement with the Member State concerned … to extend [the two year] period’. As a matter of the meaning of words, there is no extension of time without a first time (in this case the two years) and a second (extended) time: Article 50 expressly authorises either. This is in no sense implicit; in no sense does it attract the bother that Allott describes.
14

Tom wilkinson on September 9, 2019 at 11:14 am
Reply

It is totally clear in the original ruling on the withdrawal condition. State that Britain leaves with agreed conditiono between the Europrange legislature and the British parliament. If. No agreement is reached by a specified date. Then Britain must leave on that date.
No “Ifs or Buts”.
15

Roger on September 9, 2019 at 1:43 pm
Reply

I believe Michael Detmold’s analysis to be correct. There is already agreement to be able to leave in that Article 50 of the treaty allows a member state to give notice to leave. We’ve done that and the treaty is implicit in allowing us to leave and provides for the timing to be determined. It’s not any more complicated than that.

A “WA” is not necessary, nor should it be. A country can release itself from any treaty when it wishes. No treaty should bind a country to the point that it can no longer make sovereign decisions of its own.

If we can see through the dogma of the EU making out that it’s very complex, and actually leave, we will see that the EU bureaucracy is easy to be rid of. What is truly complex is our ongoing relationships which will be best handled without the bureaucracy hanging over us all, them and us, though the Eurocrats themselves struggle to think freely. We don’t need the shadow of overly burdensome EU rules and ideas prejudicing how we handle our future. That’s surely at the heart of the 2016 decision of the people.

By all means let’s have some level of basic agreement on day-to-day immediate issues and work on the rest freely as we go forward with the only constraints being reality not EU rules and philosophies. The EU seem to think we’ll still be quasi-members and tied to them. We shouldn’t be. If we weren’t members we wouldn’t think in that way, so why would we as we leave?