The Scotland Bill

Written evidence submitted by Alan Trench

1. This memorandum is submitted by Alan Trench to the Scottish Affairs Committee for its inquiry into the Scotland Bill (Bill 115) currently before Parliament. I am an honorary senior research fellow at the Constitution Unit at University College London, and was formerly research fellow in the School of Law at the University of Edinburgh. I have written widely on various aspects of devolution in the United Kingdom, including intergovernmental relations and devolution finance, and also have experience of how decentralised and federal systems work in a number of other countries around the world, including particularly Australia, Canada, Germany and Switzerland. I have acted as specialist adviser on devolution to the House of Lords Constitution Committee and the Lords Select Committee on the Barnett Formula (which reported in July 2009), and am currently specialist adviser to the House of Commons Welsh Affairs Committee. I am also author of a blog ‘Devolution Matters’, on which many posts deal with issues relating to implementation of the Calman Commission’s recommendations. [1]

2. Overall, I consider the proposals set out in the bill and the accompanying Command paper Strengthening Scotland’s Future, Cm 7973, to be a very considerable improvement on those set out by the previous UK Government in Scotland’s Future in the United Kingdom, Cm 7738. In particular, the proposal to use real tax revenues not Treasury estimates or forecasts (after a necessary transition phase) is wholly to be welcomed. However, a number of the bill’s proposals present serious technical and financial problems, which mean that the most they can be given is a cautious welcome. In particular, that welcome is dependent on seeing them as steps toward the formulation of a more effective system, rather than an end point in themselves; in my view, the bill will not succeed in creating a stable or durable basis for Scottish devolution as it stands.

3. The problems I would identify with the proposals set out in the Scotland bill relate principally to its institutional arrangements; the extent to which it succeeds in creating a realm in which there is meaningful devolved fiscal autonomy; and the failure to address issues relating to the block grant. I also have concerns about the proposals relating to borrowing powers, and about some of its constitutional provisions. These relate both to issues raised in the Calman Commission’s report but which do not appear in the bill or Command paper, and ones which were not part of the Commission’s recommendations but which are proposed in the bill.

Institutional arrangements

4. As regards institutions, the system proposed in the Command paper uses the Office of Budgetary Responsibility to carry out estimates of Scottish tax revenues and economic performance, with income tax revenues to be collected and administered by HM Revenue & Customs (HMRC), and an ‘Intergovernmental Bilateral Committee on Fiscal Devolution’ to be created as a forum to discuss shared UK-Scottish interests in fiscal and economic policy. In addition, it is proposed to use the National Audit Office to audit the amounts of tax revenue payable to the Scottish Consolidated Fund.

5. Each of these measures presents problems. The Command paper emphasises that the Office of Budgetary Responsibility is independent, and by implication able to play such an important role impartially. Its independence has of course been widely questioned, and even if it is independent in relation to HM Treasury it cannot be said to be impartial in relation to the Scottish Government as it is part of the UK Government. There is no direct involvement for the Scottish Government in relation to tax collection – it will not have, for example, the right to appoint a member of the Board of HMRC. It is offered, rather, a ‘clear line of sight’ (the meaning of which is not clear) to HMRC, supported at a political level by the Intergovernmental Bilateral Committee. The role of the National Audit Office (NAO) is welcome, but it is not for the UK Government to confer functions on it as it is accountable to the House of Commons, and strictly speaking its staff are officers of Parliament not civil servants. Moreover, it may be appropriate for Audit Scotland to be involved in the task of auditing Scottish tax receipts as well as NAO.

6. A better approach would be to review the role of HMRC. Its role should clearly be to serve both UK and Scottish Governments, if it is to collect taxes payable to both bodies. That should include appointment by the Scottish Government of at least one member of its board. Moreover, there should be accountability to the Scottish Parliament for both HMRC and the Department of Work & Pensions (DWP), to the extent that DWP has a role in tax collection or administration in Scotland. (The working of tax credits means that such a role is substantial.) That accountability should have a statutory basis, and the bill should therefore be amended to create a basis for it.

7. The Command paper avoids setting out the way that issues of UK-level tax policy affecting devolved taxes will be handled. These could be significant; changes in allowances and reliefs, in the rates of tax and in the tax banding structure, will affect the overall revenue from the devolved tax powers. When the UK Government makes such changes, there will be an obvious issue of working out how to compensate Scotland for the revenue foregone as a result, in accordance with the ‘no detriment’ principle stated in the Command paper. But, in addition to that, there needs to be some mechanism for the Scottish Government to be involved actively and at an early stage in such decisions.

8. These need to go beyond the existing arrangements for intergovernmental relations, whose shortcomings are well known. [2] In this respect little has changed since the present UK Government came to office, and in some respects this has got worse – not least because of the abolition of the post of Director-General for Devolution in the reformed Cabinet Office structure. The importance of such issues means they cannot be left to the rather haphazard working of intergovernmental relations as that presently operates. The Command paper’s proposals create the risk that either the Scottish Government will be consulted only at a relatively late stage (through the Intergovernmental Bilateral Committee), or simply informed of them as a fait accompli (as presently often happens with changes to the block grant), which will limit the extent to which the Government becomes genuinely accountable. Achieving such early consultation requires both a fundamental difference of outlook and approach from HM Treasury and HMRC, and institutional mechanisms to ensure that this manifests itself. I am not convinced that the proposals in the bill and Command paper will be sufficient to do so.

The extent of devolved fiscal autonomy

9. I should clarify what I mean when I refer to ‘creating a realm of meaningful devolved fiscal autonomy’. In systems of finance in federal systems, it is common for taxing powers to be divided between two or more levels of government, and for some of those tax bases to be shared. This does not, in itself, mean that each government is subject to constraints imposed by the other in how it uses that power. Such power also does not depend simply on the number of taxes devolved or the limitations on the amount of revenue those taxes may generate. Rather, the question concerns the extent to which the Scottish Government and Parliament may autonomously make decisions in relation to devolved tax matters. The principle at issue here is of genuinely ensuring devolved control over tax functions that are nominally devolved. From the point of view of securing real fiscal accountability, this is essential; unless one level of government is clearly responsible for decisions taken in relation to devolved tax ‘room’, it cannot be fully accountable for those decisions. Where responsibility for a decision is shared, accountability also becomes blurred and, among other things, there is a serious risk of blame-shifting.

10. The system proposed in the Scotland Bill and Command paper fails to do this adequately. There are a large number of areas where UK Government decisions will affect devolved fiscal decisions or revenues. These include allowances, reliefs and tax rates and bands, as noted above. This is not an absolute constraint, and there are sound practical reasons to ensure a degree of UK-wide commonality on such matters, particularly if there is to be a single agency responsible for the collection and administration of all taxes. However, such arrangements should be resolved by negotiation and agreement, in which HMRC would act impartially as the agent of both UK and Scottish Governments, rather than unilateral determination by one tier of government.

11. The constraint on progressivity is another significant factor here, and in my view the approach taken by the Holtham Commission here (to allow the devolved administration to choose how many percentage points of tax to charge on each UK tax band, subject to a limit of a three point variation from the UK level) is preferable. [3]

12. Another is the question of the limited (and pro-cyclical) nature of the tax base that will be available to the Parliament. This is appreciably narrower than that recommended by the Calman Commission, as a share of income tax on savings and dividends has been omitted, for no very good reason, as well as the aggregates levy and air passenger duty. As important as the revenue that a particular element of the tax base might generate is the interest it creates in increasing the size of that base (whether by fiscal or other policy levers). It would be desirable to increase the size of that base, in two respects. One would be the restoration of the share of savings and investment income. The other would be at least partial devolution of excise duties on alcohol, tobacco and gambling. (By ‘partial devolution’, I mean a splitting of the tax base so that both UK and Scottish tiers levy charges on this, as has been done for income tax.) The ability to set differing amounts of such ‘sin taxes’ would enable the Parliament to decide how to address serious social problems, as well as raise revenues from them. (Debates about minimum pricing for alcohol in Scotland illustrate the point well.) While the Calman Commission considered and rejected the devolution of these taxes, its reasoning on this point was based in concern about tax competition and spill-over that can be managed in other ways, especially if such taxes are partially not wholly devolved.

13. A further option for further fiscal devolution would be capital gains tax in relation to land and real property (not shares or other intangibles). This would be consistent with the principle that taxes relating to land should be devolved.

14. In the longer term, a much further-reaching redesign of the UK’s tax system may be called for, clarifying the relationship between UK-level taxes and functions, and devolved ones. This is an area on which I hope to be able to carry out further work in due course.

15. The system proposed in the bill would create a system substantially less under devolved control than the taxing powers of provinces in Canada, cantons in Switzerland or states in the USA. In such countries, the lower level of government raises 60-70 per cent of its total spending from its own tax base, rather than the 30-35 per cent contemplated here. It would create greater autonomy than enjoyed by German Länder or the regions and communities in Belgium (where reform of federal finance to create a measure of fiscal autonomy has been central to the inability to establish a federal government since the June 2010 elections). In Australia, the states raise about 55 per cent of their spending, but given how the system of equalisation works there it is debatable whether what is proposed in this bill is greater than the effective fiscal autonomy of Australian states.

Mechanisms for grant reduction, and the block grant more generally

16. It is a matter for considerable regret that the Command paper leaves the question of how the reduction from the block grant to allow for the new tax powers to be resolved later on, despite the importance of this issue to the working of the proposed arrangements. In my view the failure to set out a proposed mechanism for this is a grave failing on the part of the advocates of the bill and Command paper, as its effect is simply unknowable at this stage. This is hugely disappointing, given that the UK Government has had 18 months to decide how to respond to the Calman Commission’s recommendations. Moreover, the reason given for this (the uncertain state of the public finances at the present time) is unconvincing – future public finances are always unpredictable. For my part, I can do little more than point to the careful and thoughtful discussion of this issue in the report of the Holtham Commission in Wales; the same principles apply to Scotland as to Wales. [4]

17. Equally regrettable is the refusal of the UK Government to take on wider issues of reviewing the working of the block grant. The problems of the present arrangements are very considerable, and increasing. Quite apart from issues of the extent to which the present arrangement delivers UK-wide equity, a block grant which depends on allocating to the devolved governments proportional shares of changes in spending in England on ‘comparable functions’ implicitly ties devolved public services to those in England. It is consequently hard for the devolved parts of the UK to strike out in substantially different directions in public policy.

18. Funding of higher education is a good example of this – the adoption of deferred variable fees in England increased the resources available to universities in England and Wales, without triggering increased funding for Scotland through the block grant. Scotland’s choices have been to fund universities out of its block grant (and cut other areas of spending), allow higher education funding to decline (and so let its universities fall behind England), or to follow the English lead. It has chosen the first of those, but found it hard to sustain the level of funding necessary. [5] That problem will only become worse with the changes recently proposed by the UK Government following the Browne Review and Spending Review, involving a severe cut in the teaching grant to universities in England (which will be passed through the Barnett arrangements) as well as increased funding for universities through higher tuition fees (which will not).

19. Creating a block grant that makes Scottish public services less dependent on those provided in England is far from straightforward, but it would be possible to improve the present arrangements, and to do so at little or no cost to the UK Exchequer. The Lords Select Committee on the Barnett Formula (which I advised) made a number of suggestions about institutional structures to redress this issue, which may be of use. [6]

Borrowing powers

20. Two approaches to the question of borrowing powers are possible: direct borrowing by the Scottish Government on the bond markets, and borrowing regulated through HM Treasury subject to prior conditions and limits. Direct borrowing in the bond market has many attractions, but also raises the very serious issue of whether there would be an implicit UK Government guarantee of such borrowing so that Scottish creditors would expect the UK Government to cover Scottish debts in the event of default. A significant recent book highlights the political and economic dangers of such a bail-out, and the difficulties of controlling borrowing by state-level governments if there is an implicit guarantee from the federal/ central state. [7] Given the constraints on devolved fiscal autonomy noted above and the heavy reliance of the Scottish Government on the UK Government’s block grant, there are strong reasons to ensure that the UK Government is able to control the overall amount of devolved borrowing and so its own liability.

21. That said, the levels set out in the bill and Command paper are low. In particular, the £500 million borrowing limit for cumulative current spending is very tight. It amounts to only 2 ½ per cent of Scottish Government spending in 2008-09, and is subject to further limits when used to cover mismatches between estimated and actual revenues within a year. It is hard to see how this will be adequate to cover actual losses if tax revenues should suddenly decline. Unless these limits were changed, the Scottish Government would be compelled to cut spending on public services quickly in order to remain within these statutory limits, if there should be downturn in the wider economy (which itself would be due to factors beyond the Scottish Government’s control), or for some reason in the collection of tax revenues.

22. The capital borrowing limit is also low, amounting to the cost of only one or two large-scale infrastructure projects. Such a limit should put the Scottish Government in a position to make a choice between the various ways of funding such projects – whether from current revenues, PFI/PPP type structures, or own borrowing. A low limit for such borrowing will narrow the choice available to the Scottish Government to fund such schemes, even when its own revenues are sufficient to enable it to cover the costs of such borrowing.

23. In relation to borrowing, HM Treasury is given extensive powers. Not only will borrowing have to go through the Treasury, but it will have active powers to control and limit such borrowing. This is consistent with a view of the Scottish Government as a subordinate one, not one entitled to make its own decisions in accordance with its own mandate and accountability to the Scottish Parliament. In this respect, the proposed arrangements contradict the Calman Commission’s principle of adding financial accountability to existing devolved political accountability. This provides a strong constitutional basis for such borrowing, even if carried out through HM Treasury, to be subject to only such legal constraints as are absolutely necessary for carrying out the reserved function of UK macroeconomic management. There need to be adequate institutional arrangements – including for scrutiny of HM Treasury’s actions by the Scottish Parliament – to ensure that is the case. However, there is no provision for this in the Command paper.

Transitional issues

24. The Command paper indicates an extended and phased transitional period, extending until 2016. Given the scale of adjustment necessary on the administrative level, this appears to be an appropriate period to ensure that the new arrangements will indeed be capable of functioning when introduced. The Committee will not reminding about the problems at HMRC revealed by recent disclosures concerning operation of the Scottish variable rate, or its new computer system. What is important is that there is no slippage in the timetable for implementation, so that the new powers are available as planned and particularly from 2016. The fact that the timetable is now published will help with that; the Parliament might also wish to ensure that it is able to take evidence periodically from HMRC to ensure that this is in fact complied with.

Constitutional issues

25. Three of the bill’s constitutional changes seem particularly worthy of note. All three of these are matters that have been added in the Command paper, and do not result from recommendations of the Calman Commission. First, there is the power to enable partial suspension of Acts of the Scottish Parliament pending their consideration by the UK Supreme Court under section 33 of the Scotland Act 1998, rather than requiring the whole Act to be suspended pending determination by the Court (clause 7 of the bill). There is no parallel to this in the devolution legislation for Wales (passed only in 2006, of course) or Northern Ireland. Second, there is the power to confer legislative competence on the Parliament (by ‘un-reserving’ matters in Schedule 4 or 5) for a temporary period only (clause 10). Third, there the provisions giving UK Ministers the power to act to implement international obligations, even when these relate to a reserved matter (clause 23). The explanation offered in the Command paper is essentially that such action may be convenient for practical reasons – but that is already an option, by means of simultaneous joint action by UK and Scottish Ministers. Many such obligations concern devolved matters, in a range of areas including particularly criminal law and justice.

26. In the first two cases, the objection is essentially the same. At first glance, each of these proposals appears to make the devolution arrangements more flexible, and therefore more workable in practical terms. Certainly, the flexibility of the 1998 legislation has been a feature that has contributed to its effective working and to the minimisation of ‘devolution issues’ considered by the courts. [8] The ability to apply the constitutionally-defined devolution settlement in a flexible way has been valuable. However, these measures risk undermining its constitutional status, and encouraging UK Government departments to regard devolution as something that is wholly flexible. That would result in devolution considerations receiving even less consideration at UK level than they presently do. Ignorance, incomprehension and neglect of devolution matters is a significant, and perhaps growing, problem, within the UK Government. Further, they risk an attempt by the UK Government to control in detail the activities of the Scottish Government or Parliament, implying a degree of micro-management that is inimical to the principles of devolution set out in the 1998 white paper Scotland’s Parliament (Cm 3658). These proposals would be better if they were omitted altogether. If that is not possible, clear and preferably statutory explanations of the circumstances in which they are to be used, the role of the Scottish Parliament and Government in their use, and the more general principles to underpin their use are needed.

27. The provision relating to international law raises further problems. It would already be possible to obtain the stated desired outcome (a single action implementing international obligations both in Scotland and in England and Wales), by joint action by Scottish and UK Ministers. International relations are a reserved matter, and there is therefore no question that the UK Parliament has sole power to legislate for the implementation of such agreements. Scottish Ministers are further prevented from any action that breaches UK international obligations (section 58 of the 1998 Act). This clause creates no requirement for agreement on the proposed action between the two sets of ministers, nor even any requirement for consultation with Scottish Ministers about proposed UK action. The potential scope of such agreements is unlimited but broad. This therefore amounts to a substantial restoration to UK control of any matter that is the subject of an international agreement, whether or not the UK Parliament has legislated about it. The necessity for this is not clear, nor is it clear whether this measure is in fact proportionate to the problem which it purports to address.

Concluding remarks

28. The Scotland bill is a flawed piece of legislation. Those flaws relate both to what it does contain, and what it does not. Most of them are part of its design, and are not such that they can be redressed during its consideration by the Scottish Parliament, or at Westminster. Rather than representing the end of a process of reviewing and revising Scottish devolution, it will require further revisions in the coming years (possibly even before it wholly comes into effect). The Committee, and the Parliament, need to be aware that these proposals present significant problems, will require continued active engagement by both government and parliament to be workable, and will be only a staging post on the route to a more durable and workable system.

February 2011