This post, written by George Peretz Q.C. of Monckton Chambers and James Webber of Shearman & Sterling, assesses the UK’s proposed revisions to Article 10 of the NI Protocol.
Introduction
Subject to the qualification that it applies only to “measures which affect that trade between Northern Ireland and the Union which is subject to this Protocol“, Article 10 of the Protocol on Ireland/Northern Ireland (“the Protocol“) applies EU State aid law in its entirety to UK subsidy measures. Measures caught by Article 10 are subject to the suspension obligation, prior approval by the European Commission and supervision by the Court of Justice of the European Union (“CJEU“) – as if the United Kingdom had remained a Member State.
The meaning of the “affect that trade” qualification is obviously critical. However, its meaning is wholly unclear. Since services are not “subject to the Protocol“, subsidies affecting only services are unlikely to be caught (unless they have an indirect effect of helping goods suppliers): but the concept nonetheless creates the potential for aid to beneficiaries elsewhere in the UK to be caught by the requirements of the Protocol, given the low threshold the equivalent “effect on trade” concept is usually regarded as imposing in EU law (and the low standard of evidence traditionally accepted as sufficient to overcome that threshold). The significant legal uncertainty this “reach back” creates has been extensively commented on by State aid experts and has now reached the High Court in the British Sugar case (See R v Secretary of State for International Trade ex p British Sugar, not yet reported filed under Claim No: CO/1034/2021, which alleges that aid supposedly paid to Tate & Lyle Sugars Ltd (who only refine sugar in in London) engages the Protocol). That uncertainty has been made worse by substantial disagreements as to the extent of “reach back” as between guidance issued by the UK government and by the Commission (discussed here).
While one can see, given that the Protocol for many purposes has the effect of placing Northern Ireland within the customs union and the single market for goods, that the EU would be concerned to ensure that there was no possibility of subsidised Northern Ireland goods flooding into the EU, it is nonetheless unprecedented, and raises significant democratic issues, for decisions in areas as politically important and sensitive as tax and spending to be subject to prior supervision by a foreign power and oversight by a foreign court. Those problems are aggravated by the fact that the “affect that trade” concept is likely to extend that authority to at least some fiscal decisions that principally concern Great Britain – notwithstanding that GB is on the other side of a customs border and has agreed with the EU, in the Trade and Cooperation Agreement (“TCA“), the most extensive (albeit reciprocal) subsidy control measures ever contained in a free trade agreement. This aspect of Article 10 is often referred to as “reach back”.
The UK Government published a command paper, entitled “Northern Ireland Protocol: the way forward” on 21 July 2021 (“the Command Paper“). The Command Paper contained brief suggestions for replacing Article 10.
The suggestions are brief and still need to be worked up. However, as explained below, in our view they are capable of providing a fairer, more reasonable and sustainable basis for dealing with subsidy competition between Northern Ireland and the EU than Article 10. In particular, it is important to recall that Article 10 was negotiated in October 2019, at a time when it was entirely unclear what, if anything, would be agreed between the EU and the UK in relation to subsidies in their future relationship agreement and what subsidy control policy (if any) the UK would be likely to adopt after Brexit. As the UK government points out, that uncertainty has now been replaced by extensive UK commitments on subsidy control in the TCA and by detailed legislative proposals by the UK government in the Subsidy Control Bill, likely to be enacted later this year.
Command Paper
The suggestions for replacing Article 10 are contained in three paragraphs, short enough to be fully quoted here:
“63. The Protocol at present means that Northern Ireland remains part of the EU’s subsidy control framework in certain areas. This is in part a result of the fact that the Protocol was negotiated in 2019 when neither the UK nor the EU knew the nature of commitments to be made on subsidy control in the future Trade and Cooperation Agreement or indeed whether it would be agreed at all.
64. We now know that there are comprehensive and robust commitments in place on subsidy control, based around shared principles underpinning both the UK and EU regimes agreed in the Trade and Cooperation Agreement. These are being further strengthened through the UK’s Subsidy Control Bill which is currently before Parliament. These arrangements provide a more than sufficient basis to guarantee that there will be no significant distortion to goods trade between the UK and EU, whether from Great Britain or Northern Ireland, thus making the existing provisions in Article 10 redundant in their current form.
65. Nevertheless, given that Northern Ireland producers would continue to have some privileged access to the Single Market in this model, we would be prepared to establish enhanced processes for any subsidies on a significant scale relating directly to Northern Ireland – for example enhanced referral powers or consultation procedures for subsidies within scope, to enable EU concerns to be properly and swiftly addressed.”
The proposal has two basic parts:
1.To replace Article 10 with the subsidy control provisions in the TCA
Title XI, Chapter 3, TCA contains the subsidy control provisions. These have been analysed in detail elsewhere (see here and here). However, the key points are:
- The TCA has a common definition of subsidy. This is essentially the same as the definition of State aid in EU law, except with respect to effects on competition and trade.
- Subsidies must be assessed using a common framework of principles, encompassing concepts of good subsidy policy. Subsidies must be proportionate, must pursue a defined public policy objective, must balance that objective against adverse effects on competition etc.
- Each of the UK and EU must offer remedies in their domestic courts to affected third parties, including aid recovery when subsidies have been granted in breach of the requirements of the TCA.
- Each of the UK and EU commit to transparency of subsidies paid, to permit scrutiny and allow private parties to seek judicial protection.
- At a state to state level, there are novel and extensive “rebalancing” measures, including rapid tariff retaliation if either party considers that subsidisation will cause a significant negative effect on trade.
Each side has procedural autonomy. The EU retains its State aid system of block exemptions and prior notification, the UK is free to design a new system, as recently set out in the Subsidy Control Bill. That system anticipates that the Competition and Markets Authority (“CMA“) will play a significant role offering advice on compatibility with the subsidy control principles, with subsidy decisions of public bodies overseen by the Competition Appeal Tribunal (“CAT“). Both are highly respected institutions that no neutral observer could plausibly argue lack capacity or independence for their anticipated roles.
The subsidy control provisions of the TCA, as implemented by the Subsidy Control Bill, are sufficient to provide each party with comfort that the other will not use subsidies to compete unfairly. That fact makes it much harder to justify the need for EU law and EU institutions to retain State aid jurisdiction over Northern Ireland in order to achieve that objective.
Removing Article 10 would also resolve the “reach back” issue, whereby EU law and jurisdiction could apply to subsidy measures in GB, overlapping with the provisions of the TCA. The interaction between the two schemes is not discussed at all in the TCA and is creating damaging uncertainty that needs to be urgently removed. Clause 48(2)(a) of the Bill effectively disapplies UK subsidy control rules to any measure that has been given “in accordance with” Art.10, but, given the imprecision of the “affect that trade” concept, the dividing line remains highly uncertain in many cases.
2.In respect of aid of “significant scale” to Northern Ireland producers, a new procedure that appears to be aimed at allowing the EU greater involvement in the UK subsidy control decision making process.
While the logic of replacing Article 10 with the subsidy control provisions in the TCA is compelling, the Command Paper implicitly anticipates the principal difficulty – convincing the EU that its interests are sufficiently protected – particularly in relation to subsidies that could significantly benefit goods suppliers in Northern Ireland, who are free to export to the EU without any possibility of tariffs. Article 10 provides the EU with full control by imposing EU law, procedure and courts on the UK. As such, any revision will involve the EU relinquishing some of that control and a strong case will have to be made that the replacement is an acceptable alternative. That case needs to have two elements. First reassuring the EU that its concerns can be met in a different way; second convincing the EU that the problems thrown up already (and likely to be thrown up in future) by Article 10 make an alternative desirable.
The Command Paper’s strategy for this is to offer the EU “enhanced referral powers or consultation procedures for subsidies within scope, to enable EU concerns to be properly and swiftly addressed.“
Consultation could mean that the Commission has a formal consultation or advisory role for aid with a genuinely significant effect in Northern Ireland – similar to the CMA under the Subsidy Control Bill. This would be very helpful in building trust and allowing the Commission to input on subsidy decisions before they become disputes under the TCA. It would also respect UK independence as the final subsidy decision would not be subject to prior notification and approval but would be for the UK authorities (albeit, in cases where the Commission raised serious objections, in the knowledge that action could well be taken under the TCA). We also see no reason why the Commission could not be given standing to challenge such decisions before the CAT, if it so wished: that would give the Commission a way of bringing a subsidy that it felt was prohibited or failed to take proper account of the agreed subsidy principles before an independent court.
It is less clear what the Command Paper means by “referral powers”. It is very unlikely that the Paper is envisaging that approval of certain subsidy/aid decisions could be referred to the Commission. Another possibility, more likely to be acceptable on the UK side, is that EU State aid law could be retained in Northern Ireland but applied by UK authorities and courts (something like the model in the EU/Ukraine Association Agreement), possibly with referral mechanisms to the CJEU in the event that questions of law arise – though much would turn on governance arrangements. Further possibilities, within the proposed UK subsidy control regime, might be to provide for wider mandatory referral to the CMA of subsidy measures in Northern Ireland than is generally contemplated under the Subsidy Control Bill, or to even provide for referrals to the CMA at the request of the Commission.
In any case, the Command Paper suggests that these enhanced procedures should apply to aid granted to “Northern Ireland producers“. This is narrower than in Article 10, since it removes the effect on trade concept and focuses on undertakings in Northern Ireland. The EU may also wish to extend this to deal with GB producers who make substantial supplies in Northern Ireland of goods that are likely to flow into the EU. However, one obvious difficulty here is that the UK would presumably need reciprocity – i.e. similar protection in respect of EU producers making substantial supplies in Northern Ireland at risk of flowing into GB. This would be difficult to dovetail with existing EU State aid law and procedure. It is also notable that it refers to “producers” – presumably of goods rather than applying more widely as Article 10 does to banks or other service providers that may impact goods markets.
Both limitations make sense in the context of the TCA and the Northern Ireland Protocol together. The Protocol only applies to goods trade (and electricity) and the TCA covers subsidies in the economy as a whole. Put together, that suggests that it would be possible to develop balanced proposals along the lines set out in the Command Paper, with an enhanced role for the EU in respect of goods producers genuinely active on the market in Northern Ireland, but otherwise relying on the level playing field provisions of the TCA.
Conclusion
It is fair to say that the Command Paper was, as a whole, not greeted with any great enthusiasm by the EU, though notably it was not rejected out of hand either. It is also well-known that there are considerable disagreements between the UK and the EU as to the future operation of the Protocol. However, whatever view may be taken by the EU of the UK government’s general stance towards the Protocol either in the Command Paper or more widely, its specific proposals on Article 10 should, in our view, be considered seriously.
The sequencing of the negotiations between the UK and EU meant that Article 10 was agreed before the subsidy control provisions of the TCA (just as the trade and customs provisions were agreed before the zero-tariff provisions of the TCA). The EU considered that it needed to insist on the application of EU State aid law in Northern Ireland as a necessary protection for its internal market given the open north/south border: there was no alternative way at that time for the EU to ensure that it was protected from distortion to competition and trade via subsidisation of goods flowing south over that border.
Now that the TCA is in place and the Subsidy Control Bill before Parliament, there clearly is an alternative way to regulate such subsidies. Moreover, this has been accepted as sufficient for the EU to agree a free trade agreement with the UK.
It is therefore quite reasonable for the UK to seek to replace Article 10 with something more proportionate based on the provisions of the TCA – indeed it is a shame that this was not done as part of the TCA negotiations, as the present authors both suggested at the time. This would have been a more natural point for these revisions to occur.
Since Article 10 is favourable to the EU, it is also realistic that the Command Paper anticipates replacing it with more than just the TCA provisions. Although the precise proposals need to be worked up, enhanced consultation for the Commission – especially if prior to aid granted in Northern Ireland – would provide the EU with significant additional comfort that aid decisions were consistent with the TCA and therefore not distortive of competition and trade between the EU and UK.
In respect of State aid / subsidy control at least, the Command Paper is a helpful contribution and, we would argue, should be taken seriously by the EU, whatever its views of the Command Paper as a whole.