Response to ‘The Financing of the Scottish Government’ Inquiry by the Scottish Affairs Committee

Preface

1. This is a response by David Phillips to the Scottish Affairs Committee’s inquiry in to the Financing of the Scottish Government. David Phillips is an Associate Director at the Institute for Fiscal Studies (IFS) and leads its work on devolved and local government finances. However, the IFS has no corporate views on the issues covered in this response (or indeed any other issue): all views are those of David alone, and he is solely responsible for any errors or omissions.
2. David was also a co-author of the independent report on block grant adjustment mechanisms commissioned by the UK and Scottish Government’s as part of the 2023 Review of the Scottish Fiscal Framework. Except where that report or previous reports co-authored with David Bell and David Eiser are directly cited or quoted from, all views are those of David Phillips alone, and not necessarily representative of the views of his co-authors.
3. This response deals with aspects of questions (1) to (4) in the terms of reference for the inquiry. Answers to questions (1) and (2) are more most detailed.

Summary

4. Question 1. The Barnett formula has largely served its intended purpose of reducing year-to-year negotiation over funding levels, at least for Scotland. But it leads to arbitrary funding levels, has been partially superseded in Wales and Northern Ireland, and should be reformed. Scotland would likely lose from reforms in the short-to-medium term but could be offered full or partial transitional protection. Both if the Barnett formula is kept or if it is replaced, more information on the operation of the funding system, and assessments of relative funding levels and funding needs should be published.
5. Question 2. Reforms to and indexation of borrowing and reserves limits as part of the 2023 Scottish Fiscal Framework Review were welcome but could have gone further. In particular, limits could have been indexed to the revenues and spending at risk, which is likely to grow faster than inflation, and the ability to borrow a modest amount (perhaps 1% of Resource DEL) for resource purposes other than forecast errors should be considered. With regards to the calculation of block grant adjustments, no method can fully satisfy all of the Smith Commission principles for the Fiscal Framework. The chosen approach (indexed per capita), prioritises the ‘no detriment’ principle over the ‘taxpayer fairness’ principle, as desired by the Scottish Government. It also largely protects the Scottish Government from revenue or social security spending shocks affecting Scotland in line with the rest of the UK (rUK), but provides no protection for asymmetric shocks. It would be difficult to provide greater protection for the Scottish Government without potentially weakening its financial incentives and responsibility.
6. Question 3. The provision of additional resource borrowing powers, and/or the ability to defer or carry-forward any late in-year changes to funding would provide the Scottish Government with both more financial certainty and flexibility. Devolved governments can raise a dispute with HM Treasury if they believe the Barnett Formula has been applied incorrectly or the block grant otherwise calculated incorrectly, with HM Treasury Ministers ultimately deciding whether any change is made. There is a separate process for either the Scottish or UK government to raise disputes with other elements of the Fiscal Framework, such as block grant adjustments and compensation for ‘spillover effects’, with no change to default arrangements if agreement cannot be reached. Putting funding arrangements on a statutory basis should be considered, allowing the operationalisation of the Barnett Formula (and any replacement) and broader Fiscal Framework to be challenged in court.
7. Question 4. Wales and Northern Ireland now have ‘needs-based’ factors to supplement the basic Barnett Formula but it seems unlikely HM Treasury would agree to such an approach for Scotland currently given its relatively high funding levels. There is almost certainly scope to learn from other countries, including in relation to funding allocation methods, wider fiscal frameworks and the statutory underpinning of subnational fiscal systems. However, any reforms will need to reflect the UK’s political context.  

Question 1. Appropriateness of the Barnett Formula

8. The Barnett Formula was introduced 1978 (and began operating for Scotland and Northern Ireland in 1979 and for Wales in 1980) and is named after Joel Barnett, the then Chief Secretary of the Treasury. Prior to that, during the 1960s and 1970s, public expenditure plans for Scotland, Wales and Northern Ireland were settled by negotiation within the wider public expenditure framework, on much the same basis as for other government departments. Because such negotiations were often difficult --- especially during periods of public spending restraint such as the late 1970s --- and because there were referendums on devolution to Scotland and Wales taking place in 1979, it was felt that continuing with this approach was undesirable and that a more ‘mechanical’ approach to budget allocation would be beneficial: hence, the Barnett Formula.

9. The Barnett Formula calculates only the change in funding for the Scottish Government and other devolved governments each year. The funding for year t is then equal to funding in year t-1 plus the formula-calculated change in funding. 

   The aim of the formula is to provide the same cash per person change in funding for Scotland as is budgeted for comparable spending responsibilities in England. At Spending Reviews, these are calculated by summing up changes resulting from changes to each Whitehall department’s departmental expenditure limit (DEL), separately for capital and resource DEL. For each department and spend type, the resulting change in the Scottish Block grant is calculated by:

     Where the comparability percentage is the share of a department’s spending that relates to responsibilities that are devolved to the Scottish Government, and the appropriate population proportion is either Scotland’s population as a share of England’s (if responsibilities are also devolved to Wales), or England and Wales (if the Whitehall department is responsible for spending in Wales). Comparability percentages are calculated using pre-existing spending patterns rather than planned spending changes, which may not be in line with these existing patterns. As a result, the change in funding per resident in Scotland may not actually match the change in comparable funding per resident in England.

     At other fiscal events, the formula is typically applied on a programme-by-programme basis. Each programme is assigned a 100% or 0% comparability percentage depending on whether or not the equivalent responsibility is devolved to the Scottish Government.  

10. The Formula has significantly reduced the need for annual negotiation in relation to the Scottish Government’s core block grant funding. This has actually been more true since the advent of devolution in 1999 than prior to it; during the 1980s and early-to-mid 1990s, the formula was often bypassed, especially for Northern Ireland and Scotland, to provide additional funding for public sector wages, in the context of public sector employment being relatively higher than in England. Since devolution, the Scottish Government has had to fund public sector pay rises from within its Barnett-determined core block grant.
11. Recent formula-bypasses for Scotland have typically been for specific funding streams – such as rail investment funding, city deals, “levelling-up” funding, and post-Brexit agricultural funding. Several of these are being brought back within the scope of the Barnett Formula – including rail investment funding, post-Brexit agricultural funding, and the UK Shared Prosperity Fund as it is re-purposed for core local government funding. This illustrates the trade-off the UK government faces between reflecting variation in the need for / benefits of funding for particular services or investments across the UK nations, and minimizing bespoke calculations and negotiations over specific funding streams.
12. There have been more significant changes to the operation of the Barnett Formula in Wales and Northern Ireland. In particular, changes in funding in Wales and Northern Ireland are now calculated using a so-called needs-based factor. This currently multiplies the standard increase for Wales by 1.05, with this factor set to increase to 1.15 if funding per person falls to 115% or less of English levels. For Northern Ireland, the needs-based factor is 1.05 if funding per person is over 124% of English levels, and 1.24 if funding per person falls to or below this level. In both cases, the ‘funding floors’ (of 115% and 124%) are based on analysis initially carried out by the Holtham Commission, commissioned by the Welsh Government to examine its funding arrangements, and its’ relative spending needs. The ‘needs-based’ factors are designed to slow down any convergence in funding to English levels, and then stop funding falling below this relative floor.
13. These arrangements for Wales and Northern Ireland reflect one of two fundamental problems with the Barnett Formula itself: it takes no account of the relative spending needs of the devolved nations, relative to England, or changes in these relative needs. Indeed, if population grows at the same rate in the devolved nations as in England, the series of equivalent per-person increases in funding for the devolved nations as in England generated by the Formula, will lead to a convergence in funding per person, irrespective of the needs of the populations of each nation.
14. There are no official estimates of the spending needs of the devolved nations relative to England, which is why the Holtham Commission’s estimates have been utilized. These suggested that relative funding was at or slightly below relative needs in Wales around 2010 (although it has since risen somewhat), and that funding was soon set to fall below relative needs levels in Northern Ireland. This is why there was pressure to reform (or replace) the Barnett Formula in Wales and latterly Northern Ireland, which has been partially addressed via the needs-based factors.
15. In Scotland, the situation is somewhat different. HM Treasury estimate that comparable funding per person is the highest of any UK nation at 126% of English levels during the current Spending Review period. Conversely, while needs are estimated to be higher than in England, higher incomes and lower poverty than in Wales and Northern Ireland mean that needs are estimated to be lower than in those nations. The Holtham Commission provided tentative estimates of spending needs per person being 105% of England’s as of the late 2000s. Even allowing for a much higher weight on population sparsity than Holtham’s analysis (based on spending patterns in England) would be unlikely to push the figure much above 110-115%.
16. Thus, while the Welsh Government and NI Executive have, at different times, considered the replacement of the Barnett Formula with a needs-based approach to be financially beneficial, that has not been the case with the Scottish Government. Scotland highly likely receives a share of funding substantially above its relative share of needs – helping pay for free university tuition, free personal care, and a number of other entitlements residents of England do not enjoy. (Scotland’s higher income tax rates have, to a large extent, simply offset weaker underlying tax base growth, with the small net increase in revenues helping pay for more generous benefits and lower council tax).  
17. The other major problem with the Barnett Formula is that it does not properly account for changes in the relative populations of Scotland (or Wales or Northern Ireland) relative to England. This is because while the changes in funding each year are calculated using the latest population figures, the baseline funding levels to which these changes are applied to are not updated to account for changes in populations.
18. The devolved nations benefit from this ‘flaw’ in the Barnett Formula because their populations have typically grown less quickly than in England. This means that the increase in funding per person calculated by the Barnett Formula is higher than in England once one accounts for population growth.
19. Boileau and Phillips (2023) shows how population growth, inflation and real-terms changes in spending in England interact to determine both the absolute growth in funding per person in Scotland, and relative growth in funding per person compared to England. This shows that all else equal; faster population growth leads to lower absolute and relative funding per person in Scotland; faster inflation leads to lower absolute and relative funding per person in Scotland; and bigger real-terms increases in spending in England lead to higher absolute but lower relative funding per person in Scotland. In essence, the long-run relative level of funding per person in Scotland compared to England depends arbitrarily on relative population growth, inflation and overall funding growth, which is hard to justify.
20. The main projection from this analysis, based on population projections as of 2020, and spending in England growing in line with the OBR’s long-run projections of GDP growth plus inflation, shows funding per person in Scotland growing by 0.2 percentage points less per year than in England over the next 30 years. This would see funding per resident fall below 120% of English levels in the mid-to-late 2030s, and below 115% of English levels in the late 2050s.

Figure 1. Main projections for Scottish Government’s absolute and relative funding per person, 2027–28 to 2057–28 (Boileau and Phillips, 2023)

21. This convergence would be faster and deeper if (a) Scotland’s birth rate and immigration rate are higher than projected, (b) overall spending growth (whether due to inflation or real-terms spending growth) in England was faster – both things which the Scottish Government says it would like.
22. In the longer-term, the Scottish Government may therefore benefit from the replacement of the Barnett Formula with a more rational approach that takes account of spending needs and population. Reductions in funding in the short-to-medium term could be avoided or at least reduced through the operation of transitional protections.
23. Development of a needs-based formula would be both technically and politically demanding. Spending needs cannot be directly observed but must instead be inferred (or assumed) based on observable socio-economic and geographic characteristics. Decisions would need to be taken on the characteristics to account for and the weights to apply to them – which in turn should depend on the range and quality of services for which relative spending needs are being assessed for. Formulas would need to be updated as the characteristics of the different nations and the drivers of spending needs change and as the balance between different services (for which patterns of relative need may differ) in total spending changes. Each of these decisions is likely to be contentious, so there may be a role for an Independent Commission to advise on the formula. This issue is discussed briefly in response to question 4 below.
24. With both a potential new system, and the existing Barnett Formula, the UK government should be as transparent as possible about the factors underlying funding decisions and the resulting funding levels. It is not currently meeting this benchmark.
25. The Statement of Funding Policy and Block Grant Transparency Reports published by HM Treasury do provide important information on how funding changes made using the Barnett Formula have been calculated – but they omit certain key information and in the case of the Transparency Reports are often published with a significant lag. The Statement of Funding Policy should include more information on the spending included and excluded from calculating comparability factors – and show reconciliations with other published spending figures. The Transparency report spreadsheets could also be made more user-friendly, making it easier to select different vintages of Resource and Capital DEL plans to see how spending plans have changed over time.
26. The new needs-based factors and funding floors for Wales and Northern Ireland mean that regularly publishing information on relative funding levels (including how these have been calculated) is vital – and this should include Scotland. If the new ‘floors’ are to avoid coming as arbitrary as the outcomes of the basic Barnett Formula, the spending needs assessments the floors are based on should be updated and improved over time. The resulting estimates and full methodologies should be published.

Question 2. The 2023 Fiscal Framework Review

27. The 2023 Fiscal Framework Review made several changes to the borrowing and reserves powers of the Scottish Government. 

 a) The original ‘Scotland-specific shock’ clause was removed, with the £600 million in annual borrowing to address forecast errors previously only available once a Scotland-specific shock had been declared made the new standard limit for borrowing to address forecast errors (up from £300 million). 

 b) The resource drawdown limit was abolished, meaning the Scottish Government can draw down in full in a single year any capital and resource reserves held in the Scotland Reserve.

 c) All annual and aggregate borrowing and reserve limits have been indexed according to forecast inflation as measured by the GDP deflator since 2024–25.

28. These changes were an important step in the right direction – inflation had already eroded around 20% of the real-terms value of the existing limits by 2023–24, and would have otherwise eroded them further. Difficulties in forecasting mean that the previous £300 million limit for forecast errors in normal times would likely have been regularly exceeded, with capped reserve drawdowns exacerbating this situation. The doubling of this normal limit, together with the uncapping of reserve drawdowns provides important additional fiscal flexibility. 

29. However, rather than link the limits to inflation, it would make more sense to link them to the amount of revenue and social security spending at risk, which will typically grow faster than inflation. If there were concerns about linking borrowing and reserves limits to variables partially under the Scottish Government’s control, limits could be linked to tax and spending forecasts for the rest of the UK (as with the block grant adjustments). 

30. There is also a case to widen the scope for which resource borrowing powers can be used. Currently, they can only be used to offset reductions in devolved tax revenue or increases in devolved social spending as a result of forecast errors, or reconciliation payments for past forecast errors. However, there may be unexpected shocks to public service spending, or shortfalls in revenue or increases in social security spending that are forecast in advance. The Scottish Government cannot borrow to smooth the resulting financial pressures, although it can utilize reserves.

31. In light of this issue, myself, David Bell and David Eiser have recommended that the Scottish Government be allowed to borrow a small amount – perhaps initially 1% of its Resource DEL per year, subject to an aggregate cap of 3% of Resource DEL – for resource purposes other than forecast error. This would give it more scope to address short-term revenue shortfalls or spending pressures for reasons other than forecast errors, with the aggregate cap preventing it from borrowing indefinitely

32. Such a limit for Scotland (and the other devolved nations) would have very little impact on the UK’s overall deficit or public debt. Indeed, even much higher limits would be only a rounding error for the UK’s fiscal forecasts and trajectory. However, we concluded that substantially higher limits could be seen as unfair to England. This is because there is no England-only borrowing: any borrowing by the UK government is either for UK-wide purposes or for spending that generates funding for the devolved governments via the Barnett Formula. Even though the borrowing would be repaid by the devolved governments using future funding, it may be deemed unfair for residents of devolved governments to be able to bring forward more spending via borrowing than those in England. Capping borrowing at a relatively modest amount would provide further important fiscal flexibility while minimizing any perceived fairness issues.

33. We also suggested two further elements of a regime for discretionary resource borrowing to help allay concerns about fairness and scrutiny:

a) Requiring any discretionary resource borrowing by a devolved government to be funded through bonds issued directly by itself, rather than by borrowing through the UK government. This would ensure that costs of servicing the debt reflect the perceived risks of holding devolved government debt, and limit the scope for any negative spillovers onto UK taxpayers.

b) Requiring the devolved governments to notify their respective legislatures whenever they plan to utilise their discretionary resource borrowing powers, enabling the rationale for that decision to be robustly scrutinised. 

34. While the Fiscal Framework Review made significant changes to borrowing and reserves powers, it largely maintained the status quo in relation to the calculation of block grant adjustments (BGAs) to account for tax and social security devolution. The one change to block grant adjustments made was to increase the BGA for the Crown Estates – this reduces (albeit not fully) the extent to which Scotland benefits twice-over from increases in spending funded by Crown Estates revenues, and particularly revenue from offshore windfarm licenses.  

35. Apart from the Crown Estates and a few other small BGAs which are fixed in cash-terms, the BGAs for devolved taxes and devolved social security benefits are updated each year using the indexed per capita (IPC) method. Under this, the BGA is changed in line with the change in revenue or spending per person in the rest of the UK (rUK) and the change in Scotland’s population. This was being used on an interim basis prior to the 2023 Review and will continue to be used going forwards. 

36. This approach insures the Scottish Government against revenue or spending shocks that affect its devolved revenues or social security spending in the same manner as in the rUK. This is because any fall in revenue that is mirrored elsewhere in the UK will be offset by a reduction in the revenue BGAs, while any increase in revenue will be offset by an increase in the revenue BGAs. However, the Scottish Government gains or loses in full if its revenue per person grow by more or less (or fall by less or more) than in the rUK. Similarly it gains or loses financially if its social security spending per person grows by less or more (or falls by more or less) than in the rUK. This provides the Scottish Government with an incentive to take action to boost revenues and constrain social security spending, but means it also bears in full the risk of any idiosyncratic change in devolved revenues or social security spending. 

37. By indexing the BGAs to Scotland’s rather than the rUK’s population growth, the IPC method also means that Scotland’s revenue BGAs grow less quickly than they otherwise would, given Scotland’s population typically grows less quickly than the rUK’s. However, this means that if Scotland boosts its revenues by attracting more migrants, in turn boosting its population, its revenue BGAs will increase – this weakens the financial incentive to boost migration. 

38. There are many other possible ways of updating the BGAs each year, with different methods implying different risks and incentives for the Scottish and UK government. The Scottish and UK governments’ commissioned myself, David Bell and David Eiser to undertake an independent analysis of the different mechanisms, the risks and incentives they entailed, and the extent to which they were consistent with the ‘Smith Commission’s principles for the fiscal framework. 

39. The key principles were:

a) There should be no detriment to the Scottish or UK governments’ budget simply as a result of the initial transfer of tax and/or spending powers (‘no detriment from the initial decision to devolve’). In defining this principle, the Commission also stated that the BGAs should be ‘indexed appropriately’, suggesting this principle should be interpreted dynamically too.

b) The devolved Scottish budget should benefit in full from policy decisions by the Scottish Government that increase revenues or reduce expenditure, and the devolved Scottish budget should bear the full costs of policy decisions that reduce revenues or increase expenditure (‘economic responsibility’).

c) Changes to taxes in the rest of the UK, for which responsibility in Scotland has been devolved, should only affect public spending in the rest of the UK; changes to devolved taxes in Scotland should only affect public spending in Scotland (this has often been referred to as ‘taxpayer fairness’, although that terminology was not explicitly used by the Smith Commission).

d) The UK Government should continue to manage the fiscal risks and shocks that affect the whole of the UK for the newly devolved revenue streams and spending responsibilities (‘UK economic shocks’).

40. We concluded that no method of calculating the BGAs fully satisfies all of the principles set out by the Smith Commission. The fundamental tension is between the ‘taxpayer fairness’ principle and the ‘no detriment’ principle. The interaction between the BGAs and the Barnett Formula used to allocate funding to the Scottish Government means these two principles are incompatible.

41. To see this, consider what would happen if the UK government increased income tax rates in rUK and used this to increase spending on the NHS. Under the Barnett Formula, the Scottish Government would receive a population-based share of the increase in English NHS spending. To stop Scotland benefiting from a tax increase only applying in rUK, i.e. to satisfy the ‘taxpayer fairness’ principle, the income tax BGA (a deduction from the block grant) would also have to increase by a population share of the increase in tax revenues in rUK. That way, the increase in block grant funding via the Barnett Formula and the increase in the income tax BGA would exactly offset.

    But under this method of indexing the income tax BGA, even if Scottish income tax revenues per person grow at the same percentage rate as in the UK, the Scottish Government would, over time, lose more and more from income tax devolution. This is because Scottish income tax revenues per person are lower than those in the UK – so if revenues per person grow at the same percentage rate, their lower absolute level means that the extra amount raised per person in Scotland would be less than the additional amount raised per person in rUK. Scottish revenues could therefore be reasonably expected to fall further and further behind the income tax BGA, reducing the Scottish Government’s funding and violating the ‘no detriment’ principle. 

    Our report reviews a wide range of different options for how to calculate and index the tax and welfare BGAs over time. We find some methods better satisfy certain principles, while others better satisfy others. There is therefore a trade-off between the different Smith Commission principles.

42. By continuing with the IPC method, the revised Fiscal Framework has prioritised the ‘no detriment’ principle over the ‘taxpayer fairness’ principle, in the long-run likely benefiting the Scottish Government to the tune of billions of pounds per year. Scotland will continue to benefit from the implicit redistribution of income tax revenues from rUK to Scotland via the Barnett formula.

43. A number of alternative BGA mechanisms have been proposed. These take into account the implications for revenue and spending growth of factors such as: differences in the structure of the Scottish and rUK tax bases at the point of devolution; and differences in the rate of change of population age structure between Scotland and rUK post-devolution. These approaches generally come closer to achieving the ‘no detriment’ principle in the period after devolution than the IPC approach, since they control for factors which are known about with reasonable certainty before devolution happens, and which the Scottish Government has relatively little influence over. However, such approaches would generally violate the ‘taxpayer fairness’ principle to a greater extent than the IPC approach, since they would imply even larger effects on the Scottish budget of tax changes in rUK for taxes which are devolved in Scotland.

44. There may be a case in principle to incorporate some element of fiscal insurance explicitly into the BGA process, to protect the Scottish Government from shocks or longer-term factors largely outside it’s control that lead to differential revenue or social security spending growth per person than in rUK. Other countries often have some degree of risk sharing for such idiosyncratic factors using full or tapered floors and ceilings on revenues, and the government is shortly to reset the English business rates retention system to redistribute revenues between English councils following 13 years of retention and resulting revenue divergences.

45. Implementing any of these approaches in practice would be subject to a significant challenge though, as explained in our independent report: 

    “Changes in tax and social security policy post-devolution in both Scotland and rUK mean it would be difficult to estimate what revenues and spending would be in Scotland if policy had remained in line with rUK. But [floors, ceilings and] resets of limits have to be set on the basis of policy being the same in Scotland and rUK, otherwise the Scottish Government would be compensated for tax cuts (and spending increases) and penalised for tax increases (or spending cuts).

    There are two stages to adjusting revenues and spending for divergences in policy post-devolution: adjusting for the mechanical effects of differences in policy given the tax base and spending needs pertaining; and adjusting for the behavioural effects of policy divergences, which can affect tax bases and spending needs. 

    To do the first, one would need to calculate what Scottish revenues and social security spending would be if the policy in place in rUK applied in Scotland. For taxes where the only things that have diverged post-devolution are tax rates and bands, such as income tax, this should be feasible. This should also be the case for benefits where all that has changed are payment amounts. However, for taxes where there have been changes to taxbases, and benefits where there have been changes to eligibility criteria and assessment processes, such as disability benefits, this would be a much more difficult challenge. 

    Adjusting for behavioural effects is also difficult. For example, if the tax base or number of people eligible for a benefit has grown more or less in Scotland than in rUK post-devolution, to what extent is this driven by differences in policy, versus being driven by other underlying socio-economic factors. This is an important question to address, because while we may want to insure the Scottish Government against risks associated with the latter, we would want it to bear the behavioural as well as the mechanical effects of its policies on its revenues and spending.  For example, if the Scottish Government were to have a substantially higher tax rate on incomes above £125,000 (a highly responsive group of taxpayers), we would want it not only to gain revenues as a result of the higher tax rates, but also to lose as a result of behavioural responses undertaken by taxpayers to reduce their tax liabilities (such as reducing their work effort, engaging in greater tax avoidance or evasion, or migrating). If the Scottish Government does not bear these costs, its incentives are skewed towards setting higher tax rates and more generous benefits policies than would otherwise be the case. 

    However, even after a policy is implemented, one cannot know for sure what the behavioural response to it is – one can only estimate it statistically, and such estimates are subject to both measurement error and methodological difficulties. For a tax such as income tax, even relatively small differences in estimates of the scale of behavioural response can mean differences of tens of millions of pounds in revenue. Agreeing the size of the behavioural adjustments to make to estimates of the tax revenue capacities of the devolved governments would therefore likely be very politically difficult. It is for this reason that the 2016 Fiscal Framework suggests that behavioural ‘spillover’ effects of one government’s decisions on the revenues or spending of other governments should only be compensated for in exceptional circumstances. 

    Thus, implementing insurance via the BGAs in a way that is consistent with the principle of ‘economic responsibility’ is likely to be difficult in practice. It will prove difficult to adjust for even the mechanical effects of policy divergences for taxes and benefits where there have been major changes in the design post-devolution; and adjusting for the behavioural effects of policy divergence would require agreement on uncertain behavioural elasticities.” 

46. The use of the IPC method can therefore be seen as a practical compromise between different Smith Commission principles, and provides insurance for common but not idiosyncratic shocks affecting Scotland’s devolved revenues and social security spending. It would be worth monitoring revenue and spending trends though to see whether there is evidence of significant-enough divergence that makes introducing some component of insurance for idiosyncratic shocks worthwhile given the challenges and potential distortions to Scottish Government incentives it would entail.

Question 3. Funding certainty, coordination and challenge

47. The enhanced borrowing powers discussed in paragraphs 31 – 33 would provide the Scottish Government with more certainty over the availability of funding (if the UK government made in-year reductions to planned spending in England), as well as more flexibility to respond to shocks.
48. In the absence of this, the UK government could commit to allowing the Scottish and other devolved governments to defer in-year cuts in funding or carry-forward in-year increases in funding outside normal Reserve limits.
49. Funding guarantees set at or above the level of the Barnett-determined (or in future, other formula-determined) block grant, as were used in 2020–21, would not be an appropriate long-term solution. This is because they would imply Scotland and the other devolved nations would each year receive an amount that was equal to or greater than the amount implied by application of the Barnett (or other funding) Formula. Allowing borrowing or deferral would provide in-year funding certainty, while avoiding this systematic bias towards ‘over-funding’ the devolved governments.
50. I am not best placed to discuss the degree of coordination between the Scottish and UK governments on in-year fiscal changes. Evidence from interviews undertaken by myself, David Bell and David Eiser suggests that coordination and information sharing with stronger during the height of the COVID-19 pandemic than before or since, but these interviews were undertaken in Autumn 2021.
51. The Scottish and other devolved governments are able to raise a dispute if they are unhappy about the calculation of the block grant and operation of the Barnett Formula. The dispute is raised directly with HM Treasury, which decides whether to make changes to its initial plans. If it does not or the devolved government remains unhappy, it can raise a dispute to the Finance Interministerial Standing Committee (F:ISC), made of representatives of the UK and devolved governments. The F:ISC is a consultative rather than decision-making body though, with final decision-making power resting with HM Treasury Ministers.
52. The Fiscal Framework agreement sets out a different dispute resolution process for issues related to the calculation of BGAs, policy spillover effects and other elements of the Fiscal Framework agreement. This escalates disputes from officials, to senior officials, to Treasury and Finance ministers, with technical input potentially sought from the Scottish Fiscal Commission and Office for Budget Responsibility, which will be published. If no agreement can be reached on a dispute, rather than HM Treasury’s view prevailing, the default arrangements under the Fiscal Framework would prevail (for example, no compensation for a spillover effect). However, it is possible to escalate the dispute to the standard dispute resolution processes – and in that case it is unclear whether the ‘default’ or ‘HM Treasury’ position is applied if no agreement can be reached. This should be clarified.
53. There may be a benefit in putting funding arrangements, including the operation of the Barnett formula (or any replacement formula) on a statutory basis. Legislation would allow the devolved governments (and potentially others) to legally challenge the operation of funding arrangements, and mean that changes in devolved funding policy are subject to greater parliamentary scrutiny. Such legislation would need to be written carefully so that the intentions of the funding regime were clear – courts should be asked to determine whether a clear set of rules are being followed, not asked to determine a fair or appropriate funding allocation or regime (which is ultimately a political decision).

Question 4. Lessons from elsewhere 

54. It is always important to see if lessons can be learned from policy elsewhere: funding arrangements are no exception.
55. There are several differences between funding arrangements for the Scottish Government and those of the Welsh Government and NI Executive. The most significant are the additional tax revenues (and compared to Wales) social security powers devolved to the Scottish Government, and the ‘needs-based’ factors and funding floors used for Wales and Northern Ireland.
56. It seems unlikely there is an appetite to reduce the degree of fiscal devolution to Scotland by removing or scaling back existing tax powers.
57. Legislation allows for the devolution of corporation tax to Northern Ireland but not Scotland or Wales, but this has yet to be enacted. Corporation tax, being subject to significant behavioural response (including profit shifting) is a poor candidate for devolution, and if it were to be devolved a formula apportionment basis (as in the US or Germany) rather than a separate accounting basis (as legislated for in Northern Ireland) would make more sense.
58. Notionally, all benefits including the State Pension are devolved to the Northern Ireland Executive. However, its system almost entirely mirrors the system in place in Great Britain, and is funded directly by HM Treasury rather than via BGAs. This devolution is more notional than real and reflects historic circumstances.
59. A funding floor may be attractive to the Scottish Government, but given the relatively high levels of funding Scotland receives, the case for introducing a needs-based factor and funding floor now is much weaker than for Wales and Northern Ireland.
60. Looking overseas, it is notable how varied funding arrangements for devolved / sub-national governments are. Each responds to the constitutional and political contexts of the country, and in many cases also reflects a series of reforms over time – albeit typically less piece-meal than in the UK. Any system for Scotland and the wider UK must reflect our political and constitutional context, or be aligned with constitutional reforms that can command widespread political support.
61. Areas to consider other countries’ experience include funding allocation arrangements (such as the independent, advisory Australian Commonwealth Grants Commission), wider fiscal frameworks including fiscal insurance and borrowing regimes (such as Spain or Germany’s systems) and statutory underpinnings (such as Germany’s Finance Constitution). Experts from the OECD or the countries in question will be best placed to advise on the operation of their systems. The Local Government Information Unit has been undertaking a review of subnational finance systems, which although focused on local government also discusses regional/state financing arrangements for several countries, and may also be worth examining.