The judgment of Foxton J in R(British Sugar) v Secretary of State for International Trade[2022] EWHC 393 (Admin) is the first judicial consideration of
- the impact of Article 10 of the Ireland/Northern Ireland Protocol, which (as discussed here) applies EU State aid law to all UK measures that affect trade in goods or electricity between Northern Ireland and the EU; and
- the post-Brexit UK subsidy control regime (at least in its transitional phase pending enactment of the Subsidy Control Bill, which consists of applying the subsidy control principles of the UK/EU Trade and Cooperation Agreement (“TCA”) as domestic law by virtue of section 29 of the EU (Future Relationship) Act 2020).
Being the first judgment does not necessarily make it interesting: but in this case not only does the judgment throw light on how the UK courts may approach future litigation under those regimes, but it also contributes to the wider issue of selectivity (or “specificity” in TCA jargon), namely the controlling principle of both the EU State aid and the TCA subsidy control regimes that ensures that those regimes apply only to measures that in effect target particular businesses or classes of businesses for favourable treatment and do not apply to general measures such as low general corporation tax rates.
The measure complained of was an autonomous tariff quota (ATQ) applicable to imports of raw cane sugar into the UK after the end of the transition period. The effect of the ATQ was to allow 260,000 tonnes of raw cane sugar to be imported duty free into the UK (in addition to duty-free imports from certain African, Caribbean and Pacific countries and from the EU) on a first-come first-served basis: tariffs after exhaustion of the quota were £28/100kg, save where free trade agreements applied (so tariffs were zero on imports with EU origin).
An ATQ open to anyone might not look at first blush like a promising candidate to be counted as a State aid or a subsidy: it looks like a general tax measure. Indeed, there is no EU case in which an ATQ has been said to involve State aid – though, as the judgment points out, that could also stem from the fact that in the EU such decisions are for the EU, and EU measures do not fall under the State aid rules.
The argument that this ATQ was selective/specific turned on the fact that the UK market for refined white sugar (the stuff that you put in your tea or sponge cakes) has certain peculiarities. That market has two main players: British Sugar (which uses UK-grown sugar beet) (“BS”) and Tate & Lyle (“T&L”) (which uses raw cane sugar imported from countries with warmer climates than the UK’s). (Some other refined white sugar is imported from the EU.) That fundamental difference between BS’s and T&L’s business models explains the very different views those businesses have historically taken on such issues as sugar tariffs and subsidies for domestic raw sugar production – differences of view that have frequently profited lawyers and generated a certain amount of case-law before the EU courts. It also helps explain T&L’s public position as one of the few large UK businesses to support Brexit.
The essence of BS’s complaint was that the ATQ would – as a matter of undisputed economic reality – benefit T&L almost exclusively. There just are no other significant UK importers of raw cane sugar, and nor are there likely to be. Moreover, BS was able to point to a volume of evidence that showed that the UK government was well aware of that fact, had extensively engaged with T&L, and had T&L’s position very much in mind when formulating the ATQ. Those familiar with the current UK government’s attitude to business will not be astonished to read in the judgment accounts of internal civil service e-mails expressing the need to protect a “national icon” that was “pro-Brexit” and concerns about a “Ministerial backlash” if T&L was not protected, due to its “lobbying reach”. However, subsequent e-mails noted that it was “best to put [those concerns] to one side”, and other e-mails recorded concerns about the impact of not having an ATQ for jobs at T&L and on competition in the UK refined sugar market, and that the effect of not allowing an ATQ would be to increase sugar imports from the EU.
State aid and Article 10 of the Protocol
Selectivity
The problem with BS’s argument in relation to State aid (as applicable under Article 10 of the Protocol) was, however, that it was unable to say that a zero tariff on all sugar imports would have been selective. That led it into a logical quagmire, since it followed that an ATQ set way above T&L’s possible imports would also not have been selective (as the effect would have been precisely the same as a zero-tariff regime). In the end, BS was driven to rely on the intention of the measure (referring to the material discussed above) rather than on any aspect of its design as such. But at that point its case crashed against the principle that the selectivity of a measure is to be ascertained by objective features (such as its design) and not by reference to underlying intent: thus, the court rejected its attempt to rely on the principle on Joined Cases C-106/09P and C-107/09P Commission v Gibraltar and UKthat a measure whose design reveals it to be selective (in that case by introducing a corporation tax based in part on size of premises so as to produce the result that offshore companies with no premises were exempt – a result that could be deduced from the objective facts of the scheme) on the basis that, here, there was nothing in the objective structure of the scheme that led to the conclusion of selectivity. Further, applying the well-known three stage test set out in Joined Cases C-20/15P and C-21/15P World Duty Free, BS was unable to show that there was a difference in treatment of undertakings in a comparable legal and factual position: BS was not in a comparable position to T&L because it did not import raw sugar, and any other importer of raw sugar would face the same regime as did T&L.
The court therefore held that Article 10 did not apply to the ATQ because there was no selectivity and it was not State aid.
Effect on trade
That conclusion makes the next part of its judgment – on the application of the “affect trade” test in Article 10 – obiter. There, the court held that that test had to be applied differently to the “effect on trade” test in Article 107(1) TFEU. It based itself on the point that the EU unilateral declaration referred to the need for there to be a “genuine and direct” link between the measure at issue and an effect on trade in goods or electricity between Northern Ireland and the EU did represent a meaningful qualification of the Article 107(1) approach when applied to Article 10.
Its reasoning is, however, highly questionable.
- First, as I argued in detail here, it is doubtful that the words “genuine and direct” amounted to any meaningful qualification to the Article 107(1) approach: a qualification is not meaningful if the contrary proposition would never be relied on, and since no Commission decision or court judgment applying Article 107(1) would ever expressly rely on an effect that was “pretend” or “indirect”, it is hard to see what the qualification adds.
- Second, the court was wrong to regard the statement by the Commission (in its notice to stakeholders, which I discussed here) to the effect that the EU declaration added nothing to the established Article 107(1) test as inadmissible: that approach ignores the fact that (as I also discussed) Article 12 of the Protocol gives the Commission (and the Court of Justice of the EU: “CJEU”) the same position in relation to Article 10 as they have in EU Member States under EU law – and given the Commission’s role in the enforcement of the State aid rules, it is simply wrong to regard its official statements on matters of interpretation of the law as inadmissible (which is not to say that they are binding).
- Finally – precisely because its judgment on the point was obiter – the court was able to avoid putting any flesh on the bones of what a qualified approach to the test meant in practice: and it specifically refused to rule on the UK government’s attempt to get it to endorse the UK government’s theory that Article 10 applies only where the measure has a “first order” effect in Northern Ireland or where its “secondary effects” are likely to be “channeled” towards identifiable undertakings in Northern Ireland. The fact that the court appears to have realised that articulating and applying, in the context of Article 10, any meaningfully different “affect trade” trade test to that used in Article 107(1) is likely to be treacherous ground is itself a warning against any attempting to build on its obiter holding on the point, not least because the ultimate authority on the interpretation of Article 10 is (under Article 12) the CJEU (a point the High Court, interestingly, fails to refer to).
Subsidy control provisions of the TCA
In relation to the subsidy control provisions of the TCA, BS suffered similarly difficulties in relation to “specificity” as it did in relation to “selectivity” under State aid rules – though, notably, the High Court looked only at WTO, and not EU, authority in discussing that issue. The court noted that different treatment between comparable importers could be a subsidy but also that a general system of preferences would not (see the Panel report in Canada – measures affecting the Automotive Industry at §10.162) and that tariff quotas are an established feature of the WTO framework. In the end, however, as with its discussion of selectivity, the decisive points against specificity were that BS, as a non-importer, was not in any comparable position to T&L, and T&L was not treated any differently to any actual importer (see §146 of the judgment).
Conclusion
The facts in British Sugar are unusual and the case is unlikely (whichever way any appeal might go) to generate many further challenges to tariff quotas (particularly not in the EU system, where tariffs are an EU competence not subject to the State aid rules). However, the court’s robust approach to the question of intent and insistence on looking at the question of “design” and hence selectivity/specificity in an objective way and without focus on detailed internal evidence of exactly what ministers intended to achieve is likely to point against too great a reliance on internal evidence of intent as opposed to a more objective analysis of the scheme at issue in its context. The judgment is also a sign of the willingness of the UK courts to engage with WTO authority in the subsidy context. As for the court’s analysis of “affect trade” in Article 10 of the Protocol, for the reasons above its judgment should, in my view, be approached with caution, and with the awareness that (as I pointed out here) any serious dispute about the application of that test in a case where it was the decisive issue would be likely to end up, one way of another, in the CJEU.
GEORGE PERETZ QC
22 March 2022