Official Report 667KB pdf
Our next item is an evidence session with the Scottish Fiscal Commission on the Scottish Government’s budget for 2025-26. This is an opportunity to consider the commission’s forecast for spending on devolved social security benefits and how that is reflected in the social security budget for 2025-26. I welcome Professor Graeme Roy, the commission’s chair; Professor Francis Breedon, who is a commissioner; and Michael Davidson, who is head of social security and devolved taxes. Thank you for accepting the committee’s invitation. Before we move to questions, I invite Graeme Roy to make some opening remarks.
Good morning. I am grateful for the opportunity to speak with you. In my opening remarks, I will provide a broad overview of our latest forecasts and analysis, including assessment of the implications for the Scottish budget.
Following the United Kingdom autumn budget, the Scottish Government has had a significant increase in its block grant, based on United Kingdom Government spending plans. There is a large uplift in capital spending in 2025-26, and the resource budget—out of which social security is funded—is growing in both 2024-25 and 2025-26.
However, the overall increase in next year’s Scottish budget is relatively modest, due to the weakening net tax position. The real-terms growth in resource funding between this year and next year is estimated to be 0.8 per cent. Once we account for social security commitments, resource spending in other areas is falling in real terms by 0.3 per cent. That reflects social security’s taking up a growing share of the budget.
For next year, we forecast overall social security spending of £6.9 billion, which is up from £6.2 billion this year. That increase is a result of the annual uprating of payment amounts and continued increases in the number of people who receive disability payments. It does not include the additional costs, from 2026-27, of mitigating the two-child limit.
In 2025-26, social security spending is forecast to be £1.3 billion higher than the corresponding funding that will be provided by the UK Government. That gap is a result of the Scottish Government’s long-standing approach to social security, which can be thought of in terms of the following four factors.
First, the Scottish Government has introduced new payments for which there is no UK equivalent. The most significant of those is the Scottish child payment.
Secondly, policy and delivery changes have been introduced for disability payments, with the aim of increasing take-up and improving the experience of the people who apply.
The third factor is the more expansive approach to some replacement payments—for example, the expansion of the pension-age winter heating payment to all pensioner households next year.
Finally, there are policies that seek to mitigate UK Government decisions, such as the existing mitigation of the bedroom tax and benefit cap through discretionary housing payments and the planned policy on mitigating the two-child limit. As I mentioned earlier, our forecasts do not yet include any costs for the policy to mitigate the two-child limit. We were told of those plans on the Thursday before the budget and, because that was very late in the process, we were unable to provide a full costing. We have highlighted the issue as a fiscal risk, and we will provide an update to those forecasts on 7 January.
Our illustrative analysis suggests that social security spending could increase by around £150 million in 2026-27 and rise to more than £200 million by 2029-30. However, as I said, we will provide a formal costing and update on social security on 7 January, and we will write to the committee to provide information on that.
Thanks very much. That has been helpful. I move to questions.
Professor Roy, at the Finance and Public Administration Committee just two weeks ago, you highlighted four specific risks for the Scottish budget—namely, the growth in public sector pay, the on-going demographic issue affecting the labour market, the national insurance changes that will have to take place in various portfolios and—as you mentioned just then—the social security gap of £1.3 billion over and above the block grant. Helpfully, you have also just said that, on 7 January, you will provide projected costings for the change to the two-child cap. Given those substantial risks across the economy, do you feel that the social security situation is dominating, in terms of the longer-term projections?
We would not make a judgment call on whether a particular issue was dominating. Ultimately, it is the Government that balances the budget, so it is its discretionary choice to prioritise some areas of the budget, whether they be pay, social security or the health service, over others.
What we have highlighted is simply this: once you set the policy and lock in a commitment to spend certain amounts in certain areas, particularly in areas such as social security that are demand led, that will have implications elsewhere in the budget. Therefore, that £1.3 billion figure is quite useful, as it highlights not only the opportunity cost but the scale of the relative investment when it comes to having the sort of different system from that in the rest of the UK that the Government has been very clear about wanting. Essentially, it measures the scale of that policy commitment.
From a public finance point of view, though, that will have implications, and that is what we mean when we talk about fiscal risks. We would not say that it is dominating, but it is a choice that is made and resources have to be found for it.
I presume, though, that it is an increasing factor, given the Scottish Government’s policy choice to mitigate the two-child cap, and it will have implications not only for the budget that has just been published but for subsequent years. Therefore, the aspect of that risk must be quite substantial in future budgets.
Essentially, what we have highlighted in respect of the additional commitments falls into two parts, the first of which relates to the choices that are being made on replacement payments that increase the level of commitment beyond the BGA, and the second of which relates to where there is no equivalent block grant adjustment flowing through, and where the Scottish Government has decided to make a particular commitment. That is where the £1.3 billion figure comes in. When we publish our projections on the change to the two-child limit, we will see that as an additional commitment on top of that.
It means that social security’s share of the budget is increasing relative to other parts. That is an entirely appropriate policy choice, if that is the policy agenda that you want to put in place, but it comes with the opportunity costs of either raising taxation to fund it, or changing investment plans elsewhere.
I think that there is a distinction to be made between a risk—that is, an uncertainty, or a surprise that comes along and knocks your budget off course—and a deliberate policy course that you have chosen. There is a certain amount of risk with social security, because it is demand led and there will be uncertainty about the spend. However, that is different from choosing to spend a chunk of money on it, which is a deliberate choice.
All countries are facing similar pressures, with the demand-led situation, the demographic change in the labour market and so on. How easy is it to make predictions in respect of the UK system of social security, which will obviously have implications for Scotland? Is it broadly in line with other countries when it comes to the share of the budget increasing?
Obviously this is important in Scotland, too, but a big uncertainty in all the major economies is disability payments and, in particular, mental health related payments. We have seen a surge in mental health issues, particularly among young people. We do not know for certain, but in the UK, it seems—surprisingly—to correlate with the uptick in disability payments. It has become a big source of uncertainty and a true risk in all social security budgets over the past few years.
You cite countries including Germany, Switzerland and New Zealand in your sustainability statistics. One of the interesting dilemmas that we face in Scotland is how we react to increasing social security spend in the UK. If the Scottish Government makes a different policy choice, it will have implications for the block grant adjustment and the overall gap—that is the critical point. How is that gap funded? When you come to do your next fiscal sustainability report in due course, will you look at statistics that show that the trending increase in the proportionate share of the budget spend on social security will continue to increase, and probably substantially? Is that correct?
It is worth unpicking a few things from that, because there is quite a lot in there that drives the result. From a public finance point of view, the growth in disability payment inflows and case loads, which is common across the UK and Scotland, has an impact on the amount that we are spending in Scotland, but it also has an impact on the BGA. If we look at the growth in social security spending in our forecast, we see that the net position has not changed significantly relative to last time, even though we have assumed a higher case load.
When we talk about risks, they are risks that are shared and it is a broader question for UK public finances and for society about what is driving the concerning inflow of people into disability payments. That is less of a public finance question and more a question about whether something is going on in society. As Francis Breedon said, with young people’s mental health in particular, that is quite worrying and has big implications.
However, from a public finance point of view in Scotland, to the extent that the growth is shared between us and the rest of the UK, that increases spending but also increases the BGA. What really matters from a public finance point of view is where that potential difference comes through. As I highlighted, it comes through a mixture of where the Scottish Government has, in its policy agenda, chosen new and different payments or payments to mitigate UK Government decisions, and the creation of a system that is designed to be more supportive and to encourage uptake.
You will see in our forecast that a lot of that growth starts to level out, but that is uncertain. That is a risk about what we think could happen in the long term.
From your comparative studies against other countries, can you point to any evidence that shows why the UK, and Scotland in particular, has a higher incidence of people claiming benefits because of mental health issues? Is there any evidence that you can point to that helps with that?
It is very early. As I tried to say, there is a bit of a disconnect between data on disability payments and data on health. The health issue is, arguably, really big in Europe and the United States, and I am not sure that the situation there is different to what it is here.
Simultaneously, we are seeing an increase in disability payments in the UK. Logically, you would think that the two things are connected. I do not have a very careful comparison, but I do not think that we are seeing such big increases in disability payments elsewhere. That does not mean that the health issue is less elsewhere—it could just mean that the connection between it and payment is weaker in other systems, in particular the American system.
If you look at things such as the proportion of spend on ill-health payments, you will see that the UK has historically had lower levels than other countries in Europe. Yes—there has been growth, but the level effect is, perhaps, less. Francis Breedon is right. We are seeing similar trends across many high-income economies: it is a concern not just in the UK, but more broadly.
The broader point, which is probably the main interest of the committee, is that we talk in terms of public finances and we show what is happening to the case load that is coming in, but the most important question is about what, upstream, is driving that demand and that case load, because that is where there is concern. The statistics on mental health issues among young people are concerning, particularly those on young women. Those are linked to issues around social media, anxiety in schools and so on. With the child disability case load, there is inflow. There is a real issue upstream, which is less a social security question and more about what is happening more broadly in society to drive the increase in demand.
09:45
I presume that that is very difficult to measure.
Yes—and there is lots of speculation about what the cause could be. We see the case load, but in thinking about what factors are potentially driving it, we move into much broader areas of public policy around—
How you pay for it.
Yes. It is a really interesting question.
There is obviously a lot of interest in the issue, so several members want to come in with supplementary questions.
I will come back to child disability payments in a wee minute, but first I want to go back a step.
When you talked about uncertainty and changes in the labour market, that highlighted to me something that I read recently in a Centre for Social Justice Foundation report on economic inactivity in Scotland. There is concern that just short of 85,000 people in Scotland are economically inactive, but not because of health issues, care duties or retirement. Can you shed any light on why the figure is so high in comparison to the figure for the rest of the UK and on what impacts it is having?
Again, that is one of the issues where we are working with correlations rather than causal effects, but it feels as if it is related to what we have just been saying. The issues in health that we have been talking about have impacted on participation, as well. From our point of view, purely with our fiscal hats on, that makes the situation doubly costly, because we not only have a health problem but a lack of revenue to spend on that health problem. Participation is a significant economic issue for Scotland.
On that, the people who consider themselves to be inactive due to health issues are not included in the figure that I gave. That figure includes people who consider themselves to be inactive for other reasons, which is what I am concerned about. They are inactive not because of health issues, caring for a family member or retirement, or because they have gone on to further education. We do not know why they are inactive.
That relates to a big issue about labour market statistics and their rigour and what they capture. When we talk about the labour market, the inactivity measure is largely the residual bit that we cannot really explain. You can ask a person whether they are in work, and the simple answer will be yes or no. Then you can ask, “Are you looking for work?”, and the answer is yes or no. That gives you information on employment and unemployment.
There are people who would answer no to both those questions and go into the category of economic inactivity. You are right to say that that is a big catch-all. When people are asked and probed about the reasons why they are not in work or not looking for work, some will be clear and say that it is because of a health issue, because they are in full-time education or because they are caring for somebody. Everybody else then has a complex mix of those different elements, and their situation might be changing and evolving. They largely make up the residual category.
The broader point that I think you are getting at is an important one. In the past, we have not really had to worry too much about that, because the focus has been on employment or unemployment, and that residual category has largely been relatively small and has not really had a big impact. However, as Francis Breedon said, when there are high correlations between that inactive group and inflows to disability payments, it becomes much more important to understand that group. That is a really important question to focus on.
That is excellent—thank you very much for your answers.
Good morning. I am really interested in this line of questioning. The demand-led spend that any Government makes is potentially open-ended for the longer term and must be a risk. It is right for the Fiscal Commission to step away from the policy decisions on that, but to map out the financial exposure of the public purse.
That is helpful because, for example, we can see that the mitigation of the bedroom tax is now baked into Scottish Government spend for the longer term and, soon, winter heating mitigation will also be baked in. We look forward to your statistics on the two-child cap, which will soon be baked into future spending, as well.
On the issue of a risk register for Scottish finances, does the Scottish Fiscal Commission take a longer-term look? There are fiscal framework implications in terms of policy spillover from the next steps that a UK Government might take in relation to its spending decisions. There has been a lot of rhetoric and comment about the cost of disability payments in England and the rest of the UK, and the UK Government is wrestling with that. There is quite a lively debate about how much it is possible to humanely squeeze those payments to get lower expenditure in England, and about whether that is appropriate, policy-wise. I would say that it is not, but that is not the point.
The point is that, if the UK Government goes down one road in the rest of the UK and—rightly or wrongly—gets many people off of disability payments, the policy spillover will have a direct financial consequence for the Scottish budget because we will receive less in block grant adjustments. Does the Scottish Fiscal Commission map out the implications of UK Government policy and draw to the Scottish Government’s attention future potential implications for the Scottish budget, be they in relation to disability payments or anything else that a UK Government might wish to do?
We talk about fiscal risks, and some of the risks that you highlight are exactly the sorts of things that we will articulate and flag up to the Government. It is not our job to speculate on what the UK Government could or should do, but we highlight that, if the UK Government takes decisions in devolved areas or, increasingly, in broader social security issues, that has potential implications for the Scottish public finances.
There are two recent good examples of the point that you make about UK Government decisions having an impact on the block grant adjustment in a way that has a direct impact on Scottish Government public finances and, therefore, the choices that are made. The pension-age winter heating payment is a really good example. The UK Government’s decision to means test its winter fuel payment had a direct and immediate impact on the block grant adjustment, which means that the Scottish Government has had decide whether to continue with the payment, with the result that the net cost becomes bigger, or whether to follow the approach that is being taken elsewhere in the UK.
The other aspect that is quite interesting and involves another risk concerns the policy choices in Scotland to mitigate decisions that are taken at UK level, such as on the two-child limit. That does not have a direct impact on the block grant adjustment, but the Scottish Government has proactively taken a policy decision to mitigate that, in accordance with its policy ambitions. That, in turn, has an implication for the public finances.
That is why we talk about issues around fiscal risk. One risk, which you have highlighted, is demand in Scotland, which is uncertain: a policy is set, and we then wait to see what happens in terms of fluctuation. However, there is another risk coming through around the public finances, and that concerns choices that are taken not only in Scotland but in the rest of the UK. That is just the nature of the fiscal framework.
It is worth highlighting that there is a tendency in one’s mind to link the social security BGA with social security spending in Scotland—we are probably as guilty of that as anybody else—but, in fact, any tax and spend decision in the UK has implications for the Scottish budget, and it does not really matter whether the decisions are social security decisions, tax decisions or other spending decisions.
It is true that, because of the way that the framework is set up, it is natural to compare social security spending in the rest of the UK with that in Scotland. However, as I said, we need to get away from saying that that is all it is for. The BGA is a lump of money that Scotland gets across the board, and we do not need to link the two things in our minds.
It might not matter in those terms, in relation to what you say about the aggregate position of the Scottish finances, but it certainly matters in our scrutiny role here, when we are trying to work out the long-term implications of UK decisions for policy decisions that the Scottish Government makes, as we are trying to do today.
Professor Roy articulated the position very well, but what I am trying to get at is that the winter fuel payment change was like a bolt out of the blue. It would not have been reasonable to have forecast what the risk was in relation to that, because most people did not expect the UK Government to do that. However, there is an expectation that there could be a squeeze on spend on disability payments in the UK.
Does the Scottish Fiscal Commission identify the risks of potential UK policy changes—because those are in the public domain and are being floated, if you like, proactively—then cost some of that and give it to the Scottish Government, or does the Scottish Government say, “Here are three things that have potential policy implications for Scotland—go and cost up our fiscal risk in relation to them”? Does any such proactive work happen, Professor Roy?
Our remit is to forecast for Scottish policy choices. Our remit is tightly defined by our legislation: the Scottish Government will give us policy choices, which, essentially, we then model. We have a broader role within the public finances, whereby we can take the BGAs, make a comparison, and highlight things such as fiscal risk. However, it is not in our remit to forecast that the UK Government might decide to do something differently—on disability payments, for example—then calculate a mock BGA. The Government could do that itself, as part of its planning work. Through our remit, we have to focus solely on Scottish Government policy choices, rather than on those of the UK.
Briefly, should that be in your remit, Professor Roy? The Scottish Government works up the numbers. Others might say, “Well, it would work them up—it’s marking its own homework, so perhaps the Fiscal Commission should be doing that.” Should it be in your remit?
I will say one thing. The UK side really belongs to the Office for Budget Responsibility. We have to be careful about what we say in relation to the OBR. You could transfer the question to the OBR: should it cost up policy ideas for us to analyse? That would be how it would work. The relationship is slightly different.
So, it could be a joint space.
Good morning. I am keen to understand the problems that you guys face in trying to forecast—or, as I preferred to call it when I used to do it many moons ago, guesstimate. I am looking at your December 2023 and December 2024 forecasts. For 2024-25, the forecast has come down by £59 million. I accept that, because you know more of the facts and are closer to that timeline. However, you have increased the forecast in 2025-26 by 1 per cent; in 2026-27, by 3 per cent; in 2027-28, by 4 per cent; and in 2028-29 by another 4 per cent. What underlying factors influence you in deciding to uplift your future forecast?
There are probably three ways to think about how revisions happen to the forecasts. Some of them net off—we will come back to that discussion about the interaction with the BGAs—and others widen the position.
The first set of factors that we have to contend with and think about is changes in the uprating of the social security payments. In a high-inflation environment, because of the linking of payments to inflation rates, payment levels increase. That is natural when a social security system is linked to prices. From a public finance point of view, that also leads to an increase in the BGA, so the net effect is cancelled out.
A second set of factors is the growth in the case load for social security payments. As we have just discussed, some of that is because of common trends across the UK and internationally, particularly in disability payments. People flow into those payments because of something else that has happened upstream. Again, all else remaining equal—if there is that same growth—there are impacts on the BGA as well as on payments.
Inflation and the case load effect are two reasons why we have increased our forecasts over the time period, because more data has come through.
Then there are discretionary decisions to make changes in the policy environment. For example, since December, we know that changes to things such as pension-age winter fuel payments and the two-child limit are coming down the line. That means that, when we look at one forecast relative to where we are now, the policy has changed, which leads to a potential increase in expenditure.
10:00The final bit is about what happens to the revisions to our forecasts if we see data that is different from what we assumed. I can give an example of the type of thing that we look at. We have to make judgments about what we think will be the level of demand from inflows and about what we think that payments such as the average payment for a social security benefit will be.
We also have to think about outflows, including what proportion of people who come in to receive a payment flow out at the other end. We track that data and have a great relationship with Social Security Scotland and with the Scottish Government, which provide us with up-to-date information all the time that we are constantly able to look at.
We might come on to this: one thing that we see driving some of the uplift does not come from changes in inflows or payments—which we have actually revised slightly downwards—but from fewer people flowing out of the system. We may come to that as part of the broader discussion, and it is one of the main reasons why we have increased our forecast in some key areas.
I accept the need for uprating because of inflation. Regarding the case load increase, are you seeing any of the benefits starting to level out? Where is that levelling out happening across the board, and how will the extra £1 billion that you have built in even itself out? Is there any indication of when that will happen?
We are now very much in that phase. When we came to see you last year and the year before, we said that we needed to wait and see. We have made lots of judgments or, to use your word, guesstimates of where things might go, but we do not get that stability until we see the data coming through.
The Scottish child payment is pretty stable now. We know the number of children who are eligible and we know the take-up rate, which is pretty stable, so that forecast does not move around too much.
We are still in a really interesting phase with the disability payment as it starts to roll out. The really good thing is that we no longer have to make judgments or look at what is happening at the UK level to try to work out Scottish figures. We now actually have Scottish data to look at and can see what is going on. Michael Davidson might want to say a bit more about what we are seeing in the data.
The volume of applications for adult disability payment is stabilising. We thought that that number would increase, which it has done, but it is now beginning to stabilise. The outflows element is really interesting. We thought that the number of people who move out of the payment or whose payment is reduced following a review would be far higher than we are seeing in the data. That is one of the areas of uncertainty that we talk about in our report: we still think that there will be a higher rate of outflows than we are seeing at the moment, but if that does not happen, that will have an impact on expenditure further down the line.
About 2 per cent of payments are reduced or stopped when they come up for review, but we were assuming that that figure would be about 5 per cent, and that is significantly lower than what is happening with personal independence payments. That is a policy choice. The Government has been clear that it wants what it views as a light touch. I am in no way saying that that is a bad thing, but the idea is very much one of support, rather than demanding information from people, and that is one area where we see an increase in spending.
As Graeme Roy said, inflow levels have stabilised. We are looking at whether having fewer people who stop receiving a benefit will have an impact on future inflows. Our understanding of the previous PIP system from the way that it operated in England and Wales was that people who have their award stopped will have a period when they do not receive it but will then make a new application and flow back on. We are looking at whether there might be a future reduction in inflows because people stay on the payment for longer, rather than flowing off and then flowing back on. If someone stays on the payment for longer and is paid for more months, the costs are higher.
We are now getting really good data from Social Security Scotland about that type of stuff. Outflows are the last piece of the story and we are in the early days because we have been seeing that data for only a few years. That is why outflows naturally cause a big uncertainty in our forecast.
I am a member of the Economy and Fair Work Committee, and we are always complaining that, sometimes, there is not enough data. You have said that you are getting data from Social Security Scotland, but are there any other types of data that you do not have access to but which would make your life easier?
One of the benefits of having Social Security Scotland delivering these payments at a devolved level is that we have administrative data that is unique to Scotland and which we can really track and use. Very early on, we highlighted some issues that we had with the data that Social Security Scotland was collecting, but, after working with us in order to understand our needs and to be able to inform our work, it has made really useful improvements, and the quality of the data is now really high. That means that we can model the payments and change things in real time.
There is another really useful thing that Social Security Scotland gives us. As part of the round of discussions that we have when we pull our forecasts together, we get a chance to meet its directors and the people who are delivering the payments, and they tell us what is happening. We might see something in the data and ask, “Why is this?”, and they will be able to say, “Well, it’s because we made a push to encourage applications over that three-month period. You might see a spike there, but it’s not going to continue, because we are easing off on that.”
As for our asks when it comes to data—and this comes back to Roz McCall’s point—one of the real weaknesses in the UK is our labour market data. As you will see from our report, it has become so poor that we cannot use it in our main forecast, and it is not a great place to be in if we cannot use the main official UK labour market data. That really matters. Because response rates have been so low and because the confidence in that data is really poor, once we start getting into really tricky areas—that is, once we go beyond people who are in or are actively looking for work to look at those people who might be inactive, and, indeed, inactive for quite complex reasons—if we have really poor-quality data at a UK level, it becomes a real struggle for us to comment on what that might mean for Scotland and the modelling. If I were to highlight one area for improvement, that would be a huge area to work on.
I will leave it at that. Thanks very much.
I will bring Roz McCall back in.
My apologies to the panel—I should have said hello earlier.
I would like to revisit a number of issues. I am pretty sure that you have already answered these questions very eloquently, but I just want to home in a little bit on the child disability payment.
I note that we have a 60 per cent increase in case load, while the figure for the rest of the UK is 47 per cent. Both figures are rising, but why is it rising so much in Scotland? Can you give us any insight on that? First and foremost, why is it rising at all, but why are we at 60 per cent in Scotland?
You are right. We are seeing a significant increase in case load and applications, with the figure slightly higher in Scotland than it is in the rest of the UK. A large part of it comes back to Francis Breedon’s point about the trend that we are seeing in many countries of an increase in young people, particularly young women, having mental health and mental wellbeing issues, which is leading to an increase in case load and applications. There is a general trend in that respect.
As for why that is happening, my honest answer to you is that I do not know. It is one of the real issues here; we see the data—that is, the case load—and we can all speculate on why that might be the case. It could be the rise of social media, it could be issues in schools, it could be legacies of Covid or it could have something to do with the wider challenges of poverty and inequality in the broader economy. We can all speculate, but a hard causal link between the increase and a particular factor is still uncertain. Various think tanks and researchers have started to speculate and do some studies on what the link might be.
Something that partly explains the slightly bigger increase in Scotland relative to the rest of the UK is the different delivery system in Scotland. We have always thought that, if you have quite a different delivery system that is very much about supporting people through the process, being less intrusive, being more inclusive and encouraging people to apply, it will lead to a higher case load over time. There has been a deliberate attempt by the Scottish Government to have quite a different delivery system for social security, and that might explain some of the differential that we thought would happen with regard to the case load in Scotland being higher than that in the rest of the UK.
Okay, so just to clarify, it is not about Scotland’s specific health or social issues; it is more that the delivery system has been more inclusive.
We speculated at the time that that would be one of the reasons why. As I said, we still do not understand what is driving the overall trend, so there might be some reason why Scotland is seeing it more than the rest of the UK, but that would be speculation. We certainly have not seen any evidence to let us make a judgment call that we think that something unique is somehow happening in Scotland that is leading to the higher inflows relative to the rest of the UK.
That leads me on to data. My colleague, Mr MacDonald, has already asked about the data concern and I accept your answer to him that things are better now because there has been an increase in the data that you receive from Social Security Scotland. Is there any more that we can do on data to make sure that you have the information that you require, especially when it comes to the anomaly with the child disability payment?
From a public finance point of view, the data that we get from Social Security Scotland and what it tracks that is valuable are the case load and the elements that people cite in the applications. That is helpful for us. The bit that is missing—and it largely starts to move outside our remit—is much further upstream: what is happening in schools, what is happening with young people’s mental wellbeing and what is happening more broadly in society? It is difficult to see what is going on there.
That is not just a Scottish issue, but a genuine UK issue about understanding wellbeing among children and the factors that lead to the decline in mental wellbeing and, therefore, increases in disability payments.
That is helpful.
Now that we have better data, to what extent have clear trends emerged in the child disability payment? Will that give more certainty with future forecasts?
Again, that comes back to the answer to Gordon MacDonald’s point. Now that we have the data and trends, the level of uncertainty is diminishing. It will never go away, but it is diminishing in the sense that we can see that applications have risen relative to what they were and that they have stabilised. On things such as average payments, one of the amendments that we have made is that we have revised down what we think the average payment would be. We made a judgment call about what we thought the value of average payments would be, but we have seen an inflow, although it has not been at the high end of payments. The average payment has come down, so we have adjusted our figure down. As we get more years of data, the average becomes more stable and gives us the trend.
If we take a step back, what we are seeing with the child disability payment is higher case load. We think that the spike in applications will start to level off at the higher level, so the case load will be more stable, but the average payment has come down. All else remaining equal, that should give a bit more confidence about the stability of the forecast.
I will now bring in Paul O’Kane, who joins us remotely.
I am going to ask some questions about the adult disability payment. We have already touched on a number of different elements in that space, but I am keen to know whether Professor Roy or others can elaborate on the reasons for increased disability spending across the UK. We have heard about some of the different aspects in relation to support for children, but how well understood is the picture on the increase?
I can go through a couple of things. One of the factors in areas where we have seen an increase in spending is inflation, with uprating leading to the increase. When you see the numbers going up and our forecast for adult disability payment rising over the next five years, you would expect that to happen, even if case load did not change, simply because of inflation. There is an inflation element to all that.
The second part is about picking up on some of the issues that we have discussed around the case load and the rising inflows and, in the Scottish case, the lower outflows.
In the past, we have talked about the cost of living crisis potentially having a short-term impact on inflows, because people are facing significant hardship. Based on the evidence that we have seen in the past, we know that that leads to more people moving into that part of the social security system than into other areas. That situation should start to ease in time, and it is possible that that might be an area where there will be more stability.
10:15The big issue that we come back to is that of what the underlying driver, or the source, is of people flowing into disability payments. Before the budget, the Institute for Fiscal Studies did some interesting analysis in which it summarised the outcomes in relation to where people are on disability payments.
Two interesting points emerged from that. The first is the fact that a greater proportion of adult disability payments is going to younger adults, relative to the pre-pandemic position. The second point, which is related to that, is the fact that, relative to the pre-pandemic position, a greater proportion of disability payments is going to people who say that the reason why they require that support is that they have some form of mental or behavioural disability. Disability payments are going to more people with those types of challenges, rather than to people with the more traditional disabilities that one might think of, such as respiratory disease or mobility issues.
Thank you. That was a helpful answer that sets in context some of the other issues that we are discussing.
An issue that we have touched on previously is the uncertainty surrounding the unique Scottish effect on adult disability spend. You said that you think that we now have better data on that, and data that is demonstrably Scottish. I am keen to understand what patterns you think are emerging from that.
That question summarises some of the questions that we have already answered. The judgment call that we made before adult disability payments were rolled out was that we thought that there would be a higher case load as a result of the new system. Ultimately, that was the policy objective—the Government wanted to improve the process for people who applied for such payments and to encourage people who might have been excluded in the past to apply. That would lead to an increase in case load.
The forecast in that regard was always uncertain because, as Francis Breedon said, it was based on a judgment and the experience with payments that had been rolled out in the UK. That judgment has largely been borne out. As we mentioned, we have made some changes on average payments and inflows and outflows, but that judgment has come through. There are now more people in Scotland on adult disability payment than would have been the case if we had followed the trends that have been evident in the rest of the UK on PIP.
As we said, that gives us a bit more of a baseline, and we can start to look at potential changes around that baseline. As Michael Davidson said, application volumes are more stable, and we hope that there will soon be more stability in relation to outflows. We are already seeing more stability in relation to average payments. Once we have that, we will be able to have a much clearer view of the Scottish effect and of the relative difference between our forecast and the block grant adjustment.
That is right. At this stage, the big change has been in the delivery style, which has dominated the trends on case load. Implicit in Paul O’Kane’s question was the issue of whether there are greater levels of disability in Scotland than there are in the rest of the UK. We will be able to return to that. At the moment, the numbers that we are concerned about are dominated by the different approach to delivering such benefits.
That is helpful. The other areas that I wanted to cover have been touched on, so I will not take up any more time. I hand back to you, convener.
Thank you, Paul. The next question comes from Marie McNair.
Good morning. The UK Government is planning to reform health and disability benefits. How might that affect the Scottish budget?
This comes back to the deputy convener’s comments. From a public finance point of view, the obvious way that that would happen is that, if any decisions were taken at UK level to change disability payments—let us focus on those—that led to a relative increase in case load or in payment, that would lead to a higher block grant adjustment, which would lead to more funding flowing into the Scottish budget. If any decisions were taken by the UK Government to potentially scale back on average payments or on case load, that would lead to a smaller block grant adjustment relative to everything else, and that would lead to more pressure on the Scottish budget.
The classic example that we have seen is the pension-age winter heating payment, whereby a decision by the UK Government to change eligibility for the winter fuel payment, with its impact on case load, had an immediate impact on the block grant adjustment, which then had an impact on public finances through the Scottish budget.
I want to look at that. There will probably be a reduction in disability capability assessments, but we will see what happens. Obviously, the budget is also burdened by the decisions that we have taken to mitigate Westminster policies such as the bedroom tax and the benefit cap. How much is needed to mitigate those policies? Is there any indication that the UK Government will scrap them, or do you anticipate that that burden on the budget will stay constant?
The good thing about our remit is that we can just model Scottish Government decisions and we do not have to speculate about UK Government decisions. As much as the deputy convener’s encouragement for us to do that is welcome in terms of supporting the SFC, it is an area in which, thankfully, we do not have to do too much.
We have an interesting figure in our report—figure 5.8—in which we break down the £1.3 billion differential. There are areas in which there is a block grant adjustment and a different delivery system is leading to higher expenditure relative to the funding. There are also areas where the payments are unique to Scotland, including, largely, the Scottish child payment and payments to mitigate the bedroom tax. We forecast that, next year, those payments will total just over £600 million.
There is an interesting question about mitigation policies. It is not just about the Scottish child payment. There was a decision to tackle child poverty, so we have introduced a completely unique payment in Scotland relative to the rest of the UK. Things such as mitigating the bedroom tax and the two-child limit, which we might come on to, are deliberately targeted at addressing decisions at UK level that the Scottish Government does not agree with, and those social security commitments add to the burden on the budget.
I have a quick add-on question to give me clarity. ADP and PIP have exactly the same criteria—
Pretty much.
—but there is financial divergence. Is that purely down to political choices?
It is down to the delivery system, which is a policy choice. The Scottish Government has been very clear, throughout all its work on social security, to take a different approach from that taken by the UK Government and the Department for Work and Pensions. It is much more about person-focused support for customers, so people are encouraged and helped to apply, and people who might be at the margins of applying are encouraged to do so. It is a much more inclusive and supportive process.
I mentioned the phrase “light-touch reviews”, which the Government talks about. That is very much about the fact that, if you are on a payment, the review process is meant to be less intrusive; it is meant to be about asking people whether they still need the payment. The individual can then tick a box and say that they do. It is less intrusive. That is part of the deliberate policy choice to have quite a different social security delivery system from that in the rest of the UK.
The criteria are a bit different. Admittedly, the differences are mainly minor, but the terminal illness criteria are quite clearly different.
Yes, terminal illness is the one area in which there is a bit more of a difference, but that is only a fairly small part of the case load. The points that Graeme Roy mentioned are the main drivers of the difference in spending.
I move on to the winter fuel payment, which the UK Government announced would be means tested. There was a call for people impacted by that to take up pension credit. What is your expectation of increased take-up of pension credit in light of what has happened?
We included an increase in pension credit take-up—a behavioural adjustment—when we did our costing for the policy for this financial year. We included an assessment of the behavioural change for people who are now no longer able to access the universal winter fuel payment. We assumed that more people would apply for pension credit and therefore have access to it. That has pretty much been borne out by the data; we have seen an uplift in people applying for pension credit.
We assume that there will still be more people applying for pension credit into next year, even with the Scottish Government’s policy change, in part because it has highlighted that as an issue. We think that people are now much more aware that it is important to apply for pension credit, because it is a gateway to the winter fuel payment, and, secondly, because there is still a higher payment for people who are in receipt of pension credit relative to the universal payment from the Government. That is the long answer to your short question. The short answer is yes, we think that there will be an increase in uptake in pension credit applications as a result of the winter fuel payment policy.
I very much support that.
What is the knock-on effect of the Scottish Government’s policy on the affordability aspect of the pension-age winter heating payment? Is there an impact there?
It is one of those things where there is a positive and a negative effect on the public finances. It is clear that the Scottish Government’s decision to have some form of universality in there is an additional commitment relative to the UK commitment, and that costs money. Our forecast is a bit confusing. We are forecasting less than we were forecasting last year, because we were assuming that it would be universal, but we are forecasting more than we were forecasting in the autumn, when the means-tested element came in. We think that the cost to the Scottish Government of extending the universal bit is around £70 million.
That is on the negative side, from a public finance cost perspective, but there is a potential indirect feedback loop that would provide some additional funding into the Scottish budget, which is that more people in the UK moving to pension credit would lead to an uptake in eligibility for the winter fuel payment at the UK level, which would lead to a higher block grant adjustment and more money flowing into the Scottish budget. More people shifting their behaviour at the UK level will have an indirect impact on the Scottish budget.
That is a huge knock-on effect.
But the scale of that effect is fairly small, because even with the pension credit take-up, our forecast is only for £2 million per year in 2025-26. In the grand scheme of things and the overall scale of the social security forecast, even if there are larger shifts in the pension credit take-up than we forecast, we do not think that there would be a large impact overall.
Do not worry—I will not be soliciting you to consider any further remit for the SFC in my next question.
It would be helpful to get some information on the record from you. The committee will have to return to scrutinising the Scottish Government’s efforts to mitigate the two-child cap. We do not underestimate the challenge for you in ensuring that you have all the relevant information to project what the costs of that might be—not just in the short term but in the longer term. I will not list off potential factors from my notes, but it would be helpful if Professor Roy could say what those various factors and any related challenges might be. That would give us a steer for future scrutiny of the matter.
10:30
There are several things to note. We will set out all the different assumptions and elements when we publish our report on 7 January.
The first thing that we have to think about is how many families will potentially be impacted, which is not as easy as looking at the number of people who are on universal credit who have more than two children, because some of those families are already exempt. There is a question about how many families there are, and how many families are subject to the two-child limit. That is the first element.
The second element is the average number of children—there could be three or four children—and the potential effect of that. We also have to factor in the two-child limit only kicking in from April 2017, which adds complexity, because it means that the proportion of families who are subject to the two-child limit will increase over the forecast horizon as more children are captured by the policy.
The final bit—where we have to make a big judgment call—is the potential behavioural effects of the policy. What might the effects be on employment if someone is in receipt of an additional payment? Does the payment have a material impact on behavioural response?
The good thing is that we have quite good statistical data. We do not have the administrative data that the Government will need to deliver the policy, but we have quite good statistical data on the number of families who are involved. For example, we can make judgment calls about the number of children, and we have made judgments about behavioural effects of the Scottish child payment, which is not too dissimilar to some of the effects of the two-child limit. We are confident that we have all the bits of information that we need to make a robust forecast. Those are the big-ticket items that we have to make a judgment call on or look at the data for.
I do not want to ask too much more, given that a lot more information will be put into the public domain on 7 January, which is not very far away. However, I saw in my notes that there could be a potential implication in relation to the benefits cap. The Scottish Government mitigates the UK benefits cap in some respects—in relation to discretionary housing payments, for example—and it tries to avoid clawback on that. Is that another cost factor that might have to be taken into account? If the DWP claws money back because of that mitigation, that would have to be projected into the overall cost for us, which could be even higher.
Michael Davidson might want to explain some of the detail about what we do. We follow a similar approach to the one that we take with the Scottish child payment, which is to assume that it lies outside of the overall benefit cap, mitigating the two-child limit. That is for the two Governments to agree on, but we take what the Scottish Government’s policy is. My starting point is that we would take that as the two-child limit position, given that that is what we have done with the Scottish child payment. However, you are right that there is risk in what might happen when it comes to delivery.
I have a very technical question for Professor Roy. When the cabinet secretary announced in the budget statement her intention for the two-child limit policy to come in, she intimated that it would take time to be implemented. She said that it would be dependent on the Scottish Government being able to access good-quality information from the DWP. If that does take time, might it mean that, although you will produce your best estimates on 7 January based on the statistics, there might have to be a revision further down the road?
I am confident that we will produce a robust estimate based on the statistics that we have and the clear instruction from the Government that the policy will come into effect in 2026-27. Of course, if there were to be any change to that—for example, if, for whatever reason, the policy did not come in until halfway through 2026-27, or if it was delayed by a year—that would have an impact on our forecast. We will forecast on the full financial year, starting in 2026-27. However, you are right: if there is any slippage or delay, or even any acceleration, for example, in 2025-26, that would change the number. We will be very clear in saying that 2026-27 is when we think that the policy is coming in.
Indeed, it is fair to say that one of our biggest sources of forecast error in the past is that things have not arrived at the announced date. With any new system, that is always a big source of forecast error.
Given what you have said, I think that it is fair to say that you feel that you have strong statistical data and that the Scottish Government has provided the administrative data that you require to develop a robust costing. We look forward to your report in January.
Will you tell us how far ahead you are able to make forecasts? The two-child limit policy affects only children born since 2017 and will not be fully rolled out until 2035. Once a policy is introduced, it can be very difficult to withdraw it, so we need to make decisions for the long term. We do not know what will happen with the UK Government’s policy position on the issue, but if the Scottish Government were to take that on for the long term, how far ahead are you able to forecast? We need to be able to understand the long-term implications of any policy decisions.
You are right. One thing that we have to take account of is the fact that the number of families who are impacted by the policy will evolve over the forecast horizon. Following the policy’s introduction, and as children get older, a greater proportion of families will be subject to the two-child limit through the forecast.
On the broader point about what we do in the longer term, we will take the Scottish Government’s policy as given and we will model that. If the policy is to mitigate the two-child limit, we will take that and assume that the two-child limit remains at UK level and that the Scottish Government chooses to mitigate it. With our modelling, the data and the number of families impacted, we will assess that over the forecast horizon. Our forecast horizon on the budget is five years, so we will go to 2029-2030 in our main analysis.
There is a broader piece, which allows me to flag some more publications from the Fiscal Commission that will come in due course. In those, we start to look at the long-term fiscal sustainability work—that is, at projections over decades. That is not to give direct forecasts of what might happen; it is just to show that there are trends that could have an impact.
Our next report in the spring will be about ill health and implications of the costs to Scotland’s public finances of our ill-health record, and, if we can turn that round, what the potential savings are that we might make, in time. That is where we might get into some of the longer-term speculative issues around social security and the like.
The report in January will give us forecasts only for the next five years. Is that correct?
Yes. The two-child limit policy was introduced too late for us to include it in our main headline table. We will do two things. We will provide details about how we have made our forecasts and the total expenditure on that policy, and then we will update the social security table, so that you have revised figures on the totality of spend and the totality of the net position.
Will you give an indication of when the longer-term forecasts—that is, those that go beyond five years—might be incorporated into your work programme and when we can get that information? Are we likely to get that information by next summer? You are doing other work; I do not know whether you are able to programme that in now.
What we will publish in spring is part of our fiscal sustainability series. This year, we are looking at issues around health and ill health. That will not go into specific policy areas, but it will talk about the spending commitments that are being made in Scotland. That will show, relative to the rest of the UK, that we can tie that to the health record in Scotland. We can then look at what happens as we age to see, as commitments continue, what the potential public finance implications of that might be, and, if we can turn the ill-health record around, what the potential long-term savings might be.
Thank you. I have no doubt that the committee will come back to that.
In the light of what you were saying about forecasting and ill health, does the introduction of means testing for the winter fuel payment have any effect on ill health, especially in Scotland, which is considerably colder than the rest of the UK? Would your forecasting look at the effect and impact of that on ill health?
We would not look at the cost benefit of a policy, because that lies outwith our remit. That is the sort of thing that you would expect the Government to do—in essence, to look at the potential health, economic and social effects of a policy choice. We model the public finance costs and revenues that result from a decision, but we would not do a cost benefit analysis of a policy. However, you would hope that policy makers would look at the long-term—or even short-term—implications of policy choices, particularly those that impact vulnerable people.
That is great—thank you very much.
You will be pleased to know that that concludes our evidence session and our public business for today. I thank each of you for attending. In January, the committee will take evidence from the Cabinet Secretary for Social Justice on the budget.
Before I close the public session, Bob Doris would like to make a comment.
You will be glad to hear that this is not related to the evidence session. I would like to put something on the record. I have given the convener advance warning.
On 31 October, I suggested that Mr Balfour had questioned in the Scottish Parliament why we were building Scotland’s social security system and that he had said that it might be cheaper to simply run it from Westminster. I suggested that that was Mr Balfour’s consistent position. Mr Balfour quite forcefully disagreed, and I offered to apologise if I had misrepresented his parliamentary contributions. You will have guessed what is coming next, convener. Having checked the Official Report, I cannot find the position that I characterised in Mr Balfour’s parliamentary contributions. I can find that only in a blog that was written outwith the Parliament.
I therefore apologise for inaccurately representing Mr Balfour’s stated position in his contributions to the Scottish Parliament. I am disappointed that he is not here as I put that on the record, but I will reach out to him personally.
Thank you very much for that, Bob. I am sure that Jeremy will appreciate that.
The committee now moves into private session to consider the remaining items on the agenda.
10:42 Meeting continued in private until 11:01.