Briefing paper from the Committee adviser Dr Christopher McCorkindale to the Committee, 23 November 2021
Briefing paper from the committee adviser Dr Christopher McCorkindale, for the Committee's The UK Internal Market Inquiry. 23 November 2021
Even before the Subsidy Control Bill was introduced into the UK Parliament, devolution has been impacted by the issue of subsidy control in important ways. First, in the face of disagreement during common framework negotiations about whether subsidy control was a reserved or devolved matter, section 52 of the Internal Market Act 2021 explicitly reserved the ‘[r]egulation of the provision of subsidies which are or may be distortive or harmful by a public authority to persons supplying goods or services in the course of a business’ in each of the devolution statutes. Second, both the Scottish Parliament and the Welsh Parliament refused legislative consent to the Internal Market Bill (the Northern Ireland Assembly did not hold a formal legislative consent vote) which was - as with the EU Withdrawal Bill (consent withheld by the Scottish Parliament) and the EU (Withdrawal Agreement) Bill (consent withheld by the Scottish Parliament, the Welsh and the Northern Ireland Assembly) - nevertheless passed without any consequent amendment. Whilst there might be an argument that, with the article 50 clock ticking and a ‘no deal’ default position waiting at the end of the article 50 period, the EU Withdrawal Bill constituted a justified exception to the rule that ‘the UK will not normally legislate with regard to devolved matters [nor to adjust the boundaries of devolved competence] without the consent’ of the devolved legislatures, the case is far less clear in so far as the EU Withdrawal Bill and the Internal Market Bill were concerned. Not only was devolved competence adjusted (with regard to subsidy control), but this adjustment was done by imposition rather than with the consent of the devolved legislatures. There are three reasons why this is of constitutional relevance to the Subsidy Control Bill:
Schedule 3 of the Subsidy Control Bill makes explicit that the principles (Chapter 1 of Part 2, via paragraph 6), prohibitions and other requirements (Chapter 2 of Part 2, via paragraph 7), and exemptions (Chapter 3, via paragraph7) set out in the act apply to devolved primary legislation and that (with some exceptions) challenges to devolved primary legislation on those bases may be raised in the Court of Session (or, respectively, in the High Court in England and Wales, or the High Court in Northern Ireland). It has been a feature of devolution that, unlike Acts of (the UK) Parliament, devolved primary legislation is subject to judicial review and may be struck down by the courts where it exceeds the legislative competence of the Scottish Parliament (Scotland Act 1998 (SA98) sections 29-30A). Challenges to the validity of Acts of the Scottish Parliament (ASP) may be raised by the UK or Scottish Law Officers in pre-enactment proceedings at the UK Supreme Court (SA98 section 33) or in post-enactment proceedings raised by any party with a “sufficient interest” in the matter at the Court of Session. However, it is a novel feature of the Subsidy Control Bill that it creates a new route to challenge, and ground(s) of challenge to, the validity of an ASP that is self-contained in the substantive legislation and not made by amendment to the Scotland Act. Given the constitutional function and status of the Scotland Act, as well as the democratic quality of ASPs, there are good constitutional reasons - not least, the intelligibility of the devolution scheme as a whole - why limits to the validity of devolved primary legislation should be contained within the Scotland Act itself.
In a widely read blog post, George Peretz QC has argued that these provisions depart from the norm in a second, more substantive, way. This, he said, is because they:
…represent a significant departure from the general position, set out in AXA General Insurance v Lord Advocate [2011] UKSC 46, that judicial review of devolved legislation is not generally possible on the general common law grounds of irrationality, unreasonableness, or arbitrariness (2)
As Peretz acknowledges, there is no explicit direction in Schedule 3 as to the grounds or intensity of review that should be applied by the Court of Session where ASPs are challenged in this way. This creates an area of uncertainty. On the one hand, the reason why the Supreme Court held that ASPs are not subject to the traditional common law grounds of review – why courts, in Lord Hope’s view, should ‘intervene, if at all, only in the most exceptional circumstances (3) – is the democratic quality of the Scottish Parliament: ‘firm[ly] rooted in the traditions of universal democracy’, ‘draw[ing] its strength from the electorate, and benefiting from the ‘depth and width of the experience of its elected members and the mandate that has been given to them by the electorate (4). On the other hand, there is an expectation, in Schedule 3 of the the Subsidy Control Bill, that meaningful review should be exercised (including recovery where a subsidy is found to have been unlawfully granted). It is questionable whether an application of the extremely high threshold for review set out in AXA would meet this task. Furthermore, the proposed limits on the legislative freedom of the Scottish Parliament to legislate directly to grant subsidies are themselves contained in a UK Bill that, if enacted, would enjoy the highest legal authority. In light of recent, and somewhat restrictive devolution jurisprudence from the UK Supreme Court, it is difficult to determine whether those limitations are explicit enough for the courts to apply traditional grounds of common law review to ASPs, or to apply a greater intensity of review to that which is set out in AXA. An opportunity to test this argument might arise, for example, in relation to clause 12(b) read alongside Schedule 3 (5). That is to say, whether the ‘view’ of the devolved legislature, that any subsidy granted by way of an ASP is consistent with the subsidy control principles, is a reasonably held, is exercised for a proper purpose and that takes account of all relevant (and no irrelevant) considerations.
Legislative asymmetry, of course, is a feature (indeed, perhaps the defining feature) of the UK constitution in the devolution era. It flows from the fundamental principle of parliamentary sovereignty and finds explicit expression in the devolution statutes (e.g. section 28(7), Scotland Act 1998). In the context of subsidy control the key legislative asymmetry is that between Acts of [the UK] Parliament that are not subject to judicial review under Schedule 3 and devolved primary legislation that (as we have seen) is subject to such review. The former position, that UK primary legislation is beyond the scope of judicial review is consistent with UK constitutional principle and with the treatment of UK primary legislation on the EU-UK plane by virtue of the EU-UK Trade and Co-operation Agreement (TCA). However, the latter position (that devolved primary legislation is subject to judicial review) is not necessary as a matter of constitutional principle. The justification for review seems to hang on ‘practical’/policy-orientated considerations, inter alia, to guard against anti-competitive subsidies made through primary legislation and to prevent a ‘subsidy race’ between the UK’s constituent parts. What is produced, on the other hand, is a legislative opportunity (whether taken or not) for the UK Parliament to shield anti-competitive subsidies, or subsidies that impact negatively on Scotland, Wales and Northern Ireland, from strong-from review (as opposed to the voluntary and non-binding nature of the CMA referral system that applies to Acts of [the UK] Parliament) and in a way that (whether exercised or not) advantages one constituent part of the UK over the others. Nor is this position required by the terms of the TCA, which is silent on competition within and across the UK.
There are two important constitutional considerations that arise from the Subsidy Control Bill as it cuts across executive competence.
As with legislative asymmetry, executive asymmetry is (a more limited) feature of the devolution scheme. Whereas the Kilbrandon Commission took the view that powers of intervention by UK Ministers to ‘prevent adoption of policies considered to be inconsistent with the maintenance of the essential political and economic unity of the UK,’ and regarded such a power as ‘an essential feature of a non-federal constitution’ (6), the powers of intervention conferred upon UK Ministers take a different focus. The UK Government retains three powers to veto, repeal or require action on the part of the Scottish Ministers that might impact adversely on the UK’s international obligations, defence or national security, or the operation of the law in reserved matters (7). More than two decades later these powers have never been exercised.
Under the Subsidy Control Bill there are additional powers by which the UK Government might intervene with regard to the exercise of executive power by Scottish Ministers. Clause 55 allows the Secretary of State to ‘call in’ subsidies awarded by public authorities where there is a risk that the subsidy will not meet the subsidy principles and requirements or where the subsidy is likely to negatively impact competition and/or investment within the UK. Clause 60 allows the Secretary of State to make a post-award referral of subsidies to the CMA where the same risks apply. There is no equivalent power on the part of the devolved administrations. It is unclear why, if the ‘call in’ and referral powers are designed to protect competition and/or investment within and across the UK, these powers should vest only in the Secretary of State and not also in their devolved counterparts. As these powers stand, there is no reason of constitutional principle why the Secretary of State should have a power, for example, to ‘call in’ a subsidy granted in Scotland that they believe will have a negative impact in England, but where, for example, Scottish Ministers have no right to call in or refer a similar subsidy in England that they believe will have a negative impact in Scotland.
Further asymmetry includes the power of the Secretary of State to issue guidance about the practical application of, inter alia, subsidy control principles and requirements to which a public authority ‘must have regard’ when ‘giving a subsidy or making a subsidy scheme’ (clause 79). The Bill requires the Secretary of State to ‘consult’ with ‘such persons as the Secretary of State considers appropriate’ if a decision is made to issue guidance. However, it is arguable that, given the effect of guidance on primary and secondary law-making powers in the devolved sphere, there should be a duty to issue guidance and that the devolved administrations should be drawn out from the general category of ‘persons as the Secretary of State [might] consider appropriate’ in order to create a duty of meaningful consultation - or, to go further, a duty of co-operation - with the devolved administrations before such guidance is published.
These asymmetries seem to defeat rather than reinforce the policy objective of creating a level playing field across the UK. If, as noted above, the success of a new subsidy regime depends upon buy-in from the devolved administrations, then these asymmetries present some cause for concern.
A further consideration is the extent to which the Subsidy Control Bill fits a more general post-Brexit pattern of UK primary legislation that – in the face of legislative dissent – nevertheless constrains the scope of existing and future devolved executive power. Neither EU law (as it applied pre-EU withdrawal) nor the TCA as it applies now have required subsidy control measures and enforcement to apply internally to the UK. These internal requirements are ‘gold plated’ by the subsidy control principles (F and G) in Schedule 1 and require Scottish Ministers (as well as the Scottish Parliament and public authorities), and their UK and devolved counterparts, to consider the impact of subsidies on and across the constituent parts of the UK. Furthermore, by virtue of clause 1(7), all (past and future) ASPs and Scottish Statutory Instruments (but not Acts of [the UK] Parliament, where a contrary intention is expressed in the relevant statute] are to be read as being subject to the subsidy control requirements contained in the Bill. The Bill therefore cuts across devolved competence in significant ways. Added to the market access and spending provisions contained in the Internal Market Act 2021, as well as their as yet unknown interplay with common frameworks and with the keeping pace powers conferred on Scottish Ministers by section 1(1) of the UK Withdrawal from the European Union (Continuity) Act 2021, there is likely to be some period of uncertainty (reinforcing the need for clear and workable guidance, made in consultation or co-operation with the devolved administrations) in delineating the boundary between devolved and reserved matters or the extent to which devolved matters are subject to reserved control and intervention. In this space of uncertainty there is scope for significant political and constitutional disagreement as well as (potentially contentious) recourse to the CMA referral and judicial review mechanisms referenced above.
Chris McCorkindale
References:
1. Thomas Pope, ‘The subsidy control bill does not guarantee post-Brexit state aid success’ (01 July 2020) Institute for Government, available at https://www.instituteforgovernment.org.uk/blog/subsidy-control-bill.
2. George Peretz QC, ‘The Subsidy Control Bill and devolution: a balanced regime?’ (15 July 2021) EU Relations Law, available at https://eurelationslaw.com/blog/the-subsidy-control-bill-and-devolution-a-balanced-regime.
3. AXA General Insurance v Lord Advocate [2011] UKSC 46 at para 49 per Lord Hope.
4. Ibid.
5. The UK Withdrawal from the European Union (Legal Continuity) (Scotland) Bill – A Referance by the Attorney General and the Advocate General for Scotland [2018] UKSC 64; Reference by the Attorney General and the Advocate General for Scotland – United Nations Convention on the Rights of the Child (Incorporation) (Scotland Bill) [2021] UKSC 42. For commentary see Mark Elliott, ‘Devolution in the Supreme Court: Legislative supremacy, Parliament’s ‘unqualified’ power, and ‘modifying’ the Scotland Act’ (15 Oct 2021) Public Law for Everyone blog, available at https://publiclawforeveryone.com/.
6. Report of the Royal Commission on the Constitution 1969-1973 (1973) Cmnd 5460 at para 75.
7. Scotland Act 1998 s 58(1), 58(2) and 58(4).