Social Justice and Social Security Committee

Meeting date: Thursday, December 22, 2022

Official Report 644KB pdf

---

Budget Scrutiny 2023-24

Our next item is an evidence session on the Scottish Government’s 2023-24 budget with the Scottish Fiscal Commission. I welcome to the meeting Professor Graeme Roy, Professor David Ulph and Michael Davidson, who is the SFC’s head of social security and public funding. Thank you very much for accepting the committee’s invitation.

There are a few points to mention about the format of the meeting before we begin. Members who are attending remotely, please wait until I say your name before speaking. Colleagues who are in the room should indicate to me or the clerk if they wish to ask a supplementary question, and committee members who are online should use the chat box or the WhatsApp group. Before we move to questions, I invite Graeme Roy to make opening remarks.

Good morning, convener and committee members. Thank you very much for the opportunity to come along and give you some evidence on our forecast, which was published last Thursday. The report comes at a time when the near-term outlook for the Scottish and United Kingdom economies has weakened. The purchasing power of household incomes is anticipated to fall by the largest amount since Scottish records began in 1998. Inflation should peak at around 11 per cent by the end of this year, which will outstrip earnings growth across the economy. High inflation and the recession will affect everyone, but there will be particular pressures on lower-income households, in part because they spend a larger share of income on essentials such as energy and food.

Higher inflation feeds through into our forecast of social security spending, because the majority of payments are increased each year by inflation. The rate being applied in April 2023 is 10.1 per cent, which is the level of the consumer price index from September 2022, and that accounts for more than £400 million of the increase in our forecast for spending for 2023-24.

We forecast that total social security spending will be £5.2 billion in 2023-24, rising to £7.3 billion in 2027-28. The increase in expenditure comes from more people receiving payments and, on average, people receiving higher payments over time. The overall difference between the funding that the Scottish Government receives for social security from the UK Government and its social security spend is estimated at £776 million in 2023-24, growing to £1.4 billion in 2027-28. There are uncertainties around those forecasts, which I am sure we will pick up in the evidence session when we come to it. One of the main drivers of the difference is the Scottish child payment, with forecast spending of £442 million in 2023-24. The Scottish child payment is a major part of the Scottish Government’s tackling child poverty delivery plan, but, as it is a new benefit, there is no associated funding from the UK Government.

There are changes to disability benefits, which gradually increase spending over the forecast horizon. Most significant is the adult disability payment, on which we estimate that spending will exceed the equivalent funding by £208 million in 2023-24, rising to £659 million in 2027-28. As I said, that additional spending is more uncertain than other parts of our social security forecast, as it is based on our assumptions and judgments of the impact of Scottish Government reforms.

To reduce the uncertainty in our forecasts, it is important for us to receive timely and detailed data on disability benefits in Scotland, and I wrote to the convener last week to provide the committee with an update on our access to such data.

Thank you very much. We move straight to questions from members. Paul McLennan will kick us off.

Good morning, witnesses. It does not seem that long ago since you were last in front of us, and there are still the same issues. You touched on a couple of issues right at the start, one of which was the cost of uprating, which is included in the block grant adjustment. To what extent does inflation create a risk to the Scottish budget?

The second issue is the current level of uncertainty in inflation forecasts. You talked about the inflation rate being 11 per cent and it impacting people on lower incomes. Will you say a bit more about food inflation, in particular? Food inflation is predicted to be high, and it is already high. I ask Professor Roy to answer first.

I will give you a high-level answer, then I will bring in my colleague to pick it up. You are right: one of the big differences in doing the forecast this time compared with previous years is the effect of inflation and, in particular, how it feeds through to the uprating of social security payments. Figure 5.1 in our report shows an increase of £3.6 billion, of which about £1.2 billion comes through from inflation. It adds quite a significant amount of the additional expenditure that we think will happen in the years ahead.

On the risk to the Scottish budget, I note that, because of the way that the fiscal framework works, the social security benefits that have an equivalent benefit in the rest of the UK are being uprated by an equivalent amount, and, via the BGA, that feeds through to the Scottish budget. On that basis, there is not a risk in the sense that the significant uplift that is happening has an equivalent in the BGA.

There are two caveats to that, which are worth highlighting. First, inflation is having an impact not just on the social security element of the budget; it is obviously having an impact on the broader Scottish budget more generally. That is probably shown most starkly if you look at the totality of spend. We think that that will go up by about £1.7 billion, but that is only about £279 million in total once inflation is accounted for. The amount of capacity elsewhere in the budget to potentially cope with additional changes in social security is being squeezed, so risk potentially exists there.

Secondly, the elements of the social security payments that do not have an equivalent in the BGA, such as the Scottish child payment and additional changes that the Scottish Government has made to existing benefits that do not have an equivalent part in the BGA, are not factored into the forecast. Figure 5.10 shows that, by the end of the forecast, those elements total about £185 million.

That is the broad answer to your first question, but I will hand over to David Ulph to add anything and perhaps to pick up the question about food inflation.

Let me make a few additional points. To get this in context, uprating for inflation accounts for only £28 million of the £776 million estimated funding gap for 2023-24, so the proportion of the funding gap that is accounted for by uprating for inflation is quite small for next year. For 2027-28, the corresponding figures are that, out of a total funding gap of £1.4 billion, uprating accounts for only £185 million. The amount that is attributable to uprating is very small, both next year and through into 2027-28.

A useful figure that you might want to bear in mind is that every additional 1 per cent increase in inflation adds just over £10 million to the funding gap in the year in which that happens and in every subsequent year after that. An important aspect of uprating for inflation is that it has a persistent effect. Once you uprate the level of benefit, that higher benefit applies for ever more.

We have not broken down the amount of uprating that is attributable specifically to food. Michael Davidson, do you have any figure for that? No. Paul McLennan is right that it is a very serious component.

Anecdotally, we have seen various forecasts and estimates that say that food inflation is probably nearer 15 per cent than 11 per cent and that the rate for other essentials is probably nearer 20 per cent than 11 per cent. We need to dive a little bit deeper into the impact on lower-paid people, in particular, because, proportionally, they have to spend more on food than anybody else. We must be aware of that when considering the issue.

The important point is that there are different inflation rates for different people in the population. It depends very much on what prices are going up and what proportion of the budget that accounts for for those people. In this case, because inflation is largely driven by energy prices and that is having a knock-on effect on food prices, those are very high proportions of the budgets of poorer people. That is why it really matters for poor people. Their effective inflation rate is much higher than the average.

Is the Fiscal Commission considering doing something more specific on that in the future? You said that, proportionately, it is a higher part of their income. Could the commission do a deeper dive into that and, for example, say that there is a more specific impact for people in a lower income decile? Again, that really impacts on the work of the committee. I know that we are talking anecdotally, but could the commission do a more specific piece of work on that?

There are other bodies such as the Institute for Fiscal Studies that produce studies on that—we cite the IFS’s work in our reports. We try to highlight distributional breakdowns of the effects in our reports, but, because a lot of our stuff is at high-aggregate level, it does not necessarily drive quite a lot of the forecasting that we do. However, whenever we think that it is relevant, we try to bring that in. Graeme Roy, do you want to add something to that?

I agree that the effects of inflation are a challenge, and I have a couple of points to make on that. First, David Ulph and Paul McLennan are right about the different inflation effects that people face. There is also something about the proportion of that to people’s income and, therefore, how much they can adjust their behaviours. If someone has high earnings and high savings, they can flex their expenditures and behaviours much more easily than someone who does not have a high level of income. It is an issue.

As David Ulph was saying, because we look at the macro picture in our forecasting, we tend not to go down to that level. There is a gap in UK statistics about different inflation rates across the country, which is an issue. There is an increasing amount of work going on to look at different inflation rates across individual household types, but we do not have different inflation rates for different parts of the UK, so there is not a published inflation rate for Scotland or for the rest of UK.

There is another thing. Not only do we know that there are different price levels in inflation for households, but, even within Scotland, energy costs in rural areas are much higher than they are in other parts of the country, which adds to the complexity point that Paul McLennan has raised.

The SFC has talked about forecasts of increasing numbers of applications for disability benefits. Are there similar trends in Scotland to those in the rest of the UK?

There are a couple of things to say on that. Recent trends are towards increasing numbers of applications; members probably saw that quite a lot recently with the significant increase in personal independence payment applications in the whole of the UK. Broadly speaking, there are similar trends in that context between Scotland and the rest of the UK. Figure 5.5 in our report is quite a helpful chart that shows the jump in the number of applications for PIP in Scotland. Broadly speaking, we see the same increase in inflows over that time period. I will bring in David Ulph on the work that we have been doing to think about what might be driving that.

We have been thinking about the range of factors that might be driving that. In our report, we identify three sets of drivers. One driver is the cost of living crisis, which operates in two ways. First, people with cost of living pressures are much more likely to apply for whatever money might be available for them, so that tends to increase the number of applications for any type of social security. If people think that they are eligible for disability benefits, they will apply for them in cases where they might not otherwise have done.

The other effect of the cost of living crisis is that it can create mental health problems for people who are trying to work out how to cope with that crisis, and that will trigger a disability claim for mental health reasons. Therefore, the cost of living itself is one of the factors.

09:15  

Another factor is the lengthening backlog in the national health service. The backlog in the NHS means that people who have a disability and are waiting for treatment are not getting that treatment, so a mild or medium condition can turn into a very serious condition that entitles them to disability benefits.

The third factor is the long-term impact of the pandemic through some of the effects of long Covid, particularly on mental health, which is a major component of disability benefit, but also through other channels. We know from the work that we have done that about 40 per cent of people who claim disability benefit do so for mental health issues, so the mental health effects of the cost of living crisis and the long-term impact of the pandemic are a serious component of what is driving up claims.

That is extraordinary. Does that change the estimate of how much more ADP will cost compared with continuing with PIP?

There are a couple of things to consider when making a judgment on that. In the increases in the number of PIP applications across the UK, how much of that is additional demand and how much of that is demand from people who are already eligible and are now taking it up? Our assessment at the moment is that it is mostly new demand. It is mostly, as David Ulph said, linked to things such as legacy effects of the pandemic. On that basis, we think that the increase that will come through will, ultimately, feed through to ADP, because it will be mirrored in the UK and kind of proportionate, so a BGA will flow behind that.

That does not change our assessment that the changes that the Scottish Government is applying around ADP to make it a different system will unlock a potential increase in payments and eligibility from people who would already be eligible for that. We made a slight adjustment in our forecast for the short term because we think that there is a fixed number of people who will do that, but it does not really change our assessment that there will be additional people claiming ADP in the future, even with the spike up in the number of PIP applications that we have seen in recent months. The PIP increase is new demand and, in ADP, the extra funding element is from changes in the new system.

I have a brief supplementary question. Thank you for answering our questions so far and for what you have shared in advance of the meeting. I want to pick up on the points that you made on mental health and the backlog in the NHS. Can you give a proportion of the increases that are attributable to those two factors?

Unfortunately, we cannot. We do not have enough data for a long enough period of time to be able to break that down. The other problem in trying to break it down is that, for many people, there might be multiple factors at work in driving what is happening. Of those three elements that I discussed, the increase in uptake might be because of cost of living pressures and mental health issues and other problems. It is quite hard to disentangle the three factors and say exactly how many people are coming out from each. We do not have the data yet to track down to that level of detail.

Do you expect to get the data? Is it gathered?

We can look at the Department for Work and Pensions data on that, which is a very rich data source. We can look at that and track it, so that gives us our baseline number of claimants and we can see how far that goes up. That is essentially what we are talking about here. We are not talking about the new people coming in because of Scottish Government reforms; we are talking about an increase to the baseline. We can do some further analysis, but it will be difficult because there are many people for whom more than one factor will be in play.

We track supplementary data—perhaps not on the social security payments side but things such as labour market data. Particularly at a UK level—as long as we assume that there is no difference between Scotland and the UK in the spike—in time, you can track to see whether there has been an increase in inactivity and people saying what the reason for the increase in inactivity has been, whether it be for ill health or early retirement and so on. Once we see changes in the labour market, we can start, in time, to get more evidence on whether there is an increase in people saying they have ill health conditions, which we can map across to see whether that is correlated with the social security data. However, as David Ulph said, it is uncertain at this time. All that we see is the spike and, at this stage, we do not genuinely understand what is driving it.

Good morning, panel, and thank you for coming to the meeting.

It may be too early to ask about this, but one of the forecasts that you or your predecessor made was that there would be a higher uptake of ADP compared with PIP. I think that the Scottish Government has budgeted on that, because the system is meant to be kinder, fairer and smoother. Are we seeing that trend—more uptake of ADP compared with applications for PIP at the DWP—or is it too early for that trend to show yet?

In short, it is too early. Obviously, there has just been the pilot and the initial roll-out. We will get our first real cut of the data in the spring of next year probably, and we will start to see the broad trends in that. However, even then, the data will not be robust enough for us to start to use it in our forecasts.

We have mentioned the uncertainty. We are at a really tricky point in the forecasting period at which we have a completely new approach to the benefits coming through the system, but, until we start to get data on take-up, the type of people involved, the in-flows, the age profiles and the reasons for people claiming, and until we get a decent track of that, we will not know whether the things that we think might happen have really come through.

In short, it is too early for that, but I hope that we will get that in the next year or so.

That is helpful. I was about to ask roughly when we should look for that. Will we be able to do some analysis to say that, if we still had the DWP PIP scheme, we would reckon that X number would have got awards but, because there are ADPs, the number is higher or lower than that? Will you be able to do that, or will that be too difficult to dig down into?

I will go first. David Ulph or Michael Davidson might then want to come in.

At a crude level, that could be done to an extent. The block grant adjustments could be compared with how much is being spent on ADPs, and the approximate difference between the two will show that. However, what we really need is quality data. That is exactly what we are talking about: it all depends on the quality of the data. Are we seeing differences in take-up by people with different types of conditions, by age, or by gender? In all those things, it ultimately comes down to the quality of the data. We have pressed so much on having robust data as soon as possible because that will let us answer those questions and start to change our forecasts if we need to do that, depending on take-up and eligibility.

The Office for Budget Responsibility is involved in the process because it forecasts the block grant adjustment, which, to some extent, would have happened had the Scottish Government not made the reforms. Over the past year, we have managed to bring our forecast on that much closer to the OBR forecast. The problem is that its methodology is very different from the methodology and the data that we have, and we have not been able to mimic exactly what it is doing. In some ways, it is helpful to have two different people using two different approaches to the same thing. The fact that we are now narrowing that gap means that we are starting to get better at forecasting.

Essentially, what would have happened had the reforms not been made is a counterfactual question. We are trying to forecast a relatively hypothetical number there, but the fact that the two methodologies are starting to converge gives us some reason to think that we will be able to do that better in the future.

You mentioned the funding gap. I think that £881 million was forecast, but that figure has dropped slightly. Is there any specific reason for that? That is probably a question for Professor Roy.

There are a couple of issues. We will maybe come on to this when we have a discussion about what we mean by a funding gap. In some ways, the phrase “funding gap” is not always entirely helpful, because it creates the impression of a hole and a gap. Actually, none of that spending is ring fenced by the BGAs—it is what the Government chooses to spend on priorities across the entire suite of public spending and from looking at the revenues coming in. We will maybe come back to that. It is really important to be clear about what we mean by funding gaps in that context and about the uncertainty there.

On the difference between the BGAs and the amount of spending on that area, I think that that has narrowed by about £106 million in 2023-24. On the key reasons for that, we have discussed PIP and ADP and the fact that some of the increase in ADPs that we were thinking of is being captured by what is happening on PIP in the United Kingdom. There is an associated increase in the BGA relative to what we thought was going to happen because of the increases in PIP across the rest of the UK.

There is a small, modest issue around the child disability payment. We do not yet have data to update our forecast from last year, but the OBR has increased its forecast for the child disability payment across the rest of the UK. That leads through to a higher BGA. That is the other key reason why the gap between the two has decreased.

Although we have not updated our forecast because of those data issues, the work that we have done suggests that we think that, broadly, the outturn data is in line with what we forecast. That is why we have not revised up or changed our forecast on the CDP.

We will move to questions from Pam Duncan-Glancy.

With the exception of one small question, my questions have largely been answered. I noticed that, in the data—forgive me if I am misreading this—there are assumptions that there will be no increases in the winter heating payment or child winter heating assistance and that there will be less spending on best start grants. Can you say anything about what led to those assumptions?

Did you mention the pensioner winter heating payment?

Yes.

Does Michael Davidson want to come in on that?

Yes. We made assumptions around what the Scottish Government’s policy on uprating will be. There is a group of benefits for which uprating is applied and there is a statutory process for that, but the approach is more on a year-to-year basis for the other benefits. There have previously been decisions. The decision on child winter heating assistance and the winter heating payment has been to apply uprating in 2023-24, when those payments are made. However, we assume that, following that, the policy will revert back to not uprating those in future years.

09:30  

The overall expenditure that we forecast for the winter heating payment stays level because, roughly speaking, the number of people whom we expect to receive it will stay constant. Once the payment level has increased in 2023-24, we have kept it fixed for the rest of the horizon. Obviously, if policy changes, we will factor that into future forecasts, but that is our understanding of what the policy is for that benefit at the moment. I think that that also applies to child winter heating assistance.

Thank you. Just to check, you do not think that there will be an increase in child winter heating assistance applicants.

Broadly speaking, there could be some increases in child winter heating assistance applicants, but that very much lines up with our child disability payment forecast. Although that increases a bit towards the end of the horizon, the increase there is more related to the uprating than the actual number of children getting it. I think that that is right—I am just checking my caseload. There is some increase in the caseload for child disability payments and for child winter heating assistance as well, but that increase is fairly marginal. We have rounded the expenditure to £5 million. There are probably increases within that, but the rounding to £5 million probably makes it look flat.

The reason why I asked that question is that that felt a little inconsistent with some of the information in the previous conversation about the increases in disability payments and why that would not be seen to be driven through the child winter heating assistance. From our cost of living conversation, it would appear that more people might become eligible for some of the winter heating payments as a result of that situation. However, you are working on the assumption that that will not be the case.

Yes. The winter heating payments are a combination of benefits and low-income benefits. Although there is a slight increase in unemployment in our economic forecasting, a lot of the people who already have low incomes will be eligible through their low-income benefits rather than through becoming unemployed. We think that the effect is quite marginal.

That is helpful. Thank you. Those are all the questions that I had on that theme.

We have covered a lot of these questions, but I want to revert back to a point about financial management. Where do you think that the Scottish Government is in developing that? Graeme Roy, you spoke about “the funding gap” and hinted at what you thought about the use of that phrase. Where is the Scottish Government when it comes to managing that?

It is not for me to comment on how the Scottish Government is managing that. We do the forecasts and set out the position, but it is for the Government to manage that.

I will say a couple of things. The phrase “the funding gap” is quite challenging because it creates the impression that there is a gap when, in fact, what the Government is doing is setting out the totality of the budget, the different elements of spending, what it is doing on the revenue side and the balance when it comes to making choices on social security, public services and tax more generally. We know that policies such as the Scottish child payment, which are new policies that do not have an equivalent BGA, have to be funded through taxation or through adjustments in public services in other areas. Out of the funding difference, that will come to about £450 million by the end of the forecast period.

There are all the other benefits that have a BGA aspect. If a different approach is adopted and having a different system leads to higher take-up and higher average payments, that will have to be paid for. That comes back to the point about what that means in terms of the balance of tax and the balance of public services. I guess that the point—this is not to answer your question but to go as far as I am prepared to go—is to say that, essentially, the Government needs to plan on that basis, just as it would do with all our forecasts, and say, “This is what the independent assessment is of potential funding coming down the line.” It must plan for that and think about its taxes, about its public spending in other areas and, ultimately, about the outcomes that it is trying to achieve. The totality of all of that is really important.

That is helpful.

An important issue to think about when we are talking about managing this area of spending is that the distinction between social security and other areas of Government spending is that social security is essentially a demand-led form of payment. Once the policy and the criteria have been designed, the rates have been set and Social Security Scotland has implemented those policies, the amount of money that goes out the door will depend on the number of people who turn up and claim it. That is not something that you can control. That is why it is so important to think ahead and to understand, albeit from our highly uncertain forecasts, what the potential implications will be for the long-term funding of social security, because it is not something that can be managed in the same way as health or education spending.

That is a very good point. The projection that there will be £106 million less of a gap since we last met is important, but it is a data question, which I know we will come on to later.

I will touch on forecast accuracy. You have highlighted that the forecast error went from 4 per cent down to 2 per cent, and you mentioned two factors. Could you say more about that? Looking at the years ahead, how difficult or how much easier will that process be? You have touched on how you assess Government policy when you look at this. How is that working? Could the Government work more closely with you in that regard? What is the relationship in that respect? I know that there is meant to be a process of stepping back, but could there be closer working between you and the Scottish Government on the forecast element of things?

You are right—on forecast accuracy, the forecast error went from 4 per cent down to 2 per cent. The key reason for that is that, when we first do our assessment, we use provisional data. Later on, when we got the fully audited data from Audit Scotland, that identified a slight issue with the provisional data. Once that was adjusted for, the forecast error came down to 2 per cent.

To be honest, we get such issues all the time; they are to do with differences between the provisional data and the audited data. Ultimately, it is the audited figures that are the most important. It is just a timing issue—when we do our forecast evaluation report in the summer, the audited data is not ready. Audit Scotland said that the error that it found was isolated, so it will not be a recurring issue. That broadly explains that.

With regard to your broader point about future uncertainty, I touched on that a wee bit in my answer to Jeremy Balfour. Particularly with adult disability payment and child disability payment, we are in a very challenging period in relation to the forecast element of the new payments that are being set up, in that we know that the policies are being rolled out but we do not yet have the ex-post data to enable us to see the effect.

Over a number of years, in consultation with the Government and by drawing on evidence about what has happened in the past, we have made our own judgments about what we think might happen with the changes in policies, but there are margins for error in that. Only once we start to get the data coming through will we know what the additional take-up is, what is happening with additional eligibility or what is happening with the payments. In some ways, we just have to be patient; we will start to know about that. That brings us on to the data point. The better granular data we get and the sooner we get it, the sooner we can start to make those assessments.

As far as the relationship with the Government is concerned, we have close engagement at an official level with the Scottish Government and Social Security Scotland about the data. Since we wrote in the summer, we have had conversations with them about the data and the types of information that we need. During this forecast round, we had helpful discussions with senior colleagues in the Government on the qualitative side, on what their sense is of how things are going and whether there are any areas where they are finding something completely different to what they thought and what we thought. Those sorts of positive and constructive discussions are helpful but, ultimately, it will be when we get the data that we will be able to properly assess the situation and to see whether the programmes are having the transformation that is hoped for.

I would like to take some of those questions a bit further, as some of the questions that I had intended to ask have already been answered. What further data do you think that you need from Social Security Scotland? You mentioned that data that you get from the DWP allows you to make further assumptions about the impact or where the spikes are coming from. Do you expect that Social Security Scotland will gather similar data? Have you asked for that? As you know, we will hear from Social Security Scotland shortly, so we have a timely opportunity to indicate if there are any areas that you think that it might need to consider.

I will give a quick answer before handing over to David Ulph, who has probably been working on this area slightly longer than I have. We do not yet have the data. Obviously, ADP is just being rolled out, but CDP has been out since November and we do not have the data that we need to change our forecast there. We do not have information such as average payment award, nor do we know things such as the number of new clients and inflows coming in. We need that information in order to be able to assess what is going on there.

In the summer, we wrote to Social Security Scotland—we copied in the committee—about our statement of data needs and what we need in that regard. Quality of data, including on the transition period, is really important so that we can see whether those policy changes have made a difference.

I will hand over to David Ulph to give some broader reflections on the work around data and what we need.

To go back to child disability payment, one of the issues that we talked about last time was that of collecting data on the sex of the child. That is not being done systematically by Social Security Scotland. Data about the sex of the child is provided only through the equalities monitoring part of the application; it is not done as part of the application process itself. That may not matter in terms of the amount of award that Social Security Scotland gives people, but it might matter if there are certain types of conditions that are more prevalent among boys than among girls. Being able to get data on the sex of the applicant helps us to understand whether such trends are there and helps us to better forecast the extent to which there will be higher average awards because of the composition of the group of children who receive the benefit.

As we get enough data coming through the equalities monitoring form, we will need to do a piece of work to see how well that matches up with other forms of data, such as the data that is collected by the DWP through its process in England. We need to find out whether there are systematic biases in the application process and whether, for example, people are choosing not to use the equalities monitoring form or are systematically doing that when it is a boy rather than a girl. If there was a systematic bias there, that could really affect our future forecasts.

If what comes through on the equalities monitoring form broadly matches up with what would have happened on the application form, over a period of time it might not matter so much that we are not getting that information through the application form. We can use what is there through the equalities monitoring portion. However, we will not know that for a period of time. Unless Social Security Scotland were to change its policy, there will be an issue about that form of data that will persist.

We could ask about that. Do you know why Social Security Scotland made that decision?

I think that it would be up to Social Security Scotland to answer that question.

That is fair.

Professor Roy, on the data points that you raised, has Social Security Scotland said that it will be able to collect that data?

09:45  

We will get an update in the spring. In part of our conversation with Social Security Scotland, it said that we will get basic data for CDP and ADP in the spring. As I said, we are in a challenging period in which, even on ADP, because it has only just started, we do not know whether what we will get is genuinely reflective of the longer-term trends. The position on CDP will be slightly better, because it will have been running for a longer period of time.

On the timing issue, it will probably be another year or so before we can be much more confident in what we say about ADP, but the data that we get in spring will be very important. We will know then how granular it is, how comparable it is to the DWP data and how the transition has been handled. We will be able to track people coming through and moving from one benefit to another to see whether there have been any changes. That data release in the spring will probably be very important in enabling us to see how robust the information is and how helpful it is in assisting with the forecasts.

Are there any obvious differences that you already know about between the data that is collected by the DWP and the data that is collected by Social Security Scotland, other than those that we have just discussed?

Broadly, the differences are those that we have discussed. As I said, once we get the data, the key thing will be to track it through. Information on the average payment, payment flows and so on will really help with the forecasting because that will enable us to say, “Look, this is what’s been captured, so if we make these assumptions about cost of living, take-up or eligibility, we can start to adjust that.” Once we get the data, we will be able to make an assessment of that.

An important issue here is not just the collection of the data but its publication, because we like to base our forecasts on published data so that if somebody else wants to try to replicate our forecasts, they can do so. That is all part of our mission to be very transparent about how we produce our forecasts. We always tend to rely on published data rather than data that is given to us through the back door, as it were. Therefore, it is important that Social Security Scotland does not just collect the data but finds tools and techniques to publish it. The DWP has a well-developed tool called Stat-Xplore, which it uses to make its data available. If Social Security Scotland could develop a similar tool, that would be really helpful to us.

At the moment, are you relying on data that is not published?

Again, that is a matter that you would have to pursue with Social Security Scotland, but at the moment we do not see that coming along.

I can answer that. Our understanding is that the volume of data that Social Security Scotland will publish on CDP will increase, but we are not sure whether the data that it provides in March will be published at the same time as it is able to provide it to us for the forecasting.

That is helpful. Thank you.

On a similar theme, on page 91 of your report, you state:

“We are concerned that the CDP data dissemination issues will carry over to adult disability payment ... statistics.”

What conversations have you had with Social Security Scotland in that regard? How reassured are you? Obviously, Social Security Scotland will be giving evidence later this morning. Should we pursue that issue, or are you content with what you are hoping to get?

We are broadly saying that the issues relating to accessing the equality data that we need on child disability payment are the same as those relating to adult disability payment. However, the magnitudes are much bigger, which is why the matter is really important from a funding point of view, rather than from a social security point of view.

As I said, our statement of data needs still stands. The key thing is still the request for data that we made in the letter that we wrote in the summer. We do not have that data, so we have not been able to update our forecast. We have been engaging with Social Security Scotland and the Scottish Government to impress on them the need for that data, and we have had constructive conversations with them over this period, but we do not have the data. We are told that we will get updated basic, useable data in March, but until we get that, we do not really know. As I said, we have written a statement of data needs, and any support in strengthening that request for data would be very welcome.

Professor Roy, I want to bring you back to the point that you made at the start of the meeting about the fiscal framework. I am interested in your thoughts on the fiscal framework. Part of the issue relates to the complexity of the existing fiscal framework, and I know that there are on-going discussions between the Scottish Government and the UK Government on the matter. What are the commission’s asks from those discussions? Given what you said about part of the issue being about the fiscal framework, what would make your life simpler in relation to your work? What are your asks? Obviously, there is a lot for the UK Government and the Scottish Government to discuss in relation to social security and so on, but what changes to the fiscal framework would make things easier for you?

First of all—you would expect me to say this—ultimately, it is for the two Governments to negotiate and decide on the fiscal framework. A broader point that I would make—

I am asking you to avoid the politics. What would be your message to the two Governments? Without going into specific details, what are the key things that you would like them to discuss?

One of the things that I, as a member of an outside organisation in all of this, have been struck by is the uncertainty that exists in fiscal and economic forecasting, although I think that we now have a much better handle on that. For good reason, I do not think that we thought too much about that back in 2016, when the system was being designed and the framework was set up.

The review was always designed to be quite timely so that, five or six years after the framework was agreed, we could look back and consider how it was coping. Are the levels of borrowing and forecasting—all those sorts of things—designed as we hoped or thought they would be? For me, a key part of the fiscal framework review involves thinking about questions such as, “Is this how much uncertainty we thought there would be?” and “Is this the moveability we thought there would be?” Part of the reason for that is that we have been through quite an exceptional economic period. Back then, who would have thought that we would have Brexit, the cost of living crisis, war in Ukraine and a global pandemic?

There is lots of economic uncertainty, but we now have a much better handle on things. Our forecast evaluation reports were really quite good in showing how things relating to our income tax forecasts, our economy forecasts and social security were potentially moving around. My advice is that, when people work on the fiscal framework review, they should look at our evidence and consider, given the moveability and how that is impacting on potential projections for funding, whether we have the right tools and the right flexibilities to manage the situation and whether we have the freedom to make the adjustments that we would like to make. However, ultimately, it is for the Governments to decide whether that is the case.

That is really useful.

On the last point that Graeme Roy made, it is striking that—we have made this point in some of our reports—the borrowing limits and the limits relating to what can be paid into and drawn down in the reserves do not vary with the size of the budget.

I have raised that issue with the committee, as other members know. It is very interesting and heartening to hear that.

My second observation is that a learning process is taking place. All our discussions with the Scottish Government suggest that it is starting to learn how to think about managing some of the uncertainty that Graeme Roy talked about. When a big reconciliation will be coming along, it can look ahead and think about doing various things with various elements of funding to try to land that particularly well. A learning process has been taking place, and we are all getting a bit smarter at doing this stuff.

I appreciate that. Thank you.

I thank the witnesses for providing evidence this morning. It has been very helpful, especially in advance of our next session. I will briefly suspend the meeting to allow us to set up for our next panel.

09:54 Meeting suspended.  

10:00 On resuming—