BRITISH SUGAR PLC (R on the application of) v SECRETARY OF STATE FOR INTERNATIONAL TRADE

This application for judicial review seeks to challenge the decision, taken under Part 1 of the Taxation (Cross-border) Trade Act 2018 (the 2018 Act) and given effect by the Customs (Tariff Quotas) (EU Exit) Regulations 2020 (the Regulations), to provide for an autonomous tariff quota (ATQ) for raw cane sugar, such that no import duty was payable on the first 260,000 metric tonnes (mt) imported for reg.

The challenge is brought by the Claimant (British Sugar) against the Secretary of State for International Trade (the Secretary of State). The application was originally brought on the basis that the decision to for the ATQ had been the Secretary of State’s. The application was resisted on the basis that the decision had been taken jointly by the Secretary of State and Her Majesty’s Treasury (The Treasury). However, on 17 January 2022 the Government Legal Department (GLD) wrote to British Sugar stating that it had come to GLD’s attention in the course of preparing the Secretary of State’s skeleton argument that the decision in question had not been taken by the Secretary State jointly with the Treasury, but by the Treasury on the Secretary of State’s recommendation, with the decision as to the form of the ATQ (particularly as to whether it would involve licensing or be operated on a “first come, first served” basis) having been taken by the Secretary of State for the Environment, Food and Rural Affairs (SSE). Sensibly, the parties were able to agree that the hearing would proceed on the basis that the challenge was brought to the Secretary of State’s recommendation to the Treasury.

The Interested Party (T&L) is the only refiner of raw cane sugar in the United Kingdom (UK) on any appreciable scale. It is common ground that, to date, it has imported well over 99% of the raw cane sugar which has benefited from the ATQ.

British Sugar contends that the Secretary of State’s decision to recommend the ATQ was unlawful on two grounds:

Ground 1 alleges that the ATQ constituted unlawful State aid to T&L, contrary to Article 10(1) of the Protocol on Ireland/Northern Ireland (the Northern Ireland Protocol), a protocol to the Agreement on the withdrawal of the United Kingdom from the European Union of 24 January 2020 (the Withdrawal Agreement).

Ground 2 alleges that the ATQ constituted an unlawful subsidy to T&L, contrary to Chapter Three of Title IX of Part II (the Subsidy Control Provisions) of the Trade and Cooperation Agreement between the UK and the EU (the TCA).

It will be necessary to say a little more about these two regimes, and their interrelationship with each other, in due course. Before doing so, I will set out my findings as to the underlying facts.

The UK is what is known as a “deficit country” so far as refined sugar is concerned, which is to say that it cannot meet the entirety of demand within the UK for this product from sugar refined domestically. In broad terms, the evidence before me established that:

50% of the demand for refined white sugar was met by British Sugar, which refines sugar extracted from domestically-grown sugar beet at four sites in England. By virtue of being domestically-grown, no tariff applies to the sugar beet used by British Sugar.

There is no market for imported sugar beet. However, sugar beet imported from the European Union (EU) benefits from a zero tariff under the TCA.

25% of the demand is met by T&L, which is the only refiner of cane sugar in the UK of any appreciable scale, and produces refined sugar using raw cane sugar imported from tropical and semi-tropical countries outside the EU at a factory in London.

The remaining 25% of the demand is met by importing refined sugar from the EU, where it is mostly produced by refining locally-grown sugar beet (on which no tariffs are paid by the refiners) but also from raw cane sugar originating in certain French départements and régions d’outre-mer.

S.1 of the 2018 Act provides for the imposition of import duty “by reference to the importation of chargeable goods into the United Kingdom”. S.8 provides for the Treasury to make regulations establishing and maintaining a customs tariff in force. S.8 provides:

As is apparent, s.8(8) contemplates that there will be a standard case and a series of departures from that standard case.

The first departure, s.9, is for preferential rates which are lower than the standard tariff as a result of “arrangements with the government of a country or territory outside the United Kingdom.” There are zero rate tariffs in operation under this section by reason of arrangements between the UK and:

ACP countries (African, Caribbean and Pacific countries first given duty free access on sugar imported into the EU, subject to a quota limit, under the Lomé Convention 1975 and its successors); and

the EU by virtue of the TCA;

S.10 provides for preferential rates which the UK applies unilaterally to goods “originating from an eligible developing country”. Under this section, Regulations 3, 6 and 12 of the Trade Preference Scheme (EU Exit) Regulations 2020/1438 provide for a trade preference scheme to be known as the GSP or Generalised Scheme of Preferences for Least Developed Countries, and imports within this scheme are subject to a zero tariff.

S.11 deals with quotas and provides:

S.11(2)(a) addresses tariff quotas which result from bilateral or multilateral arrangements, and s.11(2)(b) addresses unilateral tariff quotas. If the UK-Australia FTA is ratified, then this will give rise to a tariff quota under s.11(2)(a) in its first 8 years of operation. It was agreed before me that the effect of s.11(7) is that it is for the Treasury to make ATQ regulations having regard to the recommendations of the Secretary of State, save that the decision whether the ATQ should be subject to a licensing or allocation system is a matter for the SSE or another Secretary of State.

S.19 provides for regulations making provision for full or partial relief from a liability to import duty. Under s.19(2)(c), such provision may be made by reference to “the purposes for which goods are imported.” Under Regulation 20 of the Customs (Reliefs from a Liability to Import Duty and Miscellaneous Amendments) (EU Exit) Regulations 2020, relief has been granted from the standard rate of import duty on both raw cane sugar and beet sugar which have been imported for the purposes of refining.

Finally, s.28(1) of the 2018 Act provides that “in exercising any function under any provision under this Act” the relevant minister or government department “must have regard to the international arrangements to which Her Majesty’s Government in the United Kingdom is party that are relevant to the exercise of the function”. British Sugar says that the Northern Ireland Protocol is such an international arrangement, and, in the alternative to its argument that the ATQ was an unlawful contravention of the Northern Ireland Protocol, submits that the Secretary of State’s decision to recommend the ATQ to the Treasury was unlawful because the Secretary of State failed to have regard to the terms of the Northern Ireland Protocol in making her recommendation (it being common ground that the Secretary of State concluded, on the basis of advice, that Article 10 of the Northern Ireland Protocol was not engaged).

While the UK remained a member of the EU and during the Transition Period, the tariff applicable to imported cane sugar was set at EU-level, in the form of the Common External Tariff (CET).

In anticipation of the expiry of the Transition Period provided for by Article 126 of the Withdrawal Agreement (the Transition Period), consideration was given by the UK Government to the tariff provisions which should apply to a range of imported goods as part of the UK’s independent trade policy.

Documents obtained by British Sugar under Freedom of Information Act (FOIA) requests show that there were Government Departments in contact with T&L over this issue from at least early 2019. A letter from T&L to the Department of Environment Food and Rural Affairs (DEFRA) of 23 January 2019 refers to prior discussions about the post-Brexit tariff position so far as imported cane sugar was concerned, and it is clear from the letter that T&L had originally lobbied for a rather different tariff regime for raw cane sugar than it understood the Government to be proposing. T&L expressed concern that other operators might seek to access any ATQ, either an existing beet refiner (i.e. British Sugar) or a new small-scale entrant, which would threaten T&L’s viability. T&L was pushing for as large an ATQ as possible, limited to refiners.

On 8 March 2019, at a point in time when it was necessary to anticipate the possibility that the UK might leave the EU without a trade deal, the Secretary of State published details of a temporary tariff regime (TTR). The TTR provided for a tariff rate quota (TRQ) for raw cane sugar and raw beet sugar, with the raw cane sugar for refining quota being 260,000 mt. An internal DEFRA note of a meeting with T&L on 13 January 2020 referred to T&L's "contentment" with a TRQ of 260,000 mt.

However, on 23 January 2020 the UK Parliament passed the European Union (Withdrawal Agreement) Act 2020, enabling ratification of the Withdrawal Agreement, and on 6 February 2020 the Secretary of State announced the launch of a public consultation on the development of a new post-Brexit tariff regime, referred to as the UK General Tariff (UKGT). That consultation was conducted between 6 February and 5 March 2020. The consultation papers did not refer to a proposal to introduce an ATQ for raw cane sugar.

An email from an unidentified official within the Department for International Trade (DIT) of 5 May 2020 referred to concerns expressed by DEFRA that “the sugar tariff will lead to a certain UK manufacturer going under [i.e. T&L]”, continuing “this industry is a national icon and is pro-brexit”. The DIT recorded that it was not opposed to an ATQ for raw cane sugar, and that “[n]ot doing something is likely to lead to Ministerial backlash, due to the lobbying reach of this company”.

Emails exchanged by officials within the Treasury over the period 5-7 May 2020 discussed DEFRA’s statement as to the effect of a sugar tariff on T&L, noting “it is true that the company concerned (Tate & Lyle) was pro-Brexit, has an iconic brand and has political links, but best to set that to one side”. A further email of 7 May 2020 noted DEFRA’s argument that an ATQ “provides access to cheaper cane for T&L but also avoids ACP nations from losing their preference entirely and facing a significant cliff edge on exports”. DEFRA officials continued to discuss this issue during May, noting that an ATQ of 260,000 mt as previously discussed would continue to satisfy a redacted entity, which was clearly a reference to T&L. Officials from the DIT and the Treasury were involved in similar exchanges, a feature of which was a view that T&L would not be happy with a 260,000 mt ATQ, but would want a higher figure.

It is clear that T&L, with the support of DEFRA, was pushing for an ATQ substantially higher than 260,000 mt. The Department for International Development (DfID) held a different view. A memorandum of 14 May 2020 referred to the need for the “ATQ volume … to be large enough to enable the refiner to scale up production, but not too large”, suggesting that the figure DIT and DfID had put forward was “an initial estimate of the lower bound of the amount required for the refiner to scale up production, providing the least opportunity for ACP supply to be displaced and for adverse effects on negotiation capital with FTA partners”.

On 14 May 2020, an email was submitted to the Secretary of State referring to DEFRA’s proposed ATQ of 500,000 mt and the DIT proposal of 260,000 mt. The memorandum expressed the view that, without access to cheaper raw cane sugar, the viability of T&L was in question “causing ACP producers to lose their market”. The figure of 260,000 mt was recommended, essentially for the same reasons as those in the DfID memorandum of the same date. An email of 14 May 2020 records that the Secretary State accepted the 260,000 mt recommendation, to be kept under review, and recommended that “Defra might engage with T&L on what they could realistically live with”. A Treasury email of 15 May 2020 also reflected a concern that “if the ATQ is too low and [T&L] decide they are not viable, and close down, we lose productive capacity in the UK”.

The position of DIT and DfID as regards the amount of the ATQ appears to have won the day. On 19 May 2020, the Secretary of State published a ‘Summary of Responses’ to the Consultation, and a document entitled ‘Government Response and Policy’ (the UKGT Document). The UKGT Document recorded (at Section 3 page 17) that the Government “has also at this time established an autonomous quota to allow for a set volume of raw cane sugar to enter the UK tariff free, in order to balance support for UK producers and to maintain preferential trade with developing countries”, referring to a proposed ATQ of 260,000 mt. The following day, British Sugar wrote to the Secretary of State expressing concern at the introduction of the ATQ without consultation. In response, the Secretary of State said that the Government had not yet decided to adopt the ATQ, although it was minded to do so, and that there would be a process of consultation on whether to enact the ATQ. In his witness statement, Mr Mason of T&L said of this announcement that “it was not a complete surprise to us that the UK Government said it was proposing an ATQ be established”.

It is clear that both T&L and British Sugar were in contact with the Government, seeking to advance their own interests so far as the status of raw cane sugar for refining under the new UKGT was concerned. An email note of a meeting between British Sugar, the Rt Hon. Greg Hands MP and DIT officials on 1 July 2020 records British Sugar expressing their disquiet at the proposed ATQ, with Mr Hands MP recorded as replying that “260k tonnes is ‘substantial’ but ‘not as much as Tate & Lyle would have wanted’”. An email exchanged between various Government departments including DEFRA, DIT and the Treasury on 8 July 2020 stated that:

An internal paper of 10 August 2020 discussed the sugar market generally, and referred to the decision of DIT and the Treasury to create an ATQ to remedy the problem facing “the sole cane refiner” while “also trying [to] keep the trade preference for ACP/LDC [Least Developed Countries] suppliers”, noting the link between these objectives: “[o]f course, in order to have any market at all for ACP/LDC suppliers, the UK must retain a viable refining industry”. A DIT paper to Mr Hands MP of 12 August 2020 expressed the view that the ATQ would allow T&L to “increase its production which D[EFRA] believes will replace some of the EU’s share of the domestic market”. The paper also noted:

A further round of public consultation took place between 14 September and 5 October 2020. At the same time, various Government Departments provided their input, the contents of which are before the court as a result of FOIA requests. Those responses refer to the need to ensure the viability of T&L, the potential implications for sugar beet refiners and demand for EU beet, the resistance of ACP producers to the ATQ (which would divert imports from their own tariff free but more costly sugar cane), and the fact that if T&L did not survive, there would be no market for ACP or LDC cane sugar at all. It is fair to say that a range of justifications for an ATQ are expressed by officials in different departments, albeit ensuring the viability of T&L is a common theme. Benefits which it was thought that preserving T&L would achieve included increased competition and diversity of supply to the benefit of consumers, protecting jobs, ensuring a market for ACP and LDC cane suppliers, making T&L more competitive when “competing with EU and beet-produced white sugar”, displacing EU white sugar in the UK market and encouraging smaller scale refiners to enter the UK market.

DEFRA set out its post-consultation position on the size of the ATQ on 16 October 2020, in a submission to the Secretary of State (Rt Hon George Eustice MP), Rt Hon. Victoria Prentis MP and Lord Goldsmith of Richmond Park. That submission referred to a consensus between officials of DEFRA, the DIT and the Treasury in favour of an ATQ of 390,000 mt, which amount “would not be expected to harm the UK beet sector as it can already compete against sugar produced at global prices” and stated that it was “EU sugar exports that are most likely to lose out”. The figure of 390,000 mt was said to reflect the amount necessary “to allow EU sugar exports [which averaged 362,849 mt] to be fully replaced”. It noted that without a significant ATQ, it was likely that T&L would cease to be viable with significant job losses, the LDC trade with the UK in cane sugar could be extinguished, and there would be a significant loss in negotiating capital in any FTA negotiations with Australia. However, an email of 29 October 2020 suggested that the SSE did not support the 390,000 mt figure, arguing that the 260,000 mt had featured in the TTR because in a no-deal scenario, EU-produced refined sugar would face considerable tariffs. It was suggested that the size of the ATQ in a scenario in which there was a trade deal between the UK and the EU should be lower than 260,000 mt. The submission noted:

On 23 October 2020, the Treasury set out its position on the size of the ATQ following the consultation process in a submission which went to the Chancellor of the Exchequer. It noted the competing interests of British Sugar and T&L, and T&L’s request for an ATQ of 710,000 mt. After considering the interests of producers and consumers, the paper recommended an ATQ of 500,000 mt, which would “boost the efficiency and long-term viability of the UK cane refining sector, secure the UK as an export market for raw cane sugar from developing countries …, create employment opportunities in a less advantaged part of London … [and] [offer] diversity of supply to the UK”. The paper noted that DEFRA also supported the 500,000 mt figure but DIT and the Foreign, Commonwealth and Development Office (into which DfID had been subsumed) (FCDO) wanted a figure of 260,000 mt. The Chancellor said that he did not have a strong view, and he asked DIT and DEFRA to reach an agreed position.

On 3 November 2020, the then Minister of State for Trade Policy, the Rt Hon Greg Hands MP, decided to make a recommendation on the Secretary of State’s behalf to the Treasury to introduce the ATQ at a level of 260,000 mt.

The submission made to the Minister of State on 2 November 2020 for the purpose of reaching his decision was lengthy and detailed:

It noted that other departments (specifically the Treasury and DEFRA) had pushed for a larger ATQ volume, but that this had been resisted “to protect negotiating capital and the interests of developing countries”. The reference to negotiating capital is a reference to the fact that a lower figure would allow the UK to “trade” tariff-free access for raw cane sugar above that point with sugar cane-producing countries in trade negotiations.

The reason why a lower ATQ would protect developing countries was that such countries were already benefiting from tariff-free imports under a scheme available to sugar cane producers in the ACP group of states, who would see that advantage eroded by the amount of any ATQ.

It recorded the views of T&L both in the consultation “and in previous conversations with various Government Departments” that the current market situation (one in which refiners using domestic sugar beet paid no import duties whereas T&L did so unless buying ACP raw cane sugar which had higher production costs) rendered their business “unsustainable”. T&L said that an ATQ of less than 260,000 mt would not allow them to carry on their business, and would be “devastating” for it, an assertion which the authors of the submission found credible.

It noted that the effect of T&L ceasing production would be that “there would be no raw cane refiner in the UK, removing the market for developing countries, potentially resulting in there being a single domestic sugar producer (British Sugar) in the UK and disincentivising any free trade agreement [FTA] partners with cane sugar production [capacity] (Australia and, eventually, Brazil) to be interested in paying for cane sugar access in a deal”.

It noted that, on the basis of T&L’s submission, an ATQ of 260,000 mt would ensure its survival, and that the DIT had been “contacted by another company which is looking at setting up a raw cane refining process in the UK, albeit on a much smaller scale to T&L”, who had made it clear that for this to be viable, it needed an ATQ.

It noted that maintaining a raw cane refining capacity was necessary to secure the “UK’s wider diplomatic and trade objectives of importing raw cane sugar from ACPs and LDCs”.

It recorded that the Treasury and DEFRA did not think the ATQ would negatively impact the domestic sugar beet industry, because prices were set by parity with the costs of imported refined sugar from the EU.

It recommended setting the ATQ at “the lowest volume possible which should keep a refiner in operation”.

It recorded DIT’s view that there was no need to notify the ATQ to the EU Commission on the basis that it constituted State aid for the purposes of Article 10 of the Northern Ireland Protocol.

On 13 November 2020, an email to the SSE recorded that DEFRA had considered various options, including a split ATQ (100,000 mt at 0% and 160,000 mt at 15%) and a 260,000 mt ATQ, and was recommending the latter. The SSE confirmed he was content to proceed with the 260,000 mt figure.

On the basis of Mr Hands MP’s recommendation, the decision to introduce the ATQ was taken by the Financial Secretary to the Treasury, the Rt Hon Jesse Norman MP, subject to collective agreement, on 27 November 2020. Mr Hands MP and Mr Norman MP sent a “write around” letter to the Prime Minister on 1 December 2020 seeking the agreement of the EU Exit Strategy Committee, the SSE and the Health Secretary to an ATQ of 260,000 mt to be allocated on a “first come, first served” basis. The ATQ was recommended for essentially the same reasons as those which had been set out in the submission to Mr Hands MP. In summary:

As I have stated, the issue of whether the ATQ should be available on a licensing or “first come, first served” basis was one for the SSE. The SSE, the Rt Hon. George Eustice MP, approved the proposal for a “first come, first served” ATQ at some point prior to 7 December 2020. A letter from the Prime Minister’s office of 15 December 2020 records a request by the Secretary of State for Health for confirmation that the ATQ would not lead to a reduction in the price of sugar, and stated that the DIT and Treasury had confirmed that “the best available evidence suggests this to be the case”. The Regulations were made jointly by the Treasury and the SSE on 15 December 2020. On 16 December 2020, the Government published its response to the Consultation Process and announced its intention to introduce the ATQ.

At 11.00pm on 31 December 2020:

The Regulations (and other regulations under Part 1 of the 2018 Act) took effect. They imposed:

a standard fixed tariff of £35.00 per 100kg for raw cane sugar;

a reduced duty of £28.00 per 100kg for raw cane sugar imported for the purposes of refining; and

an ATQ of 260,000 mt for sugar imported for refining which was tariff free.

Alongside these provisions, raw cane sugar could also be imported tariff-free or at a tariff less than the standard-fixed tariff:

where it was imported from LDCs as a result of provision made under s.10 of the 2018 Act;

where it was subject to a free-trade or preference agreement under s.9 of the 2018 Act; and

where it was subject to a quota contained in an international agreement given effect under s.11(2)(a) of the 2018 Act.

The Northern Ireland Protocol came into effect.

Modifications to the 2018 Act effected by the Taxation (Post-transition Period) Act 2020 intended to give effect to the Northern Ireland Protocol came into effect.

The provisions of the TCA came into effect at 11.00 pm on 31 December 2020.

On 15 January 2021, British Sugar wrote to the Secretary of State asking her to provide sufficient data to enable it to assess whether the ATQ was compatible with Article 366 of the TCA, and expressed its view that the ATQ was incompatible with the Northern Ireland Protocol. The Secretary of State replied on 26 February 2021, asserting that the TCA did not apply and that there was no incompatibility with the Northern Ireland Protocol.

On 15 January 2021, British Sugar communicated its decision to apply for a review of the ATQ under Article 369.5(a) of the TCA and s.29 of the European Union (Future Relationship) Act 2020 (the 2020 Act). By way of brief explanation:

Article 369.5(a) of the TCA provides that an interested party may apply for a review by a court of tribunal of “the grant of a subsidy by a granting authority” or “any relevant decision by the granting authority”.

S.29 of the 2020 Act provides that domestic law has effect with such modifications as are required for the purposes of implementing the TCA in law, so far as such implementation is necessary for the purposes of complying with the international obligations of the UK under the TCA.

Ms Demetriou QC advanced five propositions which the Court was asked to accept on the basis of this material, and, at my request, Mr Robertson QC provided a document setting out the Secretary of State’s response. I address the five propositions in turn.

I accept that an ATQ of 260,000 mt first surfaced in early 2019, when the Secretary of State was formulating recommendations for the TTR, and that this followed discussions with T&L. I do not accept, however, that the figure of 260,000 mt came from T&L (and I accept the evidence of Mr Mason of T&L that the 260,000 mt figure did not originate from T&L). Thereafter, as set out above, officials in various departments argued for different figures, some higher and some lower than 260,000 mt, which to a degree reflected the different areas of interest of those departments. The ATQ of 260,000 mt was finally adopted by way of a compromise between those competing positions, and the competing policy concerns which they, in turn, reflected.

I should also record that just as T&L was in contact with various Government departments with a view to getting its point-of-view across, I am sure British Sugar did likewise (and indeed there was some evidence to this effect before me). I did not find it all surprising that rational economic actors should seek to persuade the Government to adopt the tariff regime best suited to their own businesses.

This proposition is challenged by the Secretary of State who argues as follows:

The position is more nuanced than either formulation acknowledges:

There can be no doubt that a liberalisation of the existing tariff regime (which imposed a significant tariff on imported cane sugar save for cane sugar from the ACP and LDC countries, which had higher production costs) was seen as necessary to ensure that T&L remained viable. While many of the official documents preferred the formulation “the UK raw cane sugar refining industry” to referring to T&L by name, the reality (as everyone knew) was that T&L was the UK raw sugar cane refiner.

Ensuring the viability of T&L was seen both as a desirable end in itself, and as essential to the pursuit of Government policies: preserving jobs, preserving competition and diversity of supply in the UK market (both by ensuring the survival of T&L and encouraging other cane-refining capacity to enter the UK market), providing a market for ACP and LDC raw cane sugar and a market for the raw cane sugar of countries with whom the UK might wish to negotiate an FTA (which would not exist if T&L ceased to operate).

Those additional benefits were not some remote or collateral or consequential benefit of the survival of T&L (in the manner that “Imperious Caesar, dead and turned to clay, might stop a hole to keep the wind away”), but the other side of the same coin.

The decision to achieve those ends through an ATQ rather than a zero tariff, and the decision to set that ATQ at 260,000 mt, reflected a compromise between conflicting policies:

A zero tariff decision, or an ATQ of greater than 260,000 mt, would have been more favourable to T&L, but damaging to ACP and LDC producers, and damaging to trade negotiations with other raw cane producing countries who might be expected to trade something for tariff-free access to the UK market.

An ATQ of less than 260,000 mt was seen to run the risk that T&L would not be viable, and would exit the raw cane sugar refining market with the adverse consequences in [43(ii)] above.

The ATQ was established, and set at a volume of 260,000 mt, to achieve the objectives identified in the preceding paragraph.

While the Secretary of State was aware that an ATQ might encourage more sugar refining capacity to enter the UK market, and saw this as a potential beneficial outcome of an ATQ, there was an awareness that, certainly in the first year or so of its operation, the overwhelming likelihood was that the overwhelming majority of the ATQ would be used by T&L. Had there been a contrary view, then there would have been too great a risk of the ATQ of 260,000 mt not ensuring T&L’s viability. That finding is borne out by the evidence as to the use of the ATQ in its first year of operation, which (according to the evidence of Ms Claire Vince of the DIT which I accept) has involved five importers (including T&L) taking the benefit of imports under the ATQ, but well over 99% of the imports which have benefited from the ATQ going to T&L.

The Secretary of State’s expectation that the ATQ would overwhelmingly be used by T&L reflected the fact that T&L was the only raw cane sugar refiner in the UK, and it would have been a costly exercise for any another refiner to establish itself on any scale in that market, which on the basis of prevailing market economics would not have been attractive. So far as this issue is concerned:

British Sugar had applied for permission to adduce expert evidence on the issue of whether it would be “practicable” for another refiner to begin operations in the UK so as to be capable of benefiting from the ATQ, and had obtained an expert report intended to show that starting up a raw cane sugar refinery in the UK would not have been an economically rational decision.

That application was resisted by the Secretary of State. At a hearing before Mrs Justice Collins Rice, Mr Johnston for the Secretary of State resisted the application, on the basis that “that evidence simply does not address a point on which the Secretary of State has joined issue in this case”, and that the issue of whether it would have been economic for another refiner to enter the market was “not relevant” (transcript of 14 October 2021, p.29). Later Mr Johnston summarised the Secretary of State’s position as being that “it is in principle possible for anyone who wishes to import raw cane sugar to do so, that is the beginning and end … of the Secretary of State’s case” (p.39), and that “the Secretary of State … simply is not fighting on the terrain” of whether “as a question of fact … no other entity would actually enter the market” (p.43).

Mrs Justice Collins Rice refused to give British Sugar permission to adduce the expert evidence, holding (at [2021] EWHC 3472 (Admin), [12]) that British Sugar could advance its arguments on the basis of Dr Carr’s witness evidence and legal submissions.

I am willing to accept British Sugar’s submission that it would not have been economically rational for an operator to enter the UK refined sugar market following the introduction of the ATQ for the purposes of refining raw cane sugar on any material scale. Dr Carr’s evidence was that establishing a raw cane refining capacity in the UK would involve a new entrant in costs of £100 million to £200 million, to establish a refining capacity and develop customer relationships in the UK.

However, I am not persuaded that this was the result of any feature of the ATQ, still less that the terms of the ATQ had been designed to achieve this effect. The “barriers to entry” to which Dr Carr referred were not those imposed by the tariff system of which the ATQ formed part, or the ATQ itself.

British Sugar argued that there were two features of the ATQ which were a positive inhibition to new entrants to the market.

First, what was said to be its temporary nature, which is said to have made it “yet more commercially unviable” to enter the raw cane sugar refining industry. I do not accept this suggestion (although I would note that, on the basis of Dr Carr’s evidence, it is said that market entry for a new cane refiner is commercially unviable regardless of this feature):

I accept that the fact that a measure is introduced on a time limited basis may make it “de facto selective”: see for example Joined Cases T-239/04 and T-323/04 Italy and Brandt Italia v Commission ECLI:EU:T:2007:260, [66] (judgment of the then Court of First Instance (CFI)).

However, the ATQ was introduced as a measure which would continue to operate until positively amended, rather than a measure which was temporary by its terms (as would be the case, for example, with legislation with a sunset mechanism). It could, of course, be amended, but that is true of any measure taken by or under the auspices of a sovereign parliament. Benjamin Franklin may be right that the only certainties in life are death and taxes, but as recent history has shown, the level of both can vary significantly over time.

While the Government’s response to the consultation exercise published on 16 December 2020 provided that the ATQ was to be reviewed after 12 months, the possibility of future review is inherent in any tariff policy. In this case, the Government was embarking on a new tariff regime for the first time for nearly 50 years. There are clear parallels with the observation of the Court of Justice in Case 596/19 P, Commission v Hungary ECLI:EU:C:2021:202, [58]:

Further, the justifications offered in the Government’s response were supportive of the fact that there would be a long-term supply of tariff-free raw cane sugar for refining into the UK market. Thus, the response noted that one reason for setting the ATQ at 260,000 mt was the desire to preserve a negotiating position which would allow more tariff-free raw cane sugar to be imported through FTAs providing “permanent, preferential access”. It also expressly referred to the Government’s belief that the ATQ would help ensure a viable cane sugar refining industry in the UK, maintain a diverse sugar market and provide an “opportunity for investment into the UK’s raw cane sugar refining industry”.

British Sugar’s own position (as set out in its letter to the Secretary of State of 20 May 2020) was to seek “an assurance that this quota is for one year only”, which would be a somewhat surprising position for it to have taken if a time-limit on the ATQ was a feature which had the effect of benefitting T&L.

While not relevant for present purposes, I should record that on 21 December 2021, the Secretary of State announced a decision to maintain the ATQ in force subject to a further review before 31 December 2024.

Second, the amount of the ATQ, which was said to have been set at the level believed to be the minimum necessary to ensure T&L’s viability, with the result T&L that would be fighting hard for every tonne. As to this:

The ATQ was set at the level it was for the reasons set out at [43] above. The 260,000 mt figure was not intended to inhibit market entrants.

The “first come, first served” nature of the ATQ exposed T&L to the risk of being unable to obtain the volumes of raw cane sugar necessary to ensure its viability.

It is clear from Dr Carr’s evidence that British Sugar itself “has no will or ability to utilise this asset, regardless of the level of tariff on raw sugars”.

I am not persuaded, in the light of Dr Carr’s evidence as to the scale of capital input necessary to establish a rival cane sugar refining capacity at any scale in the UK and the innate advantages which T&L enjoyed as the established cane refiner in any event, that the size of the ATQ was in practice an inhibition to new market entrants. On the contrary, I am satisfied that an ATQ set at a higher level would have been more, not less, useful to T&L, which is no doubt why T&L (who might be thought to know its own interests best) lobbied for that outcome.

I agree that the advice to Ministers was consistently to the effect that the ATQ was likely to reduce the volume of refined sugar imported from the EU.

British Sugar maintains two distinct, but related, challenges to the ATQ:

First, that it constitutes unlawful State aid as a matter of EU law, and engages Article 10 of the Northern Ireland Protocol.

Second, that it constitutes an unlawful subsidy under the TCA.

As will be apparent below, there is a considerable degree of commonality of analysis in the issues raised by these two arguments, but it is important to recognise the different legal regimes which apply, and the reasons for them.

As is well known, the Northern Ireland Protocol is intended to reflect the fact that, while the United Kingdom as a whole has left the EU, Northern Ireland remains in the European Single Market for Goods (the Single Market) constituted by Articles 34 to 36 of the Treaty of the Functioning of the European Union. Reflecting that fact:

As with the other constituent parts of the Single Market, it is the European legal order which determines what constitutes State aid in this context (Article 10 and Annex 5 to the Northern Ireland Protocol).

In relation to Article 10, the institutions, bodies, offices and agencies of the EU have the powers conferred on them by EU law.

In practical terms, this includes the EU Commission in respect of such powers as are conferred on it by EU law in relation to State aid, and the Court of Justice of the European Union.

By contrast, the subsidy control provisions under the TCA have (to an appreciable extent) a different legal pedigree, having been derived in part from the rules of the World Trade Organization (WTO) which regulate issues relating to the freedom of trade between signatory sovereign states rather than members of a particular trading block. Reflecting this:

The language of the subsidy control provisions is, to a significant extent, drawn from WTO sources.

The dispute resolution regime provided for by Article 375 of the TCA is for an arbitration tribunal or dispute resolution by the Dispute Settlement Body under the WTO Agreement.

Article 516 provides that “the interpretation and application of the provisions of this Part shall take into account relevant interpretations in reports of WTO panels and of the Appellate Body adopted by the Dispute Settlement Body of the WTO as well as in arbitration awards under the Dispute Settlement Understanding”.

There is no role for the European Commission, nor any obligation to seek the sanction of the European Commission before implementing a subsidy.

I shall return to the significance of these two distinct regimes when considering the issues which arise as to the interpretation of the Northern Ireland Protocol below.

British Sugar does not argue that T&L has received a direct transfer of aid (or subsidy) from UK government resources, but that the effect of the ATQ has been to relieve T&L of an expense which it would otherwise have had to pay. In both contexts, British Sugar relies on the tariff for raw cane sugar imported for the purpose of refining of £28/100kg which would be payable for imports of non-ACP and LDC sugar in excess of the ATQ. British Sugar does not contend that:

the £28/100kg tariff (a reduction from the standard tariff of £35/100kg effected under s.19 of the 2018 Act);

the zero tariffs payable on ACP imports by virtue of regulations made under s.9 of the 2018 Act;

the zero tariffs for LDC imports imposed under s.10 of the 2018 Act;

the reduced tariffs for smaller quotas of raw cane imported from Cuba and Brazil under legacy arrangements entered into by some EU members prior to their accession (so-called CTX countries); and

any zero tariff raw cane sugar which may come to be imported by virtue of the UK-Australia FTA, whether under a tariff quota by virtue of s.11(2)(a) or a general zero tariff under s.9;

Ms Demetriou QC also accepted that if there had been no tariff at all on raw cane sugar for refining, there would be no State aid or subsidy for T&L, even though T&L would benefit from the overwhelming majority of raw cane sugar for refining imports benefiting from the zero tariff, and the financial benefit to T&L would have been significantly greater than that enjoyed as a result of the ATQ. British Sugar’s position in the event that an ATQ had been set at a level above the realistic volume of imports was unclear. However, while this would have involved T&L benefiting from an ATQ as a matter of form, the position as a matter of substance would have been the same as if there had been no tariff.

The effect of these arguments, therefore, was that the decision to set an ATQ at a level below the volume of raw cane sugar for refining which T&L would import, and to set it below that level not to assist T&L, but to benefit the ACP countries and preserve negotiating capital in FTA discussions with third party countries, would have had the effect of converting something which was not State aid or a subsidy into something which was. That, at first sight, would be a somewhat surprising outcome, and any argument which would lead to that conclusion merits close scrutiny.

There is no dispute that the effect of Article 10 and Annex 5 to the Northern Ireland Protocol is to make Article 107 of the Consolidated Version of the Treaty on the Functioning of the European Union (Article 107) applicable to the UK “in respect of measures which affect that trade between Northern Ireland and the Union which is subject to this Protocol”.

Article 107(1) provides:

Article 107(2) then identifies a number of measures which “shall be” compatible with the Single Market, but there was no suggestion that any of these mandatory provisions applied in this case. Article 107(3) identified a series of measures which “may be” compatible with the Single Market. However, the decision as to whether or not the measure in question falls within one of the categories in Article 107(3) is for the European Commission, not the Member State. Article 108(3) requires a Member State to notify to the European Commission “in sufficient time” any proposal to grant State aid, and the Member State is required not to put the measure into effect until this procedure has resulted in a final (and favourable) decision from the Commission.

Reverting to Article 107(1), there are a number of conditions for its application, and disputes about whether a particular measure constitutes State aid generally proceed by considering them in turn. However, as Lord Woolf MR noted in R v Customs and Excise ex parte Lunn Poly and Bishopsgate Insurance Ltd[1999] 1 CMLR 1357, [25]:

Lord Justice Clarke at [75] made a similar observation, and noted that that global question was to be approached “on a broad pragmatic basis in the light of the policy underlying the Treaty” (see also [74]-[75]), by which he meant the analysis needed to be “realistic and in touch with the facts” (R (on the application of Professional Contractors Group Ltd and others) v Inland Revenue Commissioners[2001] EWCA Civ 1945, [50]). Lord Justice Clarke also emphasised at [50] that “[i]t must surely be the substance of the matter and not the form that controls whether or not there is State aid”. That observation is a constant theme of the EU case law, generally when noting that a measure which is State aid as a matter of substance cannot escape the strictures of Article 107 because it is given effect by a particular form. However, as Lord Justice Clarke observed, the converse is equally true.

It is common ground that the question of whether or not the ATQ constitutes State aid as a matter of EU law is a matter for the court. As Lord Woolf observed in Lunn Poly at [24], that question is an issue of “precedent fact”.

Bacon on State Aid [2.02] states that:

there must be aid in the sense of an economic advantage;

the advantage must be granted directly or indirectly through State resources and must be imputable to the State;

the measure must favour certain undertakings or the production of certain goods (‘selectivity’); and

the measure must be liable to distort competition and affect trade between Member States”.

I gratefully adopt that formulation. In this case, the argument concentrated on (c) and (d), and there was no suggestion that if British Sugar could overcome those hurdles, and also the Secretary of State’s argument that State aid through tariff measures did not fall within Article 10 of the Northern Ireland Protocol, Ground 1 might nonetheless fail on the basis of requirements (a) and (b).

Bacon on State Aid notes at [2.119] that:

The parties relied on this distinction here, and agreed that the ATQ was not de jure selective. An example of a measure which was held to be de jure selective was the contributions made by France towards the payment of the pensions of staff of France Télécom under a scheme put in place when the former state-owned entity was opened up to private investment in Case C-211/15 P Orange (formerly France Télécom) v Commission ECLI:EU:C:2016:798. The General Court’s conclusion that the measure was “selective in that it concerned only France Télécom” was upheld by the Court of Justice, which held at [53]:

However, there was a dispute between the parties in this case as to the proper approach to be applied in determining whether the ATQ was de facto selective. In summary:

British Sugar contends that it is sufficient to show that the terms of the measure had been drafted for the purpose of favouring certain undertakings, and contended that this was the position here.

The Secretary of State contends that it is necessary to follow a three-stage test set out in decisions such as Joined Cases C-20/15 P and C-21/15 P Commission v World Duty Free Group and others[2017] 2 CMLR 22, [54]-[60] (the World Duty Free test):

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 the 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.

If, contrary to its submissions, it is necessary to apply that three-stage test, then British Sugar contends that doing so leads to the conclusion that the ATQ is de facto selective in any event.

In support of its argument on the issue at [69(i)] above, British Sugar relied on the decision in Joined Cases C-106/09 P and C-107/09 P Commission and Spain v Government of Gibraltar and United Kingdom [2012] STC 305. That was a case in which the Government of Gibraltar had introduced a tax regime which distinguished between offshore and onshore companies, which the Commission had determined to constitute State aid. The Gibraltar tax regime was then amended to adopt a single regime, which on its face did not distinguish between onshore and offshore companies, but which used criterion for determining the applicable tax rate (such as number of staff on the payroll, the size of property occupied) which did exactly that. The Commission determined that these measures amounted to State aid, a decision overturned by the General Court on the basis that the Commission had failed to apply the three-stage World Duty Free test. The Commission challenged that conclusion on appeal ([49]), arguing that:

That argument was upheld by the Court of Justice who held at [88]-[93]:

I accept that Commission v Government of Gibraltar provides support for the argument that it is not necessary to apply the three-stage World Duty Free test not only in cases where the measure is de jure only applicable to certain undertakings, but where the terms of the tax have been designed so as to direct the benefit of State resources to particular undertakings. Gibraltar was a case in which the ‘normal’ or ‘reference’ system had been conceived as an exercise in State aid. As the Court of Justice observed of the Gibraltar case in C-596/19 P, Commission v Hungary ECLI:EU:C:2021:202, [49]:

However, this is not such a case. The only features of the ATQ which British Sugar pointed to in this connection were the fact the Government had announced its intention to review the ATQ after 12 months and the 260,000 mt figure. As to this:

I would note that the 12 month review was not a term of the ATQ itself, although I do not think that of itself would have been sufficient to reject British Sugar’s argument based on Commission v Government of Gibraltar.

However, I have already rejected on the facts British Sugar’s argument that the effect of the announcement on 16 December 2020 that the ATQ would be reviewed after 12 months or that it would be for 260,000 mt was to make the ATQ selective in T&L’s favour.

There is nothing in any of the material before the court to suggest that the decision to review the ATQ after 12 months or the fact that the ATQ was limited to 260,000 mt rather than a higher figure was designed to favour T&L at the expense of potential new entrants to the market for refining raw cane sugar.

I have accepted, however, that the Government’s expectation when introducing the ATQ was that the overwhelming majority of the ATQ would be used by T&L, not because of any term of the ATQ which was designed or in fact had this effect, but because T&L was the only raw cane sugar refiner operating in the UK. I am not persuaded that a measure which, on its face, is of general application falls to be treated as a de jure selective measure merely because, for reasons which are extraneous to the fact, terms or design of the measure, there is only one undertaking operating in the relevant market.

The only authority I was referred to in relation to this issue was Case T-14/96Bretagne Angleterre Irlande (BAI) v Commission ECLI:EU:T:1999:12, CFI.

In that case, in 1993 the Basque Government and the Provincial Council of Biscay had entered into an agreement with Ferries Golfo de Vizcaya SA (FGV) (which was intending to start a ferry service between Portsmouth and Bilbao) to purchase 26,000 travel vouchers to be used on the route, which were priced at above the market rate, and has also agreed to cover losses incurred by FGV in the first three years of its operation.

When the Commission indicated its intention to investigate this (1993) agreement as State aid, it was replaced in 1995 by a new agreement between FGV and the Provincial Council of Biscay alone, for 46,000 travel vouchers priced at below FGV’s published price, which figure was said to be based on the estimated demand for travel by particular groups in the Basque region. Other objectionable features of the 1993 agreement were dropped. As a result, the Commission investigation was terminated.

BAI, which operated a ferry service between Plymouth and Santander, objected to the Commission’s decision to terminate its investigation. The Kingdom of Spain, intervening, argued that “there was only one operator in a position to provide the transport services demanded by the provincial authorities for the benefit of people residing in their territory” ([56]).

The CFI held that the fact that the payments to FGV had been in return for travel vouchers did not prevent them constituting State aid ([71]). The Court was not persuaded that the 1995 agreement was “in the nature of a normal commercial transaction” ([75]-[76]) or that there was a “real need” for the Provincial Council to purchase the number of travel vouchers in question ([79]).

I am not persuaded that the BAI case assists British Sugar on the issue at hand. It was a case in which the State resources in question were supplied to FGV, and FGV only, under an uncommercial contract between the two. It was not the case that the vouchers were available to any shipping company operating on the Portsmouth-Bilbao route. This was, therefore, a de jure benefit, and a very clear case of state resources being used to support a particular operator.

It follows that I accept the Secretary of State’s argument that it is necessary to approach the issue of selectivity in this case by reference to the three-stage World Duty Free test.

Ms Demetriou QC argued that the ‘ordinary’ or ‘normal’ tariff regime is the £28/100kg tariff applicable to raw cane sugar imported for refining purposes. That involves a rather formalistic approach to the identification of the ‘normal’ regime, and, as I have noted above, the logical consequence of the argument would appear to be that any departure from the £35/100kg tariff is a departure from the ‘normal’ tariff.

A particular difficulty with applying this part of the World Duty Free test in the current context is that what differentiates the application of what British Sugar contends to be the ‘ordinary’ or ‘normal’ regime from the derogation is not any feature of the importer, the use to which the raw cane sugar is to be put or the source from which it is acquired. It is simply whether the raw cane sugar in issue forms part of the first 260,000 mt of non-preference raw cane sugar imported in the relevant annual period or not.

By contrast, when discussing the first limb of the test in World Duty Free, the Court of Justice emphasised the differential treatment of operators inherent in a derogation from the ‘normal’ regime:

At [57], the Court of Justice suggested that the distinction between the ordinary system and the derogation involved a differentiation “between operators who, in the light of the objective pursued by that ordinary tax system, are in a comparable factual and legal situation”.

At [67], the Court of Justice suggested that there needed to be “differences in the treatment of operators, although the operators who qualify for the tax advantage and those who do not are, in the light of the objective pursued by that Member State’s tax system, in a comparable factual and legal situation”.

At [77], the Court of Justice referred to “two categories of operators [being] distinguished and … subject, a priori, to different treatment, namely those who fall within the scope of the derogating measure and those who continue to fall within the scope of the ordinary tax system.”

Similarly, I note that Mr Justice Bourne in R (Enterprise Managed Service Ltd and another) v Secretary of State for Housing, Communities and Local Government[2021] EWHC 1436 (Admin), [101] observed that:

On this issue, Mr Beal QC for T&L referred me to the decision of the Court of Justice in Case C-524/14 P, Commission v Hansestadt Lűbeck ECLI:EU:C:2016:971, in which the Commission challenged a schedule of tariffs applicable to airlines using Lűbeck airport as State aid. That argument was rejected. The Court emphasised (at [54]) that “[i]n order to determine whether a measure is selective, it should therefore be examined whether, within the context of a specified legal regime, that measure constitutes an advantage for certain undertakings over others; which are, in the light of the objective pursued by that regime, in a comparable factual and legal situation”. At [59], the Court observed:

It is also well-established that the form of the measure under consideration – for example that it is worded as an exception – is not itself decisive in identifying what the ‘ordinary’ or reference system is: Case-203/16 P Dirk Andres v Commission [2018] ECLI:EU:C:2018:505, [91]-[92] and [104].

Applying these principles here:

The form of the ATQ is not itself decisive.

British Sugar argues that the operators in a “comparable factual and legal position” and who are subject to the “ordinary tax system” are “all producers who refine sugar from cane and those who refine it from beet”.

There is, of course, essentially only one sugar refiner importing raw cane sugar into the UK for the purposes of refining (T&L) and only one operator who refines sugar from beet (British Sugar).

British Sugar does not pay the £28/100kg tariff which it contends is the ‘ordinary’ or ‘normal’ tariff regime, because it does not import raw cane sugar

Anyone else looking to import raw cane sugar would be in the same position as T&L: it would not pay a tariff to the extent that it was able to benefit from the 260,000 mt on a “first come, first served” basis or acquired raw cane sugar from a preferred producing country, but otherwise would.

There is EU case law on which T&L, in particular, relied which has held that beet and raw cane sugar refiners, “cannot be regarded as being in a comparable situation as regards the organisation of the market for sugar and, more particularly, the availability of supply on the EU market” (Case T-103/12T&L Sugars Ltd v Commission ECLI:EU:T:2016:82, [112]-[113]). It is not clear to me how far such findings in other contexts necessarily carry across to the issues at hand. However, even if the `ordinary’ or ‘normal’ regime is widened to encompass all raw sugar imported for the purposes of refining, once again British Sugar does not pay any tariff on raw beet sugar – it does not import beet, but even if it did, sugar beet imported from the EU has a zero tariff.

Similarly, anyone else looking to import raw beet sugar would be in exactly the same position as British Sugar (paying no tariff on sugar beet imported from the EU).

In these circumstances, I am unable to accept British Sugar’s argument that the ATQ involves a derogation from an “ordinary” or “normal” tariff position which, in practice, no one else (and none of those alleged to be in a “comparable and factual position”) other than T&L will pay. The difficulties with applying this aspect of the World Duty Free test themselves suggest that the issues raised in this case are not in State aid territory.

Essentially for the reasons set out in the preceding section, I have concluded that it does not. All that can be said is that someone who imports raw cane sugar for refining will benefit from the ATQ to the extent that the 260,000 mt has not been consumed at the stage of the relevant importation, but not otherwise. As the Court of Justice noted in World Duty Free, [59]:

On my findings, this issue does not arise, and there are risks in approaching the issue on what (on my findings to date) would be a wholly artificial basis. If, however, I had found in favour of British Sugar on the first two parts of the World Duty Free test, I would have had difficulty with Mr Robertson QC’s argument that any selectivity arose “from the nature or general structure” of the tariff system.

It is well-established that the concept of ‘State aid’ does not cover tax measures that differentiate between comparable undertakings where that differentiation flows from the nature or general structure of the system of which the measures form part: e.g., C-159/01 Netherland v Commission [2004] EU:C:2004:246, [42]-[43] and Joined Cases C-78/08 – C-80/08 Ministero dell’Economia e delle Finanze v Paint Graphos Sarl [2011] ECR I-7611.

This third stage of the World Duty Free test focusses on the organising principles or structural features of the relevant system of which the measure forms part. For example, a tax system in which rates are set by reference to the level of earnings of individuals, the turnover of companies or the profits of enterprises, may by its very structure differentiate between taxable subjects dependent on their earnings, turnover or profits. Similarly a tax on carbon emissions may differentiate by reference to the quantity of carbon emitted, and a landfill tax by reference to the volume of waste disposed of. The effect of the third stage of the World Duty Free test is that differences in the treatment of comparable undertakings under the subject measure which follow from these organising principles are not treated as selective for State aid purposes. This will be so even if the organising principles in question operate progressively or regressively in order to give effect to some underlying political policy, for example:

in the case of progressive taxes on wealth or profits, a redistributive agenda;

in the case of regressive taxes, an incentivisation agenda; and

in the case of taxes on carbon emissions or linked to the degree of polluting activity, an environmental agenda.

Stage three of the World Duty Free analysis might also justify what on their face appear to be differential treatments by way of an anti-avoidance measures which are designed to achieve comparable treatment within the terms of the regime, and thereby give effect to the organising principle of the measure by that means. If, for example, a system of personal taxation seeks to levy employment and income taxes on workers, a measure intended to prevent avoidance of such taxes through the use of corporate structures does not constitute State aid for that reason (as in R (on the application of Professional Contractors Group Ltd) v Inland Revenue Commissioners[2001] EWCA Civ 1945).

That does not mean, however, as Mr Robertson QC sought to argue, that it is possible to miss out the intermediate stage of an organising principle or structural feature of the relevant system, such that it is enough for the differential treatment of comparable operators not to constitute State aid if it has been adopted to serve some legitimate social purpose (e.g. to maintain employment in a deprived area, to support a particular region or minority community etc), rather than to give effect to the purpose or internal logic of the tax regime. In Case T-55/99Confederacion Española de Transporte de Mercancías (Cetm) v EC Commission ECLI:EU:T:2000:223, at [52]-[53], the CFI noted:

A similar point is made in Bacon on State Aid,[2.134]:

In responding to this point, Mr Robertson QC pointed to the decision of the Court of Justice inCase C-53/00, Ferring v Agence Centrale des Organismes de Sécurité Sociale (Acoss) [2001] ECR I-9067. In that case, a French measure imposed an additional tax on sales on direct sales by pharmaceutical manufacturers to retail pharmacies which was not imposed on wholesalers. That additional tax was imposed in order to compensate for a distortion of competition which arose from the fact that wholesale distributors were under certain public service duties which were not imposed on the manufacturers. The Court held that this did not constitute State aid. It is true that the Court observed at [17] that:

However, the Court did not decide the case on the basis that the tax on pharmaceutical manufacturers was part of “the logic of the system”, but on the basis that the exemption of wholesalers from the tax “may be regarded as compensation for the services they provide and hence not State aid” ([26]-[27]). I note that Bacon on State Aid at [2.80] addresses the Ferring case under the heading “Payments for public services”.

In this case, if the ATQ otherwise satisfied the requirements of State aid, neither the existence nor size of the ATQ followed from the “internal logic” or structure of the UK tariff regime, but from free-standing policy choices.

For the reasons I have set out above, I am satisfied:

that the issue of selectivity in this case falls to be determined by applying the three-stage World Duty Free test; and

applying stages one and two of that test, the ATQ is not selective.

I am also satisfied that this conclusion reflects the substance or reality of the position. While I accept that the overwhelming likelihood when the ATQ was introduced was that T&L would import the overwhelming majority of raw cane sugar within the ATQ limit, that was not the result of the terms or the design of the ATQ. As the Court of Appeal noted in R (Professional Contractors Group) v IRC[2001] EWCA Civ 1945, [38] “a mere propensity for a measure to favour one sector rather than another cannot amount to selectivity”. In that case, as in this one, the lack of uniformity in the practical effects of the measure “arose not from any selectivity but simply because that is ‘how things are’”. T&L is the only refiner at any scale of raw cane sugar in the UK.

By contrast, British Sugar’s State aid argument involves a number of difficulties and unrealities, including those outlined in Section C2 above. Ms Demetriou QC contended that it was the Secretary of State’s argument which did not accord with the substance of the position, positing a position in which the Government had agreed to reimburse T&L for any raw cane sugar tariffs it incurred, or imposing an ATQ which was limited by its terms to imports by T&L. However, whatever the final status of such measures would have been as a matter of substance:

Both would have been instances of de jure selectivity.

It would have been the very terms of the measures themselves which would have precluded any other operator from benefiting.

In any event, the examples are entirely artificial in a market in which, for reasons which are wholly extraneous to the ATQ, T&L was the only realistic importer of raw cane sugar at an appreciable volume for sugar refining purposes.

The Secretary of State raises two further arguments in response to British Sugar’s Article 107 argument, each of which raises an important issue as to the interpretation of the Northern Ireland Protocol:

whether provisions in the UK tariff regime can ever fall within the State aid provisions of the Northern Ireland Protocol (the Application of Article 10 to Tariffs Issue); and

the proper test for determining what impact the measure in question has or is liable to have on EU-Northern Ireland trade for Article 10 to be engaged (the Effect on Trade Issue).

The preliminary steps in both of these arguments overlap to some degree.

It was agreed that interpretation of the Northern Ireland Protocol is to be approached on the basis of Articles 31 and 32 of the Vienna Convention on the Law of Treaties 1969 (1980) (Cmd 4140) (VCLT), which themselves codify customary international law (Republic of Ecuador v Occidental Exploration and Production Co[2006] QB 432, [33]-[34]).

The general rule of interpretation is set out in Article 31:

It will be apparent that this is a predominantly textual rather than teleological rule of interpretation (as the International Court of Justice observed in Territorial Dispute (Libya/Chad) (1994) ICJ 6, [41]). The reasons for this approach were noted by Mr Justice Simon in Czech Republic v European Media Ventures SA [2008] 1 All ER (Comm) 531, [15]-[19]. In particular, at [17] Mr Justice Simon noted:

Article 32 contains “[s]upplementary means of interpretation” as follows:

Mr Robertson QC further supplemented these provisions by reference to Marchiori v The Environment Agency[2002] EWCA Civ 03, [58], in which the Court of Appeal applied the principle in dubio mitius (“where a Treaty provision is ambiguous, the interpretation which is less onerous to the State owing the Treaty obligation is to be preferred”) to the EURATOM Treaty.

Recitals (21) to (24) to the Northern Ireland Protocol provide:

Article 1(2) provides that “[t]his Protocol respects the essential State functions and territorial integrity of the United Kingdom”.

Article 4 provides:

Article 5 of the Northern Ireland Protocol addresses the issues of customs duties and movement of goods:

Article 5(3) provides that “[l]egislation as defined in point (2) of Article 5 of Regulation (EU) No 952/2013 shall apply to and in the United Kingdom in respect of Northern Ireland”.

Article 5 applies the provisions of EU law listed in Annex 2 to the Protocol “to and in the United Kingdom in respect of Northern Ireland”.

Article 5(6) provides:

reimburse duties levied pursuant to the provisions of Union law made applicable by paragraph 3 in respect of goods brought into Northern Ireland;

provide for circumstances in which a customs debt which has arisen is to be waived in respect of goods brought into Northern Ireland;

provide for circumstances in which customs duties are to be reimbursed in respect of goods that can be shown not to have entered the Union; and

compensate undertakings to offset the impact of the application of paragraph 3.

Article 10 provides:

The provisions of Union law listed in Annex 5 to this Protocol shall apply to the United Kingdom, including with regard to measures supporting the production of and trade in agricultural products in Northern Ireland, in respect of measures which affect that trade between Northern Ireland and the Union which is subject to this Protocol.

Notwithstanding paragraph 1, the provisions of Union law referred to in that paragraph shall not apply with respect to measures taken by the United Kingdom authorities to support the production of and trade in agricultural products in Northern Ireland up to a determined maximum overall annual level of support, and provided that a determined minimum percentage of that exempted support complies with the provisions of Annex 2 to the WTO Agreement on Agriculture. The determination of the maximum exempted overall annual level of support and the minimum percentage shall be governed by the procedures set out in Annex 6. 3. Where the European Commission examines information regarding a measure by the United Kingdom authorities that may constitute unlawful aid that is subject to paragraph 1, it shall ensure that the United Kingdom is kept fully and regularly informed of the progress and outcome of the examination of that measure.”

Annex 5, to which Article 10(1) refers, includes among the provisions of EU law referred to:

Articles 107, 108 and 109 TFEU

Article 106 TFEU, insofar as it concerns State aid

Article 93 TFEU

In support of his interpretation of the Northern Ireland Protocol, Mr Robertson QC relied upon the terms of the Political Declaration setting out the framework for the future relationship between the European Union and the United Kingdom of 19 October 2019 (the Political Declaration):

Paragraph 4 provided:

Paragraphs 17 and 18 provided:

Finally, Mr Robertson QC also referred to a letter from the Prime Minister to Mr Michael Barnier of 2 October 2019 and a speech given by Mr Barnier on 27 January 2020. However, no serious attempt was made to explain why these materials were admissible under Article 31(2) of the VCLT, and I am not persuaded that they are.

Mr Robertson QC for the Secretary of State argues that decisions by the UK to impose, vary or abolish import and export tariffs are not capable of falling within Article 10(1) of the Northern Ireland Protocol, even if they otherwise meet the requirements for State aid.

First, Mr Robertson QC argues that “it is plain from the text and context of the [Northern Ireland] Protocol that ... the UK should have an independent trade policy following the end of the Transition Period”, referring in this connection to:

Article 4 of the Northern Ireland Protocol, which recognises that Northern Ireland is part of the customs territory of the UK and will fall within the scope of trade agreements concluded between the UK and third countries or the UK’s Schedule of Concessions annexed to the General Agreement on Tariffs and Trades.

Recital (21) to the Northern Ireland Protocol, which is to similar effect, and which expressly provides that Northern Ireland “will benefit from participation in the United Kingdom’s independent trade policy”, and Recital (22) with its reference to Northern Ireland’s integral place in the UK’s internal market.

Paragraphs 4, 17 and 18 of the Political Declaration, set out above, with their references to “the sovereignty of the United Kingdom and the protection of its internal market, while respecting the result of the 2016 referendum including with regard to the development of its independent trade policy”, respecting “the integrity of … the United Kingdom’s internal market” and “recognising the development of an independent trade policy”.

The difficulty with these arguments is that, in common with previous agreements intended to address what the historian George Dangerfield once referred to as the “Damnable Question”, the Northern Ireland Protocol, and the documents which form part of the Article 31 VCLT interpretive matrix, have something in them for both sides:

Article 4 qualifies the provision relating to the inclusion of Northern Ireland within UK FTAs with the words “provided that those agreements do not prejudice the application of this Protocol”, which would include Articles 5, 8 and 10 (which apply a substantial body of EU law including the law on State aid and the EU Customs Code “in respect of Northern Ireland”).

Recitals (21) and (22) are followed immediately by Recitals (23) with its reference to the need for Ireland’s rights and obligations under the EU’s internal market and customs union to be fully respected.

Paragraph 4 of the Political Declaration also referred to the need to be “consistent with the integrity of the Single Market and the Customs Union”, and paragraph 17 to the need to facilitate “the integrity of the Union’s Single Market and the Customs Union”.

The reality is that the signatories’ goals were potentially inconsistent, and the precise balance drawn between them falls to be ascertained from the text of the Northern Ireland Protocol itself.

Further, there is one provision of the Northern Ireland Protocol which does expressly recognise that activity by the UK in the customs sphere is capable of engaging Article 10 and the EU State aid regime: Article 5(6) which makes certain activities by the UK in relation to reimbursement, waiver or compensation of duties levied by the UK under provisions of EU law “subject to Article 10”.

Second, Mr Robertson QC points to the fact that tariff measures were not capable of falling within the EU State aid regime when the UK was a Member State, and he suggests that it would be surprising if the Northern Ireland Protocol had enlarged the scope of the EU State aid regime. As to this:

It is correct that before the Withdrawal Agreement, a provision in the EU tariff regime could not constitute State aid for the purposes of Article 107.

This was because the tariff for goods entering the EU was set at EU level (the CET), and the revenues paid, or foregone, under that regime were EU revenues. For this reason, such attempts as were made to challenge EU tariff provisions under State aid rules failed. In Joined Cases 213/81-215/81 Norddeutsches Vieh-und Fleisch-Kontor Herbert Will v Bundesanstalt Fur Landwirtschaftliche Marktordnung[1982] ECR 3583, a challenge to the allocation at Member State level of an ATQ for frozen beef and veal negotiated by the EU on State-aid grounds failed, the Court of Justice observing at [22]:

However, this issue is capable of arising in relation to Northern Ireland in respect of its presence within the Single Market, because any State aid would not derive from the resources of the EU, but from the resources of the UK (at which Article 10(1) of the Northern Ireland Protocol is squarely aimed).

Further, under the EU law, the mere fact that a measure is of a type which was not itself subject to harmonisation under EU law does not mean that such a measure cannot fall within Article 107. As the Court of Justice observed in Case C-596/19 P Commission v Hungary ECLI:EU:C:2021:202, [32]:

In short, the focus of EU State aid law is on the effect of benefits provided from State resources to an operator, whatever the field of State activity in which the measure is enacted.

Third, Mr Robertson QC submits that if UK tariff decisions were capable of falling within Article 10 of the Northern Ireland Protocol, the result would be “manifestly absurd or unreasonable” because it would (or might) prevent the UK from pursuing the independent trade policy which the withdrawal from the EU was intended to permit. This was because:

The difficulty with this argument, in the context of a treaty between two sovereign entities who might reasonably have been expected to have had different concerns and priorities in the course of negotiations, is that the issue of what might be said to be manifestly absurd or manifestly unreasonable will lie to a considerable extent in the eye of the beholder. On the basis of the arguments in this case:

To the extent a particular tariff measure constitutes State aid as a matter of EU law, and the requirements of Article 10 of the Northern Ireland Protocol are otherwise satisfied, I am not persuaded that these fall outside Article 10. There is nothing in the language of the Northern Ireland Protocol which would narrow the scope of the EU State aid doctrine in this way. To the extent the Northern Ireland Protocol says anything on the subject, it recognises in Article 5(6) that certain UK activity in the tariff and customs sphere can engage Article 10.

While I am not persuaded that the Northern Ireland Protocol is ambiguous in this respect, the consequences of this interpretation are nowhere near as dramatic as Mr Robertson QC has suggested. There has (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. British Sugar’s argument, therefore, is not one which, if it succeeded, would preclude the UK from operating an independent trade policy at all. On the basis of my conclusion as to the status, for EU State aid purposes, of an ATQ offered on a “first come, first served” basis, the scope for UK tariff measures to engage EU State aid rules will be narrower still.

Finally, the scope for decisions by the UK in the tariff sphere to engage Article 10 of the Northern Ireland Protocol may be further narrowed depending on the resolution of the Effect on Trade Issue, to which I now turn.

As I noted at [60] above, for the purposes of applying the EU State aid rules in the context of the Northern Ireland Protocol, it is the effect which the measure in question has, or is liable to have, on trade between Northern Ireland and the EU which is relevant.

Concern that Article 10 might give rise to a significant constraint on the independent trade or industrial policy of the UK was an important issue in the negotiations which culminated in the Northern Ireland Protocol. This issue was the subject of two unilateral declarations by the EU and the UK respectively: the Unilateral Declarations by the European Union and the United Kingdom of Great Britain and Northern Ireland in the Withdrawal Agreement Joint Committee on Article 10(1) of the Protocol of 17 December 2020. These involved:

A short but potentially important declaration by the EU (the EU Unilateral Declaration) stating:

A declaration by the UK taking note of the EU Unilateral Declaration.

It is common ground that these declarations are admissible in the interpretation of Article 10(1) of the Northern Ireland Protocol, under Article 31(2)(b) of the VCLT.

In addition, I am satisfied that the provisions of the TCA are also materials which fall within Article 31(2)(b).

British Sugar also sought to rely on the terms of the European Commission Notice to Stakeholders, Withdrawal of the United Kingdom and EU Rules in the Field of State Aid of 18 January 2021 (the Withdrawal Notice) which:

provides that “for the purposes of applying Article 10 of the IE/NI Protocol, the notion of `effect on trade’ in that provision has to be read in light of the same notion in Article 107(1) of the Treaty on the Functioning of the European Union”;

observes of the EU Unilateral Direction:

states that “the following measures would likely be considered to affect trade between Northern Ireland and the Union ….. Aid to a manufacturer in difficulty if its goods are available for sale in Northern Ireland”.

I am not persuaded that the Withdrawal Notice is admissible when interpreting the Northern Ireland Protocol, and Ms Demetriou QC did not seriously attempt to argue otherwise. The Withdrawal Notice was not a statement by a party to the Northern Ireland Protocol at all, still less one acknowledged by the other party. I would note that the (equally inadmissible) Command Paper presented to Parliament by the Chancellor of the Duchy of Lancaster (The Northern Ireland Protocol, December 2020 CP356) expresses a very different view – that it is not the case that “state aid rules will apply to Northern Ireland as they do today”, and stressing the requirement for a “genuine and direct link to Northern Ireland” before EU State aid law is engaged. It is not unusual for different views to be held as to the legal effect of a treaty, whose ultimate effect depends on the interpretation of the instrument according to well-established principles.

British Sugar argues that the terms of the EU’s Unilateral Declaration simply restate EU law, without in any way modifying it. Certainly, there are a number of features of the EU Unilateral Declaration which echo established features of EU State aid law as set out in Commission Notice, On the notion of State aid as referred to in Article 107(1) of the Treaty on the Functioning of the European Union (2016/C 262/01) (the State Aid Notice):

I accept that merely hypothetical effects on trade are not sufficient under EU law generally (the State Aid Notice, [192] and [195]).

While there are references in the State Aid Notice to circumstances where an effect on trade is presumed ([190]), the State Aid Notice provides at [195] that the required effect on trade is not to be presumed.

The State Aid Notice at [195] refers to the need for a “foreseeable” effect, without using the further qualifier “real”, but I am not persuaded that the word “real” necessarily adds to the requirement that a hypothetical effect is not sufficient.

However, there appears to be no equivalent under EU law to the reference in the EU Unilateral Declaration for the requirement that the effect on trade has a “genuine and direct link” to Northern Ireland, and the State Aid Notice at [191] suggests that certain measures can have “an effect on trade between Member States even if the recipient is not directly involved in cross-border trade”.

The fact that the Commission will have “due regard to Northern Ireland’s integral place in the United Kingdom” does not, of itself, assist the Secretary of State in this application, when determining the threshold question of whether the Commission’s role has been engaged at all (whatever effect it may or may not have on the activities of the Commission at that point).

I see no reason why the EU Unilateral Declaration should not be given purpose and effect when interpreting and applying Article 10(1), and in doing so, I am satisfied that it is necessary to have regard to the EU Unilateral Declaration as a whole. For these reasons, I am persuaded that the Effect on Trade Issue is not to be approached solely by reference to the general body of EU law. The EU Unilateral Declaration was a matter of obvious significance to the signatories to the Northern Ireland Protocol and was produced in a context in which words were important, and every word was inevitably going to be subject to the closest scrutiny.

This conclusion provides a further answer to the arguments raised by Mr Robertson QC on the Application of Article 10 to Tariffs Issue (see Section D3 above). It also ensures that the two distinct legal regimes in the Northern Ireland Protocol and the TCA (referred to at Section C1 above) are not wholly subsumed within each other, such that (for example) issues raised by Article 366 of the TCA as to whether a subsidy is justified would inevitably have to be referred to the Commission under Article 107(3), that it would be the Court of Justice rather than an arbitration panel which would resolve disputes, and EU law rather than the “relevant interpretations in reports of WTO panels and of the Appellate Body” which would fall to be applied. To this extent, the Secretary of State’s argument that the EU Unilateral Declaration provides a more exacting test as to whether there has been a relevant Effect on Trade better coheres with the treaty regime between the EU and the UK as a whole.

This conclusion also reduces the scope for what, on British Sugar’s case, might be thought to be the unattractive and improbable result that an indirect and very limited effect on trade in Northern Ireland required the UK to reverse the ATQ in its entirety and to recover the tariff “foregone” on the entire quantity of raw cane sugar imported under it, whether referable to the alleged effects on Northern Ireland-EU trade or not.

In any event, whatever the precise legal consequence of the EU Unilateral Declaration, it is an important reminder of the need to consider any suggestion that a UK measure falls foul of Article 10(1) because of its effects on Northern Ireland-EU trade with some care, and to test generalised assertions to that effect.

The Secretary of State asks the Court to go further, and to hold that the Effect on Trade test in the EU Unilateral Declaration can only be satisfied when the Effect on Trade is either a first order effect of the impugned measure, or where “the likely effect of the measure is to ‘channel its secondary effects towards identifiable undertakings or groups of undertakings’ in Northern Ireland’”. This latter formulation picks up language from the State Aid Notice at [115]-[116], addressing a very different issue – the position of “Indirect Advantage” when a benefit from State resources provided to Undertaking A is said to amount to (indirect) State aid to other undertakings, for example those “operating at subsequent levels of activity”. I am not persuaded that it would be appropriate to lift a test formulated in a different legal context and apply it under Article 10(1). In any event, in circumstances in which the parties to the Northern Ireland Protocol have not gone further than the language set out in the EU Unilateral Declaration, and in which the issue may have to be considered in a wide range of factual scenarios, I do not think it would be helpful or appropriate for me to seek to formulate an alternative test. In particular, I can see that there may well be cases in which the directness of any effect is less important than its scale. Instead, I shall simply apply the requirements of Article 10(1), with the benefit of the EU Unilateral Declaration, to the particular facts of this case.

The effect of the evidence as to the refined sugar market in Northern Ireland is as follows:

There is no trade in raw cane sugar between Northern Ireland and the EU.

No refined sugar (whether from beet or cane) is produced in Northern Ireland.

The ATQ will not affect the price of refined sugar, whether in Great Britain or Northern Ireland, which will in practice be set by the price of imported EU refined sugar, given that the UK will remain a sugar deficit country.

EU Rules of Origin mean that it will be uneconomic to seek to export refined sugar produced from raw cane sugar from Northern Ireland to the EU.

Sugar refined outside Northern Ireland is sold and consumed there (it is, no doubt, one of the cheering staples of Fermanagh and Tyrone). Dr Carr estimated that in 2019/2020 some 46,000 mt of refined sugar entered the Northern Irish market, 32,000 mt from British Sugar, and the remainder from T&L or from the EU.

Mr Mason of T&L said that in that period, 4,080 mt of T&L refined sugar had been sold into Northern Ireland, of which 665 mt entered Northern Ireland as a result of indirect sales made through retailers in Great Britain who had themselves purchased from T&L.

T&L expects sales of its refined sugar into Northern Ireland to drop as a result of the arrangements put in place by the Withdrawal Agreement, because of non-tariff barriers to entry in the form of additional regulatory requirements and the risk of reclaims of EU customs duty (or for businesses other than retailers or wholesalers, the need to pay that duty upfront and recover it).

This evidence does not establish the “genuine and direct” effect on trade referred to in the EU Unilateral Declaration. It is indirect, involves at best very small volumes and is premised on the unproven assertion that the ATQ is liable to lead to the displacement of EU refined sugar from Northern Ireland notwithstanding the non-tariff barriers to which Mr Mason has referred. To allow British Sugar to impugn the ATQ and require T&L to pay the equivalent of £28/100kg on the entirety of the ATQ raw cane it imported on this evidence would be to permit a barely discernible tail to wag a very large dog. I am satisfied that the EU Unilateral Declaration was intended to prevent precisely this kind of argument.

British Sugar’s second ground of challenge is brought under the Subsidy Control Provisions of the TCA. Article 366 of the TCA prohibits subsidies which conflict with the principles set out in that Article. If the ATQ does constitute a subsidy for the purposes of Article 366, the Secretary of State did not seek to advance an argument before me on the issue of whether the ATQ would fall foul of the Article 366 principles. Rather, the Secretary of State contends that the ATQ is not a subsidy for the purposes of the TCA in the first place.

The definition of a “subsidy” is to be found in Article 363(1)(b) of the TCA which provides:

It will be apparent that a number of the concepts engaged by Article 363(1)(b) are similar to those which arise under EU State aid law, although the immediate source of these provisions of the TCA appears to be the WTO Agreement on Subsidies and Countervailing Measures, 1994 (the WTO Agreement). Article 1.1 of the WTO Agreement provides:

government revenue that is otherwise due is foregone or not collected (e.g. fiscal incentives such as tax credits).”

However Article 363(2) includes a provision which has no equivalent in the WTO Agreement, and which bears a significant resemblance to the three-stage World Duty Free test of EU State aid law:

However, there is WTO jurisprudence which adopts a broadly similar approach when determining whether a particular tax treatment constitutes a subsidy. The Appellate Board in Brazil: Certain measures concerning taxation and charges AB-2017-7, AB-2017-8 (13 December 2018) at [5.195] suggested that the issue of whether a tax measure involved a foregoing of revenue otherwise due was to be approached as follows:

The TCA does not include an equivalent of Article 2.1 of the WTO Agreement. That provides:

The principal areas of dispute in this part of the case are as follows:

whether the ATQ involved “the foregoing of revenue that is otherwise due”;

whether the ATQ was specific, benefiting T&L over other actual and potential sugar refiners; and

(as between British Sugar and T&L) whether the ATQ has, or could have, an effect on trade or investment between the EU and the UK.

Issues (i) and (ii) are very closely related, and it is convenient to consider them together.

British Sugar says that:

the Government has foregone revenue from T&L which was otherwise due in the form of the tariffs which would have been payable if there had been no ATQ;

because that revenue was foregone, T&L was treated “more advantageously than others in a comparable position within the normal taxation regime, namely the other UK sugar refiner (which is also a potential cane sugar refiner)” (i.e. than British Sugar); and that

this differential treatment was not justified by a principal inherent in the design of the tariff system.

I accept that, in appropriate circumstances, tariff exemptions can constitute government revenues foregone, and there are decisions of panels or appeal boards operating within the WTO framework which have so held. For example, in United States - Tax Treatment for “Foreign Sales Corporations” (US – FSC) AB-1999-9 (24 February 2000), the Appellate Body in that case endorsed the submission of various parties that tariff exemptions can constitute government revenues foregone. At [90], the Appellate Body held:

However, it does not follow that every departure in a tariff regime from the standard (or “bound”) rate amounts to a subsidy. In Canada – Certain Measures Affecting the Automotive Industry Report of the Panel WT/DS-139/R, WT/DS-142/R (11 February 2000), Canada had submitted that there could only be a subsidy when the revenue “foregone” exceeded the amount which would otherwise have been payable. The Panel rejected that argument at [10.161] to [10.162]:

It is to be noted that the Panel recognised that general preferences did not engage Article 1.1(b) of the WTO Agreement, in contrast to the position where some importers of the relevant product benefited from the preference and other “importers of the same product from the same country” did not. No doubt for that reason, ATQs (under which no or a lower tariff is payable on a certain quantity of imports, and duty or a higher level of duty thereafter) are an established feature of the WTO framework and do not, of themselves, engage Article 1(1)(b). In European Communities: Bananas AB-2008-8; AB-2008-9 (26 November 2008), [335], the Appellate Body observed that “[i]n contrast to quantitative restrictions, tariff quotas do not fall under the prohibition in Article XI:1 and are in principle lawful under the GATT 1994, provided that quota tariff rates are applied consistently with Article I.”

Applying this guidance to the facts of this case:

I do not accept that the tariff of £28/100kg constitutes the “standard” or bound tariff. I have outlined the complex tariff regime so far as imports of raw cane sugar are concerned at Section B2 above. I note that in Brazil – Certain Measures Concerning Taxation and Charges AB-2017-7 and AB-2017-8 (13 December 2018), [5.162], the Appellate Body observed that “‘a domestic tax system may be so replete with exceptions’ that the rate applicable to the general category of income would no longer represent a general rule but, rather, an exception. In seeking to identify a general rule and an exception, ‘a panel might artificially create a rule and an exception where no such distinction exists.’” In my view, that is the effect of the distinction which British Sugar seeks to draw in this case between the standard rate for importing raw cane sugar for refining purposes of £28/100kg and the zero rate which applies to the first 260,000 mt imported for that purpose (whoever the importer is).

In any event, the tariff-free status of the ATQ did not involve treating T&L “more favourably” than British Sugar. Not only was it open to British Sugar to import raw cane sugar under the ATQ if it wished to do so, but no duty applied to imports of the equivalent raw material required by British Sugar.

This was not a case in which the ATQ was offered on a basis that only one importer of that product could benefit from it. T&L was treated in the same way as anyone else seeking to import raw cane sugar for refining purposes.

For these reasons, I am satisfied that the ATQ is not “specific” for the purposes of Article 363. Had I reached a different conclusion, I would not have accepted the Secretary of State’s argument this was justified by a principle inherent to the design of the tariff system, for the reasons set out in Section D2(4) above.

For the purpose of this part of its argument, British Sugar relied on the decision of the WTO Appellate Body in United States – Countervailing Measures on Certain Hot-Rolled Carbon Steel Flat Products from India WT/DS436/AB/R, AB-2014-7 (8 December 2014). In that case, a public body (NMDC) was selling iron ore at below market prices to Indian manufacturers of processed iron products. This was a subsidy within Article 1.1(a)(1)(iii) of the WTO Agreement: “a government provides goods or services”. India contended that this could not meet the requirements of specificity under Article 2.1 of the WTO Agreement, because the subsidy was available to and used by all users of iron-ore in India, with the result that there was no comparable entity who was being treated differently from the purchasers of iron ore from NMDC. That argument was rejected by the Panel, whose decision was upheld by the Appeal Board. The Appeal Board found as follows:

A measure is capable of being de facto specific for the purposes of Article 2.1(c) without the need to show that only a limited number of those eligible enterprises use the subsidy programme ([4.380]).

It is not necessary, for this purpose, to establish that there is discrimination in favour of “certain enterprises” against a broader category of similarly situated enterprises ([4.390]).

The provision of subsidised goods can be specific even if “the inherent characteristics of the subsidized good limit the possible use of the subsidy to a certain industry” and access to the subsidy “is not further limited to a subset of this industry” ([4.398]).

However, I am not persuaded that this decision assists in this case:

First, the context of the decision (government supplies of goods at below market prices) is a very different one from the tariff (essentially taxation) measure in issue here. The existence of a subsidy for the purposes of Article 1.1 was established by the provision of goods by a public body at a subsidised rate. In this case, by contrast, the question is whether there has been a subsidy because a certain quantity of certain imports has not been taxed.

Second, the Hot Rolled Carbon Steel Case was not addressing the particular issues which arise when the subsidy is said to have taken the form of “foregoing revenue which is otherwise due”. That raises particular complications, in order to avoid the wholly artificial outcome that any failure by a sovereign body to tax an activity which is susceptible to being taxed, or to tax it at a higher rate, constitutes a subsidy. The need to identify the ordinary or normal benchmark which represents the revenue foregone is a particular feature of arguments relating to this form of subsidy, and it raises its own analytical challenges.

Third, there is no equivalent of Article 2.1 (still less Article 2.1(c)) of the WTO Agreement in the TCA, and by contrast, there is a provision (Article 363(2)) which stipulates how the issue of specificity is to be approached in the context of tax measures and the suggestion the tariff regime involves government revenue being foregone. The application of the passages of Hot Rolled Carbon Steel Measure decision on which British Sugar relies in the context of the taxation and tariff regime could have very dramatic and improbable consequences, making any tariff applicable to products used for a particular type of industrial activity “specific”. It is no doubt because of the particular susceptibility of differential tariff or tax regimes to arguments that they constitute State aid that EU law developed the World Duty Free test, and that the terms of the TCA dealing with such measures essentially adopt that test.

Fourth, the Hot-Rolled Carbon Steel Measure decision addressed an argument by India that a government supply of goods available to all users of those goods in India could never be specific. That argument was rejected, but the Panel and Appellate Body were very far from holding that a measure on goods which applied to all users of those goods was inherently specific. Article 2.1(c) – which was in issue in that case – was simply one of a number of indica that a measure, although not de jure specific, might be de facto specific ([4.369]).

I can deal with this issue shortly. It is clear from the material which I have outlined in Section B3 above that the Government concluded that the effect of introducing the ATQ would be to displace a significant quantity of EU refined sugar from the UK market. The particular issues relating to Northern Ireland discussed in Section D4 above do not arise in the present context. In these circumstances, I am not surprised that the Secretary of State did not seek to argue that, if the ATQ otherwise satisfied the requirements of a subsidy under Article 363(1)(b), the requirements of sub-sub-sub-section (iv) were not met.

Such an argument was advanced by T&L, on the basis that sub-sub-sub-section (iv) would only be engaged if there was a material effect on trade in raw cane sugar for refining, rather than in the trade for refined sugar. However, there is nothing in the TCA which supports this argument. A subsidy is “specific insofar as it benefits …. certain economic actors … in relation to the production of certain goods or services”. If British Sugar’s challenge had been upheld, it would have followed that T&L had been benefited over others in relation to the production of refined sugar. A material effect on trade in refined sugar would clearly have fallen within sub-sub-sub-section (iv).

For these reasons, and in spite of Ms Demetriou QC’s formidable submissions (formidable in their content, but also, it must be said, formidable in their consequences), British Sugar’s challenges to the ATQ fail.

I will invite submissions from the parties as to what consequential orders should be made.