Response to Consultation on Future Funding

David Phillips

1. Introduction

This is a response by David Phillips, a senior research economist at the Institute for Fiscal Studies (IFS). David has led the IFS’s work on devolved finance issues. He is also a member of the Welsh Finance Minister’s Tax Advisory Group. However, the views and opinions expressed here are those of the author only. The IFS has no corporate views.

This earlier work may also be useful to the committee’s work. Please see the following publications:

http://www.ifs.org.uk/publications/7711

http://www.ifs.org.uk/publications/7484

http://www.ifs.org.uk/publications/7442

Please note this response focuses on the following areas:

· The trade-offs between different types of funding arrangements

· The practical issues involved in moving from

As a result, responses are not provided to all questions. And it does not make recommendations about which funding system should be used. Ultimately, that is a political question, involving bargaining between the nations and governments of the UK, and political views on the appropriate degree of redistribution across the UK and how to trade-off risk-sharing/insurance versus fiscal incentives.

2. What are the main weaknesses in the current financial settlement for Wales?

The Welsh Government currently receives the bulk of its money from the UK Government in the form of a block grant. This block grant is calculated each year as the prior year’s grant plus (or minus) an increment calculated using the Barnett formula and based on changes in “comparable” budgets in England. In addition, the Welsh Government receives income from non-domestic rates (NDR), and Welsh local authorities receive money from council tax. Non-devolved spending (such as welfare) is the responsibility of the UK government.

From 2018-19, powers over and revenue from Stamp Duty Land Tax (SDLT) and Landfill Tax (LT) will be devolved to Wales, and the block grant will be reduced accordingly. There is also the potential for income tax to be partially devolved, subject to a referendum, which would necessitate similar adjustments to the block grant. As shall be discussed later, adjusting the block grant is a complicated issue which has profound implications for the level of funding the Welsh Government will see in future, and the financial risks and incentives it will face. At the same time, the Welsh Government will acquire some borrowing powers (for both capital and current purposes) subject to strict rules.

A number of elements of this framework attract criticism:

1) That the Barnett formula is arbitrary and takes no account of relative needs (or revenue contributions for that matter).

2) That the result of the use of the Barnett formula is an underfunding of Wales, and in the long term, a “Barnett squeeze”.

3) That the specific way in which the Barnett formula treats devolved NDR revenues is flawed (http://www.ifs.org.uk/publications/7442 ) and has what appear to be unintended consequences. Whilst these flaws benefit Wales (from this year), Scotland and Northern Ireland during periods in which the Local Government budget in England is doing poorly relative to overall “comparable” departmental spending, the flaws would cost Wales (and Scotland and NI) if the Local Government budget were to do relatively well from budget settlements.

4) That information on the application of the Barnett formula is not readily available, and that application of the formula is controlled solely by the Treasury (which makes decisions about what spending is deemed ‘comparable’, for instance).

5) That a system largely based on block grant funding gives little financial incentive for the Welsh Government to boost economic performance, and hence boost revenues (or reduce welfare costs). Proposals for tax devolution will address this concern to some extent.

6) That the proposed borrowing powers are too limited and highly constrained, both on the capital side (see CIPFA’s submission) and on the current side (where borrowing will be available only for forecast errors).

There is disagreement about some of these issues.

For instance, the UK Government says that it thinks that Wales’ level of funding is within the appropriate range suggested by the Holtham Agreement, and has said it will consider a ‘funding floor’ to prevent any further convergence if agreement is reached on devolving income tax. It is also pointed out that the Barnett squeeze does not operate (at least to the same extent) when spending is being cut, and when there is lower population growth in Wales (as is currently the case).

But each is an issue worth considering by the Committee.

How can these be resolved?

Some issues can, in principle, be resolved relatively easily. For instance, further information about the application of the Barnett formula (or any replacement formula) could be published alongside each fiscal event (such as Budget or Autumn Statement). In addition, the Barnett formula (or any replacement formula) and associated Statement of Funding Policy could be managed by an independent body, rather than the Treasury.

Similarly, the flaws in the way the Barnett formula treats devolved NDRs can be corrected for by a simple change to the technical workings of the formula.[1]

However, some of the issues reflect trade-offs between different objectives. For instance, a needs-based approach to funding would cushion Wales from any adverse budgetary effects from increases in its relative needs – such as weaker economic performance or a more rapidly ageing population –, but would also mean Wales would not gain from any reduction in its relative needs. The latter would blunt the incentives the Welsh Government has to take action to reduce relative needs and to boost economic performance. Depending on whether the needs based formula also took into account revenue-raising capacity, it could also blunt incentives to boost revenue growth from devolved taxes.

Addressing one issue (making the funding more responsive to changes in relative needs) may therefore make other problems worse (lessening the incentive of the Welsh government to take action to reduce spending needs). These trade-offs between risks and incentives are at the core of any funding arrangement for sub-national governments.

How will upcoming changes in relation to: a reserved-powers model for Wales, and the St David’s Day devolution proposals and Smith Commission recommendations, impact on future funding arrangements of the Welsh Government?

The response to this question focuses on adjusting the block grant as a result of further devolution, and draws on work being undertaken as part of assessing the Smith Commission proposals.

Clearly when additional revenues or spending are devolved, adjustments have to be made to the block grant.

In principle at least, the first year adjustments to the block grant are relatively straightforward to implement, although the calculations involved may be complex. When devolving a tax, the block grant is reduced by the amount of revenue being transferred. When devolving responsibility for an additional area of spending, the block grant is increased by the amount the UK would have spent in Wales on that area.

Given the complex calculations that may be required in order to estimate these quantities, it would be preferable if the UK Government and Welsh Government agree on a methodology, and publish detailed information on the calculations. This will allow proper external scrutiny. If agreement on a single method cannot be reached, then both parties should publish detailed information on their methods, and resulting calculations. The OBR and an equivalent Welsh fiscal commission should assess and, if appropriate, sign off these calculations. Information should also be published on the mechanism by which ultimate agreement is reached (and what that agreement entailed). Consideration should be given for an independent institution to act as an arbiter.

Calculating the adjustment to the block grant in subsequent years is more conceptually complex. One cannot simply continue to deduct or add an amount equal to the revenues or spending devolved. Doing this would remove any incentive for the Welsh Government to boost tax revenues or limit expenditure growth (including through discretionary tax rises, or spending cuts): any change in revenue or spending would simply be cancelled out by an offsetting change to the block grant adjustment.

There are a number of methods for adjusting the block grant, and each has different properties. The attached presentation sets out three possible mechanisms, and their advantages and disadvantages (including worked examples of how they perform under different scenarios).[2]

The following summarises the findings (with advantages in green, and disadvantages in red):

• Fixed % adjustment to block grant (i.e. block grant reduced by the same % in each year as in 1^st year)

– Easy to understand and can be implemented using Barnett formula

– But Wales bears risk of UK-wide shocks it is ill-equipped to bear

• Index to % change in rUK revenues

– Insulates Wales from UK-wide shocks but still growth incentive

– In spirit of “no detriment simply from devolution” as Wales is left no better off or worse off in the long term if revenues grow at the same % rate as in rUK

– Wales affected (a bit) by rUK decisions on taxes that in Wales are the responsibility of the Welsh Government

• Index to £s p.p change in rUK revenues

– Also insulates Wales from UK-wide shocks while still proving growth incentive

– Wales in principle unaffected by rUK decisions on taxes that in Wales are the responsibility of the Welsh Government

– But Wales loses out unless its revenues grow more quickly in % terms than rUK

From a Welsh Government perspective, the last option would worsen the Barnett Squeeze. Indexing block grant adjustments to the percentage change in rUK revenues has attractions in general.

However, if yield from a particular tax were expected to grow less quickly (or more quickly) in Wales than in the rest of the UK due to underlying economic factors, indexing the block grant adjustments to revenue growth in the rest of the UK may not be appropriate. This may be the case for Stamp Duty, where the lower property prices in Wales mean that revenues might be expected to grow less quickly in Wales (as fewer properties will be subject to the highest tax rates even if property price growth kept up with growth in England). This issue has prevented agreement on the method to adjust the block grant after the 1^st year of devolution of Stamp Duty (and Landfill Tax) in the case of Scotland. This suggests two options:

1) Assess each tax on a case-by-case basis to decide what method should be used to adjust the block grant. However, this risks a zero-sum bargaining approach by the devolved and UK governments where each side pushes for a system that is likely to benefit them for the particular tax in question.

2) Combine revenue from all taxes and make a single adjustment. Because the bigger taxes (like Income Tax) are not so prone to such problems of differential revenue growth, doing this may make it easier to use a single principles-based adjustment mechanism.

It is also worth discussing the ‘no detriment’ principles suggested by the Smith Commission for Scotland to see if they are relevant for Wales.

The ideas behind the two ‘no detriment’ principles seem sensible at first glance:

· Neither government should gain nor lose simply as a result of the decision to devolve revenues or spending responsibilities (1^st no detriment principle);

· Each government should bear the risks and reap the benefits of their own policies, and not win or lose from knock on effects from the other government’s policies. Related to this, it seems reasonable that changes in taxes in rUK for which the Welsh Government has responsibility in Wales, should not impact the amount spent for the benefit of Wales (together, the 2^nd no detriment principle).

The difficulty arises in practice, when considering how to implement the principles.

It does not seem possible to design a block grant adjustment mechanism that satisfies the need for transparency and ‘automaticity’, and at the same time, satisfies both of these no detriment principles. For instance, indexing the block grant to the % change in equivalent revenues/spending in rUK seems to satisfy the spirit of the 1^st no detriment principle, but does not fully satisfy the 2^nd no detriment principle. Conversely, indexing the block grant the £s p.p change in equivalent revenues/spending in rUK satisfies the 2^nd no detriment principle, but does not seem in the spirit of the 1^st (as it has an effect similar to the “Barnett squeeze”).

There is a further issue with the 2^nd no detriment principle: taken at face value, it implies that where there are knock-on effects from one government’s decisions on the revenues or spending of the other, compensating transfers should take place. The calculation of such transfers would be difficult, however, involving complex calculations and modelling; seemingly minor and technical differences in assumptions may lead to very different answers. This aspect of the no detriment principles therefore leaves much scope for disagreement – which could cause difficulties for inter-governmental relations. Difficult negotiations would likely be required in such circumstances, which would mean the system was not “mechanical”, and may lead to a lack of transparency.

This would suggest restricting the circumstances in which such transfers take place to the most significant and obvious examples of “knock on” effects; and requiring full information to be published by both the UK government and Welsh government on assumptions and modelling undertaken during negotiations about what compensating transfers should take place. Again, assessment and sign off of costings by the OBR and an equivalent Welsh fiscal commission seems worthwhile; as does investigation of whether an independent body can act as an arbiter.

It is also worth noting that while the “no detriment” principles may seem intuitively appealing, they are not a central feature of the fiscal frameworks of other countries, perhaps because of the difficulties of implementing such principles in practice.

What financial and economic information is needed by the UK and Welsh Governments to provide support for future funding arrangements?

The data requirements to implement future funding arrangements will depend upon precisely what those are.

Tax devolution will require forecasts and outturns data for revenues from devolved taxes in Wales and equivalent taxes in the rest of the UK (or England). The forecasts of those revenues will require improved data and forecasts on the Welsh economy – housing volumes and prices, landfill volumes, employment, earnings, profits and other incomes etc. It would also be useful to have more detailed statistics on the devolved taxes (e.g. income tax collected by income range or tax band; housing market transactions by stamp duty band, and revenue per band).

Any move to a needs-based formula for allocating the block grant will require data on the indicators that enter that formula. The Holtham Commission showed that it is possible to base a formula on a relatively small set of indicators which would reduce data requirements.

As discussed above, information on how the Barnett formula (or any replacement formula) has been applied at each fiscal event (and the resulting budgetary consequentials) should be published to allow proper oversight and scrutiny.

Are there any issues the Committee should be aware of in relation to developments on the issues of convergence, underfunding and Barnett reform?

I would suggest the Committee examine two contributions to these issues.

First, a recent paper by myself, which looks at the Barnett Formula’s treatment of devolved NDRs:

(http://www.ifs.org.uk/publications/7442)

Secondly, a paper by Jim Cuthbert (actually from 2001) which provides some analytical results on the extent to which the Barnett formula leads to convergence at given levels of nominal spending growth, and different levels of relative population growth:

http://strathprints.strath.ac.uk/52628/

The proposal to put operate a funding floor for Wales by adjusting the Barnett formula so that Wales gets 115% of any increase in per-person comparable spending in England (if needs were assessed to be 115% of those in England) should be seen in the light of the paper by Mr Cuthbert. In effect, if population growth were slower in Wales, such a mechanism would lead to funding in Wales moving quite away above that floor.