At the end of July, the Competition Appeal Tribunal (CAT) handed down judgment in The Durham Company v Durham CC,1 its first (and so far only) judicial review application under the Subsidy Control Act 2022 (SCA22).
The question decided: was there a subsidy decision?
The only issue on the application was whether there was a subsidy decision at all: Durham CC (the Council) accepted2 that if there had been a subsidy decision,3 it had not even attempted to comply with its duties under section 12 SCA22 (the duty to consider application of the subsidy control principles) so any such decision would have to be quashed. Although there does not seem to have been any argument on the point, the judgment appears to settle that in a case where there is a dispute over whether a measure is a subsidy falling under SCA22, the CAT is the right forum for resolving that dispute (the answer “no” effectively meaning that it has no jurisdiction). Since the position in the Durham case (that the authority will have proceeded on the basis that there was not a subsidy and therefore have failed to address the section 12 duty at all) is likely to be typical of cases where the authority denies that there is subsidy decision, in most such cases a “yes” answer to the “is it a subsidy decision?” question will result in the decision being quashed without further ado.
Background
Before turning to the issues, a quick summary of the facts. The Durham Company (the Company) provides waste collection services to businesses in parts of England and Scotland. In doing so, it competes with local authority waste collection services. Local authorities that are waste collection authorities (WCAs), as the Council is, are required to collect commercial waste in their area from any business that requests it, for which it must recover a reasonable charge. WCAs are also required to collect domestic waste, for which they are not allowed to charge.
Over the past few years, the Company has brought a number of legal actions in different areas of law to deal with what it sees as the disadvantages that it faces in competing with WCAs in collecting commercial waste: In the area of VAT, it tried to judicially review HM Revenue and Customs’ position that charges made by WCAs are not subject to VAT, and while the UK was still in the transition period it tried to claim damages for breach of the State aid rules.
In the present case, the core of the Company’s claim was that the Council was allowing its domestic waste collection activity, which inevitably required an infrastructure of equipment, staff, and property, to subsidise its commercial waste activities, resulting in the Council’s commercial waste collection being charged at subsidised prices. The difficulty it faced was how to translate that claim into an argument that there was a subsidy decision under the SCA22.
Subsidy scheme?
The CAT’s approach
First, however, the Company had to overcome an argument by the Council that the decision complained about was under a subsidy scheme made before SCA22 came into force.
The Company claimed that the relevant subsidy decision was taken on 31 March 2023, when the Council set its charges for commercial waste collection for the next financial year. However, the Council relied on the fact that in 2020 it had reviewed and decided how commercial waste collection should be charged for, so that (if there were a subsidy) the 2020 decision would have been a “subsidy scheme”. As a subsidy scheme made before the coming into force of SCA22 on 4 January 2023, any decision made under it (such as, according to the Council, the charging decision on 31 March 2023) would be excluded from SCA22 by section 48(1)(a).
In finding for the Company on that point, the CAT held that in order to count as a “scheme” a measure had to have an element of fettering subsequent decisions. Since the 2020 decision merely charted a course for future years, and left open the possibility that the Council might take a different approach, it did not count as a scheme: The CAT emphasised that “there is nothing in the 2020 Scheme Decision to suggest that the Council was in any way binding itself to follow the 2020 Scheme Decision in future years” (original emphasis).
Comment
The approach taken by the CAT contradicts the approach taken by the Department of Business and Trade (DBT) in its statutory guidance, where it distinguishes between a subsidy and a subsidy scheme as follows (emphasis added):
- 2.28 A subsidy is a binding commitment with a specific beneficiary. … A subsidy may include contingencies (e.g., a payment being made on a certain date if a target is met) but these contingencies should leave no discretion to the public authority if the relevant conditions are met.
- 2.29. A scheme, on the other hand, is usually not a binding commitment and may give a public authority a substantial degree of discretion in deciding exactly which possible subsidies under the scheme should be given. It can have any number of beneficiaries, who need have no connection to each other besides their receipt of subsidies under the scheme.
It will be seen that, for DBT, a decision that binds the authority to what it does in future years (or binds it contingently on external events) is not a subsidy scheme but a subsidy tout court: In its view, the existence of a substantial degree of discretion does not, as the CAT holds, tell you that the decision is not a subsidy scheme: rather, it tells you that it is a subsidy scheme (and not a subsidy tout court).
In my view, DBT’s approach is supported by the text and structure of the SCA22, while the CAT’s view is not – and therefore is, in my view, incorrect.
It is significant that the CAT’s approach is analogous to – though the CAT, perhaps deliberately, does not mention the analogy – the case-law on what constitutes an aid scheme under EU State aid law, where the relevant legislation7 provides that an aid scheme exists only where no further implementing measures are required for individual aid to be granted on the basis of the act constituting such a scheme.8 In Commission v Belgium and Magnetrol, the Court of Justice held that that a “further implementing measure” was one where there was “a degree of discretion on the part of the … authority adopting the measures in question, allowing it to influence the amount of the aid, its characteristics or the conditions under which that aid is granted” and contrasted that with the case of a “mere technical application of the act providing for the grant of the aid in question”. The EU test therefore, like the test propounded by the CAT, has the effect of removing from the scope of an aid/subsidy scheme any measure that leaves the authorities subject to it with any significant discretion as to whether to grant subsidy, the amount of subsidy, or the terms of the subsidy.
However, section 10(1) SCA22, which defines “subsidy scheme” as “a scheme made by a public authority providing for the giving of subsidies under the scheme”, contains no reference to the absence of “further implementing measures”. That difference is deliberate and important: it reflects the fact that the role of “aid scheme” in EU State aid law is quite different from the role of “subsidy scheme” in SCA22. The crucial difference is that, in SCA22 the concept of a subsidy scheme is also used to allow granting authorities to do what, in EU law terms, would be analogous to the creation of a block exemption. In EU law that power is strictly reserved to the Commission, as the sole decision-maker under Article 107 TFEU. Under SCA22 it is given to all public authorities to cover their own subsequent subsidy decisions, and, by section 10(2)(b), to “primary” authorities (central and devolved ministers and some others) to cover other authorities’ subsidy decisions. Examples of such schemes already made include the Research, Development and Innovation Streamlined Subsidy Scheme, the drafting of which, indeed, closely resembles the drafting of a block exemption in the EU system – and certainly does not “bind” any authority covered by it to grant subsidy of any particular amount to anyone, or to do so on any particular terms (subject to the point that, if the subsidy does not fit within the scheme, the authority will have to conduct a section 12 assessment).
Indeed, the existence of section 10(2)(b) shows that the CAT’s analysis cannot be right, since (absent some other statutory power to do so) no primary authority could ever make a scheme that bound other authorities: all that a section 10(2)(b) subsidy scheme can so, absent another statutory power, is to enable other authorities, as and when they wish, to give subsidies that fit within the subsidy scheme without having to undertake a section 12 assessment.
Why did the CAT get this so wrong (as I believe it did)? It seems to me that behind the CAT’s analysis lies an unspoken concern that, because section 48(1)(a) removes from the scope of the section 12 duty any subsidy granted under a subsidy scheme made before 4 January 2023, it is necessary to define “subsidy scheme” restrictively (particularly as a pre-2023 subsidy scheme would not have been assessed under section 12 or subject to any other requirements under SCA22).
Of course, that problem is one that will get less significant over time, and it certainly should not now dictate an interpretation of key SCA concepts that makes a nonsense of those concepts. But, in any event, there is another way out of the difficulty, which is to emphasise what in my view is a requirement of a subsidy scheme, namely that it be precise enough in what it covers to be capable of being rationally assessed under section 12 for compatibility with the Schedule 1 principles (as any subsidy scheme has to be, subject to the various general exemptions). To take an example: if a public authority announced on 2 January 2023 that it proposed to give, or permit other authorities to give, “subsidies that create jobs in disadvantaged areas”, that decision is incapable of being sensibly assessed under the Schedule 1 principles and so, in my view, could not count as a “subsidy scheme”: but if the announcement set out a sufficiently clear definition of who could be considered for subsidy, and sufficient clarity as to the purposes, size, and terms of the subsidies that could be given, to be capable of section 12 assessment, then in my view there is no reason why it should not be regarded as a subsidy scheme.
In the Durham case that would mean that it would have to be clearly established what the 2020 decision actually was (there appears from the judgment to have been some doubt about that) and that it was sufficiently precise to have enabled a section 12 assessment (had that been an issue at the time). But if those hurdles were jumped, then, in my view, the Council should have succeeded on that point.
Subsidy?
The CAT’s approach
The Council did, however, win on the concept of whether there was a subsidy at all. It is interesting that, in contrast to its approach to the question of “decision”, the CAT here was alert to the need to look at the text and structure of SCA22 and not reflexively to follow EU authority (indeed, issuing a general caution that EU authority could be no more than persuasive and that attention should be paid to differences between EU law and SCA22).
The critical question was whether, under SCA22, it was impossible for a granting authority to give a subsidy to itself: that is to say, whether, in order for there to be a subsidy, it was necessary for the recipient to be a separate legal person. The CAT concluded that the answer to those questions was “yes”. First, looking at the text of SCA22, both “public authority” and “enterprise” were defined as “persons” (sections 6(1) and 7(1)), and the essence of the definition of “subsidy” was that financial assistance moved from one person (the authority) to another (the recipient enterprise). Second, the requirement that a subsidy must be given from public resources (section 2(1)(a)) meant that there must be a subtraction of public resources when a subsidy is given – which could not happen when the subsidy moved around the same person; and similarly, the provision in section 2(5) that a subsidy is given when the recipient has an enforceable right to it made no sense when the recipient and the grantor were the same legal person.
Third, the CAT argued that SCA22 deliberately avoided the term “undertaking”, preferring the term “person” (an enterprise being a person or group of persons engaging in economic activity). It considered that that strongly indicated that SCA22 did not follow the EU “functional” approach to the question of what was an “undertaking” in State aid and competition law: and that such an approach would have had the disadvantage of requiring the CAT to engage in “fine and subjective, and (potentially at least) dangerously arbitrary” distinctions between activities of the Council qua authority and qua enterprise, noting that the Council’s pricing of commercial waste collection was inevitably informed by its position as collector of household waste.
The CAT went on to give other reasons why there was no subsidy.
- Because the Council could neither undercharge nor overcharge for commercial waste services (because of its duty to statutory duty to charge a reasonable rate and its common law duty not to charge for an improper purpose, i.e. to raise revenue), the supposed cross-subsidy (really a decision as to cost-allocation) could not confer an economic advantage on its commercial waste collection operation, since a reduction in charged cost meant a reduction in charged prices.
- It was not convinced that the activity of commercial waste collection by the Council was economic activity at all: it was not decisive that a charge was made.
Comment
The CAT’s rejection of the idea that a subsidy can be granted by one part of a public body (acting as a public authority) to another part of the same body (carrying out economic activity) is, as it recognised, a departure from EU State aid law12. There is no doubt that applying State aid law in those circumstances is difficult (though perhaps not quite as difficult as the CAT thinks), so the absence of that difficulty from SCA22 will be welcomed by government and local authority lawyers.
The CAT expressly left open13 the question of whether a grant by a public authority to a separate person that it controlled (for example, a wholly–owned subsidiary) was capable of being a subsidy, though it noted that “matters may very well be different” in such a case.14 That situation is probably much more common than the situation in Durham, not least because section 4(2) of the Localism Act 2011 requires local authorities carrying out activities for a commercial purpose to do so through a company (and there are many other similar statutory provisions). It is hard to see that the CAT’s reasoning applies to prevent there being a subsidy in such a case: it is trite that a wholly-owned company is a separate legal person from its owner, and given that any company will have to prepare separate audited accounts, determining what the flow of financial assistance is from authority to separate company is generally bound to be easier than chasing funds round different parts of the same authority. So the point decided by the CAT may well be of limited practical application. Its further points – as to the absence of economic advantage and the potential absence of economic activity at all – are also specific to the particular facts of the case.
That said, the CAT’s willingness to depart from EU precedent and to focus on the language and structure of SCA22 on its own terms as a new regime follows a similar warning (in relation to the subsidy control provisions of the Trade and Cooperation Agreement) in R (British Gas Trading) v Secretary of State for Net Zero. Despite the CAT’s failure to take that approach when considering the definition of “subsidy scheme”, that is doubtless a marker for the future.
