简报

2022 六月 15

British Sugar legal briefing

Lorand Bartels, Francesca Triggs and Fabian Bickel

Introduction

In February 2022 the High Court ruled on the decision of the Secretary of State for International Trade (the Trade Minister) to remove tariffs on certain quantities of raw sugar cane imports. We first reported on this case last August when British Sugar brought these proceedings on the basis that the measure effectively amounted to a state subsidy for its competitor, Tate & Lyle Sugars Limited (T&L). As we said then, this case marks a precedent-setting period in UK trade: this is the first time that courts have scrutinised and applied the subsidy provisions in the Northern Ireland Protocol to the UK-EU Withdrawal Agreement (NIP) and in the UK-EU Trade and Cooperation Agreement (TCA). Beyond that, an independent subsidy regime has often been presented as a key benefit of Brexit – this case sheds some light on the extent to which this remains constrained by international agreements.

Background to the tariff regime at issue

Both British Sugar and T&L produce refined sugar (satisfying approximately 50 per cent and 25 per cent of all UK demand respectively), but whereas British Sugar uses homegrown sugar beet for its refined sugar production, T&L uses imported raw sugar cane. T&L is the only importer of significant amounts of raw cane sugar for refining in the UK. The remaining 25 per cent of the UK’s refined sugar demand is met by imports from the EU.

In this case, British Sugar challenged a zero-duty autonomous tariff quota (ATQ) of 260,000 metric tonnes of raw cane sugar, distributed amongst importers on a “first come, first served” basis, which the Trade Minister opened in the process of creating a post-Brexit tariff regime. Once this volume has been imported, a duty rate of £28.00/100kg applies to imports for raw sugar cane if the purpose of importing it is to refine it, and otherwise the standard rate of £35.00/100kg applies. Beyond that, as was already the case before Brexit, raw sugar cane can be imported duty free from Least Developed Countries (LDCs) and under certain Free Trade Agreements (FTAs).

As explained in the judgment, it was widely accepted that T&L was not viable without the ATQ due to the high external tariffs in place, as well as the fact that prices for refined sugar mainly depend on the price levels of EU imports of refined sugar (for more on this see the government response to the Sugar ATQ consultation). The ATQ was therefore seen as an important means to keep the UK’s main sugar cane refiner from exiting the market in order, among other things, to ensure competition and diversity of supply in the UK sugar market, although it was kept low in order to preserve negotiation capital for FTA negotiations with countries that produce raw sugar cane. However, the legislative process also revealed that unnamed officials supported T&L due to it being “a national icon” and “pro-Brexit” (para. 20 of the judgment).

Unsurprisingly, British Sugar opposed this ATQ and sought to challenge it, arguing that it constitutes an unlawful subsidy. At the time of the decision, the applicable rules in relation to subsidies were contained in the NIP and TCA which apply in parallel. Art. 10 NIP applies the EU’s State aid rules (including Art. 107(1) of the Treaty on the Functioning of the European Union) to subsidies that have an effect on trade between Northern Ireland and the EU, and the TCA requires the establishment of an independent subsidy control regime under which subsidies should meet certain principles where they have a material effect on the trade and investment between the EU and UK (see Art. 366 TCA).

British Sugar argued that the ATQ is in violation of the relevant provisions in both the NIP and TCA. Both claims were rejected by the High Court, which found that the ATQ was not “specific” (NIP) or “selective” (TCA) and therefore does not constitute a subsidy under either treaty. In addition to these findings, the High Court commented on more general issues of interpretation in relation to the subsidy regime. We will discuss the following key points

and their context in the UK’s new subsidy regime in this briefing:

• first, the judgment recognises that an ATQ can, in principle, be a subsidy, discussing the application of subsidy rules under the NIP and TCA;
• second, it sheds some light on the application of the “selectivity” or “specificity” tests in the context of assessing the circumstances where a tariff measure will constitute a subsidy (the key requirement that was not met in this case); and
• third, it explores the substance of the “effect on trade” tests contained in the NIP and TCA and confirms that challenges in national courts can be based on the NIP and TCA.

It is worth noting at this point that the new UK Subsidy Control Act 2022 (SCA) is expected to come into force in autumn 2022. It provides the framework for a new, UK-wide subsidy control regime and replaces subsidy control provisions in the TCA as a matter of domestic law – although many of its provisions are closely aligned to those in the TCA. A new Subsidy Advice Unit is set to be established within the Competition and Markets Authority to oversee the new regime, while appeals to decisions on judicial review grounds will be heard by the Competition Appeal Tribunal (CAT). It is still unclear how power will be allocated between these entities and the Commission where a subsidy measure is subject to both of the parallel SCA and NIP regimes.

ATQs as subsidies

The legal context

In its judgment, the High Court applied the subsidy provisions in the NIP and the TCA to the ATQ and mainly considered whether the ATQ in question is selective/specific. The first conclusion that can be drawn from this is that subsidy rules can apply to ATQs. In terms of the legal background to this:

• In the context of the NIP and under EU law, ATQs can constitute subsidies provided that they are selective. No precedents exist on this specific point because the common commercial policy is an exclusive EU competence to which EU State aid rules do not apply. However, previous cases have affirmed that foregoing revenue (i.e. by a government) – which covers customs duties – constitutes a form of financial assistance and can therefore be a subsidy (e.g. Commission and Spain v. Government of Gibraltar and UK, para. 142).
• Similarly, in the context of the TCA, foregoing revenue that is otherwise due, including customs duties, can constitute financial assistance and therefore a subsidy provided it is specific (Art. 363 (1)(b) TCA). This is aligned with WTO caselaw which is relevant for interpreting the TCA (Art. 516 TCA). Specifically, Art. 363 (1)(b) TCA is equivalent to Art. 1(1)(a)(1)(ii) Agreement on Subsidies and Countervailing Measures (SCM), meaning that the same precedents apply (although differences in the tests mean that a full comparison should be made between the requirements in each). In terms of WTO case law, in US – FSC the Appellate Body found that a tariff exemption could constitute government revenues foregone, but cautioned that this requires the identification of a normative benchmark to determine what is otherwise due (para. 90). Relevant case law also includes the Panel report in Canada – Autos, which found that Generalised System of Preferences (GSPs) tariff waivers are generally not subsidies as they accord favourable treatment to all products of an importer by that country within the scope of the preferences so that no revenue is otherwise due (para. 10.162).

Application of the NIP and TCA to the external trade of the EU and UK

While the High Court’s application of subsidy rules to ATQs has clear legal backing, its approach raises the question of whether provisions in the NIP and TCA can affect the UK’s right to grant subsidies. The UK Government argued with respect to the NIP that measures forming part of the UK tariff regime are not capable of falling within the NIP’s State aid provisions, in part because this would undermine the UK’s power to determine an independent trade policy and would therefore be “manifestly absurd or unreasonable” (see para. 113 of the judgment onwards).

The High Court rejected this argument, relying on Recitals 21-23 of the NIP which highlight that Northern Ireland should benefit from the UK’s independent trade policy, as well as the need for Ireland’s rights and obligations under the EU’s internal market to be respected. This allowed the High Court to confirm its view that the NIP is not ambiguous in its application to certain UK activity in the tariff and customs sphere (see Art. 5(6) NIP), although ultimately the Government’s argument was not relevant to the identification of the ATQ as a subsidy, in particular in light of the ATQ’s specific nature as a “first come, first served” measure. The point was therefore not explored extensively, and the High Court did not explain in further detail how these provisions would be conclusive of the fact that tariff measures should apply to the UK’s external trade policy, or whether the NIP was intended to go beyond a typical FTA and also partially regulate the external trade of the Parties.

Interestingly, although the same questions arise in relation to the TCA, the High Court did not address whether its State aid measures are intended to regulate the external trade of the parties here either, but simply assumed this in its reasoning.

Many questions remain as to the full extent of the application of subsidy regimes to ATQs and other tariff measures (and whether the scope of subsidy law as compared to the previous State aid regime was enlarged as a result). As the High Court pointed out, however, that the implications of this case are not as dramatic as the Government implies it terms of the effect on the UK’s post-Brexit independent trade policy.

Selectivity under the NIP

As a reminder, British Sugar argued that the ATQ violates Art. 10 NIP as it constitutes a subsidy in breach of EU State aid rules and affects trade between Northern Ireland and the EU. The High Court rejected this, ruling instead that the ATQ was not selective – despite the fact that T&L was (effectively) its sole beneficiary.

A measure will be selective if it favours certain undertakings or the production of certain goods. Finding selectivity is relatively straightforward where funds are simply transferred to certain undertakings, but less so in cases (such as this) where the measure at issue constitutes an exemption from paying taxes or tariffs that are otherwise due.

Taking this description a step further, a measure will be de jure selective if it is designed so as to direct the benefit of State resources to particular undertakings (Commission and Spain v. Government of Gibraltar and UK). Alternatively, a measure will be de facto selective if the effects of the measure are selective. As set out in the judgment, de facto selectivity can be assessed by applying the World Duty Free test. This test for selectivity contains three limbs:

• first, it is necessary to identify the “ordinary” or “normal” regime;
• second, the court must assess whether the measure under consideration differentiates between operators who, in light of the objectives pursued by the ordinary or normal regime, are in a comparable factual and legal situation; and
• third, the court must consider whether the tax measure is justified in the sense that it follows from the nature or general structure of the system of which it forms part.

In the present case, the High Court was not convinced by British Sugar’s argument that the ATQ was de jure selective, i.e., that it had been designed to direct the benefit of State resources to T&L (which would be the case if the measure distinguished between cases based on seemingly objective characteristics, but where such characteristics were chosen in such a way as to benefit particular undertakings). Interestingly, the High Court found that the fact that T&L was the only importer of raw cane sugar was not sufficient to show that the ATQ had been designed to benefit T&L as the ATQ would in theory apply in the same way to all importers of raw cane sugar.

The High Court therefore moved on to apply the World Duty Free test in order to assess whether the ATQ was selective in its effect, with a particular focus on the first two limbs of the test. It found that a comparison needed to be made between T&L and other importers that wished to import raw cane sugar. As the ATQ was distributed on a “first come, first served” basis, and another potential importer would therefore have been in an equally good position as T&L to benefit from it, the High Court found that importers were not treated differently. The High Court also considered a wider frame of reference covering all forms of raw sugar, including raw beet sugar, and found that as no tariffs were paid on these imports, an importer of raw beet sugar would be in a comparable situation. The fact that the ATQ was formulated as an exception to the £28/100kg tariff on imports of raw sugar cane did not change the result, as the form of a measure is not decisive in this respect.

There are three key points which can be drawn from the High Court’s analysis:

1) State aid rules will not apply in a monopsonistic market provided that the measure is not de jure selective

A measure will only be selective if it distinguishes between importers in some way, by design or effect. Even if the UK decided to apply a zero per cent tariff on raw cane sugar – as it has on raw beet sugar – this would not be selective because the tariff applies to all importers equally. Although this result stems directly from the World Duty Free test, it sits uneasily with the fact that T&L is in effect the only beneficiary of the ATQ. The consequence of this test is significant, as it can therefore be understood that in a monopsonistic market – such as the market for the import of raw sugar cane – State aid rules do not apply if revenue is foregone.

2) Selectivity may be easier to establish in other forms of ATQs

While in this case the ATQ was designed on a “first come, first served” basis, there are other ways in which quotas in ATQs may be allocated (see the different Quota Allocation Methods here). Where import licences are granted based on a company’s previous years’ imports (historical importers) or through bidding, selectivity may be easier to establish. In such cases, importers are being treated differently, meaning that the identification of the relevant ATQs as subsidies would have depended on the third limb of the World Duty Free test, namely whether the difference followed from the general structure of the system.

3) Other tariff measures (e.g., under FTAs) could also be caught by the selectivity test for subsidies

The High Court’s analysis raises the question of how it would have treated other preferential tariff measures. This was not examined in the judgment, as British Sugar limited its challenge to the relevant ATQ and did not challenge other measures providing favourable treatment to imports of raw cane sugar (such as tariff-free imports from LDCs as well as under FTAs). British Sugar explained this limitation on the basis that FTAs involve “a broad raft of measures” and are not specifically intended to assist particular operators (and, therefore, are not subject to State aid measures). The High Court criticised this argument on the basis that it looked only at the intent of a measure without considering its effects, which is the more significant test under EU State aid law. At a later stage the High Court referred to the Panel’s finding in Canada – Autos that GSP preferences are generally not subsidies on the basis that “they accord favourable treatment to certain products from certain countries, and all such products from those countries receive favourable treatment” (para. 10.162).

In relation to the NIP, the appropriate question would have been whether these other tariff preference measures would have satisfied the test for selectivity under the World Duty Free test. As importers are treated equally under these regimes, they would likely not have been selective either. In line with this, the High Court stated that there had “(rightly) been no suggestion by British Sugar that the consequences of FTAs or decision to dispense with tariffs for particular goods altogether can amount to State aid as a matter of EU law” (para. 119 of the judgment).

The only potential exception to this equal treatment of undertakings or goods could have been the tariff reduction from £35/100kg to £28/100kg for raw cane sugar imported for the purpose of being refined. In that case, arguments could have been made that the normal regime was the £35/100kg tariff, and that the relevant characteristics that determine the lower tariff were chosen in order to provide one undertaking – T&L – favourable treatment. Therefore, there seem to be grounds to argue that this tariff reduction measure could have been selective and hence constituted a subsidy.

Specificity under the TCA

British Sugar also argued that the ATQ constitutes an unlawful subsidy under the TCA. The High Court rejected these arguments on the same grounds as it rejected the arguments relating to the NIP, namely that the ATQ is not a subsidy in the sense of Art. 363(1)(b) TCA as it is not specific according to Art. 363(1)(b)(iii) TCA – the equivalent of the selectivity requirement in EU law.

It is noteworthy that the High Court mainly considered whether the measure at issue was specific under the TCA (Art. 363(1)(b)(iii) TCA) and not whether the measure included the foregoing of revenue that is otherwise due (Art. 363(1)(b)(i)(B) TCA). This is interesting, as this issue of whether revenue was foregone was key to the High Court’s reliance on Canada – Autos and US – FSC for its findings on selectivity under the NIP. These cases deal with Art. 3 SCM violations, of which the foregoing of revenue forms a key part but for which no finding on specificity is needed (Art. 2.3 SCM). Although the tests for revenue foregone and specificity under the TCA and SCM are broadly equivalent, they should not be used interchangeably and will not give the same result in all cases. For example, prohibited subsidies (under Art. 3 SCM), namely export subsidies and subsidies contingent upon the use of domestic content, are deemed to be specific. While the TCA also prohibits export subsidies and subsidies contingent upon the use of domestic content (Art. 367 (8), (12) TCA), it requires the measures to be subsidies which in turn requires them to be specific (Art. 363 (1)(b)(iii) TCA). There will therefore be instances where a non-specific export subsidy could violate Art. 3 SCM, but would not violate Art. 367 (8) TCA.

Effect on trade tests

In terms of the background to the High Court’s analysis of the “effect on trade” tests, the NIP only applies to subsidies that “affect the trade between Northern Ireland and the Union” (Art. 10 NIP) and the subsidy control provisions in the TCA only apply where the subsidy has or could have a “material effect on trade or investment between the Parties” (emphasis added) (Art. 366 TCA). These “effect on trade” tests are significant thresholds. In relation to the NIP, the relevant test defines the extent to which the EU’s State aid regime applies in the UK – the consequence of a wide interpretation of the effect on trade test would be that subsidy measures mainly targeted at Great Britain with only an indirect effect on trade between Northern Ireland and the EU would continue to fall within the scope of EU State aid law (i.e., the provisions would “reach back” to affect Great Britain).

The High Court’s application of the tests

The Government stated that the ATQ displaced a significant quantity of EU exports from the UK’s sugar market. In relation to this, the High Court found:

• that there is no trade in raw cane sugar between the EU and Northern Ireland;
• that no refined sugar is produced in Northern Ireland; and
• that the ATQ does not affect the price of sugar (which mainly depends on the EU’s sugar prices as the UK remains a sugar deficit country) (see para. 132 of the judgment).

Whereas the evidence was sufficient to show that there was an effect on trade for the purposes of the test under the TCA, the High Court found that it was not enough to show an effect on trade between Northern Ireland and the EU for the purposes of the NIP.

Queries arising in relation to this application

In its analysis, the High Court did not engage with the question of the appropriate standard for the effect of trade under the TCA. This is surprising as the TCA requires the effect on trade to be “material”, making it higher than the one under EU law (see an article on this by George Peretz QC here). In the same way as under EU law, the present ATQ would have satisfied the test for an effect on trade between member states as the ATQ would have displaced EU imports (Eventech Ltd. / The Parking Adjudicator para. 67). However, the High Court did not take the opportunity to comment on whether the test under the TCA was harder to meet due to the materiality requirement.

In any case, the question of the appropriate standard has now lost relevance. This is because the Subsidy Control Act 2022, which replaces the subsidy control provisions in the TCA as a matter of domestic law, does not require the financial assistance to have an effect on trade necessarily, as in addition to this it also applies to measures that are capable of having an effect on competition within the UK (s. 2(1)(d)(i) SCA).

In relation to the NIP, the High Court suggested that in addition to the “effect on trade” test under EU law, it can be inferred from the legally binding unilateral declarations of the EU and the UK in the Joint Committee that the NIP requires a “genuine and direct link” to Northern Ireland. Applying this standard, in the present case the High Court held that no “genuine and direct link” to Northern Ireland existed. It found that the effect on trade was only indirect and de minimis, and insofar as EU importers were displaced from the sugar market in the UK this would be countered by the fact that British Sugar and T&L were subject to higher non-tariff barriers due to different trade measures under the NIP.

Overall, neither the High Court’s formulation of the substantive standard nor the subsumption were convincing for the following reasons:

• beyond the reference to “genuine and direct link”, the High Court did not provide further guidance on the requirements for the necessary link;
• more generally, it is questionable whether the effect on trade test should really be interpreted in a way that prevents these “reach back” subsidies, (see George Peretz QC’s commentary on this point), and
• it is unclear why other non-tariff measures that producers face as a consequence of the NIP but which were not at issue in this case, such as additional regulatory requirements and the risk of reclaims of EU customs duty, should be considered for this test, and why the High Court did not make any findings as to the effect of the ATQ on the market shares of EU imports of refined sugar in the UK and in particular Northern Ireland.

Judicial review of subsidy decisions and the future of UK subsidy law

Turning to our final key point, this case confirms that measures subject to the NIP and TCA can be challenged by way of UK judicial review and that – perhaps even more importantly – the High Court is willing to uphold them. Taking a step back, this is not an obvious point as the NIP and TCA are international agreements addressed to the EU and the UK. The legal basis for this is as follows:

• The TCA forms part of UK domestic law following s. 29(1) of the European Union (Future Relationship) Act 2020 (EUFRA). However, when the remaining majority of the provisions of the SCA comes into force later this year (when appointed by the Secretary of State under s. 91(2) SCA), it will replace the TCA as a matter of national law (s. 29(2) EUFRA). The consequence of this will be that whereas under s. 29 (1) EUFRA and Art. 372(1)(a) TCA subsidies could be challenged by the administrative court in the form of a judicial review, subsidy decisions will then be reviewed by the CAT (s. 70(1) SCA).
• In relation to the NIP, Art. 4(1) TCA clarifies that the provisions of the NIP can have direct effect and s. 7A of the EU (Withdrawal) Act 2018 adopts the NIP into national law. As previously mentioned, the SCA will not replace the functioning of the NIP, which means that the two systems will continue to apply in parallel.However, this conclusion may soon change. Domestic courts only applied the TCA and NIP because they were adopted into domestic law; if the Government repeals the domestic legislation, as it has proposed in s. 12(1) Northern Ireland Protocol Bill with respect to the subsidy provisions in the NIP, claims can no longer be based on a violation of the NIP.

Final practical takeaways

Many questions remain unanswered, but pending any appeal (British Sugar was granted permission to appeal the ruling on 15 March 2022) there are several practical takeaways which can be drawn from the British Sugar ruling:

• private claimants can challenge ATQs under the NIP (and, until the SCA is in force, TCA) subsidy provisions in domestic courts (provided claimants pass the “sufficient interest” test for judicial review);
• provided a measure is not specifically targeted at certain undertakings (de jure selective) State aid rules may not apply even if the market is a monopsony dominated by one single purchaser; and
• the method of quota allocation (e.g., “first come, first served” or through bidding) will be key to determining whether an ATQ may be in breach of State aid rules under the NIP and TCA.

If any of the topics discussed in this briefing are of interest to you, please do not hesitate to reach out to a member of our Trade team, or your usual Freshfields contact.