Finance and Public Administration Committee

Meeting date: Tuesday, November 14, 2023

Official Report 645KB pdf

---

Scottish Fiscal Framework: Independent Report and Review

Good morning, and welcome to the 29th meeting in 2023 of the Finance and Public Administration Committee. The first item on our agenda is an evidence session on the Scottish fiscal framework independent report and review. We are joined by David Phillips, who is associate director of the Institute for Fiscal Studies and one of the authors of the independent report, and Professor Mairi Spowage, who is director of the Fraser of Allander Institute. I welcome you both to the meeting.

We have about an hour for this session. Before we move to questions, I will set the scene a wee bit. In November 2021, the then Cabinet Secretary for Finance and the Economy, Kate Forbes MSP, confirmed that the independent report would focus only on block grant adjustments. However, she said that the Chief Secretary to the Treasury had

“agreed this will inform a review that will be wider in scope”

to ensure

“that the current arrangements are thoroughly assessed and options for reform considered, and that input is obtained from a wide range of stakeholders as part of the overall process.”

On 2 August 2023, the Deputy First Minister confirmed:

“I have now reached agreement with the Chief Secretary to the Treasury ... on a package of changes to the Scottish Government’s Fiscal Framework.”

She explained that she had judged it appropriate to concede to a review with a narrower scope than the more fundamental review that was originally envisioned

“in the interest of securing long sought practical borrowing and reserve flexibilities, and to protect those arrangements that we already have in place which work in our favour”.

I wanted to note that in order to set the scene for the questions that will follow.

One of the things that the Deputy First Minister secured is the indexed per capita mechanism for calculating block grant adjustments being adopted permanently. What are the benefits of that to Scotland? Further on in the paper, it says that that will be reviewed every 50 years. Is that what they mean by “permanent”? Do you have another view of what “permanent” might mean in this situation?

The indexed per capita method of adjusting the block grant takes account of two factors that might mean that Scottish revenues do not keep pace in cash terms with those in the rest of the United Kingdom. First, it takes account of the fact that Scotland has, initially, lower tax capacity than the rest of the UK. At the point of devolution, for example, income tax earnings per capita were 79 per cent of those in the rest of the UK. Using a percentage change in tax revenues to calculate block grant adjustments accounts for that because the percentage is based on the Scottish revenues. The other factor is that the per capita element accounts for potential differential population growth between Scotland and the rest of the UK.

The way that the method works is that it takes the block grant adjustment from the previous year and applies the percentage change in revenues per capita in the rest of the UK and the percentage change in Scotland’s population. That insulates Scotland from having a different initial level of earnings per capita and from having different population change. In recent years, Scotland’s population change has tended to be lower than the change in the rest of the UK—Scotland’s population has gone up less quickly.

The fact that the method accounts for that means that the block grant adjustment grows less quickly than it would if it had to keep pace with the overall level of growth in the rest of the UK. We think that, compared with the alternative method, which is called the comparable model, it will reduce the increase in the block grant adjustment by about £50 million a year on average. That means that Scotland benefits by £50 million, then £100 million, then £150 million and then £200 million, compared with what would happen if the comparable method was used.

It is worth pointing out that, of the methods that are on the table, this one probably best addresses what is called the no-detriment principle—the idea that Scotland should not, in expectation, lose out from tax devolution just because it has been devolved, although it can lose out if the economy performs less well or benefit if the economy performs better and, of course, it can benefit or lose from its policy changes. Of the methods that are on the table, the one that has been chosen probably best addresses that, but it addresses less well some of the other principles from the Smith commission.

A key aspect of the report is the fact that there is a fundamental tension between the taxpayer fairness principle, which says that revenue should not be redistributed from the rest of the UK to Scotland post devolution, and the no-detriment principle, which says that Scotland should not, in expectation, lose out. Those two principles are fundamentally in conflict. The agreement prioritises the no-detriment principle, which benefits Scotland financially compared with the alternative.

Mairi, the Deputy First Minister more or less said that the reason why we did not go down the road that we said that we would in November 2021 is that the UK Government said that the model was a kind of take it or leave it. Is that your understanding of the situation? Was the only room for manoeuvre that the Scottish Government had to accept this?

I suppose that it depends what alternatives were on the table. I know that the committee has had concerns in the past about not just the population risk that the Scottish budget is open to in relation to the overall population, which David Phillips explained, but the mix of our population—the extent to which it is ageing more quickly or is made up of people of working age and that sort of thing. However, those different options did not seem to be on the table. Since the fiscal framework was put in place in 2016, both methods have been calculated—the comparable method and the indexed per capita method. It seems likely that the question about the block grant adjustment in the review would be which of those was going to be chosen.

The Deputy First Minister was also looking for further flexibilities, which the Scottish Government did get as part of the agreement. It was not just about the Scottish Government securing its preferred BGA option out of the two, as there are also the further flexibilities on things such as borrowing. It looks as if the Deputy First Minister made the calculation that it was better to have the agreement confirmed after many years of it dragging on and to take the additional flexibilities that were also being offered. Given the choices that the Scottish Government is facing immediately in the current budget round, the fact that there is more flexibility in borrowing will help to some extent this year, when there is a large reconciliation to be made.

I very much agree with that. In our independent report, we looked at some of the alternative options, such as adjustments related to demographics, and there are two points to be made on that. First, it is not clear, at least in the short to medium term, that adjusting for demographics would necessarily benefit Scotland. Although Scotland’s population is ageing more rapidly, a lot of the difference is not about the working-age population; it is about the growth in the child population. I have seen some calculations that suggest that, if we adjusted for the demographic mix, Scotland could actually lose out in the short term because pensioners pay tax whereas very few children do. It is not clear that Scotland would necessarily benefit in the short term from adjusting for the age mix, although that might change in the longer term.

The second point is that some tensions in the principles come to the fore here. If we think that the Scottish Government has limited control over long-term demographic trends, we might think that adjusting for the differential demographic mix is consistent with the no-detriment principle because it will prevent Scotland from either gaining or losing from things that it has little control over. However, if we think that these things are to some extent affected by Scottish Government policy—for example, it may be looking to attract workers or businesses to Scotland—it would perhaps be less consistent with what is called the economic responsibility principle, which is that Scotland should bear the costs or the benefits of its fiscal policy.

It is reasonable for Governments, when they make decisions, to trade off those different objectives and also to take account of the objective of simplicity. The more additional bits we put into the fiscal framework, the more complicated it gets. We have also seen from the census just how big changes in the overall population and—even more so—changes in the age distribution and demographics can be. That adds another bit of volatility on a 10-year cycle when the census is updated. Arguments can be made that we should control for additional factors such as demographics or tax-based structures, but there are definitely trade-offs with regard to complexity and the other principles of the fiscal framework.

On the breadth of the agreement, I note that there was quite a strong difference of opinion on what the fiscal framework actually consists of. The view of the Scottish Government and the Scottish Parliament is perhaps that it consists not just of the technical arrangements on borrowing and block grant adjustments, but also of the wider fiscal settlement and what is devolved. I think that the Treasury took the very firm line that that is a separate issue that is to do with the settlement and not the framework. I think that the Scottish Government made a decision that, in order to make progress on the framework, those issues could be put to the side and perhaps handled in a more political manner.

It was not a negotiation between equals. Obviously, the Treasury always had the whip hand. We will certainly put the question to the Deputy First Minister, but no doubt she had to accept, more or less, the progress that has been made. Some of that progress is on capital borrowing, which will no longer be fixed at £3 billion in total and £450 million a year for capital expenditure. The limits will now increase in line with inflation. As Mairi Spowage has pointed out, that should really have been the case from the beginning, because the value of those borrowing powers has been eroding significantly over time.

I am interested in how the capital borrowing power inflationary impact has been assessed, because it is based on the gross domestic product deflator. Between 2023-24 and 2027-28, it is predicted to be 5.5 per cent. That is not the figure for this year; it is the total over the next four years. Is that in any way realistic? It would mean that the resource borrowing limit would go from £600 million in the current financial year to £633 million and the capital borrowing limit would go from £450 million to £475 million. How realistic is that, Mairi?

The calculations that we did were based on the Office for Budget Responsibility predictions at the time. As we have seen, there are issues around inflation in the wider economy and things have not evolved as lots of people expected. Inflation has certainly has not come down as quickly as we expected. On that stock of borrowing, the increases will now be linked to inflation, as they should always have been. The limits were set in 2016 in a low inflationary environment, but I think that everybody should have looked ahead and considered that they ought to be index linked because their power would erode over time.

Given the quantum of borrowing, it is not going to make a massive difference to the borrowing powers of the Scottish Government. In theory, it will have the same purchasing power in the years ahead, but it will remain tied to the sorts of limits that it has at the moment. We also need to remember that, over the past few years, the Scottish Government has used up a great deal of that borrowing limit, which may well constrain the amount of borrowing that future Administrations can do.

In your submission, you say:

“There is still a question though about whether the limits for forecast error are adequate given the risks the Scottish Government are facing ... Given that forecast error is all that this borrowing mechanism can be used for, it would seem sensible to consider that they should be extended further.”

Yes. The predicted income tax reconciliation for the current year was £712 million prior to the outturn data coming in. Self-assessment receipts surprised on the upside and the figure ended up being just below £400 million. It has not reached the limit—this is just for income tax—but it could very well have done. Modelling that the Scottish Fiscal Commission has done in the past has suggested that the £600 million limit could be reached perhaps once or twice in a decade, and that is for income tax alone. The borrowing power for forecast error is to be used for all taxes and for social security as well, so there is a question about the extent to which it is still sufficient to manage the risk that exists.

One could argue that there are chances that there will be positive reconciliations in the future and that taxes will surprise on the upside while social security may surprise on the downside, so less will need to be spent than was budgeted for. The Scottish Government also has the facility to save up and put money into the reserve and so on in order to even out some of these things, as well as using borrowing powers. However, given that this borrowing is only for forecast error, the Scottish Government may be managing a large reconciliation, essentially with six months’ notice, for the budget year that is coming. There is an argument for the limit to be increased further so that it will cover an income tax reconciliation that may come fairly regularly.

10:00  

I echo those points. One can make a case that, because the powers are constrained for use simply for forecast error, they cannot be used for other circumstances, but you could have, potentially, even an unlimited capacity to borrow for those forecast errors. That is conditional on the forecasting being done independently, which it is, by the SFC. I would be much more worried about that if it was done by the Scottish Government. The other thing that one could do if one was concerned about the potential to, in effect, defer the issues indefinitely is to say that borrowing needs to be repaid within a certain time. One could combine increases in the borrowing limit with a restriction on the time period over which those borrowings have to be repaid.

With the devolution of social security benefits, something that may not be appreciated in the short term is the difficulty in predicting what rates of disability benefits people will qualify for and how long they will stay on them. There is quite a lot of uncertainty on the social security side at the moment. Even though the disability benefits are not as demand led as the unemployment benefits, which go up and down with the cycle, there is quite a bit of uncertainty about them at the point of introduction. As the data starts to come through, it might be worth looking at whether there are big forecast errors for social security benefits.

I absolutely echo that. With the devolution of income tax, it has been so obvious. We have had a few years of that, with large reconciliations being predicted and in some cases emerging in practice, but it is important to keep an eye on the extent to which we see forecast errors for social security benefits. The Scottish Fiscal Commission does its best to forecast the demand for those benefits. Social Security Scotland wants to take a different approach and the Scottish Fiscal Commission says that that is likely to lead to higher take-up, but there is quite a lot of uncertainty about how much we might end up spending on some of the benefits. As David Phillips said, they are not as demand led as benefits that are tied to the economic cycle, but they are new benefits with new criteria and they are being delivered by a relatively new agency. It is important to keep an eye on the extent to which they generate forecast errors that will need to be covered by the facility.

David, you said in the independent report that

“relevant data used to forecast the BGAs and to determine tax and welfare receipts should also be reviewed to ensure it is sufficiently robust and comprehensive for those purposes.”

You make a number of other recommendations, one of which is about

“whether the OBR and SFC forecasts could be better tied to minimise this risk.”

One of the issues at the moment with the fiscal framework is that the forecast for the block grant adjustment is made by the OBR, which makes sense because the OBR is the forecaster for the tax revenues and social security in the rest of the UK, while, on the other hand, the revenue and spending forecasts for Scotland are done by the SFC, which also makes sense, because it is the official forecaster for Scotland and it can pick out more factors that are Scotland-specific. However, as well as Scotland-specific factors driving the differences between the two forecasts, there can also be differences in their judgments about the overall performance of the economy.

My understanding is that, currently, for the earnings forecasts, it is not the case that the SFC thinks that earnings will grow much more quickly in Scotland than in the rest of the UK, apart from maybe a little bit of a lift in the financial services industry. It is just that the underlying model or framework that the SFC uses on earnings projections and forecasts differs a bit from the OBR’s. The OBR’s current forecast is that earnings will go down below 2 per cent over three years, but the SFC took the view that that is unlikely to happen, given the situation with wage settlements and the benchmarks that guide wage settlements.

That means that part of the forecast increase in Scotland’s tax revenues relative to those in the rest of the UK is driven by differences in judgments about the overall performance of the UK economy, including Scotland, rather than expectations about how Scotland will under or overperform compared to the rest of the UK. It is only the latter part that matters for the actual final ultimate budget impacts for Scotland. What might happen is that either the OBR revises up its forecast to match the SFC forecast, in which case the block grant adjustment is higher, which brings down the budget, or the SFC has to bring down its forecast to match the OBR and the revenues.

That is one of the challenges with a system of having two separate forecasters forecasting separate sides of the equation. It is not super easy to solve that, because you do not want those guys to agree on everything—you want to avoid groupthink. You want to have some challenge and you do not want people to just copy each other. On the extent to which there can be discussions and clarity around the factors that are driving these things—that is, to what extent it is differences in Scotland’s earnings versus UK earnings compared to just overall optimism or pessimism—first, the forecasting body should have a good understanding of that so that it can think about what it is doing and, secondly, that needs to be communicated to Parliament and to the Scottish Government so that, when looking at the medium-term financial strategy and the spending review, there is an understanding of what side the risks are on.

As it stands, the risks are perhaps weighted more to the downside, in terms of lower funding, than to the upside, given that part of the growth in tax revenues is the difference in overall optimism rather than the difference in Scotland-specific optimism.

I absolutely agree with that. We should not mix up the spring experience with the autumn one—there is a much bigger gap in the spring. Things have settled down a bit and, hopefully, there is a year-round budgeting process that we can rely on to know when things will happen. The spring budget was in early March and the SFC presented its forecasts on 31 May. What I took from the evidence that it gave to the committee is not that the SFC knows what the OBR is going to do but that it expects that the OBR, in its next forecast, will probably significantly revise up its earnings forecast. That is what I took from the evidence that the SFC gave, without it actually saying that. At the end of May, the SFC thought that earnings were in a better place, and that is probably what it expects the OBR to do.

We have a much smaller gap between the forecasts now, at the end of November and in December for the Scottish budget, which are the ones that matter for setting the budget. We have seen issues when the gap has been larger—that happened with the 2020-21 experience, which generated the reconciliation that we have to deal with in the 2024-25 budget. When the gaps are smaller, it is less likely that there will be big divergences in judgment, which are caused by how the economic conditions have changed. That is part of the story about the March and the May earnings outlook, but it is always possible that the two bodies will have differences of opinion on overall economic performance and the overall earnings forecast, which is what really matters for income tax.

As David Phillips said, that is good in some ways, as it avoids groupthink—forecasters should have a healthy debate and should not always agree on things. They may well have different opinions, and they have different roles. The SFC is there to forecast the Scottish revenues and the Scottish economy to the best of its ability; it is not there to take into account how that interacts with the OBR’s forecasts for the UK and the BGAs. The SFC does a lot to help scrutiny of those things, but its role is to get the forecasts for Scotland as right as it can. These things will always be in tension.

Mairi Spowage is correct in that it is not the role of the SFC to predict the other side of the equation, but perhaps the Scottish Government’s role is to take into account both sets of forecasts and the potential issues of uncertainties and the balance of risk—is it on the upside or the downside?—when it is making budget decisions about the extent to which it should build up or draw down reserves. That is one thing that the Scottish Government can do to take account of the potentially unbalanced risks.

Lastly from me before I open it to colleagues, I understand that the figure for fines, forfeitures and fixed penalties is at approximately £30 million at the moment but that there will be a flat deduction of £25 million from the block grant going forward. Is that linked to inflation at all and are those figures accurate? We have two sets of figures—£30 million for this year and £25 million from now on, which seems to be a reduction.

That is a block grant adjustment that is taken off the Scottish Government’s funding and Scotland gets to keep the revenues collected from those. Reducing the block grant adjustment means that Scotland gets more than it otherwise would have if the figure had been kept at £30 million. By freezing that in cash terms, which I believe is the agreement, Scotland gets to keep any inflationary increases in those in full. I think that the Scottish Government will be relatively pleased with that part of the deal.

That is right—as long as we do not become a more law-abiding society, of course. Thanks very much.

The second of the bullet points on the Smith commission says:

“The devolved Scottish budget should benefit in full from policy decisions by the Scottish Government”.

Is the assumption there that, if Scotland varies from the rest of the UK, that is entirely Scotland’s responsibility and it is under our control, for better or for worse?

That is the economic responsibility principle. The way to interpret that is that for taxes or spending that is devolved to the Scottish Government, when it comes to Scottish policies, whether they are related to those specific taxes or spending items or are more general—for example, policies that might boost or harm the economy—Scotland should bear, as far as possible, the full benefits and the full costs of those changes.

It is quite hard to do that in practice. For example, let us say that one of the impacts of a Scottish policy is to attract more high-skilled workers from the rest of the UK into Scotland. As it stands in the block grant adjustment mechanism, Scotland would get to keep the higher revenues generated in Scotland and, because those people are coming from the rest of the UK, that would reduce revenues in the rest of the UK and slow down the increase in the block grant adjustment slightly. Scotland would gain a little bit more from those extra revenues, because it would gain from the higher earnings and a little bit from the impact on the rest of the UK. You can see other instances where those effects might work in opposite directions.

That is helpful. I will just say where I am trying to go with this: it is about the impact of London. As I understand it, London is an outlier, certainly in the UK and possibly in Europe. Broadly speaking, in most things, Scotland competes quite well with the rest of England, Wales and Northern Ireland—the rest of the UK, apart from London and the south-east. If London grows faster than the UK average on the economy, tax and all the rest of it, as it has tended to do, we have no control over that, but we are being punished for it.

On that, I would distinguish between the two sides of the fiscal framework. One argument that can be made is that, when we look at the block grant adjustment, we should not look at trends in revenues in the rest of the UK as a whole; we should exclude London, because by adding in London you are making the comparator not fair for Scotland. That is because, although revenues in Scotland might keep pace with revenues in the rest of the UK excluding London, if the rich in London are pulling away from the rest of the UK and from Scotland, Scotland cannot keep up with that. Conversely, if in a downturn there is a huge fall in the top incomes in London and that is dragging down revenues, Scotland is not seeing that, so why should Scotland benefit from the fall in top incomes in London?

You can make that case. However, the other side of the equation is the way that the spending side of the budget works: as that tax from London is spent, either it is spent on things that are subject to the Barnett formula, and hence Scotland gets its share of that revenue via the Barnett formula, or it is spent on things that are UK-wide, such as social security or debt interest, and on average Scotland receives a population share of that spending. If you look at just the revenue side, you can make a case that Scotland is potentially being compared to a baseline that it is hard to keep up with but, if you reflect on the spending side of the budget, you see that the money comes to Scotland through the Barnett formula or through spending on UK-wide things.

10:15  

That comes back to the tension between the principle of no detriment from tax devolution and the issue of taxpayer fairness in the redistribution of tax revenues from the rest of the UK. The issue is where you draw the line on those things. The further you draw it towards no detriment, the further you move away from taxpayer fairness, and you need to draw the line somewhere.

I absolutely agree with everything that David Phillips said. You cannot divorce the issue from the way that Scotland is funded through the Barnett formula. If you look at spending on services in Scotland compared to spending in other parts of the UK, particularly parts of England, Scotland has higher spending. The fact that revenues generated in London and the south-east will be funding spending in devolved areas in England that generate consequentials or UK-wide spending is important to bear in mind when thinking about the budget overall.

As you said, Scotland has above average spending, but the Barnett formula is designed to reduce that gap, as I understand it. On top of that, despite what you have said, it still seems to me that, given that London in recent years has been doing better than the rest of the UK, the fiscal framework is disadvantaging Scotland. I take the point that a collapse of the London economy would benefit us, but that does not seem to be the evidence so far.

I will give a quick answer and, if David Phillips wants to talk about Barnett convergence, he can do so. The Barnett formula convergence has not really happened to the extent to which it was expected, for various reasons, including the differences in population movements over the years and the impact of things such as spending cuts during the austerity years and so on, and there is still significantly higher spending in Scotland than in every part of England, and we should bear that in mind.

I agree with Mairi Spowage that the Barnett squeeze has not really happened to date. There are three factors that have been holding it up: population; the cuts in the 2010s; and Scotland benefiting from an error in the way that business rates were accounted for in the Barnett formula in the 2010s, which amounted to about £1 billion by 2015. However, the situation is going to change. We have already seen it start to squeeze down since 2020. In 2019, spending per capita in Scotland for devolved services was about 129 per cent of the average for England, and it is now about 125 per cent. I expect that, by the close of this decade it will be about 122 per cent, and will fall below 120 per cent by the mid-2030s. That squeeze will put pressure on the Scottish Government’s budget in the years ahead.

You are also right that, if it is the case that London and the south-east will pull away from the rest of the UK in the long run and it was not within the capabilities of the Scottish Government to make policy changes with the levers available to it to change that, there could be a potential for divergence and a fallback on the income tax side. I come back to the point about the two principles that were set out by the Smith commission: that, on the one hand, Scotland should not lose out from tax devolution in expectation, effectively; and, on the other hand, post-devolution revenues in the rest of the UK should not be to Scotland from devolved taxes. Those principles are incompatible. Either we need to go back to those principles and revise the position, or we should choose one to prioritise.

In this agreement, the Scottish Government and the UK Government have chosen to largely, but not completely, prioritise the principle of no detriment over that of taxpayer fairness. Hence, there is still some potential for detriment around issues such as the south-east and London pulling away from the rest of the UK and Scotland.

You are saying that, having been at 129 per cent, rightly or wrongly, we are on a decline to 120 per cent?

Yes, and, in the longer term, given current population projections, I would expect that to converge to about 115 per cent. Given Scotland’s demographics and socioeconomics, that is probably still a little bit higher than Scotland’s relative needs, but it means that many of the things that Scotland has traditionally benefited from and enjoyed, such as free tuition, free personal care for the elderly, would have to be funded by alternative means, and that the more universal social welfare state in Scotland will be more difficult to fund in the 2030s and 2040s. Even if the UK Government were to match spending needs in the rest of the UK—which is a big “if”—it would still be hard for Scotland to fund that than it has been traditionally in the past, given the Barnett squeeze.

But there has not been any assessment of needs for quite a long time I think. If needs includes rural areas, we have got big challenges with that.

David Phillips might correct me if I am wrong, but I think the last big assessment of that was done by the Holtham commission for Wales many years ago, which I think was in 2010. That suggested that Scotland had been relatively overfunded by the Barnett formula, by quite a way, and that Wales had been generally underfunded. I think that that was the last big assessment of the situation, and it would certainly be interesting to update that. That took account of things such as the higher cost of living in rural areas and the great cost of providing services in those areas, but it still suggested that Scotland was considerably overfunded compared with the sums that it would get according to a needs-based formula. It would be interesting to redo that exercise and see where we are now, given all the changes that have happened to the health, wealth and outcomes of lots of different people across the whole of the UK.

It would be particularly interesting to look at it given the rise in the elderly population and particularly the concentration of elderly people in more rural areas. The older population is where a large share of spending goes and if they are in the more rural areas where it is harder to deliver services—I think that it has become increasingly more so in rural areas—that could change some of the figures a little bit compared with, say, 2010 when the calculations were last done.

At previous meeting, the Deputy First Minister told us that both Governments were having to make compromises. Do you see that the UK has made any compromises in this?

I think that the UK has made compromises. My overall impression is that both Governments are reasonably happy with the agreement because they avoided the big risks that they wanted to avoid.

The UK Government’s compromise was on the block grant adjustment indexation where, partly because of individuals involved in the negotiations, there was a particular focus on trying to make sure that Scotland’s block grant adjustment mechanism treated population the same way as the Barnett formula does. That is what is called the comparable model. That was brought in in Wales in the fiscal framework. I think that the UK Government compromised on that. It would have wanted to put that in there because several people feel that it is unfair that Scotland is effectively protected from slower population growth on the revenue side but benefits from it on the Barnett formula side, and that there is asymmetry in there. The UK Government compromised on that one, although, as I said, you could argue that, by compromising, it made it more consistent with the no-detriment principle, which is one of the principles of the Smith commission.

The other area where both Governments compromised was on the Crown Estates. The UK Government wanted to index the Crown Estates to revenues in the rest of the UK, largely to account for the fact that, with regard to the offshore wind farm licences, if the arrangement had been fixed in cash terms at its old level, Scotland would have got its offshore wind licences as revenues and then would have got a Barnett share, or another share, of the revenues being spent from the English and Welsh licences. Treasury thought that was unfair and wanted to index it. The Scottish Government argued that its position is different because its Crown Estates are quite different to those in the rest of the UK, as they do not involve places such as part of Regent Street and the central London estates that are racing away in value, so it would not be right to index Scotland to that. In the end, the Scottish Government compromised around the sum of £40 million by 2028, which I think is roughly halfway between full indexation and fixing it in cash terms, so I think that there was some compromise there.

Where the Scottish Government would have felt that it compromised was that, although it got an improvement in its borrowing powers, the arrangement was linked to inflation rather than being indexed to the amount of revenues or spending at risk, which typically grows faster than inflation. I agree that there was a compromise there. Overall, however, the Scottish Government will be pleased it has kept the index per capita and that it has got some increases, and the UK Government will be pleased that it had some increase for the Crown Estates and that it did not have to grant substantially increased borrowing powers.

The impression that I got is that both sides are pretty happy. I think that the Scottish Government won an award for its negotiations at the Chartered Institute of Public Finance and Accountancy event.

Professor Spowage, as well as responding to that, could you also comment on where we are now with regard to Scotland-specific shocks?

I agree that both Governments will be happy. To be honest, generally it is the teams of officials who will be happy with the outcomes; I imagine that the politicians involved will just be happy that it is done. As the convener touched on, there is a question of what happens now with the agreement and any review of it in the future. I am sure that the committee will be interested in the extent to which there will be any scrutiny of future discussions or agreements on this, given that there was limited scrutiny this time. Obviously, there was a consultation that went out but there was not any discussion or consultation on the agreement, which came out during the recess and caught everyone by surprise. The scrutiny by and involvement of Parliament in what happens next is a wider question.

On the Scotland-specific economic shock, the provision that was put into the original agreement was that borrowing powers would be doubled for forecast error from £300 million to £600 million in the event of a Scotland-specific economic shock, which was defined as Scotland’s economic performance seriously underperforming that of the UK and also being very low. That provision has been removed and the borrowing limit was just increased to £600 million all the time, so there is no current provision for additional flexibility of borrowing powers in the event of a downturn that is just in Scotland rather than across the UK. There is no provision there for further borrowing in the event that that happens now.

If there was a specific shock to particularly Scottish sectors such as tourism or food and drink, would that cause us a problem?

Yes, to the extent to which a sector impacts on the income tax revenue take. For example, if hours or wages were to be constrained in the event of a Scotland-specific economic issue, the extent to which there could be increased social security payments is relevant. As we said earlier, the social security payments that are devolved are not generally linked to the economic cycle as such, but, yes, there could be an impact in that way, and in relation to spending in areas such as crisis funds and other things that local government manages.

If there was a Scotland-specific economic issue—such as the downturn in the oil and gas industry in 2015-16—it is likely that there would be downward pressure on revenues and upward pressure on spending, which could cause issues for the Scottish Government. However, you must remember that the additional borrowing flexibilities in the event of a Scotland-specific economic shock were taken account of in relation to the forecast error borrowing. The borrowing would be additional borrowing for that purpose, not just borrowing to spend more money on whatever the Scottish Government wanted to spend it on on the resource side. I am not sure that there was huge flexibility there anyway to deal with those sorts of things, given the restrictions around that borrowing.

I will add that the borrowing powers are not to address ordinary downturns in revenues; they are to address unexpected downturns in revenues. There could be scope to further enhance the powers of the Scottish Government to borrow. At the moment, for example, if, in advance it is forecast that Scotland’s revenues perform less well than those in the rest of the UK, there is no scope to borrow to smooth that, even if that is a temporary dip because of, for example, a downturn in the oil and gas industry that is expected to come or because of issues with tourism or other factors.

I think that there would be issues in giving the Scottish Government discretionary borrowing powers that are very large. I can talk about those if you want, but for a small amount—perhaps, say, 1 per cent to 1.5 per cent of the Scottish Government’s resource departmental expenditure limit, which would be £460 million, subject to some caps on the total amount and some caps on how long the borrowing can be for—either there is a case to give the Scottish Government some powers to respond not only to unexpected downturns in revenue or shocks to social security spending but things that are forecast in advance for social security spending or tax revenues, as well as unexpected shocks to public service spending. Something like 1 per cent to 1.5 per cent of the fiscal DEL would not be a threat to fiscal sustainability, and I do not think would be an issue in relation to fairness to the rest of the UK, but would give more flexibility for the Scottish Government to respond to other forms of shocks.

10:30  

In the absence of any real public discussion or parliamentary scrutiny before the agreement and the publication of the report, we are left reading priorities into the text. It seems that the Scottish Government has placed the population dynamics at a higher level of priority than other economic challenges such as low gross value added or longer-term lack of investment in research and development that might be driven by more capital spend. Is it your feeling that it has put the immediacy of those issues ahead of longer-term economic transformation, as it might put it?

I am not sure that I would go quite that far. My understanding is that, on the block grant adjustment, you are right that, focusing on population has tried to address something that would hit the Scottish budget immediately—as I said, that involves around £50 million a year cumulating for each year, and that was not accounted for. On the capital side of things, it is perhaps less the case that the Scottish Government did not want to prioritise than that the UK Government did not want to cede too much ground in that area.

This is a personal opinion—it is not based on inside knowledge or anything—but I think that the UK Government is perhaps concerned that, if it were to grant substantial additional borrowing powers for capital investment, those could be used in the short term to substantially boost spending and could be used in arguments about the constitution. I think that the UK Government would have been much more reluctant to see substantial additional capital borrowing powers, particularly in the context of the current set-up in the rest of the UK, where there are no England-only capital borrowing powers.

You are right that it has ended up prioritising short-term funding issues over long-term potential investment, but I am not sure that that was a thing that was substantially under the Scottish Government’s control to vary.

The cumulative nature of the population risk compared with the comparable method on the table—if those were the two methods that were on the table—made it, from the Scottish Government’s perspective, a no-brainer. That does not mean that it could not also try to push for increased capital borrowing powers and, as I said, the Scottish Government has extensively used those capital borrowing powers since they have been in place—up to 75 per cent of the overall cap in the current year. I guess that the Scottish Government would say that, in the negotiations, it was trying to argue for both. However, from its perspective, it absolutely had to secure the IPC method, because that cumulative erosion of the spending power of the overall budget would be a significant issue permanently embedded in the fiscal framework.

As much as other members may describe the negotiation as a process in which the UK Government has the whip hand, it involves a two-party signatory agreement where one party can withhold agreement and say that “Until we get more of what we want, we are not going to agree to it”. Although, as David Phillips described, one party might, for very sound reasons, be reluctant to go in a particular direction, it is a two-party agreement.

That takes me to the issue about how rushed some of this feels towards the end. It is a long-term process but it seems like we are now locked into what is described as an agreement for more than 50 years on the basis of trying to deal with a short-term issue in terms of borrowing capabilities to deal with a £1 billion black hole of the Scottish Government’s own making. Is that not part of the risk in the way that the Scottish Government has dealt with this? It has focused on short-term budget priorities and how it can get through this budget year, and has locked us into a process for 50 years as a result.

The fiscal framework provisions make clear that both Governments can suggest changes to the fiscal framework review that would be reviewed by the joint Exchequer committee. I imagine that this committee or future iterations of it in the Scottish Parliament would be interested in how it can make suggestions in that regard, or how its scrutiny can be brought to bear on any discussions on the arrangements. I think that it is unlikely that this agreement will lie untouched for 50 years. There are provisions to suggest changes, but there is not a point at which it would definitely be reviewed, as there was in 2016. The trigger for that was because the two Governments in 2016 could not come to a permanent agreement on the BGA mechanism. Now that they have, it does not mean that other provisions in the framework could not be looked at.

The £1 billion issue that you referred to was set out in the medium-term financial strategy and that was the revenues that the Government thought it would have access to to spend compared with the commitments that it had made. It is obviously up to it to explain which things it is not going to do to meet that challenge, and that is what it will be doing in the budget. I suppose that the increased borrowing powers made the approach to the issue a little bit easier, given that the Scottish Government is able to borrow for the whole of the income tax reconciliation now, whereas it was not able to do that before.

I do not know whether your characterisation is fair, because there will clearly be opportunities to review the arrangement in future years. The Scottish Government and the Treasury can make suggestions and changes and, as I say, I do not think that the fiscal framework will remain as it is for 50 years.

The Barnett formula was set in similar circumstances. It was thought that it might be revised within a couple of years but it has existed for decades and we are still unpicking the complexities of it. Mairi Spowage might want to come back to me on that one, but I invite David Phillips to share his thoughts first.

It is fair to say that, sometimes, arrangements that are seen as temporary can end up becoming permanent fixtures of the system. It is also fair to say that the Scottish Government would have had a eye on the short-term fiscal position when it comes to trying to reach an agreement for the current financial year. I point out, however, that might have been a typo in some of the notes, because my understanding is that it is not a 50-year review cycle but a five-year review cycle. It is five years, but not more than once in any UK or Scottish electoral cycle.

Upon which basis decisions are made, I suppose.

I am sure that a future UK Government could easily reopen those negotiations if it wanted to be more generous to Scotland.

Let me return to the issue of forecasting by the OBR and the SFC. What you have set out for us is extremely important not only for determining some of the issues within fiscal frameworks but in relation to setting budgets. We have had the OBR and the SFC in front of this committee—they were extremely professional in both cases, I have to say. Is there statistical evidence, within recent trends of the forecasting of the OBR and the SFC, or particular data that shows a greater issue of error or volatility, or is it just the overall predictions that provide some problems for us?

I understand that there is a particular difficulty with self-assessment. The initial expectation for the most recent year was about £700 million in reconciliation, and it came in at just under £400 million—a £300 million difference. As a share of overall income tax revenues, that is only about 2 per cent or so, but, as a share of the self-assessment revenues, it is probably more like 10 to 15 per cent. There is quite an issue with self-assessment.

Looking forward, I think that there is also a risk that the Scottish Fiscal Commission has quite a difficult task ahead. Do we baseline the faster growth and assume that it is now at a higher level, or will there be some reversion to the mean—will we grow faster in this year and slower in the subsequent year? There is a particular issue there.

More generally, we have seen a big issue in the economic statistics in the UK recently whereby different earning series are giving a vastly different picture of what is happening to earnings. For example, over the past few years, things such as the labour force survey, the average weekly earnings and the annual survey of hours and earnings—some of the main surveys that we use for earnings—show that earnings have grown much more slowly than HMRC data suggests. However, the HMRC data over the past couple of months has suggested much slower growth than some of those other indices. I know that the SFC can take account of the HMRC data that it gets through as a feed, but we know that that data is not a full sample.

So, there is a particular issue with self-assessment and a fundamental issue that earnings data in the UK seems to have got a lot worse recently, which is causing some issues in the forecasting process.

When you say that it has got a lot worse, is it that the trends of that forecasting have got worse, or is there just extra volatility in the earnings that is very difficult to predict? I am asking whether there is a problem with the statistics that we are using and, therefore, a difficulty for the OBR and the SFC in interpreting that data or whether something else is causing a problem.

There are definitely issues with labour market statistics currently. There has been a big drop in the response rate to surveys such as the labour force survey and the annual survey of hours and earnings—ASHE—which has made it much harder to know what is generally going on with both employment and earnings, particularly in those areas that are not so well covered by the income tax statistics, such as the self-employed and those at the bottom end of the market.

The Financial Times contained an interesting article about this a couple of weeks ago. The presentation of the data is making life very hard for the OBR and the SFC. That matters to us big time because the data is absolutely crucial not only for informing us what the current economic picture is but for informing the Government when it sets policy. Can we do anything else to help to reduce the forecasting error, which obviously has created difficulties between the OBR and the SFC, certainly over the past few years?

I think that it would be difficult to reduce the forecast error much more. There are huge challenges with all the data about earnings and the labour market, which have been well documented in relation to the ONS’s difficulties with response rates and the labour force survey, and I am not sure that those challenges will be addressed by the transformations that the ONS is trying to put in place. It is a big challenge. We have seen that even in what the Bank of England thinks about the economy and the decisions that it is making—because it is hearing very different signals about wage growth and the economy.

You must remember that the SFC has extra challenges because it does not have all the data sources for Scotland that the OBR has at the UK level. It does not have the equivalent of the average weekly earnings data to look at, to see whether it is different from the HMRC data. All that it has is the HMRC data, and it is doing its best to forecast using that. Extra data sources could possibly be made available to allow the SFC to do that sort of thing, but there are huge challenges in collecting that data at the UK level at the moment, so it is pretty difficult.

I would always advocate for more access to data for analysts and researchers. HMRC data is a particularly difficult area for academic researchers and researchers outside Government such as David Phillips and me. Access to more data would obviously help, but forecast error will always be a feature of the process. The SFC’s forecast error happens because it is forecasting for a smaller economy, so the data and the forecasts are likely to be slightly more volatile. Whatever forecast error the OBR has, the SFC will probably have a slightly bigger one just because Scotland has a smaller economy. Such errors are part of forecasting. The issue is amplified in Scotland because there is the OBR forecast error on top of the SFC forecast error. They could go in the same direction, but they might go in opposite directions; therefore, the issue is amplified for the Scottish budget.

That hits the nail on the head. In Scotland, forecasting is a difficult job because the SFC is having to deal with two forecasts rather than just one.

If we accept that there will be a degree of forecast error, we have to play that into the system. Do you know whether, when the OBR and the SFC are looking back at the predictions that they made and how things turned out in practice, there is a lot of co-operation between the two in order to assess where some of the forecast error came in? In your opinion, is that well documented?

10:45  

I understand that they work together fairly closely and will be trying—as fellow analysts and forecasters doing a very difficult job—wherever possible to learn from each other and to improve the situation. Let us not forget that the past five years or so, during which the SFC was set up, has been a mad time economically, with lots of things happening that were not expected. There have been a lot of difficulties that none of us could have predicted. However, I understand that they are working very closely, as fellow analysts, wherever possible, to share information and to understand where their models are maybe looking at things slightly differently, in order to reduce those differences.

The OBR has recently been doing some work to look at its Scottish forecasts. It has focused more on its own forecasts of Scottish revenues—as opposed to its forecasts of rest-of-UK revenues, which go into the block grant adjustment—which is what matters for the Scottish budget. That can be very useful, and both the OBR and the SFC produce forecast evaluation reports, which look at the historic performance of their forecasts and try to learn lessons from those. I am pretty sure that they also look across their two systems.

As there will always be forecast errors, it is worth thinking about not just the increase in limit that we have seen, which will help to address the fact that the forecast errors are now bigger than they were previously, but whether there is a case for going even further and potentially saying that it is important to cover any forecast error but subject to some overall cap and some time limits on repayment. I also go back to the fact that understanding qualitatively, at least, which direction any adjustment may come from is important. Then it is up to the Scottish Government to set its budgets with those risks in mind.

My own view is that, in 2021-22, when we had the reconciliation, it was quite clear that Scottish revenues were not going to outpace those in the UK by £500 million or whatever was said. It would have been possible for the Scottish Government to bank some of that money into reserves, to potentially reduce the amount that would need to be borrowed in the coming year to help to smooth things.

That is very helpful. I think that the committee would agree with the statement that forecasting is an inexact science. Obviously, we have to accept that. However, is there anything that we can do through budget analysis and fiscal framework analysis to ensure that the forecasting error is minimal?

We have talked before about the issue of the RUK forecast, which the block grant adjustment is based on, coming from the differencing-off of the OBR’s Scottish forecast from the UK forecast, but its Scottish forecast is done using a different economic model and not Scotland-specific economic determinants. The OBR, in its commentary around its Scottish forecast, does seek to adjust the Scottish share on the basis of Scotland-specific factors, and some of the work that it is doing to look at that may well improve it. However, more work could be done in that area, because the RUK forecast is differencing that off from the UK forecast, and its view of the Scottish economy not being based on a Scotland-specific model is potentially a bit of an issue.

Other approaches to the issue are being taken in other parts of the UK. Wales has made a different decision about the extent to which it wants its own fiscal institution, and it is asking the OBR to take that role—at the moment, at least. It will be interesting to see how the different approaches lead to different outcomes in things like reconciliations, forecast errors and so on.

Thank you. That is very helpful.

Good morning. I want to go back to a comment that David Phillips made earlier about capex and the accommodation that was made by the UK Government. In some respects, I am surprised. I appreciate why the Scottish Government wanted to fix on the IPC, for all the reasons that we have discussed. However, in relation to the current fiscal challenges, I am surprised that it did not push more around capex thresholds, given that there is a very real need for capital projects and given what those could have brought to the economy in the complete absence of any of the meaningful levers to grow the economy that it might ordinarily expect to have. A lot of what we are discussing is really dancing on the head of a pin in terms of the nature of the fiscal transfer and the way that things are happening in the UK.

Do you agree with my assessment? If you had been doing that, would you also have been pushing hard for increased capital borrowing powers, with the intent of using them because there is a good reason to do so in the current economic climate?

First of all, I do not know what was going on behind the scenes. We wrote the independent report and had some discussions with the two Governments in the context of that, but I do not know what conversations were taking place and how much different aspects were being pushed by both Governments. I could not say whether the Scottish Government should have pushed it more or whether it pushed it as hard as it could and it was a red line for the UK Government. I do not know. My sense is that the UK Government would be more concerned about giving ground on borrowing than it would be about what accumulate to quite large amounts on the BGA side but are, in any one individual year, relatively small amounts. I would not want to comment on whether the Scottish Government should have tried harder, because I do not know how hard it tried.

On the capital borrowing limits, you could argue that, yes, given that we know that inflation for many things on capital is outpacing the GDP deflator at the moment—that was commented on earlier—if you want to maintain the real value of the capital borrowing, linking it to capital inflation as opposed to GDP inflation might have been a start.

When it comes to what the priority is then, in terms of further increases to increase capital investment or to increase resource investment, I think that there is a question about what the Scottish Government’s real challenge is. Is the real challenge at the moment in maintaining services and maintaining the resource side of the budget, or is it in maintaining investment in the economy? Also, what is the supply-side capacity for those things? Is there the supply-side capacity to deliver substantially higher investment without driving up costs further?

I do not know the answer to those questions, but I would hope that the Scottish Government is doing some work around those things when thinking about its negotiation strategy—thinking about not just what the numbers are, but where it can get the biggest bang for its buck and where it can avoid, for example, spill-over effects that end up pushing up costs and undoing some of the potential benefits that the higher limits could give you.

That is very helpful. Do you want to come in on that point, Professor Spowage?

Yes. I agree. A lot of the challenges for Government capital spending can be about the capacity for delivery. Obviously, through Covid and everything else, there have been lots of challenges in the supply of materials and labour. The Government has to consider those things and the extent to which there is capacity to deliver on the capital spending programme that it is looking to set out. In recent years, there have been significant challenges in doing that in some parts of Scotland, so it is definitely something to consider. Otherwise, as David Phillips says, you might just push up prices further.

I am not sure that it is the same in Scotland, but, in England, capital budgets are consistently underspent because delivery lags behind planning.

Thank you for that helpful clarification.

On the point about the baking in of some of the things that we have been discussing today and the nature of the fiscal transfer system, how does what we have now compare with other fiscal transfer systems elsewhere in the world? We have previously talked about how complex the system is. Are we simply baking in complexities that will need to be managed in future—we talked about some of those earlier—or are we starting to move towards more comparable methods that are used elsewhere? I do not think that that is the case, but you are the experts.

Because the UK started with a highly centralised position and has been devolving things, rather than starting with a decentralised system and centralising certain things, the idea of there being block grant adjustments linked to particular taxes is quite unusual. I think that that reflects the starting point. Given that we are moving from a point where everything was centralised to one where things are decentralised while trying to avoid, where possible, detriment and taxpayer unfairness, it is understandable how we have ended up in that situation. That is quite unusual, and it is quite complex.

Where I think that things are less complex than in many countries is around what happens after the point of devolution. Many countries do not just have a fixed block grant adjustment that goes up in line with what happens in the rest of the country or does not respond to changes in the relative circumstances of the countries involved. They have some kind of on-going insurance or equalisation mechanism. For example, in Germany or in Scandinavia, if the economy of one part grows faster than that of the rest of the country, it gets to keep some of that, but it must hand some of it back to the central Government so that it can be redistributed. Similarly, if one part falls behind, it will get some insurance to stop it falling further behind.

When the Independent Fiscal Commission for Northern Ireland looked at the issue for Northern Ireland, it said that although it agreed, at least in the first instance, with the principles that underlie the Scottish agreement—that there should be economic responsibility and Scotland should gain or lose if its revenues grow faster or slower from devolved revenues so that it has responsibility for its policies—there should be a backstop to that. Because we know that revenue performance is driven partly by Scottish decisions and partly by things outside the Scottish Government’s control, there should be a backstop if the revenues fall too low or some kind of mechanism—I am not sure that “backstop” is the right term—to redistribute money to the rest of the UK if its revenues grow particularly fast. That is something that we do not have in the fiscal framework in Scotland. Many other countries do.

I think that, in future arrangements and future reviews, it would be worth looking at whether some form of on-going insurance or cap or ceiling would make sense in the context of the Scottish fiscal framework.

Absolutely. The Northern Ireland report goes into that in a lot of detail. The discussion about the extent to which the devolution of taxes adds to accountability and so on is an interesting one. This is something to consider that is not present in the UK system, which is quite unusual. You can totally see how we have got here. As David Phillips said, we have the centralised system, and Barnett had to remain—that was one of the agreements. Therefore, the complex position that we have ended up in is the best way to meet some of the principles that are set out in the Smith commission, albeit that they are in tension. It is completely understandable how we have got here. Some quite significant reforms would be necessary for it to be more like a federal system or a different approach.

I have one last question, which picks up on what John Mason was talking about earlier. To what extent are we baking in recognition of the tremendous pull of London and the south-east, which I think Vince Cable described as the vortex that sucks everything in? Are we not simply recognising that that will continue to be the case—indeed, will always be the case—even allowing for a technical mechanism, which would mean that, if there was a cataclysmic event, regardless of whether we are talking about Scotland or anywhere else, we would have very little capacity to do anything about it under the system that has now been devised?

As I said, the fundamental tension with regard to whether London and the south-east should be excluded from the comparison comes back to the issue of how to prioritise no detriment versus taxpayer fairness. How do you square the revenue side and the spending side of the budget? One way in which you can do that, while keeping a relatively simple mechanism that involves comparison with the whole of the UK, is to have a backstop that says that if there were to be an on-going trend such that Scotland were to fall behind, there would be some sort of reset or insurance mechanism that would prevent that from continuing indefinitely.

Equally, the flipside of that is that if Scotland were to improve relative to the rest of the UK, there would need to be a reciprocal arrangement, but such a mechanism might be a way to address the concerns that John Mason raised while having a main underlying system that is not overly complex and which is still based on revenues in the whole of the UK.

At the same time, it should be recognised that spending in Scotland is also supported, through Barnett, by the revenues that are generated in London and the south-east. It is always important to remember that as well.

We have covered all the changes that are mentioned in the summary of changes to the fiscal framework, except one. It is a fairly minor one, but it would be remiss of us not to touch on it. What will happen to the coastal communities fund? What will that mean for Scotland? We understand from the agreement that that fund will be absorbed into Barnett, and that there will be no immediate impact on funding. What are the long-term implications?

11:00  

I must own up to not having looked into that in any great detail. I can explain the principle of the change. By rolling the fund into Barnett, that will mean that it will be included in the comparability factors that are allocated to a department in England. If, subsequently, that department gets any increases, Scotland will get a population share of the increase that relates to the part that is the coastal communities fund in the rest of the UK—or in England, I should say. If Scotland were to receive a bigger than population share increase in subsequent funding, it would lose out from the mechanism. If it were to receive a smaller share, it would gain from the mechanism. It depends on its relative use of those.

What are the financial parameters that we are talking about?

I am sorry—I have not looked at that.

I am not sure either, convener; I am sorry.

On that positive note, we will end the session. I thank our witnesses—both of whom will be involved in the next session—very much for their contributions. I will call a wee break until five past 11 so that our new witnesses can be brought in.

11:01 Meeting suspended.  

11:07 On resuming—