In his opinion today, the Advocate General at the CJEU has made clear that Brexit is not a one-way cul-de-sac with no prospect of turning back. Even at the point where the withdrawal agreement has been negotiated, the UK can take stock and reflect during a “cooling off period” that continues for the two year period in Article 50, right up to the point where the withdrawal agreement is concluded by mutual agreement on both sides.

The opinion is important since it breaks the “Sophie’s choice” between approving the deal and leaving the EU without a deal. Both options present insurmountable compromises for the UK in terms of economic disruption to its trade in goods and services and loss of individual rights for its citizens. As the AG indicates, if his opinion is followed, the eventual ruling from the CJEU and Court of Session will open up a “third way” for MPs in Westminster to take the initiative of unilaterally revoking the UK’s decision to withdraw from the European Union. Interestingly, he even goes so far to say that if there is a democratic change which reverses the decision to withdraw, deprives it of its effect or removes its constitutional basis, the State “can and must” notify that change to the Council.

Importantly, in interpreting the scope of Article 50 TEU in the light of the aims of the Treaties and international law, the AG has come down firmly in favour of unilateral revocation rather than subjecting the State’s change of mind to additional hurdles, such as obtaining the unanimous consent of all 27 Member States in the European Council. He regards unilateral revocation as key to respecting the sovereign will of a State to change its mind in line with its constitutional requirements, respecting democracy and preserving rather than destroying the fundamental rights of citizens in the UK and other Member States.

In his view, the right to revoke should be subject to a few simple formalities:

(i) a formal notice of revocation setting out the reasons for the change of heart;

(ii) exercised in accordance with the national constitutional requirements (be that another election, referendum or parliamentary meaningful vote as the case may be);

(iii) within the two year negotiating period in Article 50 – i.e. before 29 March 2019 or before a withdrawal agreement is concluded by both sides (whichever is the earliest); and

(iv) in good faith and in line with the State’s duty of sincere cooperation under the Treaty. That means that it cannot be misused for tactical negotiating advantage or abused for ulterior purposes.

Conversely, the AG rejected the Council and the Commission’s arguments that they should be allowed to impose a check on potential abuse by insisting on unanimous consent. In that scenario, a single Member State would effectively be allowed to veto the UK’s continued membership of the EU and it would empower the European Council to expel a Member State against its will.

In terms of the effects of revocation, the AG is clear that the departing State maintains all of its rights and privileges right up to the point of departure envisaged by Article 50. If a notice of revocation is served before that time limit expires, those rights continue and the status quo ante is restored. Although there may be “collateral damage” in the sense of financial repercussions for the EU institutions and other Member States resulting from the negotiation process and related practical steps, that should not be a bar to revocation. In any event the costs of proceeding with withdrawal would far exceed the minimal costs generated by the revocation.

The UK Government’s intention to maintain the EU State aid rules has been clear for some time.  I briefly discussed the reasons why it has taken that view – and done so without significant opposition – in a piece I wrote here.  The intention to hold onto the State aid rules was strong enough for the Government to commit to maintaining them even on a “no deal” Brexit.  But we now have a deal, subject to the vagaries of UK politics (which I shall not attempt to predict).  What has it got to say about State aid?

There are really three sets of issues here.  The first is what might be called the planned withdrawal period – the transition period and the arrangements for handling cases that straddle the end of the transition period.  The second is the management of State aid under the “backstop”, that is to say the Ireland/Northern Ireland Protocol which it is hoped will never be operational but which will come into force if no final agreement is reached before the end of transition.  And the third is the political declaration that sets out the outline of the final relationship.

Transition

The transition regime is brutally simple: the UK remains a Member State for all purposes save those associated with having any say in legislation or decision-taking: Article 127(1) simply provides that all EU law continues to have effect in the UK.  So there will be no change in the State aid rules at all during the transition period, save that the UK will lose its vote in any State aid legislation or decision-making by the Council.

The transition period, however, will have a definite end (though what that will be has not been decided).  What happens to ongoing procedures at the end of transition?  The answer is to be found in Article 92, which provides that the Commission will remain competent to resolve any State aid investigation that has been allocated a case number by the end of transition.  The Commission will then proceed as normal either to raise no objections, find no aid, or start a formal investigation (Article 92(5)).

Further, Article 93 provides that the Commission will remain competent to initiate new State aid procedures for up to four years after transition end.  These will presumably only relate to aid granted before the end of transition (aid granted after transition end not being a matter falling under EU rules, subject to the backstop), and hence will be investigations of unlawful aid or misuse of aid.

The Commission decisions that result from these investigations will be binding “on and in” the UK: so private parties as well as the UK itself will be bound: Article 95.

One issue not dealt with is how the Commission will approach an aid notified before the end of transition but which will cover a period extending to after the end of transition.  To the extent that the measure is to be implemented after the end of transition, it will presumably fall outside the Commission’s jurisdiction.  But a measure implemented before the end of transition might well have many of its effects on competition (and many of its benefits in terms of the aim of the measure) after the end of transition.  It is not clear, at least to me, quite how the Commission is supposed to deal with such post-transition effects: but for what it is worth, it seems to me that the Commission could only properly take into account effects occurring before the end of transition (not least because the measure could be withdrawn and re-implemented the day after transition, at which point the Commission would have no jurisdiction over it.

Backstop

The existence of strong State aid provisions in the backstop should come as no surprise.  The backstop involves a customs union between the whole UK and the EU.  In a customs union, the EU is deprived of its ability to apply countervailing measures to subsidised imports.  So an inevitable quid pro quo of a customs union is bound to be cast iron guarantees that unacceptable subsidies will not be granted to UK businesses exporting to the EU.

There are two aspects to the backstop: rules that apply in Northern Ireland (more accurately, to measures affecting trade between Northern Ireland and the EU) and rules that apply to Great Britain (more accurately, to measures affecting trade between Great Britain and the EU).

As for Northern Ireland, Article 12 of the Protocol states that the State aid provisions set out in Annex 8 shall apply “to the United Kingdom … in respect of measures that affect that trade between the part of the territory of the United Kingdom to which [the Union Customs Code] applies by virtue of Article 6(2) of this Protocol and the EU which is subject to this Protocol.”

That less than luminously clear provision needs unpacking. “The territory of the United Kingdom to which [the Union Customs Code] applies by virtue of Article 6(2) of this Protocol” is Northern Ireland, excluding its territorial waters.  And the “trade … which is subject to this Protocol” is goods trade, subject to the “holding provision” for fisheries products set out in Article 6(1), fifth paragraph.  Annex 8 to the Protocol sets out (I think) all relevant State aid legislation.

The net effect is that the UK remains subject to the EU State aid rules (including enforcement by the Commission and the Court of Justice) in relation to measures that affect trade in goods as between Northern Ireland and the EU.  That provision is evidently likely to cover many measures taken by the devolved administration in Northern Ireland.  But it is also important to appreciate that that provision could well include any UK Government measure that extended to Northern Ireland such as, for example, a tax break.  It even appears to bring in measures that do not extend as such to Northern Ireland, but (for example) benefited UK businesses that also produce goods in Northern Ireland.

Meanwhile, further State aid provisions apply to measures affecting trade in goods between Great Britain and the EU (technically, trade to which Article 1 of Annex 2 applies).  Part Four of Annex 4 sets them out.  The key point is that Annex 8 to the Protocol – which, as I have said, consists of all EU State aid law – applies to the United Kingdom: Article 7(1).

But two features of this “GB regime” stand out and distinguish it from the “Northern Ireland regime” described above.

First, there is a general carve out for all UK agricultural support up to a maximum to be set by EU/UK agreement through the Joint Committee – Articles 7(2) and 8.  Failing agreement, however, the carve-out is to be suspended (Article 8, last paragraph), which would appear to put the UK in a less than powerful negotiating position.

Second, and of more general interest, the powers of the Commission are to be exercised by the UK independent authority set out in Article 9 – which will be the Competition and Markets Authority (CMA).

Article 9 of Annex 4 then sets out what powers the CMA is to have.  It is to be operationally independent and to have guarantees of independence.  More startlingly, it is to have “powers and functions equivalent to those of the [Commission] acting under [EU State aid law]”.  And its decisions are to produce in the UK “the same legal effects as those which comparable decisions of the [Commission] acting under [EU] State aid law … produce within the [EU] and its Member States.”

This is strong stuff.  The Commission has powers to declare Acts of Parliament that confer State aid to be unlawful: and where that happens, they are of no effect.  The CMA is, it appears, to have equivalent powers with the same legal effects.  That would appear to give the CMA, which performs its functions on behalf of the Crown, the power to declare an Act of Parliament to be unlawful as inconsistent with the State aid rules set out in Article 8.  This is, to put it mildly, a considerable constitutional innovation: a critic might go so far as to say that the CMA is to have the power to declare Acts of Parliament to be of no effect that it took a revolution and a Dutch invasion to remove from the hands of the Stuart Kings.  On the other hand, of course, the Commission has (since 1973) had these powers as part of the UK’s membership of the EU: and these provisions could be seen as simply “bringing home” those powers.

Article 11 requires the UK Courts to enforce the standstill provisions in Article 108(3) TFEU, to review the CMA’s decisions, to enforce those decisions and penalise non-compliance, and to award damages for breach of Article 108(3).

The CMA is, however, not left to “get on with it” subject to the scrutiny of the UK courts.  Article 10 requires it to liaise closely with the Commission: and it is required to send the Commission drafts of all decisions, and then take “utmost account” of the Commission’s opinion (which is to be given three months to opine, which will have the effect of slowing down State aid approvals during the backstop period).  Finally, under Articles 13 and 14, the EU is to have power to take “appropriate remedial measures” if it considers that the UK’s application of the GB State aid regime “threatens to seriously undermine the conditions of competition between [the UK and the EU]”.  The Commission will therefore retain significant powers to step in if it feels that the CMA is being insufficiently robust in preventing unacceptable State aid.

The backstop provisions are in many ways startling: but of course it is hoped that they will never be needed because the final relationship will resolve the Irish border problem.  So we should look at what the final relationship has to say about State aid.

Final relationship

Paragraph 79 of the political declaration says: –

The future relationship must ensure open and fair competition. Provisions to ensure this should cover state aid, competition, … and relevant tax matters, building on the level playing field arrangements provided for in the Withdrawal Agreement and commensurate with the overall economic relationship. The Parties should consider the precise nature of commitments in relevant areas, having regard to the scope and depth of the future relationship. These commitments should combine appropriate and relevant Union and international standards, adequate mechanisms to ensure effective implementation domestically, enforcement and dispute settlement as part of the future relationship.

This is not particularly enlightening.  However, the reference to “building on the level playing field arrangements provided for in the Withdrawal Agreement” does seem to indicate that the backstop model may serve as a starting point for discussion of the State aid provisions of the future relationship, though plenty of room is left for a looser relationship (as would seem appropriate if the future relationship excludes a customs union and is a form of “Canada plus”).

Conclusion

 At the time of writing, predicting the fate of the withdrawal agreement is similar to predicting the fate of a small sailing ship about to be hit by a hurricane: its survival looks open to serious doubt, but is by no means out of the question.  Possible alternative outcomes appear to include a referendum including an option to remain in the EU, application to join EFTA/EEA, or “no deal”.  What does, however, seems to be tolerably clear is that the State aid rules will, in one form or another, continue to apply to the UK.

The Government has this afternoon laid regulations before Parliament to reform the UK competition regime in the event of a ‘no deal’ Brexit. The Competition (Amendment etc.) (EU Exit) Regulations 2019 (“Competition SI”), are accompanied by an Explanatory Memorandum and various pieces of short guidance from the CMA on its post-Brexit role (consolidated here).

Before anyone gets too excited, bear in mind that many of these changes may never be implemented if the Government (as hoped) manages to negotiate a deal with the EU. If there is a deal, there will be a transition period during which everything will pretty much remain as it is now and the Government may commit to keeping the UK and EU antitrust regimes aligned over the longer term.

But given that a ‘no deal’ scenario now appears more likely than previously anticipated, the Government’s plans are of considerable interest.

The various documents were only published this afternoon and there is a lot of material. A more detailed analysis will only be possible in the coming days, but below are a handful of headline points based on a (very) quick review.

Section 60 CA98

The most striking point is perhaps that the Competition SI will repeal s.60 of the Competition Act 1998 (“CA98”) in its entirety. Section 60 currently requires the Chapter I and Chapter II prohibitions to be interpreted so far as possible consistently with the approach under Articles 101 and 102. A number of groups (including the Brexit Competition Law Working Group) had recommended that s.60 be maintained in an amended form, for example so that it only required the CMA and UK courts to ‘have regard to’ the equivalent approach under EU law. A major benefit of this would have been to make it more likely that UK and EU competition law would continue to develop broadly in parallel, minimising the dual regulatory burden imposed on UK firms.

Under the Competition SI, however, s.60 is repealed in its entirety and replaced with a new s.60A which relates exclusively to EU court judgments and Commission decisions that pre-date exit day. There will be no legislative obligation on UK courts and decision makers to have regard to EU judgments or decisions that post-date exit day. In practice UK courts and the CMA might nonetheless choose to have regard to how equivalent issues continue to be treated under EU law, but they will need to reach decisions about the extent to which that is desirable for themselves.

Where UK courts are obliged to follow EU judgments that pre-date exit day, but not those that post-date exit day, an obvious problem arises, namely how UK courts should proceed when EU law has developed since the date of Brexit. This is a general issue under the EU (Withdrawal) Act 2018 (“Withdrawal Act”) which is not limited to the competition law field. New s.60A(7) proposes a solution, as it provides that the relevant court or decision-maker may disapply the interpretative obligation if they consider that to be appropriate in the light of various criteria, one of which is a post-Brexit development in EU law.

The other criteria include differences between UK and EU markets, developments in economic activity, generally accepted principles of competition analysis or the particular circumstances under consideration. These criteria are both wide-ranging and broadly expressed, and their interpretation is likely to be the subject of considerable debate in court. The Explanatory Memorandum is perhaps somewhat optimistic in stating that this provision “will provide UK courts and competition regulators with clarity as to how Chapter I and II are to be interpreted after exit”.

Other significant points

Other key points include the following:

• EU Commission decisions making infringement findings will no longer be binding after exit day for the purpose of follow-on actions in the UK courts. An important exception is that the transitional provisions will provide that the binding effect of Commission decisions that pre-date Brexit will be maintained, even if any appeals against them are only concluded after Brexit.
• A specified list of current EU block exemption regulations will be retained (renamed as ‘Retained Block Exemption Regulations’) but new EU block exemptions introduced after Brexit will not automatically be incorporated into EU law. The CMA and Secretary of State could potentially choose to copy across any new EU block exemptions into UK law using their powers under s.6 CA98.
• The default position under the Withdrawal Act is that substantive EU law in force the day before exit day will continue in force. However, the Competition SI will immediately revoke, among other provisions, Articles 101 and 102 and the EU Merger Regulation (“EUMR”).
• One consequence of this is that the CMA will have merger jurisdiction (where the domestic merger thresholds are met) even where the European Commission is investigating a merger under the EUMR.
• As a result of this, and also the need for additional action by the CMA to enforce the antitrust and state aid rules, the CMA is anticipating that it will need to carry out a lot more work, in particular on more complex cases. It is working with the Government on secondary legislation that will facilitate this and is increasing its headcount.
• As you might expect, there is a mass of transitional provisions.

As noted above, detailed consideration of these and the many other reforms in the Competition SI will need to wait. One initial observation, though, is that the Government’s power to make legislative amendments to the CA98 derives from s.8 of the Withdrawal Act – which confers powers on Ministers to remedy situations where retained EU law will not operate effectively or will be otherwise deficient after Brexit.

Precisely what counts as a deficiency is elucidated in s.8(2) of the Withdrawal Act, and thought will need to be given as to whether all of the amendments set out in the Competition SI fall within the scope of that provision. Based on this particular statutory instrument, however, the Government appears to be interpreting the scope of s.8 quite broadly. The Government has previously justified its broad s.8 powers on the basis that they are required to address situations where legislative provisions would ‘fall down’ on exit day – such as where they attribute roles to EU institutions that will no longer be co-operating with the UK post Brexit.

Many of the amendments in the Competition SI are broader than that. It would have been technically possible, for example, to retain s.60 CA98 unamended. There is no technical reason why UK courts and decision-makers could not remain under an obligation to interpret UK competition law consistently with EU competition law. One can see that might not be considered appropriate from a political and policy perspective post-Brexit, but interventions based on considerations of appropriateness could extend far beyond those required to remedy obvious legislative deficiencies.

This is far from an academic point, as s.8 of the Withdrawal Act will provide the legal basis for hundreds of Brexit related statutory instruments that will be placed before Parliament over the coming weeks and months. So far legal challenges relating to Brexit have largely concerned issues of high constitutional principle. We are about to descend into detailed legal thickets in every area touched by EU law.

Julian Gregory is an EU and competition law practitioner who assisted the Brexit Competition Law Working Group (including Jon Turner QC) with its report on the implications of Brexit for competition law.

George Peretz QC analyses the trade consequences for both the UK and the EU if the UK leaves the EU with no deal as to the withdrawal agreement, including no transition period.

To read the full article published by Practical Law, please click here.

R. (on the application of Webster) v Secretary of State for Exiting the European Union

On 12 June 2018, Lord Justice Gross and Mr Justice Green refused permission to apply for judicial review to challenge the Brexit negotiations on the basis that the United Kingdom had, so it was argued, failed to make a valid decision to leave the European Union pursuant to Article 50(1). The Judgment has today (20 June 2018) been publicly released, and is available here.

Summary

Permission was refused (both on the papers by Supperstone J, and after a renewed application at an oral hearing by Gross LJ and Green J) on three grounds:

(1) The claim is out of time. As is well known, CPR 54.51 requires a claim form to be filed promptly and in any event not later than three months after the grounds to make the claim first arose. The UK’s notification letter was sent on 29 March 2017. The claim form, however, was not filed until nearly 9 months after the notification. No good reason had been advanced for the extension of time; indeed, Gross LJ stated:

“There is no conceivable – let alone good – reason for extending time. This is a paradigm instance of a claim needing to be made promptly and within the applicable time limit.”

(2) There had been undue delay in bringing the claim such that it would be to the detriment of good administration to hear it (see section 31(6)(a) Senior Courts Act 1981). Gross LJ explained:

“If the claimant was to pursue this claim…then it cried out for a prompt pursuit… The notion that good administration would be assisted by the grant of permission here or that the negotiations would not be derailed is with great respect wholly unreal, not least given the timetable under which the negotiations are taking place.”

(3) The claim is, in the words of the Court, “unarguable”, “doomed to fail on the merits” and “Totally Without Merit”. The Article 50(1) decision, the Court held, was authorised by Parliament in the European Union (Notification of Withdrawal Act) 2017 and was contained within the Prime Minister’s letter of notification.

Commentary

Permission could conceivably have been refused simply on the basis of the first two grounds summarised above, namely, that the application was out of time and the delay in hearing it would be contrary to good administration. These two grounds are individually determinative – that is, regardless of the potential merits, permission could have been refused on either basis alone. Nevertheless, the Court’s assessment of the relevant Article 50 decision is constitutionally interesting.

The effect of Shindler v Chancellor of the Duchy of Lancaster [2016] EWCA Civ 469 and R (Miller) v Secretary of State for Exiting the European Union [ 2017 ] UKSC 5 was as follows:

(1) The referendum was not itself the decision for the purposes of Article 50(1). The European Union Referendum Act 2015 did not say anything about what should happen if the majority of votes were cast in favour of the UK’s leaving the EU. As such, as a matter of domestic law, it was an advisory referendum: see Shindler and Miller. The availability of an expressly ‘binding’ model was well known to Parliament before enactment of the European Union Referendum Act 2015 (see, e.g., section 1(2) of the Northern Ireland Act 1998), but Parliament chose to legislate for a referendum the outcome of which would not legally require the Government to take, or to refrain from taking, a particular course of action. Nor was there anything in the 2015 Act itself to suggest that the holding of the referendum amounted to the taking of a decision that Parliament would, if it wished to do so, be legally incapable of overriding or reversing.

(2) The referendum was, however, “part of” the UK’s “constitutional requirements” for the purposes of Article 50(1), as Parliament had required a referendum to be held on the question of whether the UK should leave the EU: see Shindler.

(3) Miller concerned the other “part” of the “constitutional requirements” for triggering the Article 50 process. As is well known, the Supreme Court held (by a majority of 8 against 3) that an Act of Parliament was required to trigger Article 50; the Prime Minister had no relevant prerogative power to so act.

An issue that was never fully resolved in explicit terms in the Miller litigation remained, however: when and by whom the underlying decision to withdraw from the EU was taken. This was the legal issue at the heart of the latest crowdfunded action, which Gross LJ and Green J have now answered in Webster.

Whatever uncertainty might have surrounded the issue of when and by whom the ‘decision’ was taken in the immediate aftermath of Miller, the matter was put beyond doubt by the legislation enacted in the wake of the Supreme Court’s judgment, namely the European Union (Notification of Withdrawal) Act 2017. This authorised the giving of notification under Article 50(2). Section 1 of the European Union (Notification of Withdrawal) Act 2017 simply states:

(1) The Prime Minister may notify, under Article 50 of the Treaty on European Union, the United Kingdom’s intention to withdraw from the EU.
(2) This section has effect despite any provision made by or under the European Communities Act 1972 or any other enactment.

As Gross LJ explained, this legislative authorisation in the 2017 Act “plainly contemplated and encompassed the power to take a decision to withdraw and conferred that power expressly on the Prime Minister”.

The UK’s “constitutional requirements” (as required by Article 50(1)) were therefore satisfied: (i) primary legislation required a referendum; (ii) that referendum was held and resulted in a majority in favour of a particular result; (iii) legislative approval of that result was forthcoming; (iv) the Prime Minister was given a (statutory) power to effectively implement that result; and (v) the Prime Minister utilised that power in giving notification. It followed, in the crucial conclusion of the Court, that:

“The Prime Minister’s letter itself contains a decision; backed by the authority of the 2017 Act, that decision complies with the requirements of Miller. No additional UK constitutional requirements remained to be satisfied.”

It is important to bear in mind that Article 50 does not speak of a formal ‘decision’ having to be made: it speaks of a Member State’s having “decided” to leave in accordance with its “constitutional requirements”. Those requirements – as elucidated in Shindler and Miller – are met, in my view, by the combination of the referendum outcome plus the 2017 Act. What Webster now confirms is (i) that those constitutional requirements were, indeed, met (and comprehensively so) and (ii) that the 2017 Act is to be interpreted as authorising a decision, rather than forming the decision. There was thus no need for any other action to be taken before the Article 50 negotiations could begin, such as an additional Act of Parliament which the applicant in Webster was seeking. Gross LJ – correctly in my view – put it this way:

“I reject the argument that additional formality was required under the UK constitution or that there was any requirement for the Art. 50(1) decision to be in some separate document from the Art.50(2) notification.”

This is undoubtedly correct. Against the background of the Shindler and Miller litigation, the resolution in Parliament during the Miller litigation (in favour of leaving) and the terms of debate concerning the Bill which became the 2017 Act, the notion that the UK’s notification is invalid because no ‘decision’ to withdraw had (yet) been taken seemed remarkably hard to sustain. Indeed, the Supreme Court in Miller stated expressly that only a short Act would be required. The 2017 Act records the UK’s intention – present tense – to leave, and authorises an action (notification) which would be otherwise meaningless without that precondition having been met. There is no requirement in UK constitutional law for the Act to state in bold text something along the lines of “This is a decision” – the meaning of the words used, in the context outlined, have that effect.

What is more dubious, however, is whether it was necessary for the Court to hold that the Prime Minister took the decision. In light of the fact that the Court agreed that no formal “decision” is needed – Article 50 only speaks of “constitutional requirements” – one cannot help but think that it may have been simpler to say that it was the electorate plus Parliament (in concert) that decided to the leave the EU, rather than the Prime Minister. This would have married the concepts of “constitutional requirements” with a “decision”. Nevertheless, it is beyond doubt that a decision has, indeed, been made. The Court has therefore reached the right result. Whether that decision to leave is inherently conditional (as a matter of domestic law on account of parliamentary sovereignty) or unilaterally revocable (as a matter of EU law) are different matters, of course…

General

In April, I wrote a post (here) describing the dispute settlement mechanisms (DSMs) in the draft Withdrawal Agreement (here) as they relate to the transitional period, which is set to run from the day after Brexit day (30 March 2019) until the end of December 2020. Last month, this was followed by another post, which concerned the provisions in that draft relating to the period commencing on 1 January 2021 (here).

In contrast, as promised, this post concerns the DSMs which might be enshrined in the “future arrangements” (i.e. the treaty or treaties governing the longer term) between the EU and the UK.

These “arrangements” may take several years to negotiate and ratify. The fast-track procedure enshrined in Article 50 TEU will no longer be available. What is more, it is highly likely that the treaty or treaties will extend to matters within the powers of the Member States, in which case they will have to be mixed agreements requiring ratification by all the Member States as well as by the EU. Indeed, in a recent speech, Michel Barnier, the EU’s chief Brexit negotiator, said as much (here).

Even though according to Article 50(2) TEU the Withdrawal Agreement must “take account of the framework” of the “future relationship” between the parties, discussions between the EU and the UK about the nature of that relationship are only just beginning; and the EU has made it clear that there is no question of the treaty governing the “future arrangements” being signed, let alone ratified, until after Brexit day (29 March 2019).

As was pointed out in my two previous posts, we are still in the dark as to precisely what the DSMs in the draft Withdrawal Agreement will look like – or whether the negotiations on that agreement will even come to fruition. Still less can anyone have a clear idea of what the substance of these “future arrangements”, including the DSMs, is likely to be.

Accordingly, what follows in highly speculative – all the more so because, as explained below, the Court is set to rule in the coming months on the extent to which disputes can be settled by arbitration.

Nevertheless, some points are relatively clear.

First, in keeping with the EU’s practice the “future arrangements” will no doubt provide for a Joint Committee of the parties, one of whose functions will be to resolve differences of view between them. However, the EU is most unlikely be content with that purely political body but will insist that a court or at the very least a tribunal of arbitration be established to decide disputes which cannot be resolved amicably.

Second, the “future arrangements” are likely to provide for a range of different mechanisms depending on the subject matter and the identity of the parties.

Third, in view of Opinion 1/91 (here), it will not be possible for disputes to be decided by a body consisting of current members of the Court of Justice and representatives of the UK, as that would undermine the integrity of the Court of Justice.

Fourth, whatever precise form the DSMs take, one can expect a clear shift away from the EU law model in which (as in domestic law) individuals and legal persons can enforce their own rights in the courts towards a range of mechanisms closer to traditional international law. This would suggest that less, if any, reliance will be placed on courts than under the draft Withdrawal Agreement.

The cases most obviously suited to being heard by courts are those relating to citizens’ rights. But in any case, as explained in my previous post, the UK has accepted Article 151 of the draft, which would empower the UK courts to make references for preliminary rulings to Luxembourg until the end of 2028.

All the rest is a matter for speculation.

Courts

The first major question is: assuming that it wishes some matters to be heard by courts at all, to what extent will the EU continue to insist that jurisdiction be conferred on its Court of Justice? In its resolution of 14 March 2018 (here), the European Parliament declared that there must be “a binding role for the CJEU in the interpretation of Union law” (paragraph 12). The Government will be loath to accept this.

In its report of 3 May entitled “Dispute settlement mechanisms after Brexit” (here), the House of Lords Select Committee on the EU paid considerable attention to this issue. The Select Committee expressed concern at the prospect of the court of one of the parties having jurisdiction over disputes involving the other party or its citizens, since “justice must not only be done, but should manifestly and undoubtedly be seen to be done” (paragraph 84 of the report). Despite this, it saw the wisdom of maintaining the present system during the transitional period (30 March 2018 to 31 December 2020), because the period is relatively short and it would be “too burdensome and time-consuming” to set up another system at this late stage (indeed, it seems scarcely possible to envisage how this could be done at all).

Crucially, however, the Select Committee took the view that the jurisdiction of the Court of Justice should only be maintained for a short period after the end of the transitional period.

If that approach is followed, there would seem to be little room for the mechanism laid down in the EU’s mixed Agreement with the Ukraine (here). Under this Agreement, certain types of case are to be decided by arbitration but, in the event of a dispute as to the interpretation of a relevant provision of EU law, the panel will be required to refer the matter for a preliminary ruling to the Court of Justice, whose judgment will be binding on the arbitrators (Article 322(2)).

An alternative possibility might be to confer jurisdiction over certain categories of proceedings on the EFTA Court. Even though that body would be seen as more neutral than the Court of Justice, it is by no means clear that the Government would accept it; and the EEA-EFTA States (Norway, Iceland and Liechtenstein) would also have to consent. On this website, the reader will find a link (here) to the written evidence which Carl Baudenbacher, Michael Bowsher QC and I gave a few days ago to the House of Commons EU Scrutiny Committee for its current enquiry on the same subject. This evidence relates to the provisions in the draft Withdrawal Agreement concerning the immediate aftermath of the transitional period, but it is fully relevant here.

Of course, the third option would be to create a “bespoke” court specifically for the “future arrangements”, but that would raise considerable practical problems.

Arbitration

The second major question is: to what extent can disputes be settled by arbitration?

Needless to say, arbitration can take several forms: for instance, it can be used for “interstate” disputes (including those between a State and an international organisation), investor-State dispute settlement (ISDS) or disputes between commercial entities.

In its ground-breaking Opinion 2/15 on the Agreement with Singapore (here), the Court held that an ISDS mechanism which “removes disputes from the jurisdiction of the courts of the Member States” could not be regarded as “purely ancillary” to matters within the competence of the EU, so that the Agreement could not be ratified by the EU alone (paragraph 292). Like other Free Trade Agreements negotiated recently by the EU (e.g. those with Canada (CETA) and Vietnam), this agreement contains a provision stating expressly that its provisions lack direct effect – thereby excluding in the clearest possible terms recourse to the courts of the Member States and thus indirectly to the Court of Justice. In any case, the Court did not rule on the compatibility of such a mechanism with the EU Treaties.

That is precisely what the Court is called upon to decide in answer to Belgium’s request for a Opinion on the highly contentious DSMs enshrined in the Comprehensive Economic and Trade Agreement (CETA) between Canada on the one hand and the EU and its Member States on the other (here). As just mentioned, the provisions of CETA are expressly stated not to have direct effect. The hearing in that case has been scheduled for 26 June.

To what extent the recent judgment in Achmea (here) can be seen as a guide to the outcome of those proceedings is a matter of debate. There the Court held that provisions in an treaty between two Member States which established an ISDS mechanism were repugnant to the EU Treaties because they excluded the jurisdiction of the national courts and thus indirectly precluded the Court of Justice from delivering preliminary rulings. In CETA, the Court might distinguish Achmea on the grounds that it concerned an agreement between Member States.

Conclusion

In short, in this bizarre chess game, which could drag on for many years, some of the key rules have yet to be laid down by the Court. And then there is the politics.

Many thanks to Carl Baudenbacher and Panos Koutrakos for their helpful comments on an earlier draft of this post.

The UK has expressed its interest in joining the WTO’s Agreement on Government Procurement (as revised) (‘GPA’) (available here) in its own right for after Brexit. The letter from the UK’s Permanent Representative to the WTO is available here.

The GPA is a plurilateral agreement i.e. optional and subject to accession by WTO Member States. There are currently 19 parties to the GPA comprising of 47 WTO members. The UK is currently a member pursuant to its EU membership.

In the usual WTO format, the GPA is composed of two parts: the text of the Agreement itself and the parties’ own market access schedules of commitments (‘schedules’). As common in the WTO, most rules do not automatically apply to all procurement activities of each party; instead, the parties’ schedules determine whether a procurement activity is covered by the GPA or not. The schedules are divided into seven annexes:

Annex 1: central government entities
Annex 2: sub-central government entities
Annex 3: other entities
Annex 4: goods
Annex 5: services
Annex 6: construction services
Annex 7: general notes.

The EU’s schedule is available here.

Benefits of the GPA

A possible vacuum

There are no general multilateral obligations concerning government procurement (the WTO’s label for what is more commonly known as public procurement) under WTO rules. (There are certain specific provisions concerning State-Trading Enterprises, for instance.) Government procurement is therefore largely excluded from the basic obligations under both GATT and GATS. Absent the UK being a party to the GPA, the default position under WTO rules apply with the result that there are no mutual rights of access to public procurement. As such, without membership of the GPA, a ‘hard Brexit’ scenario (i.e. the WTO membership fall-back option, absent any deal with the EU) becomes ever harder in procurement terms, with particular consequences for UK entities looking to bid for government contracts abroad and in so doing needing to be able to rely upon GPA rights.

Scope

The scope of procurement activities covered under the GPA schedules for the EU (and the UK at present) is, however, narrower than the scope of covered procurement under the EU procurement laws. For example, market coverage access is particularly more limited in terms of below-threshold or private contracts subsidised by government; defence; and utilities. However, activities could be added to the UK’s GPA schedule if the UK wished to expand the coverage of the public procurement market access in the UK. Further, there is no most favoured nation rule regarding the coverage of the GPA, so the UK and EU (or, indeed, the UK and other GPA parties) could extend coverage between themselves without extending such coverage to other WTO members or GPA parties.

However, the initial indications surrounding the UK’s application to re-join the GPA are that the UK and EU have agreed that the UK will remain subject to the same rights and obligations currently applicable pursuant to the commitments in the current EU schedule. This is understandable: proposing to adopt the commitments would have delayed and possibly complicated matters if done as part of the UK’s application for accession rather than simply adopting the status quo of the currently-applicable schedule. First, protracted negotiations with the EU would likely have pursued (both concerning any division and ‘horse-trading’ if the UK proposed to widen access for other markets or restrict EU access). Second, such a course would increase the risk of potential negotiations with other GPA parties seeking to alter the terms of their access, though this is an inherent risk of re-application in any event. British officials have been keen to highlight that rolling over membership of the Agreement should be relatively easy, since there is an incentive for other members to retain their access to the UK procurement market too. But any negotiation in the WTO provides the opportunity to make new demands.

Obligations and requirements

The nature of the obligations and requirements under the Agreement are also less ‘deep’ in the GPA than under EU law. For example, whilst — unlike most WTO rules — the GPA provides for remedies for affected undertakings before national review bodies (including the suspension of contracts), this is significantly less stringent than those for enforcing EU Regulations and Directives. For instance, there is no standstill period or automatic suspension of award decisions under the GPA and more limited remedies post-contract (see, further, GPA Article XVIII).

Whilst that all may be true, the benefits of access to the GPA parties’ markets (including open, fair and transparent conditions of competition in government procurement in some of the UK’s most important trading partner economies) is likely to be a great attraction, especially when compared with a “no-deal” counterfactual scenario. Even if a deal were struck between the EU and UK, there is still great benefit in maintaining GPA membership in order to maintain rights to participate in procurement markets covered by other parties, including the USA and other growing markets.

To apply or not to apply?

With the Government’s aim being to maintain its current WTO commitments and rights, including the GPA for the reasons outlined, the question becomes how the UK could achieve that. The UK is currently a party to the GPA only through its EU membership. In the absence of any clear precedent, there has been some uncertainty as to whether there was a legal obligation for the UK to formally apply to re-join the GPA for post-Brexit by following the same process as any new Party to the Agreement, or whether the UK could have succeeded to all rights and obligations to the GPA in its own right without an application.

The UK has seemingly opted for the prudent and cautious approach of making that application. The accession process starts with the submission of an application for accession and has two main aspects: negotiations between the acceding member and parties of the GPA on the former’s coverage offer, and verification that the acceding member’s procurement legislation is consistent with the GPA’s requirements – for example, regarding transparency, procedural fairness for suppliers and domestic review.

One would imagine that this ought to be a relatively swifter and easier process than the accession of an entirely new party. First, UK procurement legislation is already compliant with GPA requirements by virtue of the UK implementing into domestic law the EU procurement directives. Whilst gravely over simplified, the current terms of the EU (Withdrawal) Bill would copy across those requirements into domestic law post Brexit so that domestic law would continue to be compliant. Second, since GPA accession is based on the UK’s existing GPA obligations that may not even involve the difficulties in adjusting schedules or fresh negotiations that may arise under some other WTO Agreements. Presumably most parties are content, at least for the short term, with the current arrangements. Third, the mutual interests involved of reciprocal market access continuing, at a minimum, on the current basis means that parties are unlikely to object. Nonetheless this is no doubt an opportunity to seek to take advantage of the UK’s position and in particular to require that any necessary updating be done to the UK’s schedules.

The WTO’s Committee on Government Procurement is now expected to assess the UK’s accession request when it meets later this summer. That Committee will review the draft market access offer and checklist of issues circulated with the UK’s letter.