Good morning. I am now going to begin this meeting of the Scottish Affairs Committee and our first evidence session into the financing of the Scottish Government. We have three witnesses with us today, so welcome to you all. Whether you are here in person or remotely, it is good to have you here, and we are looking forward to having a conversation with you this morning. We will all want to ask you questions, but we will try to do so in a fairly orderly fashion. Could I just ask that all questions and answers are directed through the Chair? It just helps to keep things on a formal basis. We will begin by asking the three witnesses if they would give us a very brief introduction to themselves and the work that they do that is the reason for them being here today. I will start with David Phillips, if I may. Good morning, David.

Good morning, Chair. My name is David Phillips. I am an associate director at the Institute for Fiscal Studies, and I lead our work on devolved and local government finance.

I am Claire Murdoch. I head up the fiscal sustainability and public funding work in the Scottish Fiscal Commission. We are the independent fiscal institution for Scotland, and the work that I do looks at the overall Scottish budget in terms of the funding position. We do some work looking at spending over the next five years, and we have started looking at 50-year projections of the Scottish budget as well.

My name is João Sousa. I am deputy director and senior knowledge exchange fellow at the Fraser of Allander Institute at the University of Strathclyde, where I lead on work related to the fiscal position of the Scottish Government, and interactions between the UK and Scottish Governments in terms of financing.

Thank you all for that. It is often said that the Barnett formula is very useful in so far as it allows the pooling and sharing of risks and resources to make sure that funding for the Scottish Government is stable and secure. How effectively does the Barnett formula do that?

The Barnett formula allocates funding to Scotland, as well as to Wales and Northern Ireland, but not on the basis of what is raised there in revenues. Revenues are pooled at the UK level and then allocated via this formula. That does, if you like, pool and share risks related to revenue fluctuations. If revenues were to go down in Scotland because there was a downturn in the oil and gas industry, for example, that would affect national insurance, corporation tax and the VAT receipts paid by oil companies and their workers. Because the funding for the Scottish Government via the Barnett formula does not depend on those revenues, it pools and shares that risk. However, it is only a partial pooling and sharing of risk, because the Barnett formula does not really account for the other side of the equation—not just what is happening to revenues, but what is happening to the spending needs of different parts of the country. It is a formula based on what was spent last year, and then a population share of the change in what is spent in England. What if the Scottish population is ageing more quickly? What if the English population is growing more quickly? Those sorts of needs factors are not accounted for. While there is a degree of pooling and sharing via the formula on the revenue side, there is not really a pooling and sharing of risks on the spending side. With the changes in the fiscal framework, with more devolution of tax, we have started to see the Scottish Government exposed to a bit more risk on the tax side as well, in order to give them the incentives on the tax side to grow the economy, as well as the powers to vary the tax rates. I would say that it does something to pool and share risks, but it is a somewhat arbitrary formula that does not effectively pool and share risks on the spending needs side.

Would that be as much of a concern given that there is now the ability to raise tax? Presumably, that taxation can be used to mitigate that risk.

It depends on how you think about the role of the UK Government in pooling and sharing risks around the country. Of course, if you have powers to raise tax to allow you to raise more if your spending needs have gone up, that allows you to address the need to spend, but you might also believe that, in a sharing Union, there should be a sharing of those risks across the UK as well. That has been much more of an issue in Northern Ireland and Wales, which felt that their spending needs were not being reflected by the funding levels delivered by the formula. In Scotland—and I might come on to this a bit later—the broad consensus is that, if anything, the levels of funding are higher than the levels of relative needs in Scotland. It has been much less of an issue in Scotland—a “keep our heads below the parapet”-type situation. But if you think that the Union is about sharing not just risks on the revenue side, but risks that affect spending needs, such as demographics, poverty or health, then you might say, “Let’s look at this formula from a first-principles basis.”

Presumably, that would assume that everything that a devolved Administration was doing policy-wise was absolutely to reflect those changing needs, and not because of any political priority, say.

You would not want to say, “What you are spending is your needs,” because, as you say, your spending reflects your choices on priorities. Maybe you want to be more generous on certain services and, therefore, raise taxes. Maybe you want to have lower taxes and, therefore, cut back services. Rather than base it on what is spent, a little bit like with local government in England, but also within Scotland, you would want to devise a formula based on demographic and socioeconomic characteristics that tries to estimate needs. The Scottish Government could then spend above or below that, if they were to raise more or raise less themselves through their devolved powers. Another factor that makes needs-based funding a bit tricky in practice is devising a formula that everyone can agree to.

I agree with David’s point, generally, that there is a pooling of risks to some extent. It is important also to consider that, in the way the Barnett formula is set out, it allocates spending to areas for which responsibilities are devolved, but it is also a way of ensuring that the fiscal stance as a whole is propagated to devolved areas of the UK. It depends on what the UK Government decide is the appropriate amount of spending that will happen. Therefore, given that the fiscal stance is a reserved power of the Treasury, it ensures that that is propagated in an appropriate way, or in a way that is defined and can be predicted, but it also ensures that the decisions that are taken affect not just England, but Scotland, Wales and Northern Ireland, and follow similar paths, even if the levels of and additions to spending might be slightly different from population proportions. As David pointed out, differences in population growth can and do affect the outcomes of the formula. All the needs assessments—in 1979 and 1993, and the Holtham commission in Wales in 2010—point to the relative need on a per-person basis in Scotland being more than fulfilled by the level of spending that actually happens, which is not always the case for Wales and Northern Ireland. In those countries and regions, that has led to a push for a needs-based floor, whereas that would not apply in Scotland in the same way, because Scottish spending is already at a higher level than what we think any relative needs assessment would lead us to conclude. There are a lot of difficulties in any assignment of spending on the basis of policy that is decided at UK Government Department level. Ultimately, the point of devolution is to allow for different choices and for these decisions to be made somewhere else, so you have to devise a formula for compensating and making sure that the devolved Administrations have, in some way, the ability to plan for what they want to spend on and what their priorities will be. There were previous versions of the formula. They all have their benefits and drawbacks. None of them will be perfect, especially if we want them to be simple. The Barnett formula is very simple, which is great, because it means that it is very predictable and easy to calculate, and depends on only on a few parameters, but the drawback is that it cannot take into account all the very granular changes in things such as population density, which might mean that, for example, it is more expensive to deliver some services in Scotland than it would be somewhere else less rural.

The Scottish Fiscal Commission is politically neutral, so there are certain things where we would not make any comment on what formula should or could be used, but it is probably worth pointing out that the Barnett formula is still very important in determining the size of the Scottish budget. Even with tax and social security devolution, it is still the biggest component. It is then complicated by the fact that you have had tax and social security devolution, where you have block grant adjustments, which further depend on decisions by the UK Government on tax and social security. You then have borrowing powers and some additional flexibilities. Although the Barnett formula and the way that it is calculated provide the Government with certainty on some aspects of their funding, the Scottish budget has also become a lot more variable over time and a lot more challenging for the Government to manage.

Why would that be?

The Barnett formula and the block grant that comes from the UK Government can vary as they change their spending decisions. The budget then also has the additions for tax devolution, so if income tax revenues go up, the Scottish budget has more; if it goes down, it has less. UK income tax revenues then affect the block grant adjustments, so you have that side of the equation. Then we have this complicated process, whereby we forecast Scottish tax revenues, and the OBR forecasts UK tax revenues, which are used to calculate the block grant adjustments. When we get the outturn data, an adjustment is made to the budget, which is called reconciliation, which is variable as well. You just have a lot more of these components. You are seeing similar things on the social security side as well. Instead of having the one moving piece that you had pre-fiscal devolution with the block grant, you now have several moving pieces, which are all going up and down and just need to be monitored. It is still very important, but there are other factors at play in how the budget is determined.

I have a question that is probably for João and David. Until we are at the point of convergence, which I suspect we might come to later this morning, would you say that the Barnett formula is working to the benefit of Scottish people in so far as it allows for higher levels of public spending?

We should remember that the formula affects just the changes in spending. In essence, the changes in spending are what are driving that convergence process, which over time will slightly erode that higher level of spending in Scotland, at least if current projections for spending and population growth hold true. At the moment, the level of spending is definitely to the benefit of Scotland. As João said, there have been three or four assessments over the years of the relative spending needs of Scotland compared with England. The most recent was undertaken by Gerry Holtham, which was mainly to look at the Welsh situation, but he did produce estimates for Scotland and Northern Ireland with some more uncertainty about them. At that point, in 2010, he estimated needs in Scotland to be around 5% higher per person than in England. At the moment, spending is about 20% to 25% higher in Scotland as a result of the block grant. Things might have changed over the last 10 to 15 years. If anything, the methodology used by Gerry Holtham might understate what might be implied by, for example, how the Scottish Government think about spending needs. In effect, his methodology looked at how spending was allocated within England. Unlike Scotland, England does not really have many rural and remote areas, and the Scottish Government allocate a lot of money to those areas. If you used a formula a bit more like the Scottish approach to rurality and sparsity of population, you might not get 105%; you might get 110%. It is clear that, at the moment, there is a good chunk of money—a few billion pounds a year—that the block grant is delivering on top of what a spending needs assessment would give to Scotland. The question is: in the longer term, will that sustain or will it be eroded by convergence? From a UK-wide perspective, there will be a consideration about whether that is fair to other parts of the UK, such as the north of England, for example.

The Barnett formula, by its nature, as David said, just takes the additions and calculates the share of Scottish additional expenditure. Essentially, it locks in the differences that were there to begin with, subject to any changes over long periods of time to population growth, and the fact that the additions are slightly smaller in percentage terms than they would be if you just took percentages of the additions. Essentially, what is driving the higher levels of spending in Scotland is the fact that, at the time when the formula was introduced, spending in Scotland on a per-person basis was already higher than in England. The formula does not do anything to make it lower than England. It might, at the edges, create some convergence. There was some convergence until the early 1990s, but, mostly because the population grew so much more quickly in England, and continues to do so, that convergence has been eroded. The formula itself just locks that in. It is not the formula itself, but the decisions underlying the starting point that caused spending to be higher in Scotland. Although spending is something like 20% higher on a per-person basis in Scotland, the biggest difference is in capital. It is slightly higher in day-to-day spending, but much higher in capital. You might think that there are some good reasons for the levels of capital spending that might be needed to connect more rural communities. The fact that Scotland operates and the Government support ferry services, for example, is perhaps a perfectly good reason why you might expect capital spending to be higher in Scotland, but, when you put it all together, the relative needs assessment would perhaps not produce as high a number as the formula does.

My question is to Claire, following on from her earlier evidence. The Scotland Acts of 2012 and 2016 provided greater responsibility over taxation and social security, for example. The block grant adjustment for 2025-26 is £5.6 billion, but the Scottish Fiscal Commission forecasts that spend is at £6.9 billion. The gap, therefore, is over £1.3 billion. Is the gap expected to grow in the future with recent Scottish Government announcements? If so, where is this money found from? Is it found from other devolved budgets? What fiscal risks does this bring?

There are some good questions in there. Let us talk about social security devolution. There are several reasons why social security spending in Scotland is higher than the block grant adjustments. They predominantly relate to policy decisions that the Scottish Government have made. The Government have decided to introduce new payments with no funding, the Scottish child payment being an example. They have changed their approach to delivery, so they want more people to be able to more easily access the application process for adult disability payment and child disability payment. That means that we expect to see more people successfully applying. They also have a more expansive approach to replacement payments. The Sure Start maternity grant, for example, was replaced by the Best Start grant. There are more payments, which are made at higher levels. Finally, there is mitigation of UK Government policy, so decisions that the Scottish Government have made on things like the bedroom tax, which is mitigated through discretionary housing payments, or the recent announcement about the two-child limit and universal credit. These are predominantly reasons that spending is higher on social security in Scotland. You are completely right, and we think that spending is going to be higher than the funding received from the UK Government. We have thought in the past that that would grow, and our previous forecasts had that growing quite significantly relative to the UK. Our most recent forecasts—and what matters here is what the OBR thinks is happening in the rest of the UK—do not see quite the same level of divergence in spending, which is predominantly because spending on disability payments in England and Wales is growing, and is thought to continue to grow more quickly in the future. Because of what is happening in England and Wales, where we are expecting spending to grow, the Scottish Government will receive more money through the block grant adjustments for social security. That means that the divergence in spending will not increase quite as much. If social security spending in England and Wales does not increase on the disability payments by as much as is expected, that will potentially mean that the Scottish Government have more of a gap to fill. Your final question was about how they have to fill that gap. Basically, it has to be funded from elsewhere in the Scottish budget. It represents a policy choice that the Scottish Government are making to allocate more of their resources to social security spending, which means that they spend less on other things, or they can do that through raising taxes to fund that gap. In our most recent forecast, we think that, by 2029-30, the Scottish Government are going to be spending 14.9% of their resource budget on social security spending. That is a policy choice that the Government are making.

I would very briefly say that, in 2025-26, for example, the net income tax position is forecast to be £1.2 billion. The social security position is forecast to be about £1.4 billion, which suggests that the additional income tax revenues raised are being more than spent on social security benefits, so there is a little bit of a net squeeze on the public service budgets. Q8                Mr MacDonald: The Scottish Government have called for full fiscal autonomy. How does that work in practice? Claire, do you want to have a go at that one?

I might hand that over to my colleagues. We focus very much on the current arrangements and do not take a view on how the arrangements could work. João or David might want to come in.

What full fiscal autonomy means is that, effectively, all taxes, as far as possible, would be devolved to the Scottish Government. If Scotland remained in the UK, you might have, say, the tariffs imposed on external trade still being managed by the UK Government, but all taxes otherwise would be devolved to the Scottish Government, as would all spending, including on all the benefits, such as the state pension, that are currently reserved to Westminster. Again, there may be some elements of spending, such as defence and foreign affairs, that were still pooled at the UK level, and the Scottish Government would pay, if you like, a fee to the UK Government to do those on their behalf. This would give the Scottish Government, at least in principle, more power to vary a much wider range of taxes and to potentially choose a different fiscal stance, because they would now be responsible for borrowing, and so they could choose to borrow more. The key question is: how much power would they have in practice? As it stands at the moment, if you count all the revenues raised in Scotland, including those that currently go to the UK Government, on a per-person basis, including oil and gas revenues, they are perhaps about the same as the average across the UK as a whole. Public spending, including that currently done on behalf of Scotland by the UK Government, is about 12% higher. With full fiscal autonomy, that fiscal transfer that currently exists would likely have to be removed, because Scotland is keeping all of its revenues and is responsible for all of its spending. In the absence of faster growth in revenues or a reduction in spending, that could open up a bit of a budget deficit for Scotland under full fiscal autonomy, so it would be quite a big change from the current constitutional set-up. It would most likely be much closer to an independence situation in terms of how the public finances are managed, except that there would still be some taxes, such as external tariffs, and some spending, such as on defence, managed at the UK level.

We do not know exactly how it would work. David said “as far as possible”. That is doing a bit of heavy lifting, because there are a lot of taxes that are organised on a UK-wide basis, and there is currently no way to split them without creating a whole new system. We have seen that with the assignment of VAT revenues, which has never been taken forward. To be fully taken forward under some sort of full fiscal autonomy would require agreeing to assigning VAT revenues. There are lots of problems about variability and what even constitutes Scottish revenues. Also, if you create a separate system, it would require thinking through what kind of input and output taxes would have to be collected and where, and what systems you would have in place. Similarly, where do you assign profits within the UK? Do we need to have a full transfer pricing system within companies that operate both in Scotland and in the rest of the UK? There are a lot of things to solve there. That is not to say that they are not solvable, but they require choices that might, at some point, be more or less palatable to different parts of the UK and Scottish Governments, depending on how they fall. The variability, in particular, is a real concern. One of the main reasons why VAT assignment has not been taken forward is that the variability in estimates is such that you might have to cut spending, just because your sample size says that you cannot be precise enough to estimate things properly. You might have to cut spending just because your statisticians say, “In this sample, the revenues have fallen by this much.” That is not a great position to be in, and it is not very easy to explain to the general public why they then cannot access service X, Y, or Z.

As I said, it is important to distinguish between the initial starting level of funding for the Scottish Government and the operation of the Barnett formula. As discussed, before the Barnett formula came in, Scotland was receiving a higher level of funding per person through long-standing arrangements that went back to the late 1800s—I think it was a Chancellor called Mr Goschen. That meant that, back in 1978, Scotland was getting about 30% to 35% per person more than in the rest of the UK. As João said, during the 1980s, there was high inflation, quite high economic growth and quite strong growth in the overall level of spending. The operation of the Barnett formula means that, when spending is going up a lot, you get a lot of convergence. Let me explain why. Let us say you start with £100 per person in England, and £130 per person in Scotland, which is 30% more. Let us say that, over the decades, spending doubles in England to £200, which is a £100 increase. If the population grows the same, etc., a £100 increase to Scotland puts it to £230. £230 compared with £200 is not a 30% difference any more; it is a 15% difference. During the 1980s, there was this convergence in spending, but, since then, as João said, there has not really been much convergence in spending. That means that, at the moment, spending per person in Scotland is about 20% to 25% more. Looking back at the needs assessments, that is probably a bit more than those would assess Scotland to need, relatively speaking, accounting for things like sparsity of population and so on, although there could, of course, be a lot of debate around the weights that you apply to that. Coming back to the formula itself, the issue is that, at the moment, if you say, “Let’s look at changing the formula, so that it is not just a pounds-per-head, population-based increase, but one that tries to take account of, ‘You start off with higher spending, therefore you need a bigger pounds-per-person increase to maintain the same percentage increase’,” the risk is that the UK Government will say, “Let’s look at the level of spending and open that can of worms as well.” From the Scottish Government’s more self-interested position, it might be, “Let’s look 10 to 15 years down the line, when there may have been a bit more convergence. Does it make sense for us to reopen this can of worms?” In the short term, while the formula is not that friendly to Scotland, the level of spending that you are starting with is, and so, from a Scotland-only perspective, it might make sense to wait and see.

The Goschen formula that was introduced in 1888, and which David mentioned, was indeed named after the Chancellor of the Exchequer then. The whole idea of that was to freeze the differences in levels of spending. The Barnett formula takes a different approach, in which there is some convergence if everything grows at the same rate that is used in the calculation. Probably the question to ask is whether the outcomes of the Barnett formula are fair, and then you can have a discussion as to whether the level of spending given by the starting level, and then the operation of the Barnett formula, yield what is needed for Scotland. It would be unfair to look at the formula and just think, “Is the formula fair?” without considering, as David said, the starting point. There can be different opinions and different weights that you attach to different things, and you can come to different conclusions on whether it is fair, but you would have to take into account both the starting position and the operation of the formula.

The commission would not take a view on that.

It would depend on how that was operated. If it was to operate as the needs-based elements in Wales and Northern Ireland are, that would be to the benefit of Scotland. Those needs-based elements have not assessed the level of spending for those countries; they have just tweaked how the Barnett formula operates for the changes. Because there was a concern in Wales, and latterly in Northern Ireland, that there had been quite a bit of convergence, particularly in Northern Ireland, in funding levels and that it was going to fall below that needs level, they said, “Let’s top up the increments that you get through Barnett.” Wales now gets population plus 5%, and Northern Ireland population plus 24%. That slows down or, in Northern Ireland’s case, probably stops and slightly reverses the convergence. If Scotland were to negotiate and say, “We think our needs are 10% higher, so we should get needs plus 10%,” and the UK Government were to agree to that, that would be good for Scotland. In Wales and Northern Ireland, the situation was such that there was a credible case to say that the level of spending was starting to potentially fall below that needs-based level, so the UK Government felt that there may be a case to change the formula to prevent it. Given that most of the evidence is that Scotland is some way above that needs level, they might try, but the UK Government might be more reluctant to make that sort of change in Scotland’s case, although that sort of change would benefit Scotland financially.

I am not a politician, so I am not sure of the political argument to make. If the argument was made, “We need relatively more funding because of our needs and our sparsity of population,” the UK Government might say, “You already get more funding and, compared to these needs assessments, perhaps that gap is already big enough.” Perhaps a stronger case, if you wanted to make one, would be to say that the mechanism of the Barnett formula is now different in Scotland from Wales and Northern Ireland. That is a point that you should look at. As I said, it would be a political negotiation. I would be surprised if there was quite the same openness to making that change for Scotland, given the starting point in terms of level of spending, but it is a political negotiation.

Introducing a needs-based top-up to the annual increments in the Barnett formula would increase the Scottish Government’s funding. Moving towards a needs-based assessment of the level of funding for the Scottish Government would reduce the Scottish Government’s funding.

João is nodding, so I think he is in agreement. Thank you, João.

David, you have published research that suggests that spending per capita in Scotland relative to England will fall in the next 30 years. Could you briefly explain your findings?

As João, Claire and I have said, the Barnett formula, by its design, gives a population-based increment on top of the initial funding that the Scottish Government receive. If England gets a £10-per-head increase in funding, so does Scotland under the Barnett formula. If you start with a higher level of funding, a given cash increment is a smaller percentage increase. If you start with £100 a head, a £10 increase is 10%. If you start with £200 a head, it is a 5% increase. Over time, that leads to a convergence in funding. There are other factors that operate against that. The Barnett formula does not really account for the fact that Scotland’s population is growing less quickly than England’s, so that increment of funding needs to spread across fewer additional people, which slows down convergence. We did some modelling looking at different scenarios for inflation and for above-inflation growth in spending, and the given population projections from the ONS, to look at what that might mean for the relative funding levels of Scotland compared with England. At the outset of the projections—now—it is about 20% to 25% higher. Given reasonable forecasts for inflation, spending growth and population projections, that might fall to around 112% to 115% over the next 15 to 20 years, so still substantially above England’s level, but not as much as now. The Holtham commission said that the needs in Scotland were around 105% of England’s level. 112% to 115% is still well above that. If you thought that Holtham’s methodology was underweighting things such as sparsity of population or the ill health concentrated in areas around Glasgow, for example, and that that should be more highly weighted, you might think that that 112% to 115% is starting to approach the needs-based level. That might be the point where Scotland wants to start thinking, “Do we want to open this issue up and move towards a needs-based formula?” Of course, that is all from a Scottish Government position. From a first-principles, “What is a good design?” position, you would not be doing it on such a zero-sum basis.

My follow-up question is one that you might have touched on earlier. How could a needs-based adjustment to the Barnett formula impact those projections?

If we were to say, “Let’s do what we do in Wales and Northern Ireland. Let’s not give Scotland a population-based increment, as it gets now, but, like Wales, population plus 5%,” that would slow down the rate of convergence and would mean that, rather than converging to, say, 112% to 115%, as it would on current population projections and so on, it would converge to a somewhat higher level, maybe 118% or so. I have not crunched all the numbers. It would slow down and mean that there was convergence to a higher level. The question is: would the UK Government be open to that sort of arrangement? That is a political question for the UK and Scottish Governments.

You have answered my question, which was around outlining convergence. We have heard a comprehensive explanation already this morning. Can I ask João and David—and feel free, Claire, if you have thoughts on this—whether convergence, or the “Barnett squeeze”, is an intentional design feature of the formula?

The idea behind the formula at its outset was to have some convergence, because the fact that the Goschen formula preserved those differences, plus the fact that there were top-ups on the Goschen formula outcomes, meant that the gap was growing. There was a question as to whether that was the right outcome, and there was a lot of discussion in the early 1980s between the Wales Office, the Scotland Office and UK Government Departments as to whether that was fair. The idea is that there should be some convergence over time, but it is not a drastic level of convergence. It is not something that happens in one year and you are immediately slashing things to be equivalised. But if population growth is the same, then over time, spending additions will mean that there will be convergence.

There are a couple of points there. The first thing is that, as João says, there was an intention for there to be some degree of convergence. It was also introduced as a bit of a stopgap measure at the time that they were thinking about devolution to Scotland and Wales back in the late 1970s. Joel Barnett, who it is named after, seemed to have quite mixed feelings about it. One time, he said that this formula that was named after him was like an albatross around his neck. Other times, he said that he was very proud that it had lasted so long, but it was never really expected to last this long. In terms of convergence, if you look back since the late 1970s, the absolute level of spending has probably increased about tenfold. If the Barnett formula had been working as was intended, population had grown at the same rate, and there was a 30% increase in spending in Scotland at the start—£130 compared with £100 in England per person—it would now be £1,000 in England and £1,030 in Scotland. That 30% difference would have become a 3% difference, but it has not; it has become a 20% to 25% difference. There are a couple of factors in that. First, particularly before devolution, Scottish Secretaries were quite canny at negotiating top-ups to the Barnett formula through a formula bypass. As João said, population growth has been a lot slower in Scotland than it has been in England over that period. Those two factors have slowed down convergence. There were also some flaws in how the Barnett formula treated business rates in the 2010s, which also slowed it down. Going ahead, we would expect there to be some convergence, as I said in my answer to a previous question—not to a 3% or 0% gap over time, but, given current projections, maybe a 12% to 15% gap.

Two years ago, we published a fiscal sustainability report. We looked at the Scottish Government’s budget over 50 years. We projected what would happen to block grant funding, tax revenues and the block grant adjustments, and also what would happen to the Scottish Government’s spending. We ran a couple of scenarios based on the population projections at the time. New population projections were published yesterday, so we are updating the estimates as I speak. We looked at what would happen, based on those previous population projections, if you had faster or slower population growth in Scotland. What we showed was that, if you had lower population growth in Scotland, the block grant funding would grow slightly more slowly, but your spending growth would be a lot slower. If you are looking at the whole Scottish budget, not relative to the rest of the UK, that slower population growth in Scotland could be beneficial to the Scottish budget. Conversely, if Scotland had higher population growth, you would have slightly higher growth in the Barnett formula and in the block grant funding, but much higher growth in spending in the Scottish budget, so you would have a more negative fiscal position going forward.

There is a slight irony in the situation at the moment. Because the Barnett formula does not account for differential population growth, you get more per person to spend when the population is growing less quickly. That is a challenge to one of the arguments that the Scottish Government make about, “Let’s boost population growth to boost the economy.” It will probably boost economic growth, and it might ease skills shortages, but, because the Barnett formula does not really give you more money when the population is growing more quickly, that is not necessarily good for the public finances of Scotland.

What are the different ways in which the Barnett formula could be formalised, including putting it on a statutory footing?

In some sense, it is up to the UK Parliament to decide how to put it on a statutory footing. In the same way that the Sewel convention was put into legislation, it could be added as an amendment to primary legislation, or it could be part of some other revision to the constitutional settlement. There is a question as to whether that would be beneficial. I am not qualified to say constitutionally whether that would be beneficial, but, as a user of the constitution, if you will, I cannot see that that would make a huge amount of difference. The Sewel convention is in legislation and has been bypassed before with the get-out clause of “not normally”, which is what the legislation says in terms of consent. The UK has an unwritten constitution, in which tenets of the constitution that are held to be very much fundamental are not written down. Whether it is or is not written down does not make a huge amount of difference, from our point of view. Because no Parliament can bind its own successors, even if it were in legislation, it could theoretically be repealed, so I cannot see that that would make a huge difference.

I agree with most of that. Putting it on a statutory footing might change the politics around it slightly. João is right that, at the moment, not only the operation of the Barnett formula—when it applies or not—but its existence, effectively, is in the gift of the Treasury. Now, it is in the fiscal framework agreements with the Scottish and, I believe, other devolved Governments that it exists, but those intergovernmental agreements are not enforceable in court. By putting it on a statutory basis, in principle at least, the operation or the bypass of it could be challenged in court and would require primary legislation. Depending on how it was written, that might change the politics around it. I agree with João that, because of the operation of the UK constitution, because Parliaments cannot bind their successors, and because we do not have a written constitution such as the finance constitution that Germany has, it would be mainly operating through how it changes the politics around the Barnett formula, as opposed to putting any real long-term legal constraints on the Government.

Regardless of whether it is on a statutory footing, there are elements of transparency around how it operates that could be improved, which would not require that. The Treasury publishes the block grant transparency document, which is very helpful. It does not publish it very regularly. The last version was published in July 2023. We have had quite a lot of significant changes since then. At every fiscal event, if the Treasury could clearly publish how it has operated, which elements of additional funding are being baselined for future years, and which are non-recurring, that transparency would help Parliaments here and in Scotland, Wales and Northern Ireland understand what the funding position is, and allow ourselves and other stakeholders to really understand how the Scottish Government’s funding is changing. That might lead to greater scrutiny on some of the other issues that you have been talking about in terms of the level of funding and how it is changing over time.

Would formalisation impact the effectiveness of the formula in practice? If so, how?

Formalisation of any formula—not necessarily just the Barnett formula, but also if it is replaced—could change the degree of scrutiny around the operation of the formula and how it is changed, and, as I said, it could allow it to be challenged in court while the law is in place. But the real impact would depend on the extent to which it increased the scrutiny by Parliaments, so there would be a job for people in this Committee and elsewhere in Parliament to make use of those powers. I agree very much with Claire that changing the legal basis of it is one thing, but even without doing that, there is so much that could be done to improve transparency, not just around the block grant transparency report, although that is very important, but around what is called the statement of funding policy, which sets out the rules and how they have calculated what is known as comparable spending, which is the basis on which they decide the increment in England that goes into the Barnett formula to determine the Scottish budget changes. There is a lot of scope for improved transparency, particularly in the context of the new formulas for Wales and Northern Ireland, which have some needs element, and a potential slow convergence in Scotland, which might push it closer towards needs levels in Scotland. Publish regular assessments of what the level of funding in Scotland is compared with England, and how that compares with some estimates of what funding is needed. There is a lot more scope for transparency to improve debate and the operation of the formula.

You may already have answered this, but would it be beneficial to formalise the Barnett formula or put it on a statutory basis?

I cannot see that it would necessarily do any harm. I cannot see that it would transform things either. I appreciate that that is a bit of an equivocal answer, but I would argue that some of the perhaps more pressing issues are some of the ones that Claire and David brought up—particularly the statement of funding policy and how that is calculated. The fact that the Treasury is the ultimate arbiter of what those are is the really crucial point. I do not know that that necessarily requires formalisation of it. It probably requires agreement on having some sort of third party that can arbitrate disputes, much like there are arbitrations of other disputes. That would probably be a way of thrashing out some of the disagreements about things such as bypasses of the formula or designating some things for the benefit of the whole of the UK when it might be argued that they are not.

Some funding is delivered outside of the Barnett formula—projects such as the Glasgow city region deal, which greatly benefits Glasgow and the west of Scotland, or the recently announced UK national wealth fund. To what extent is there a clear rationale for when it is and is not done? Does that in any way undermine or supplement the Barnett formula?

There are several areas where the formula has been bypassed. You have mentioned a few that are associated with economic development. The UK shared prosperity fund and the levelling-up funds also bypassed the Barnett formula. Historically, rail funding was done separately from the Barnett formula, as was agricultural funding. In each instance, because of the particular nature of these spending items, the way to assess how much we spent in different countries needed to be done on a bespoke basis. For example, the UK SPF and agricultural funding was largely to replace EU funds. They did not want a cliff edge and so decided to grandfather those allocations in. They have now, effectively, been rolled into the Barnett formula going forward. The same is true of rail funding. The city deals that you mentioned, as well as the national wealth fund, also look, to some extent, not just at needs but at opportunities. There is an element of appraisal of the potential benefits of spending in terms of certain investments. In a system of financing across a country, I think having a range of mechanisms—sometimes a standard formula, and at other times something more bespoke for things that a national Government want to allocate across the country, with a set of rules that they determine themselves—is a reasonable basis. Even highly decentralised countries such as the United States do not have a federal Government that is just completely hands-off on areas where the devolved Governments also have elements of control. What is important is that there is a degree of co-ordination and communication on spending in areas where both devolved and central Government have a role—for example, economic development, where they are both acting in that area.

There is probably more need for transparency, too, and to be very clear. Both the Scottish and UK Governments could be very clear about which parts of the block grant are coming through the Barnett formula, and which parts are non-Barnett funding. Tracking that over time can be difficult. Again, what is going to happen with that funding in the future is sometimes very uncertain. Agricultural funding has been outside the Barnett formula for a few years and is now being brought within it, but how it is working is not necessarily completely transparent and clear to everybody involved.

João, you described the Treasury as the “ultimate arbiter” of what spending is Barnett-ised and what the comparability factors are. Should other parties be consulted on those decisions? If so, who?

The Treasury is definitely the ultimate arbiter. It publishes the statement of funding policy, and what it says essentially goes. There has been a lot of debate relating to this, not necessarily within Scotland. In Wales, there has been a lot of discussion about whether HS2 should be an England and Wales or England-only project, for example. There was the Olympics case, where spending was deemed to be for the benefit of the whole of the UK and, therefore, non-Barnett-able, although that is a while ago now. There is definitely a strong case that other parties should be consulted, but at least if there is a reasonable case to raise a dispute, there should be some sort of mechanism for that to be raised. At the moment, that does not exist. It all relies on political pressure. Some sort of independent, jointly nominated panel that might arbitrate on this would be a reasonable solution to ensure that the Scottish and UK Governments can come to an agreement about any disputes that might occur in the future.

This could be another example where a statutory basis could have a role. The design of the allocation process is ultimately a political decision. Therefore, the rules need to be determined by politicians. But on the operation of those rules, I think there is a role for potentially either an independent voice or a legal voice to determine whether they are always being followed. In Germany, there is a finance constitution, and one rule is that any burden placed on lower levels of Government by higher-level Government needs to be funded. The lower levels of Government can go to the constitutional courts if they think that that has not been the case, and the court can make a judgment on that. You could have a similar situation, whether through arbitration or special courts, to decide whether the rules of the Barnett formula, which have been agreed by the UK Parliament and so on, are being applied correctly.

What more could the UK Government do to ensure that the process is more transparent?

On the transparency point, there is publication of information on how it is calculated, how much the Government are receiving, how much of that is and is not baselined, and how much is outside Barnett. That could be clearer, and it could be clearer between fiscal events in terms of how it is changing over time. It is sometimes hard to work out. Each Government might say that their funding is going up. Some will say that it is going down. It is all calculated on a different basis. The clearer that information is, the more independent voices like us can then say, “This is what we think is happening.” There are elements around the Barnett formula itself that could be more transparent, but there are also general things that the UK Government could do, which would help make the fiscal process in general clearer, such as a commitment to multi-year spending reviews, or clearer timings for fiscal events, which would mean that there is greater certainty for the Scottish Government in planning their budget. Those things would be helpful. There was also the recent announcement at the UK Budget about employer national insurance contributions, with an unclear level of funding allocated, which means that the Scottish Government have had to set their budget without including the impact of those employer NICs changes. That is probably more significant in terms of the impact on the Scottish budget than the exact operation of the Barnett formula. It is more the general certainty around how the fiscal landscape is looking, what changes are being made, and what impact they have on the funding. It is about confirming that early and publicly, and then allowing the devolved Administrations, not just in Scotland but in Wales and Northern Ireland as well, time and certainty to plan their budgets.

Transparency is often an issue, not just for the UK Government, but also, to some extent, for the devolved Governments. There is also more that the Scottish Government can do to be transparent about their own budget decisions. For example, sometime this week or next, we are expecting what they call the spring budget revision, which is a bit like the supplementary estimates at the UK Government level. They are published on a different financial basis than the budget figures that are used and discussed in general discourse, so trying to work out how the figures are changing when you have these updated spending estimates is quite difficult. All round, there is more scope for transparency, both between the UK and devolved Governments, and within the devolved Governments themselves.

I very much agree that the different bases are a huge issue that needs to be resolved. There have been some movements toward a more comparable basis from the Scottish Government, which has been welcome, but it still is a different budget basis from what the Treasury publishes. I do not really see why these publications should be referred to on different bases, and it makes it very hard for us to explain to the general public why that is. The other thing that I would mention is some of the transparency releases that the Treasury puts out. It is all well and good for us to trawl through them, but a lot of data does not mean information; it just means a lot of data. There is a role there to try to explain things a bit more and to make sure that people can understand what it is and that interested members of the public can go and find it out, instead of having to do it for a living, like we do. I am not trying to do us out of a job, but it would be nice if we could move beyond, “We had to spend this much time going through it, and these are the results,” to explaining what it is.

I want to get the words “pivot table” into this evidence. The Treasury could make its pivot tables in Excel more useful. There are simple things that just make the data that is there more accessible, not just for general members of the public, but even for us. It might sound like a very technical, minor thing, but, as João said, data is just an additional thing to wade through without the infrastructure to help you understand and make use of it.

Thank you all for those answers. The national insurance aspect probably ties in with a wider point, because not only do we have the Scottish Government in a position where they are setting their budget without clarity in that respect, but local authorities across Scotland right now are in the exact same position, making determinations in relation to potential council tax rises without having certainty about what the national insurance rises look like for them. That probably goes to the wider point in relation to the winter fuel allowance as well, where there was a systemic change in the funding that is available, with 20 or 30 minutes’ notice that it was happening. Have you sensed any appetite or change within the new Government to provide earlier and swifter clarity to the Scottish Government in that respect?

It is a good question. The change to winter fuel payments at the UK level was the first time that there had been a significant in-year change to a social security payment, which the Scottish Government chose to respond to by replicating in Scotland for this year. There are certainly signs that the whole system could become a little bit more suitable for the Government in terms of clarity around when fiscal events are taking place. We have the first UK spending review of this Parliament due in June, with a three-year outlook, and then a commitment to have that on an ongoing basis. If followed through, those should give the Scottish Government more certainty, but we have also heard commitments before about moving to one fiscal event, which has not happened. It is a bit hard for me to answer that in a politically neutral way, but there are signs that it could become clearer and smoother for the Government, which we would certainly welcome. We have made a lot of statements about how the Scottish Government should be doing more multi-year planning to give certainty to local authorities and public bodies, and there are elements where, when you have a very uncertain funding position, there is more rather than less need to do that planning. At the same time, it is important to recognise that, if you do not have the outlook from the UK Government in terms of the funding position, which is ultimately the most significant component of the Scottish Government’s budget, that is a very difficult thing to do, both practically and politically.

As well as there being formal communication between Governments, there is informal communication between Treasury and Scottish Government officials. They cannot give full details about exactly what the spending announcements are and the political policies that are being made, but it gives a little bit of mood music to help the Scottish Government understand what might be coming down the line. I did a project back in 2021 with colleagues from Fraser of Allander and from Stirling University, which asked a bit about how the fiscal framework was working during covid. The information that we got was that that sort of communication was very prevalent during the covid pandemic. There was lots of back-and-forth communication between Treasury and Scottish Government officials during that period of, perhaps, UK-wide crisis. My hope would be that, while respecting the need to keep certain political announcements confidential, there can be a degree of communication at the official level to help officials understand what is coming down the line. That potentially requires officials to withhold some information from their political masters.

Obviously there were the changes that were announced earlier in the year, and although there is better mood music and signs that there is more dialogue than there had been before, which was evident in the fact that the Scottish Government were able to plan better in terms of their autumn budget revision than they might have done had they not had any information about what was coming down the track in the autumn Budget, there are still some difficulties, which are going to be much tested in the next couple of months. One in particular is the fact that the Chancellor of the Exchequer is committed to having only one full fiscal event a year, but with the fiscal rules potentially being breached if she were to do nothing, there is a very real chance that there will be changes in the spring statement, which would come after the Scottish Government have already passed their budget for 2025-26. Therefore, that is locked in. The Scottish Government can make budget revisions during the year, but they have much fewer levers to do that in year than the UK Government have. Therefore, there is an imbalance of immediate levers. The Scottish Government might find themselves having a different Barnett settlement than they expected—either more or less, depending on the situation; in this case, we think that it could be less—and then the question is: is that at a time that is conducive to good policymaking and decision making by the Scottish Government? I would argue that it is not. The fact that the Scottish Government have more fiscal powers, as Claire explained, means that they are more dependent on UK Government fiscal events, because of the interactions between the forecasts of the SFC and the OBR. The other thing that is slightly worrisome in terms of the timing of these fiscal events is the spending review being postponed until June. Claire will probably not be looking forward to that too much, because it is likely that the Scottish Fiscal Commission will be doing a forecast either before or around the same time as that spending review. It just does not seem like that should have to happen before decisions about the next few years are communicated to the Scottish Government and the Scottish Government have the opportunity to decide what to do with the allocation. So there are some difficulties with timing that have not been resolved. Perhaps they cannot be fully resolved, but we are going to face a stern test in the next couple of months as to whether that is going to rear its head again.

That is incredibly helpful, and all of us are very conscious of that in the weeks and months to come, given the constraints that the Chancellor has in place with regard to those fiscal rules. To take a step back, you talked about the potential for a dispute mechanism to be put in place in order to resolve challenges that may or may not exist. We are months down the line from the national insurance announcement. Does that provide an argument, not necessarily for a dispute mechanism, but perhaps for a statutory mechanism to be put in place, whereby the UK Government have to provide clarity to the devolved Governments with regard to where spending sits? We seem to be sitting in this system at the moment without clarity, and it is at the behest of one party to provide that, while other parties are sitting waiting. Could a formal structure help there?

I have no comment on that question, I am afraid. Trying to put in place a formal rule here might run into issues, in the sense that the UK Government will often make changes in year for good reasons, such as topping up funding during the course of the year, so any rule on this would need to be written to allow some degree of flexibility within the year. I would need to think more about how this could work in the context of issues such as national insurance before I could give a proper answer.

There are clear challenges around this particular example, but there are also past examples where the Treasury has shown great flexibility to the Scottish Government. For example, when late changes have come through supplementary estimates that would have a negative effect on the budget, the Treasury has allowed them to be deferred. When there were drawdown limits on the Scotland reserve, and the Scottish Government had funding there and they had large negative consequentials, they were allowed to draw down more than they should have been. There are elements where, in this particular example, it is challenging for the Government, but there have also been many examples where the Treasury has shown greater flexibility. There is a bit of seeing how this pans out, what the Governments agree and what the level of funding is. While this is a clear issue for the devolved Administrations, it is also going to be an issue for local authorities and public bodies in England, so it is not a uniquely Scottish challenge.

Perhaps following a slight theme here, I want to ask you about situations where there might be disagreement about a decision to bypass the formula. Claire, does the Scottish Fiscal Commission play any role in such disputes or disagreements?

Essentially, no. There is some provision in the fiscal framework that, where there are particular disputes around spillovers, the Treasury and the Scottish Government could approach us and the OBR, but those mechanisms have never been used. There is a risk in that elements of the decisions around Barnett, in terms of what is baseline and what is non-Barnett, are essentially political decisions, and bringing an organisation such as ours into those would potentially challenge our political neutrality. We have not been asked—and, hopefully, would not be asked—on that.

Do those issues happen often?

That is probably a question that the Scottish Government would be in a better place to answer. There are always some things, but with those two major examples—HS2 and the Olympics—you are looking over a 15-year period. There are probably smaller things that are resolved behind the scenes between the two Governments without becoming public.

That is helpful. João, in one of its papers, Fraser of Allander suggested that there might be need for a third-party resolution mechanism. Do you think that is still needed?

We mentioned that in particular in relation to the statement of funding policy, because, essentially, the Treasury has all the power and there is no recourse. We at Fraser of Allander think that it would be beneficial to have some sort of mechanism, as we talked about before, whereby this could be challenged in a way that is less heated politically and, therefore, more to do with the operation and the underlying principles of it. It is beyond our scope and remit to say whether that is a priority that should happen, but we think it would be positive.

There are dispute mechanisms within the system, both for the operation of the Barnett formula and the statement of funding policy, and then for the other elements of the fiscal framework that Claire mentioned. My understanding is that those work in somewhat different ways. If it is for the Barnett formula, in effect, it goes through a process of being raised through official levels and then up to ministerial levels, through what is called the Finance: Interministerial Standing Committee, which is made up of representatives of the UK and devolved Governments. This is a consultative rather than decision-making body, and the final decision-making power on the Barnett formula still always rests with the Treasury. With the other aspects of the fiscal framework, such as the block grant adjustments or spillover effects, there is the potential for there to be disputes on both sides of the system. Therefore, rather than the Treasury’s view always carrying, if they cannot reach a resolution, it is the default rules that carry under the fiscal framework. There is a bit of a difference between how the Barnett formula is disputed—where, ultimately, if you cannot reach an agreement, the Treasury says so—and the other aspects, where if you cannot reach an agreement the default rules apply. That is where something such as an independent mechanism or the potential to see whether the operation is being applied correctly via a court, if it is made on a statutory basis, could provide a more balanced approach, on the Barnett formula side of things as well.

That is very helpful. We are now going to move on to some questions about the fiscal framework.

The fiscal framework has several mechanisms that are designed to help address risks that the Scottish Government could otherwise face. The first one, which is often not thought about, is the way that the block grant adjustments work. The block grant adjustments for tax, for example, are linked to the tax revenues in England and Northern Ireland. That means that if there is a recession that hits the whole of the UK, which means that Scotland’s revenue goes down, revenues are also likely to go down in England and Northern Ireland, and the block grant adjustments are linked to that, so they will go down as well. Scottish revenues fall, but so do the block grant adjustments. That insulates the Scottish Government from a fair proportion of the risk associated with global or UK-wide economic shocks. Of course, that is not the only kind of issue that can arise with tax devolution. There could be a Scotland-specific shock. For example, if there is a decline in the oil and gas industry that is bigger than elsewhere in the economy, that will have a particular impact on the Scottish Government’s income tax take from oil and gas workers and those supporting them. As Claire mentioned, there are also issues around the forecasts: not only might the forecasts be wrong, but the SFC and the OBR might take different positions on what is going to happen to wage or employment growth, which adds an element of noise to the process as well. To address the parts that the block grant adjustments do not do, the Scottish Government have borrowing powers and some reserve powers. When we initially looked at this, given that the cycles of the Scottish and UK economies move fairly closely together, those borrowing powers looked like they might be sufficient. That omitted the forecasting issue that Claire mentioned, whereby, even if the cycles are moving in step, because you have two different organisations making forecasts with different judgments, which adds some noise into the process, the sorts of forecast errors that you get in terms of the net tax or social security position could quite often exceed the initial borrowing levels that the Scottish Government were given under the fiscal framework. They made some changes in a negotiation in 2023, which removed some of the limits that applied. That has increased the limits in line with inflation going forward. That has definitely improved the situation, but there is a risk that it could go further in a couple of dimensions. One is that, typically, the revenue and spending under risk will go up by more than inflation. We hope that there is economic growth, and social security spending has been going up, with the rising disability benefits bill. Both of those things mean that the amount of money at risk is going up by more than inflation, and I would suggest linking the borrowing limits not to inflation but to some estimate of the funding that is under risk. It could be linked to GDP growth, or perhaps to forecast growth in the rest of the UK revenues and spending on these benefits. The second issue is that the fiscal framework gives the Scottish Government powers to borrow if there is a forecast error for tax or social security spending, but not if there is a shock to public service spending. Let us say that public sector pay deals have come in much higher than expected in Scotland, but the same is not true of England. The Scottish Government cannot borrow to cover that at the moment. They need to find in-year savings within their budget. There is a case for at least a small amount of borrowing for discretionary purposes for the Scottish Government, not just for forecast errors for tax and social security, but to cope with other shocks or other things that arise during the year. Because there is no England-only borrowing in our system, I do not think that we can have these powers being indefinite and the Scottish Government setting their own limits. For fairness to England, effectively, which will not have these borrowing powers, there should be some limits. Saying that the Scottish Government can borrow 1% or 2% of their budget up to a certain cap in total to smooth these other sorts of pressures that might hit the public service budget would improve the situation and give them more flexibility to respond, not just to shocks to taxes and social security benefits, but to public sector wage bills, flooding and other things that affect the public service budgets.

There are certain elements that give the Scottish Government certainty. Income tax is fixed at the time that the budget is set. It is fixed based on what we forecast, and the block grant adjustment is fixed based on the OBR’s forecast of UK Government revenues. As that is the largest tax, that gives the Government an element of certainty in their coming year’s budget. When the forecasts come in later, there is then a reconciliation that applies, but the Government can choose to borrow against that and then repay it over three to five years, so they can smooth the effects of that out a little bit. It is worth saying that, as David has very eloquently explained, they can borrow only for forecast error. If there is a Scotland-specific economic shock, for example, which we forecast and factored into our forecast, and we think that income tax revenues would be lower in Scotland while they remain higher in the rest of the UK, the Scottish Government would not be able to access borrowing for that. That is one potential challenge, although we have not seen it. Generally, as David said, the cycles follow broadly similarly, and the issues have more been around timing of fiscal events, or where—particularly during the pandemic—we had a UK fiscal event and a gap, some significant announcements happened, and then we were forecasting. That presented challenges. The final point that I will make is just a little separate, and it is more about the challenges that the Scottish Government have in general around balancing their budget. They have to balance their budget while using these limited borrowing powers. They cannot overspend, but they do not want a significant underspend either, because they cannot put terribly much in the reserve in relation to the size of the budget. There are a lot of decisions that the Government are having to balance as they go through the financial year, as the UK Government change their spending plans. You get funding through the block grant. Barnett consequentials will come. We also have changes to the tax forecast—not income tax, but the other taxes—as well as changes to social security spending and to the OBR forecast. All of these things are changing, and the Government are having to hit that narrow balancing point at the end of the year. It is important to recognise that that is quite challenging for the Government. In terms of how that should be resolved, David has made a few suggestions. We would not take a view on that, but it is important to recognise that it can be difficult for the Government to do that, and they are having to make decisions that give them flexibility to manage it as the year goes on.

I want to pick up on the point that you made, Claire, about the limits on Scotland’s reserve capacity. I was just wondering if there has been a recent example of the Scottish Government having to hand money back to the UK Government because it has not been spent in time.

I do not think so. Since the reserve came in, they have never hit the limit on it. Generally, in terms of how the Scottish Government manage their budget, they assume a level of resource borrowing and capital borrowing. As the year goes on, we generally see those borrowing assumptions come down, and they borrow less. Effectively, they are using their borrowing limits as a tool for making sure that they do not have too big an underspend. Before they put money in the reserve, they generally reduce the amount that they borrow, and then there is an underspend after those decisions have been made, which is smaller. In recent years, they have not fully used the reserve limits. Five years ago, they were closer to some of those limits, and it was particularly challenging when they had drawdown limits. If they put some money in it, they could draw down only a certain amount of that each year, which meant that you could end up not being able to fully fill it. Now that the drawdown limits have been removed, that is less of an issue. The scale of the reserve is predominantly used for underspends. One of the things that people talk about is if you had a year with bumper tax revenues, you could put that in the reserve and use it to fund spending in future years. It is quite hard for the Government to do that. Because of the size of the reserve, it is predominantly at capacity for underspends.

On that point, there has been a case where Wales has lost money that it has not been able to spend. That was not as part of the normal reserves. During the covid pandemic, in 2020-21, because there was a late increase in funding from the UK Government to the devolved Governments, they let them carry money forward on top of the normal reserve powers. That happened in Wales, Scotland and Northern Ireland, and allowed them to spend more in the second year of the pandemic than was spent in England, where that money was not carried forward in the departmental budgets. Wales had hoped to get dispensation to carry forward more on top of the reserves again into 2022-23. The Treasury said no, and Wales lost a couple of hundred million pounds from that, but the Scottish Government did get the money out of the door in 2021-22 and did not lose it.

Should we be concerned that the forecast reconciliations that you speak about could exceed the resource borrowing limits? For this year, some estimates are forecasting over £700 million, when the limit is £600 million. How do the Scottish Government get themselves out of that mess?

That is an estimate of the reconciliation at the moment. That is based on our and the OBR’s latest forecasts. Both could move around. Particularly on income tax, we are forecasting very large sums, and the OBR is doing the same, effectively, for the UK. Relatively small forecast errors can end up with quite big reconciliations. We did some analysis in the past that looked at the previous limits, which suggested that we could quite regularly be breaching the limits as they were at that point. We are hopefully going to redo that analysis. Given that they are increasing with inflation, but earnings growth and income tax revenues have increased faster than inflation, it is likely that they will be exceeded, as is currently projected. It is worth saying, though, that the forecast at the moment is a reconciliation of over £700 million against a borrowing limit that was £640 million in that year. That means that they can borrow the £640 million, so the gap that they will have to plug in the year will be more like £150 million—I cannot remember the exact numbers. That will then reduce the amount that they have to spend in the year that that income tax reconciliation comes in, and then they will be paying back a negative reconciliation from the borrowing over the subsequent three to five years, depending on for how long they borrow. Effectively, that reconciliation is where they received more money than they should have done in the previous year. It is not necessarily a bad thing that you have a reconciliation. It is an outcome of just smoothing income tax revenues over years, but it does create these challenges for the Scottish Government in how they are managing their budget.

I would say that there is a reasonable argument to be made that this is exactly the sort of thing that we should be worried about. Although I understand why some of these limits were put in, because you are starting a new process and creating a new framework, and you want to encourage responsible behaviour by all parties involved in disbursing the Scottish budget, it probably does not make sense to have such hard limits on this kind of thing. Certainly, they were frozen in cash terms. Now they are going up with inflation. There is a very reasonable argument for them to go up with nominal GDP, which would just take the edge off some of these hard decisions. Ultimately, the Scottish Government might be put in a position where they have to do this reconciliation at a time when there could be a shock and they would have less money to spend on public services than would be ideal. Claire is right that they probably should not have this money if these forecasts were correct, but ultimately they did, and you might be causing the wrong fiscal impulse at the time when these reconciliations come through. This illustrates the point that these limits probably should not exist, and probably should not exist for the Scotland reserve. There is a very good argument that you should allow a Parliament to make its own decisions about how it wants to smooth things across time. I understand that some of the borrowing limits are there because the Scottish Government can borrow at very preferential rates from the national loans fund, and so you want to encourage some responsible borrowing behaviour. As we have put in our submission, there are some more radical ways in which you could reimagine the fiscal framework that would allow the Scottish Government to have more limited borrowing powers from the national loans fund, as well as some other ways of borrowing that would allow them more flexibility, and probably improve the way that they can manage risk in things such as this reconciliation.

This may be a naive question, but if there was an issue, could the Scottish Government not just lift the phone to the UK Government and say, “We need a little bit of flexibility here. Can we get temporary access, or a little bit more, or a little bit of movement somewhere?” Would that not work, especially when we are hearing that the relationship is being reset?

That is how it happened during covid, but that was an emergency situation. It is more reasonable that that would be the case then, but it would be beneficial for everyone in terms of transparency, as well as predictability, and guarding against the fact that relations might not always be as good as they are, or might not improve. In the interests of predictability and guarding against uncertainty, it would be good to have more flexibility there. I do not think that the risks are as high as they were necessarily thought to be when the fiscal framework was first introduced.

One could distinguish between borrowing for forecast errors and other forms of borrowing. A limit on discretionary borrowing by the Scottish Government makes sense. Whether it is borrowing for capital spending, as they currently can, or an expansion of borrowing powers—more discretionary borrowing—for resource to cover shortfalls of revenue that are forecast in advance, as Claire mentioned, or shocks to public service spending, which I mentioned, having some limits for those makes sense. Forecast errors, provided the forecasts have been made by independent bodies such as the OBR or SFC, are not something that the Scottish Government have control over. They cannot say, “Let’s over-forecast our revenues now and get more money in. We can borrow, and this problem gets pushed down the line five to 10 years,” because they are not doing the forecast. The SFC and the OBR are. That makes it more reasonable to say, “Let’s substantially increase borrowing limits for forecast error specifically,” or even, potentially, “There does not need to be a limit for that, because this is just a bit of the operation of the fiscal framework.” What I would say, though, is that, like any borrowing, it needs to come with interest and not just be interest-free, which is something that the Scottish Government said in their submission to this inquiry. I can explain why in a second, if you would like me to, but any borrowing, whether it is for forecast errors or other borrowing, needs to have interest attached. There is a case to say, “For forecast errors, let’s just remove the limits completely,” because the borrowing is for something that is outside of the Scottish Government’s control and not something that they can influence themselves.

Good morning. I am sorry for being late. On the back of what you were speaking about there, should the Scottish Government not pay interest on borrowing when such borrowing is done to cover forecast errors?

I think that it should pay interest, and I will explain why. Let us suppose that the UK Government were to have a forecast error for income tax. At the start of the year, they set their spending plans, assuming that they will get a certain amount of income tax. Income tax turns out to be less positive, so they get less and they need to borrow that money in that year. They need to pay interest on that. In the same way, the Scottish Government set their budget at the start of the year, when they think that income tax earnings will be at a certain level. It turns out that they are not. In the short term, at least, that is not a matter for the Scottish Government. They do not need to pay it off for three years. In effect, they are getting a three-year interest-free loan on that at the moment, and then start to pay interest after those three years, once the reconciliation has been done. Already, to some extent, they are getting a better deal on addressing forecast errors than the UK Government do. Making it completely interest-free means that they get some money up front, and the fact that inflation has not taken place and that it can be put into the bank if you want to means that the value of that money up front is greater than it is later. That is what interest is there for. It is about the fact that money up front is better than money later. It works for the UK Government. That is why we need to pay interest on our debt. It is the same for the Scottish Government.

Someone has to pay the interest. If it is not the Scottish Government, it will be the UK Government paying it. I think it is very reasonable for them to be charged interest.

I was going to ask about the fact that the Scottish Government are required to run a broadly balanced budget each year and are not permitted to overspend. Can you go into some detail about why that is the case? What implications does that have on the fiscal flexibility of the Scottish Government?

Essentially, the rules are set out as part of the statement of funding policy and the fiscal framework. The Government do not have access to what we call resource borrowing for any purposes other than forecast error. David and João have talked about how the fiscal stance of the UK is a reserved matter, so the overall level of borrowing is set by the UK Government, and then the Scottish Government receive the funding based on UK Government spending. The Scottish Government then have a little bit of choice about how they set their tax policy and how they choose to spend their money. They can access the resource borrowing for forecast error for taxes and social security. Other than that, they have to balance their budget. We have talked about how there are more components of the budget that are variable, but even within the spending side of it alone, you are managing departments, spending billions of pounds and having to balance that budget. The Government have always balanced it. Any underspends that they have can go in the reserve, subject to the £700 million limit. I mentioned earlier how, when the Government are managing their budget in year, they are balancing a lot of things. One of the tools that they do use, and we see them use, is that they plan to borrow for capital and resource generally, and then reduce that as the year goes through as a way of release. That means that they reduce the size of their underspend, which frees up resource for future years. One of the things that we always like to and would hope to see at the spring budget revision is a level of transparency from the Government about where they think the budget is going to end up and what level of funding is available to support spending in future years.

I agree very much with what Claire said there. When saying that the Scottish Government balance their budget, it is important to note that that means that they are following the fiscal rules that they need to adhere to. They are not really balancing their budget each year—and João has done some work looking at this. Because they have typically been borrowing a bit to cover reconciliations for forecast errors, and for capital investment, there has been on average—João can correct me if I am wrong—around £500 million a year of net borrowing by the Scottish Government since this fiscal framework came into place in around 2016. This “we always balance our budget” is sometimes used in the context of debates around Scotland’s overall fiscal position. Of course, that ignores the fact that the UK Government borrow on behalf of the Scottish Government. Because spending is higher, but revenues are not, Scotland has a higher per-population share of the UK borrowing effect being done on its behalf. While it is true that the Scottish Government have always managed within the fiscal rules of their own operation, that has been using some of the borrowing that they can do, and using money that the UK Government have borrowed on their behalf.

It is very clear that the Scottish Government balance their cash requirements every year, which means that they can meet their requirements, but you can tell that they borrow money because they have debt that they have issued. That is perfectly fine within the rules, but they do run a small deficit every year of around £500 million. As David said, there is some work that we have done on this. In some years, they have run a surplus, and some of that additional revenue or lower spending has allowed them to spend extra money in future years. It is slightly misleading that, as is sometimes said, they always balance their budget. As David said, a more meaningful statement is that they always adhere to the fiscal rules.

If I could follow up on that, what, in your view, would be the benefits and risks of changing this restriction of following the fiscal rules? What if it were put in that context, rather than saying they have to balance the budget?

If the Scottish Government were allowed to borrow more, they could try to stimulate more activity. That does incur a risk. You can borrow from a very safe source, which you can do at the moment, but there are limits on how much and for what purposes from the national loans fund. It could be changed to allow them to borrow from the market in a less capped way, but then there is a question of with whom the ultimate responsibility would lie if fiscal policy were not to be run sustainably. This is always a trade-off in terms of risk and responsibility. That would allow the Scottish Government to perhaps have more power to influence the economy, were there to be a Scotland-specific shock—a provision that has been eliminated from the latest fiscal framework agreement, but carries risk. The appetite that people might have for that depends. It is ultimately a political decision whether to follow that. Q35            Mr MacDonald: I wonder what you think of this theory. I was a Highland councillor, and it borrowed £1.2 billion. I have just looked it up and, apparently, the UK Government Public Works Loan Board lent £22 billion to councils in Scotland as of 2023. Is that a way of the Scottish Government effectively borrowing—underfunding councils and leaving the debt problem with them?

Borrowing by local government is normalised across the whole of the UK. English local authorities have also borrowed significant amounts of money to invest in roads and schools, or in commercial property, which has been a bigger factor in some of the borrowing by English councils. I would not want to make a claim that borrowing by Scottish councils is indicative of underfunding by the Scottish Government, in particular because council borrowing is also subject to a rule that it is for capital investment. Councils cannot borrow to fund day-to-day spending, but only for capital investment. If the argument was that, to some extent, this reflects not providing enough funding to councils, that would have to be on the capital side, rather than the day-to-day side of the budget. If I might very quickly add a supplementary answer to Kirsteen’s question, there are two types of risks that are often discussed in terms of giving the Scottish Government more borrowing powers. One is risks to the UK’s fiscal situation. We have looked at that and, because Scotland, together with all the devolved nations, is pretty small in the context of the overall UK economy, it probably is not the case that giving the Scottish Government more borrowing powers—even doubling or trebling them—would really affect the UK’s fiscal credibility. I come back to this point about fairness. At the moment, any borrowing that the UK Government do is for the whole of the UK. It is either to fund UK-wide things such as defence and pensions, or to fund things that are devolved, which means that Scotland gets a share of the borrowing via the Barnett formula. That means that there is no England-only borrowing. Even if Scotland pays back what it borrows, the fact that it is wanting more borrowing power shows that borrowing power is valuable. If Scotland, Wales and Northern Ireland have local authority borrowing powers, UK Government borrowing powers and this additional devolved borrowing power that is uncapped, that is potentially unfair to England, which does not have that separate element of borrowing powers. There is that risk, not just on the fiscal side of things, but also to fiscal equity across the whole of the UK.

Does the current cap of £700 million on the Scottish reserves affect the ability of the Scottish Government to manage their finances as effectively as they might?

The first thing that I would say is that that is a higher limit than equivalent English Departments would get with the same budget, so there is an additional amount there that the Scottish Government get to borrow. Of course, the Scottish Government are not like an English Department. They are their own political decision-making body and manage a wider range of risks, especially now, with the social security benefits, which in Scotland are treated like departmental expenditure limits as opposed to managed expenditure, so you would expect there to be a higher borrowing limit. I will hand over to Claire, because she was saying that this does still pose challenges. Basically, the Scottish Government have been using their borrowing powers to help flex their reserve powers.

One thing that I should add to that answer is that the ScotWind proceeds that the Scottish Government received have, effectively, been used as a reserve. They have often planned to draw down from the ScotWind funding, and then have not necessarily drawn down as much as they planned and then used that in future years. I agree with all the points that David made. There is an element of, if the Scottish Government had a larger reserve, for example, they could have a choice about how they allocated funding. We had an example a few years ago, where the timing of our and the OBR’s forecasts meant that the gap between the income tax forecast was artificially high, and we said, “This will lead to a negative reconciliation.” In theory, if you had a sufficiently large reserve, the Government could have paid that money into it to manage that in future years. However, politically, that is quite a difficult thing to do, so there are the practical realities of how this might work, and then there are the political realities, which might be slightly different.

I cannot really see a reason why the Scottish Government—which is accountable to the Scottish Parliament as well, so there is a further interaction there in terms of the fact that you have these decision-making bodies, which could, for very good reasons, decide to allocate spending over time differently—should be forced not to do that, when it could be perfectly well argued for.

Are there any other changes to the current fiscal framework that could enable the Scottish Government to more effectively manage their fiscal position?

There is one thing where the Scottish Government are very limited in terms of their ability and that would help them plan for the future and respond to shocks, and that is the lack of flexibility that they have within the year. At the moment, the way that the Scottish budget works is that it is passed before the financial year starts. There are some revisions that can be made, but, ultimately, some big chunks of their revenues, including, for example, the income tax forecasts, are set in stone. That means that quite often they end up in a position where they are trying to allocate quite large amounts of money, not wanting, as Claire said, to leave too much unallocated, for understandable reasons, but also trying to manage the risk of changes in the year that they have not fully accounted for. Some more flexibility for changes within year would be good. Whether that is achieved through the fiscal framework itself, or some other way of doing it, it would be helpful, because what you want to guard against is having this almost stop-start way of budget setting. You want to stop having fiscal impulses when you tell people that you are going to spend this much, and then it turns out that you cannot, so you are also potentially causing projects to stop and start. That would be an improvement to the fiscal framework as a whole, or at least to the Scottish Government’s control of the budget process, which, with the grown-up Parliament that it has, they should really be allowed to do.

Was there anything in the fiscal framework that you would have liked to see agreed but was not?

I would have liked to see, in terms of VAT assignment, everyone agree that it is not sensible in the way that it was thought to be implemented. I cannot see how another few years of work are going to come to a different conclusion, other than that it does not work.

If I could add one thing on the previous question, not so specifically about the fiscal framework but more generally about the Scottish Government’s budget, our fiscal sustainability work has looked at some of the long-term challenges that the Government have. We have real changes coming through demographic change, the challenges to meet net zero, and some of the Government’s commitments on child poverty. There are really fundamental challenges facing the Government, which need to plan their budget not just one year ahead, but thinking several years ahead. There has not been the space to do that in recent years, and it has not happened. Part of that is because of how the UK Government have operated their budget, but also, the challenges that they are having in managing their budget in year mean that the focus is very short-term, rather than longer-term and looking at these bigger issues. That is one of the things that is perhaps not a fiscal framework-specific thing, but something about the way that the overall UK fiscal framework works, and the need for greater planning certainty, provided that the UK fiscal event timings happen in a way that allows the devolved Administrations to plan their budgets. Organisations such as ours can then put more pressure on the Scottish Government to deliver multi-year plans and spending settlements, so that we can see that play out and hold the Government to account on what they are doing.

I agree that this issue of in-year flexibility is a key challenge for the Scottish Government. They are right, in fairness, when they say, “Let’s say that public sector pay deals come in at more than expected.” If that happens to the UK Government, they can say, “We’re not going to borrow; we’re going to find that within limits”—they often do that to try to put pressure on negotiations—but they can always say, “Let’s go to the markets and borrow the money to pay for it.” The Scottish Government cannot do that. If their pay deals are outpacing the UK’s, they need to find that in their budget, in year, by making cuts. Saying, “You have a bit of leeway. You can borrow the equivalent of 1% or 2% of your budget in year to address these sorts of issues,” would be a big improvement. The biggest change that I would make would be to have some discretionary resource borrowing, subject to some strict limits. I would also agree with João that this, in my view, farce of saying that VAT assignment is coming down the line needs to be ended. It was repeated again in the Scottish Government’s tax strategy that, “We will continue to work with the Treasury on a basis for this.” The only way that you can really solve this is by making companies divide their accounts into what is Scotland and what is the UK. If that is the plan, say it. If it is not, can it.

That is a good question for us to ask a future panel. That concludes our questions for this morning. Thank you, all three of you, for being with us today. It has been very interesting and has given us a really good grounding on which to base our future questions to other panels. Thank you very much for that.