Welcome to the Public Accounts Committee on Thursday 11 December 2025. The Whole of Government Accounts provides a picture of the nation’s finances, combining the accounts of over 10,000 public bodies. In 2023-24, the WGA reported over £1 trillion in public spending, and balance sheets showed over £5 trillion of gross liabilities. For the second year in a row, the C&AG has disclaimed the Whole of Government Accounts due to the level of missing and unaudited data, particularly from local authorities, leading to significant gaps in data in the WGA. Our Committee has previously raised concerns about the unprecedented crisis in local audit and its impact on the WGA. Today, we will therefore be examining the progress in addressing the issues previously raised by the Committee regarding local authority audit arrangements. We will also challenge the Treasury on the usefulness of the information included in the WGA and explore the future of the WGA itself. We are very pleased to have with us the Permanent Secretary at the Treasury, James Bowler. Good morning, James.

Good morning.

Would you like to introduce yourself? And can all your team introduce themselves, please?

Certainly. To my left is the director of public spending, Conrad Smewing, and the director overseeing the Whole of Government Accounts, Andrew Cartner. Will and Rosie, do you want introduce yourselves?

Good morning. I am Will Garton, the director general for local government, growth and communities at MHCLG.

I am Rosie Seymour, deputy director for local audit reform in MHCLG.

Rosie, is this your first appearance?

No, I came in January.

Right—so you have all been here before and you know the form. That is very good indeed. Let’s get stuck in. I am sorry that I haven’t given you any warning of this question, Permanent Secretary, but I sort of dreamt this up in my sleep last night.

Oh good!

It concerns the Ministry of Defence. I have two questions. We have serious problems with how the Government’s stated aim to move from 2.2% to 2.6% of GDP is going to be paid for. It wasn’t really mentioned much in the Budget; are you able to tell us anything about that?

I will start and then the director of public spending will probably provide you with more facts. We have a spending review, which we did last summer, in July 2025, that sets out three-year spending plans for the MoD that are entirely paid for. My understanding is that that takes spending on defence under the NATO definition to 2.6%. There are then plans thereafter for the rest of the Parliament.

That is exactly right. The spending review from summer 2025 sets out all the budgets for resource DEL out to 2028-29 and for capital DEL out to 2029-30. I think those go up to 2.6% of GDP in 2027-28. In terms of how that was afforded, it is from within the overall spending envelope. The Prime Minister made an announcement—I think it was in March or something like that—that reduced spending on overseas aid to increase the spending envelope for other Departments, to allow defence spending to go up to 2.6% of GDP. The link was made earlier on to the reduction in overseas aid spending.

That is really helpful. Sorry to digress.

Not at all.

I think everyone would agree that the situation we got into with regard to local authority audit—the fact that we ended up with a disclaimed opinion on the Whole of Government Accounts because of missing data from local authorities, or data that was not complete because it wasn’t audited—was unsatisfactory. On the Whole of Government Accounts for 2024-25, what position are we likely to be in with local authorities producing audited accounts, or unaudited accounts with at least most of the data there?

I will start, and then I am sure we can pull out lots of issues. I agree with your assessment of that being unsatisfactory. There has been progress in this and for future Whole of Government Accounts in two regards. First, the level of missing data is now falling. I can confirm that in the last Whole of Government Accounts we had 211 entities that were missing data, while the 2023-24 WGA has 201. To answer your question, in 2024-25, we expect, because we have done our data sweeps—we will publish it before the summer—that there will be 145 entities missing. So it is coming down, but it is still too much. Also, we are in the process of legislating, and the second backstop has now delivered, so 98% of local authorities have submitted draft accounts for ’24-25. To answer your question directly, the data quality issues persist in local authorities. This Whole of Government Accounts has a disclaimed status from the NAO, which is understandable because of that missing data, and our expectation for ’24-25’s Whole of Government Accounts is that that disclaimed status will unfortunately continue. We want to be very open with the Committee about our expectations here. We are, I think, doing the right things in terms of recovery. There is a trade-off between timeliness and quality, and we are choosing timeliness. We are using the data that is available rather than waiting for it to be audited. My colleagues can set it out in more detail, but the answer is that we are making progress on getting accounts from local authorities, so the missing data is going down, but they remain, in certain elements, unaudited or qualified, so we expect the disclaimed status to continue.

So essentially, the large number of authorities with no data, or missing data, now become authorities with disclaimed accounts.

Yes, either unaudited and therefore disclaimed or qualified. Yes, you are right.

So progress has been made, but it is a lot slower than the progress we thought we were going to make when you last gave information to the Committee.

I think partially. Progress has been made, and I pointed out the decline in missing data. When we gave evidence at the start of the year, our expectation was that, as the accounts came in and the backstop swept up the requirement for local authorities to deliver accounts, more of them would be of a higher quality than they have ended up being. I think you are right that we were potentially over-optimistic in that regard. I do still think that we are doing the right thing to require and help local authorities to get into a better place, but absolutely, the situation in local authority audit is poor.

You were over-optimistic and the progress has not been as quick as you would have hoped, so the obvious question is: why?

Obviously we agree; the situation where disclaimed accounts in local authorities are leading to the Whole of Government Accounts being disclaimed is absolutely not acceptable. However, as James said, since July 2024 the backstop has had a really big impact. Until the year ’22-23—until recently—we had 1% of local authority accounts audited. That is now 99% and, as James said, for the year ’23-24 we have got 94%. For ’24-25, 98% have been published. There is progress. When we came before the Committee in January, we talked about two aspirations. We were deliberate in our phrasing that they were aspirations and not hard and fast targets, not because we were not ambitious and we did not have a sense of urgency, but because we knew that we could not commit. I said explicitly that I could not give a guarantee to the Committee that we would meet our aspiration to have most accounts not disclaimed by the end of ’25-26. We have done a lot of work since January, and what the auditors are telling us is just that the complexity and the size and scale of the problem is greater than they had envisaged. They have been quite clear with us that they do not think they can meet that aspiration. We still maintain our aspiration to have no qualified or disclaimed accounts by ’27-28. Again, that is hugely ambitious. We put that in the public domain to try to be transparent about what we want to achieve, but we were always caveating it and were cautious that it was not a prediction, just because the scale and complexity of the problems are so deep.

There are two issues, then. For ’24-25, how many local authorities will produce accounts that are fully audited and not disclaimed?

We will not know that—

What is your expectation?

I cannot give a prediction. We will not know finally until January ’27 for the year ’24-25—that is when the backstop kicks in—but it is good news that 98% of accounts have been published. That simply would not have been the case some years ago. If we were at 99% for December ’24 and 94% for February ’25, then I hope it is a really good number. It would be unwise of me to pin a number to the mast this morning, because there are a lot of factors outwith our control, but—

But you just pinned a number to the mast for ’27-28, so why can’t you do it for ’24-25?

For financial ’24-25—the backstop for that is February 2026—we were very clear that our expectation would be similar numbers of disclaimers to what are in the system at the moment. We were never expecting the level of disclaimers for the next backstop, which relates to ’24-25, to be significantly different from the backstop that we have just had. We put in place a recovery period over a number of years. The place where we were expecting the biggest change, and a significant reduction in disclaimers, was the backstop in January 2027 relating to ’25-26. It is that aspiration that we are now clear, from the work we have been doing with auditors, is not going to be met. We are saying that very clearly up front now, because the amount of build-back work over the last year is not in line with what was previously estimated, but we were never expecting the level of disclaimers to be very different between the February 2025 and February 2026 backstops.

I think in response to previous PAC reports you indicated not merely that would the amount of missing data come down, but that the number of unaudited accounts would come down.

We were clear about it being the January 2027 backstop date and the accounts relating to 2025-26. For the initial years, we were expecting disclaimers to stay broadly the same, and for that to be the case for the February 2026 backstop.

Getting the accounts audited seems to be the problem: local authorities are now producing data, but the accounts are not audited.

They are being audited, but hundreds are being disclaimed.

Whose fault is that? Is that an obstacle with the audit firms? You just said that they found it more complicated, but local councils are not more complicated than they used to be, are they? Is it simply that the backlog is causing the complications?

It is the process of building back. What is the source of this problem? The 2014 Act put responsibility in myriad different places: regulatory and system leadership in the FRC; code of practice in the NAO; procurement management in the PSAA; audit quality and enforcement in the FRC. That has not worked. The Public Accounts Committee has told us that, the departmental Select Committee has told us that, John Kingman has told us that, Tony Redmond has told us that, and numerous commentators have told us that. That is the source of the problem that we are trying to correct. That is why we are taking primary legislation in the first Session of this Parliament to fix it and create a local audit office that can resolve all those problems.

I will come on to that in just a second. Essentially, you are saying that it is the complication of the past few years of unaudited accounts that is now causing difficulties for audit firms coming in to try to get those accounts audited and up to date.

Yes.

Right. Is that going to cause an ongoing problem? You are being optimistic about sorting it out for 2027-28.

I am trying to get the right balance between ambition, pace and pragmatism. Clearly, in our aspiration for 2025-26 we were a bit more on the ambition and pace and did not quite land it, but we are trying to get the balance right.

Are you now saying to us that local authorities are basically getting their act together, and it is just a question of the audit firms getting their act together to deal with this backlog?

I am saying that a number of things are working. The statutory backstop is working. More money is going into the system as fees are going up—that is good. We have simplified the accounts already for the year 2025-26, and CIPFA is going to make the accounts even simpler still in 2026-27. We still have other mechanisms that will support us. The Bill, when an Act, will get rid of the requirement to have accounts signed off by a key audit partner. There are only 100 of those such individuals in the country, and that is part of the reason that we have the backlog. When we get rid of that requirement in legislation, that will have a beneficial effect.

Coming on to the local audit office, remind me when the Kingman and Redmond reviews were completed?

Kingman was 2018, Redmond was 2020-21.

It has taken a long time to do anything in response, hasn’t it?

The previous Administration’s view was that they did not accept the recommendations of John Kingman and Tony Redmond, nor those of the PAC or the departmental Select Committee. This Government do accept those recommendations; they published a strategy in July 2024 after the general election and have taken primary legislation in the first Session.

Could you explain why the local audit office is going to make a big difference? It seems that the problem is on the ground, with auditors. Why would an office make it all better?

Gareth might want to opine on this, but there are a number of problems in the market. The first is that responsibility for the audit system is spread out across a number of different bodies. There is no single owner of the system. I just talked to you about the difference between the FRC, the NAO, Public Sector Audit Appointments, the Department and the Institute of Chartered Accountants. We do not have any public sector provision in England; we do in Wales and in Scotland—there, while everything is not perfect, things are better than they are in England. Fees have been too low. Accounts are too complicated—as I talked about, the number of audit partners required to sign off an account means that progress is incredibly sticky. The Bill, when an Act in spring next year, will address all those things and will make the system much better than it is currently.

That is a very confident statement.

Well, it can’t be much worse, can it?

And obviously you have MHCLG here and the Treasury here. For the Whole of Government Accounts, we pull together 10,000 accounts across the entire public sector. The obvious point is that that is the input for the Whole of Government Accounts, and the quality of the Whole of Government Accounts therefore is determined by that. The disclaimer in local authorities goes up to Whole of Government Accounts, as you know—I just thought I would clarify that.

Will the local audit office help the Ministry to have a better understanding of where there are likely problems coming down the road in certain local authorities?

It will, undoubtedly. The lack of audited accounts makes it harder to make that judgment. I would say, however, that we are relatively good at anticipating the local authorities in which there are likely to be financial problems. Although audit is a valuable tool in our armoury, it is by no means the only way in which we assess risk in councils. We have a pretty good track record of knowing where the problems are, and a functioning audit system will improve that further still.

There is currently no direct link between the audit of an individual authority, the auditors and the Ministry.

There is if there are statutory recommendations or public interest reports; they are required to let us know. We also have regular engagement with them throughout the time, so we are able to do that, but that has to be in line with their duties as auditors. Statutory recommendations and public interest reports have to be sent to us.

So in this brave new world all local authorities will get their audited accounts to you in time. But what happens if they don’t? What are the sanctions?

I would describe it in three steps. The first thing that we do is engage, as you would expect. We pick up the phone and have a conversation. We sometimes find that deadlines have not been met and compliance has not happened for a valid reason. It may be that a cyber-attack or something else made it particularly complex, and we have a grown-up and practical conversation. Secondly—we have done this already—where local authorities are failing to meet the backstop timetable, we are completely transparent about that. We publish the list. Sunlight is the best form of disinfectant. Thirdly, there are then follow-up conversations from senior officials and Ministers about what we are doing to get back on track. Ultimately, the Department has a best value regime, which is tiered, from a best value notice to commissioners walking into the building. In extremis, we would want to make a decision on best value on a whole range of factors, but ultimately that is our sanction for a local authority that, hypothetically, will not play ball.

But there is a different degree of strength between the requirement to produce a budget—for a local authority to say what they are going to spend—and the audit of what they have spent. Producing a budget by a certain date is a legal requirement, but you do not have to produce audited accounts by a certain date. It is not the same.

The backstop dates are about exactly that. They are a statutory requirement to have audited opinions published. If they cannot do that for some reason, they have to publish an explanation of why not. As Will has explained, we will engage with them to ensure they can comply as soon as possible.

Authorities always produce their budgets on time, don’t they?

Yes. I agree that the problem is in audit, not in budget setting, but the audit process is statutory, as Rosie says, so it has to happen. It has not been happening consistently or effectively enough. That is what we are changing.

Right, so there shouldn’t be any problems?

Well, I do not want to over-promise and underdeliver. We have an ambitious programme of reform and change. We think it is pacy and going as fast as parliamentary time allows. We still have an aspiration to have no qualified or disclaimed accounts by 2027-28. That is what we are aspiring to but, as I said to the Committee in January, I cannot guarantee it. If we end up with a handful of qualified or disclaimed accounts at the end of 2027-28, I think that would still be success. It is an ambitious target—it is not a target; it is an aspiration. It is an ambitious aspiration and we are going to do our absolute best to meet it, but I think the oil tanker is turning around. It’s hard yards.

For the residents in that authority, or in the two or three authorities that do not have those accounts, that is not acceptable.

I completely agree.

Has the Treasury been helping to fund higher fees for the audit firms? That was one of the problems, wasn’t it? The whole idea of abolishing the Audit Commission, apart from everything else, was that it was going to save money and therefore fees would be reduced. Have we got fees back to a reasonable level now? Is the Treasury providing support, and will it continue to support?

We have put £49 million, over a period of years, into just that. Will will have to explain how that links into local government finance going forward, but the straight answer is yes. We have attempted to fund this, including coming out of the Redmond review. There is a question about whether you ringfence it or not when we are trying to reduce the ringfences in local government.

In this period, fees are up 150% on the last procurement that the public sector audit authority has done, so there is clearly more money going into the private part of the system. The local audit office gives us the opportunity to start to recreate public sector provision as well.

There are some quite significant reorganisations in local government coming over this period.

There are.

How is that going to complicate the issue of outstanding audits for authorities that will not exist in a couple of years’ time? Their accounts and the position they are in will then merge into new authorities.

It does add a degree of complexity, but I think it is still possible. We have seen that in the reorganisations that have happened over the last 10 or 15 years. As the Committee knows, with the exception of Surrey, our objective is to complete reorganisation by April ’28. On the two-tier system, accounts should be done in the normal way up to and including the year ’27-28, and from ’28-29 we would expect a single consolidated set of accounts. Yes, there is a degree of complexity, but I think we can handle and manage that.

So the reorganisation of authorities whose accounts are currently unaudited, and might be unaudited for the next two years, is not going to hold up reorganisation, and reorganisation will not mean that there will be an excuse for why accounts cannot be audited when we look in two years’ time.

Absolutely not, no. If a district council ceases to exist from April ’28, there is still absolutely a requirement on that council to publish accounts in the normal way for all years preceding 2020.

Then there is a difficulty, isn’t there? You are not quite sure of the liabilities of those councils, given the property speculation they have been engaging in, and where those liabilities then fall in terms of the new authorities.

I accept there is a level of complexity, but I think it is a manageable one. There is no greater level of complexity and problems than in Surrey, where we have set out a plan to move from 11 councils to two by April 2027. That obviously includes some councils in severe financial difficulty. We think there is a way through that allows the accounts to be audited in the right and proper way, and still have a consolidated set from April 2027.

Thanks for that question. The Treasury does the Whole of Government Accounts. There are 10,000 bodies, of which a big chunk of them are local authorities. We put them all together in the pinnacle of the accounting of the Whole of Government Accounts. Our colleagues in MHCLG look after local authority audit. We are working with them to try to get this back on track. In terms of the judgment and the risk, we have for the Whole of Government Accounts taken the judgment that timeliness is important, and we want the data from local authorities in our Whole of Government Accounts, even if it is slightly out of date or unaudited. That is what we have got in our performance report. We use the latest data, because otherwise you just have missing data, and how is that comparable to previous years? That is what we are doing. Our work with MHCLG is about us stress-testing its plans to get local audit back into a decent position. It got into this position for the reason that Will stated. We agree that the legislation before Parliament, which I think is in the House of Lords, to set up the local audit office and the backstops to require in statute local authorities to give their accounts by certain things—we have done two and I think there are four more to go—is the right way of doing it. In terms of what you are hearing on target versus aspiration, we do not know exactly what local authorities will give us in all those backstops. We are giving you our judgment as to what it was, and we have got to compare that. We will want to continue to require local authorities to up their game. You are seeing the pinnacle of it here in the Whole of Government Accounts. We will work intensely with local government to try to improve that. I hope I am being totally open on that. I could say, “We will absolutely do this,” but I am not in control of the 200 local authorities that do not do that. I am not their accounting officer, and I cannot require them to do a better job.

I totally agree with you, and I totally agree that an aspiration and a target are different things. What I am trying to explain is that I and the Treasury and MHCLG are entities different from local government, and therefore we cannot do their jobs for them. What we can do, and what we have done, is put it in law as to what they have to do by when. We are waiting and predicting to see what we will get in that. To your point, I totally agree. When you hear the word “aspiration”, it is right that you then stress-test what exactly that means, because that is clearly more moveable than an absolute requirement. Mr Garton set out the judgment he is taking in using those terms.

James, I was listening to your introduction, and you talked about missing data. What is the definition of missing data?

The definition means that for these accounts, you have got 10,000 entities across the public sector, and some of the data is not in the accounts.

So when you say missing, it just has not been submitted. Is that right?

It has not been submitted, and therefore we use the latest data we have. It means the data for that period from that entity has not been submitted, so you have to use the latest data instead.

In the performance report, we proxy the most recent data from that entity, but in the account itself, we do not have the data for that body. This discussion that we have just had on disclaimed data is, as everyone was saying, concerning, but we are slightly missing the glass-half-full point here, which is that the amount of missing data is falling in the WGA because of the backstop forcing local authorities to produce the accounts that have previously been missing.

That is all very well, Conrad, but at the end of the day, what is the sanction on local authorities for not populating this data? Is there any?

There is the statutory requirement on them to meet the backstop.

And if they don’t, as I said, we will engage. We have and will publish the list of those that are non-compliant. Ultimately, the Department has what we call a best value regime, which means we can issue a best value notice, conduct a best value inspection and send in commissioners to run the authority.

I am juxtaposing this against the private sector, where I run lots of companies. If I tell HMRC that I have missing data when it comes to my tax return, what do you think happens? Do you think they accept that, or do you think they say, “I’m terribly sorry, we’re going to assess you anyway”? Is there a divergence between the way in which the public sector is treated and the way in which people who are paying the taxes to fund the public sector are treated? As you know, most private businesses have had their advisers weaponised against them to help the Government to collect tax. When I hear this, I worry that there is a two-tier system here.

The best value regime actually does have some teeth. I do not think we always need it in order to make progress, but it is the ultimate weapon in our armoury. We have to operate within the rules that Parliament sets for us, though, and we do not have a stronger weapon than the best value regime. In my experience with local government, sending in commissioners to help to run an authority on a day-to-day basis is a pretty significant move—local authorities are legally independent entities—but we are working within the framework that Parliament has set for us, and that is as far as we can go.

Will, I do not want to dwell too much on local authorities, so you will be glad to know that, hopefully, this will be the last question on them, but it is a really important one. Auditing is all very well, but it is historical. I am keen to know what picture the Department has here and now of the financial health of the sector. We know from previous hearings that half of local authorities are going to have trouble balancing their books this year, and they are not allowed to produce accounts that run a deficit. How good do you think your modelling is at detecting what financial resilience is like in the local government sector, particularly in view of the very big liabilities that local authorities have on adult and children’s social care and the statutory override and so on?

That is an excellent and important question. I think it is really very good. As I said, we would benefit from a properly functioning auditing regime, but as you say, that is retrospective, so how do we have a sense of what is going on in this £83 billion system for which we find ourselves responsible? First, we have what we call the local authority financial sustainability tools—the LAST models—which are a series of models that examine local authorities’ financial positions based on a range of data, including reserves. We will look at Ofsted reports, we will look at CQC reports, we will look at the percentage of spend on statutory services, and we will come to a picture of the overall financial health of an individual local authority. That is the first thing we do. The NAO have seen those models and I think, without wishing to put words in their mouth, are relatively complimentary about them. Secondly, as Rosie said, even though we are lacking a functioning audit regime, we pay particular attention to public interest reports and VFM reports that auditors still produce. The third thing we do, and arguably the most important, although that is a contestable statement—it sounds a little bit woolly, but it is important—is gather and pick up intelligence from the sector about what is going right and what is going wrong. We are constantly engaging with the Local Government Association, CIPFA, the County Councils Network, the District Councils Network and Solace, and that often gives us a picture: “There’s a problem here.” How do I know that we are relatively competent at this? As the Committee knows, we have had a number of authorities issue section 114 notices and a number of authorities in exceptional financial support, and none of them have really been a surprise to us when they have happened. I cannot say for certain that we will carry on that record, but we feel we have a good grip of where the risks are in local government, and quite often that means we can get ahead of a problem before it materialises.

I listened to that answer very carefully, and I have gained two things from it. First, you are pretty optimistic that you know what is going on in the sector, and secondly you take steps to try to predict which authorities are going to get into trouble—we have asked you lots of questions on that before. What I am not getting from that reply is whether the health of the whole sector is deteriorating in terms of its negative balances. My hunch is that all the statutory override, children’s social services and SEND is driving a lot of authorities to become more and more indebted.

Clearly, the overall health of the sector is extremely challenging—at the beginning of the Parliament, Ministers described a crisis in local government. We are not complacent, so I am not for a moment saying that it is a bed of roses. It is tough because of rising demand and, as we all know, there are limits to how much money there is to spend. The Department is not asking for more money; it got a good spending review settlement, and core spending power for local government rises by 2.5% in real terms in every year of the settlement. Because I agree that there are big risks in the sector and it is tough, we are taking quite radical action to rebalance the system to move money around in a way that has not been done for many years, with the fair funding review. Next week, Ministers will present to Parliament, for the first time in a long time, new three-year settlements for every council in the country. That takes into account updated data on population, how much a council can raise in council tax, and the need and deprivation; it then makes an up-to-date and holistic assessment of what grant is required. So although we are not complacent, we have a very big programme of financial reform to move the £83 billion around the system to try to make it more sustainable than it is today.

There is a lot in that. Thank you for that very full answer. We have a lot to cover today, so we are going to move on from local authorities, with Catherine McKinnell.

I want to talk about the long-term trend analysis. This is probably a question for you, Andrew. Could you talk us through what the long-term analysis that you have included in the Whole of Government Accounts report is showing?

Yes, no problem. The 15-year trend analysis is one of the new things we have included in the performance report for this WGA. I think it is very helpful and very useful, and it provides better insights for us as well. When you look at the balance sheet and the income and expenditure statement, that is where you see the key movement. If we think about the long-term movement in assets for the Government, we can see that PPE has increased by £900 billion over the period. That will reflect Government investments in infrastructure assets such as rail and road. Network Rail was brought under the umbrella of the public sector, so that is reflected in there. It will also reflect revaluations. We can see a large increase in student loans, which are up by £140 billion on the balance sheet. If we look at Government liabilities, we can see that the largest increase is in Government borrowing, which has increased by 150% since the WGA was first produced. That is now the largest liability on the balance sheet: it is 40% of all liabilities. I think we have talked with this Committee before about the value that the WGA brings to provisions, particularly large provisions like nuclear decommissioning and clinical negligence. We can see that over the 15-year trend, provisions are up by £160 billion. We are continuing to make them more visible. It also shows the pensions liability in the large movements, due to the discount rate. I think that peaked at £2.6 trillion in 2021-22, and it is down to £1.3 billion in the current WGA. It helps to demonstrate those sorts of long-term trends. Finally, on income and expenditure, we can see that tax on income has increased over the period by £350 billion, which is a 65% increase. We can also see increases in social security benefits, particularly working-age benefits. That is a snapshot of what those trends are showing.

Thank you—that is really helpful. I guess the question is what you do with those insights. Have you identified key risks as a result of them? How do you use the information?

It is really interesting to look at the 15-year WGA analysis and compare it with what your attention would be drawn to if you looked at the slightly more traditional Office for National Statistics deficits-type analysis. Some things are the same, obviously. The increase in Government debt over this period is well known and understood, as is the financial crisis, covid and all those things. That is there in WGA as it is in the ONS statistics. Lots of things follow from that kind of analysis—the sensitivity of that liability to increases in interest rates, which is one of the things that is causing a lot of fiscal pressure at the moment. What WGA allows you to do is see a bigger picture. In particular, you can see the rise in assets at the same time as the rise in liabilities. The total measure of liabilities has doubled over the period from 2010. The total measure of assets has slightly more than doubled, and within that you can see that some particular kinds of asset have very significantly increased. That points you to where you should be thinking about managing the balance sheet as a whole. The most striking increase on the asset side is in the student loan asset book, which I think Andrew mentioned has gone from about £30 billion to about £170 billion. That points you to how, if you are interested in the sustainability of the public finances, you should be making the most of out of that asset base, and maximise the value of it. Those are the sorts of things that it points you to. There are also smaller things, below the level of the hundreds of billions of liabilities, that it points you to: thinking more carefully, for instance, about things like the MoD’s housing contract, which they changed recently. I could go into more detail, but that is the big picture.

I guess the question is whether you are confident. There seems to be an acceptance that this is a useful function, so clarification that this is going to continue would be helpful. Also, are you confident that the necessary action or change or behaviour learnings will be taken from these insights, and it is not just a book exercise?

That is exactly the right question, and we have discussed this previously. The way the UK system works is very much on the income and expenditure side. Parliament gets a lot of information and votes through supply estimates and all the rest of it—what you spend and what each Department is given. In the past, it has had much less focus on the balance sheet. Obviously, each Department submits its own accounts, but the purpose of the Whole of Government Accounts is to up the overview on the balance sheet side of things—the assets and liabilities, as Conrad said. We are committed to it, and it will continue. We are very keen to shine a light on all these things, as a Treasury. As for the kind of changes you are seeing, there is further to go, but in the past a Department could just start a contingent liability that might have tens of billions of costs in the future. Nuclear is a good example: decisions are taken, and nuclear has tens of billions of decommissioning costs in the future. The Whole of Government Accounts and the focus on the balance sheet have allowed a much more sophisticated attempt at governance, whereby you cannot set up a new contingent liability without notifying Parliament, in a regime where there is central expertise and a whole set of people who look at managing those things. Now, when you build a nuclear power plant, you need to understand the costs of the decommissioning and who is going to pay them from the outset. That is included. The big picture is that it is really important. I think our system over the years has focused a lot on short-term supply: “How much is MHCLG’s budget going up by? Local government’s is going up by 2.5% in real terms next year, possibly.” It has focused less on the fact that there will be £100 billion-plus of nuclear decommissioning costs over the next 50 years. So this is a really important thing to do: it is there in departmental accounts, and this brings it all together. It is worth saying that we are pretty much the only country that does this. I try not to use the phrase “world-leading”, because it is a bit naff, but we are the only country that does this. It is very difficult to do. You are accumulating 10,000 bodies into one set of accounts, and as you can hear, the local authorities are just a snapshot of the challenges you get when you try to pull all those things together. But we are trying to do it precisely to give ourselves, and you, that overview. Of course, this Committee over the years has been the biggest champion of this, and I really thank you for that. The Treasury is a champion of it, and so is the National Audit Office, so it is really helpful that you are doing this, that you have suggested we do the booklet, and that you are trying to get the use of it up, because there is probably not enough focus on those balance sheets. You have done a hearing in the past couple of weeks, I think, on clinical negligence. The Treasury wants a focus on clinical negligence. We want to be looking at the fact that that is a large future cost to the NHS. We want to be having a debate about how to manage those costs and bring them down. With my best wishes to my colleagues in the Department of Health, we want to be having that debate. That is the right thing to do for sound financial management. This Committee has been pushing us to do more and better, and we welcome that. I hope that that is a helpful answer.

It is very helpful, but it also raises the question why climate risks have not been integrated into the 15-year analysis in the most recent accounts. Can you explain that?

I am happy to talk about that one. In this WGA we have a short section on sustainability reporting and the Task Force on Climate-related Financial Disclosures, on pages 113 to 115. That has some high-level metrics on greening commitments and the Government’s response to them. The approach that we take in the UK Government to climate-related financial disclosures is that we are implementing the TCFD, which I talked about earlier. To pick up on James’s point about being the only country in the world that does WGA, the UK was the first country to commit to implementing this in the public sector. That is a phased roll-out. The first year of implementation was 2023-24; that will continue to 2025-26. Departments and organisations in the public sector are required to disclose their governance, their risks and their strategy impacts because of climate risks. We are in the process of rolling that out; it will take a few years. We cannot look retrospectively at bringing that in, because that information was not there, but one thing we want to do as it is rolled out is bring a bit more information in. In the past, we have done spotlights on Departments that have done things well or on managing material risks. For future WGAs, we are looking to bring that type of thing in on sustainability risks.

That is helpful—thank you.

Chris has a quick follow-up question, and then we will take a break.

There is a theme emerging here—I might ask colleagues to use it. You are right. There is a big focus on getting it all together, and we will no doubt talk in the second half about how we are doing that. But it is all about using it, so we are doing a whole set of things to try to make it more useful: moving, as the Committee has suggested, from the complete report to the booklet, and having the 15-year analysis so that you can pick out the trends. Also, there are plans—when we get up to speed—to make it more digitally accessible. I would say that the main thing is that this is prompting debates. The debate that we have just had on the health of local authority audit is really positive and sensible, even if the realities of it are quite tricky. I am not quite sure that we would be having this debate if it were not for the Whole of Government Accounts. It is interesting that it is coming out during this session; no doubt you have heard about it in your own Committees’ sessions. You can see the uses.

We have talked to the Committee before about how, over the last couple of years, we have been trying to get WGA back on track. We have prioritised timeliness in getting the reports out, but we still look to improve where we can. I can mention some of the things that we have done and are looking to do. In previous versions of the WGA, we have made improvements to the performance report, and we have introduced spotlights and “years at a glance”. On the missing data issue, we have introduced a lot of disclosures, and we have used proxy data for that. As James said, for this WGA, we have had the 15-year trend analysis and the pocketbook. The team at the Treasury did a Portcullis House event, which you attended, Chair, and there is going to be a newsletter to publicise that among parliamentarians. We continue to offer parliamentarians training and are happy to do so. We are doing all those types of things; we also embed it across the Government Finance Function. We do awareness sessions and talk to people about it, so that finance leaders across Government are aware of WGA and using it. Now that we have almost recovered the timetable, in the next version of WGA we want to think about the actual large product itself. We are going to make some changes to the performance report; we want to streamline it and make it a lot more visual. We want to use more infographics and QR codes to make it a lot more interactive when we put it online. We also want to use HTML formats so that it is more interactive for members of the public who want to use it. Those are a number of the proposals that we are looking to do for the ’24-25 WGA.

I am really pleased to hear that you are thinking about all those innovations, Andrew. I shall be delighted when it is online, because it will be searchable—I have been grappling with it on the tube in preparation for this morning. We are going to take a break. Unusually, I am going to warn you that I shall be asking questions on the long-term liabilities after the break. We are almost at 11 o’clock—shall we be back here by five past, as we still have a fair bit to go? As always, I warn you that the microphones will still be on, so be careful what you say. Sitting suspended. On resuming—

Thank you all very much for your replies so far. I was particularly interested to hear the innovative things that Andrew is going to do with the Whole of Government Accounts. Permanent Secretary, you pointed out that you have opened this up with that small summary book, extra training in-house and various literature. There is no point in doing these accounts unless more people are able to access them. On accessing the accounts, we need the information in a form that we can actually understand. I refer to the long-standing issue that the Committee has had with long-term liabilities in the accounts. I have read these accounts quite carefully. As I mentioned earlier, of the three big liabilities—nuclear, clinical and pensions—only nuclear liabilities have a figure for how much they increased in the year. Paragraph 1.173 on page 81 tells us that they increased by £50 billion. We are making a start—we have one out of three. Why can we not do the same with pensions and clinical negligence?

We are making a start, because we recognise that, for some users of the accounts, it is useful to see the undiscounted number as well as the discounted number, to split out the impact of the discount rate from changes in the undiscounted liability. I should say that the undiscounted number here is in real terms. It is discounted for inflation but not for the real discount rate. We are looking at doing it for pensions and clinical negligence as well. It is a bit more complicated, particularly for the pensions liability, but we have been in consultation with the Government Actuary’s Department on the best way to discount the future liability for inflation, but not for changes in the real discount rate, and to allow users to see both those numbers. The accounts are produced to be IFRS-compliant. The IFRS uses a real discount rate, and for good reasons—you need to adjust for the time value of money. If we can get these technical problems sorted out, we are hopeful that we can do something similar for the other two big long-term liabilities next year.

The ideal would be to have a bar chart with the 15 years that the Whole of Government Accounts have been going so that you could see from the bars exactly how much they have increased in each of those 15 years, and going forward too. Hopefully there could also be a little note, as we recommended, on what is being done by Government to address those large-scale liabilities.

We can definitely look at doing it in the 15-year analysis as well. The longer the time period of analysis, the more some things, such as changes in the overall price level of inflation, will dominate the change in the liabilities, so I would caution that time series analysis in nominal terms might not be the perfect thing. We can definitely also look at adding the steps that have been taken to help to manage those liabilities over time as part of the performance report.

As the Permanent Secretary has been kind enough to acknowledge, we have done a big bit of work on clinical negligence, which will be out in the new year. Hopefully that will help you in that respect.

Conrad’s answer was very good. Over the years, we have had the debate about discount rates, and in particular about the frustration of comparable data year on year when the discount rate changes. I remind everyone that, as Conrad says, for very good reasons the accounting requirements require that you take in any moment in time a spot time value of the money for the accounts when you sign them off, and you use the discount rate to do that. That is the requirement of the accounts. It can be frustrating for people because the discount rate last year will be different from the discount rate this year. We can try to show undiscounted figures. You get different answers, of course. For nuclear decommissioning, if you do undiscounted, one is going down and one is going up, so the explanation becomes more complicated. I just wanted to point that out.

I absolutely accept that explanation, Permanent Secretary, and I absolutely accept the reason why you used the discount rates and changed the actual rate, but then, of course, it has a huge effect on the actual liability.

Yes.

Coming on, for example, to pensions on page 257 of the report, there is a caution in the bottom paragraph saying that the “volatility in the revaluation gain or less is predominantly due to the change in assumptions, in particular the discount rate net of inflation and demographic assumptions.” I get that, but on that very important page on pensions, we actually see from the table that it looks as though the public sector pensions liability is going down, which is clearly nonsense, when more people are being added to the pensions schemes each year and life expectancy is going up. In a sense, it is presenting a false picture, which is why I want this constant figure, so we can see what the real-world situation is. I am not suggesting that you do not do the discounted rate.

It is a fact that there are different ways in any one time of measuring something. When you come to something like pensions, the discount rates are very specific and they change. They are basically changing on bond yields as much as anything else, and everyone knows those change yearly. I would suggest that the best way of understanding the cost of pensions—people will know about how the public sector pensions system works in this country—would be to compare it as a percentage of GDP. On your point about going up and going down, I think pensions are about 1.9% of GDP at the moment. They are expected to fall as a percentage of GDP in the next 50 years to about 1.4%. There are different ways of measuring it, but I think your point, Chair, is, “Let’s show them all and debate them and set them out,” which I really agree with. That is a totally reasonable point.

This is the last question on the long-term liabilities. It is on clinical negligence, which we have done quite a lot of work on. From our work, we have found that although, as you correctly say in these accounts, the number of claims is going down, the overall value is going up, because the value of claims in some cases is very considerable. Again, one has to be very careful how one presents the data that you are given.

I totally agree with that. It is one of the valuable things about doing these sets of accounts that you can expose the level of changes in these long-term liabilities and therefore shine a spotlight on them and help people like the NAO, this Committee and the Department of Health to work through and go, “How can we get better control of this liability?” As you say, a small number of very large claims is a big driver of that liability, and that leads you to questions about how you should try to manage those liabilities in future.

That you for those replies. Let’s leave that there for now. We have got a third of the way there on one of the liabilities; let’s get 100% next year, please.

On public service pensions and the unfunded sector, what is your view on the affordability?

As I think James was trying to say a moment ago, for the pay-as-you-go pension liabilities—the unfunded pension liabilities—I think the Treasury’s view is that probably the best way of looking at affordability is to look at the annual payments as a share of GDP, which is a good proxy for what the tax revenue would be coming in on that. In its fiscal sustainability report, the OBR does that assessment for us, and that shows the public service pension liability expenditure falling from about 1.9% of GDP now to about 1.4% of GDP in 50 years’ time. From that point of view, those pensions are affordable.

On that basis, are you confident that things should stay as they are and continue steady as they go, or should you be considering other means of paying for and funding the public service pension liabilities?

I would split it down into the liability and the means of funding the liability. The Hutton commission reforms, which came into effect about 10 years ago, were a really major change to the value of the pension liability—in particular, increasing the pension age from 60 up to the stage pension age of about 68, ending the final salary element of the schemes and making them career average instead, changing the indexation base from RPI to CPI, and increasing the contribution rate. Those were major reforms and were expected to save about £400 billion over a 50-year period. As part of agreeing those reforms, the idea was to fix that for a long period of time as the offer for pensions. On the funding side of those liabilities, the central Government pension schemes are pay-as-you-go, unfunded schemes. That is consistent with the way that the Treasury thinks about managing the balance sheet overall. The Hutton commission looked at whether you should turn that unfunded scheme into a funded one, and the conclusion that the Treasury has always come to on this is that it is not an efficient way of managing the liability to do that. In order to create the funded scheme, you would need to sell a great deal of gilts into the market and buy a great deal of financial assets. The efficiency and transaction costs of doing that would be higher than the current framework. On that side, we have a pretty settled position.

So the answer to looking at other avenues of funding is no.

I don’t think that is something that the Government are actively looking at now, no.

The point being: it is unfunded, but it is paid for from tax revenue. The liabilities are in the Whole of Government Accounts, and you don’t put tax as an asset in Whole of Government Accounts.

What Hutton did not really address was that the public sector is largely based on defined-benefit pension schemes, whereas the private sector is based on defined-contribution schemes. Are we not in effect producing an element of intergenerational unfairness, because this generation is loading a huge liability on the next generation?

I think that would be the case with either a defined-contribution or a defined-benefit pension scheme, if either of them are pay-as-you-go unfunded schemes—as in, funded out of future tax revenue. I would have to check, but I think we did consider whether it would be better to switch to defined-contribution from defined-benefit and concluded the advantage of clarity for public sector workers and the recruitment and retention advantages of offering the pension and ensuring that people have that provision in retirement outweighed the costs of change. You are right, however, that there is a difference in the structure of remuneration between the public and private sectors.

We have done quite a lot of work on nuclear, we are doing work on clinical negligence, and we will hopefully come on to pensions before too long.

I am a new MP, and I am also new on the Public Accounts Committee. I have attended four or five of these meetings now. I read in the NAO audit report that “of the 407 English Local Authorities that should have submitted audited data to the WGA, 167 (41%) did not submit data…and a further 224 (55%) submitted data that was unaudited. Only 16 (4%) of English Local Authorities” submitted adequate audited data. That comes on top of my having attended a DWP meeting about medical malpractice where I found that the accounts have been qualified for 37 consecutive years. If you put rubbish into a system, you get rubbish out. So how much does it cost to produce this? Is it actually an exercise that is worth doing until we have resolved all the issues down at local council level? I cannot see, as a businessman in the private sector, that this is necessarily a good use of taxpayers’ money. It may well trigger this debate, but I am sure that there are other ways of triggering it without necessarily having to spew vast amounts of money on a report that arguably does not paint the real picture. Can you reassure me and the Committee that we are not wasting large amounts of money producing a white elephant?

Welcome to the Committee. Thank you for the question. Let me take that head on. We are the only country that does this, but we think that is good and not bad. It is not a highly expensive thing—

What does it cost? Do we know?

It costs the time of a small team in the Treasury—not very many people. It is a good challenge, because I am reducing my budget in the Treasury and I need to cut my cloth accordingly, reducing the size of the Treasury. I said it earlier and I will say it again: the exact debate we have had on the quality of local authority finance would not be happening, I don’t think, unless we exposed it through this process. To Mr Betts’s point, don’t people who live in local authority areas around the country deserve to know that they have high-quality accounts, where they can understand the assets and liabilities of the councils that serve them? If they merge, the quality of accounts can help to do that merger. The fact that we are having that side of the debate—where we ask, “Look at the size of clinical negligence over the next 50 years. Look at the size of nuclear decommissioning. Is there an intergenerational unfairness in pensions? What are we doing about local authority things?”—for the cost of a relatively small team in the Treasury, I think provides you and us with value. You are the Committee that pushes for better value for money in public spending and the Treasury is the Department that does that on your behalf, so our objectives are entirely aligned. Entirely because of that, with the pretty small cost of pulling this together, we are asking some pretty fundamental questions that are at the base of value for money. You do not have to save much money off £125 billion for nuclear decommissioning, or the £58 billion provision and £25 billion contingent liability on criminal negligence, to make this pay for itself. It is a really good challenge, but I hope we are walking straight towards—

So we don’t know what the exact cost is. When we look at ’23-24 where, to Chris’s point, you made some aspirational statements, it does not seem that we have made the progress that we hoped to make, and today we have heard more aspirations. I just question how this Committee will know you are going to make the progress that Will has said he aspires to make. How can we think that, having not done it once, you will do it next time?

I think it is a really good challenge. I will not rehearse this, but we think we are making progress. The point I would make in defence of the Whole of Government Accounts is that, of the 10,000 bodies we have put together, we are struggling with the local authority area, because the data is not coming through in enough quality. I genuinely think you would not even be having that conversation—I do not think it has been pursued extensively in the Housing, Communities and Local Government Committee—without this. While it might be awkward for us to have aspirations and things, the fact that you are pushing to allow—to Mr Betts’s point—a better service for people in local authorities in understanding their financial management, I think is a good thing. We could get rid of the Whole of Government Accounts and I would not have to come here every year to answer these difficult questions—

No doubt you would be here for another reason.

I will see you in lots of different guises, but you take my point, I hope.

I am still struggling to see, but again, what it bears out to me is the complete separation between the private sector and the public sector. This would not be the way in the private sector. If I ran a public company or a private company with an audit report like this, I would end up in prison, probably.

The equivalent would be trying to amalgamate the entirety of the private sector into one set of accounts. This would be saying—

Yes, but for your individual companies—

Yes, but I am saying this is amalgamating all of them.

I know what it is, but if you put rubbish in at the bottom, you get rubbish out at the top. It is a bit like these local councils—it is a very good question. If you amalgamate all the local councils that are all a mess, you just get one big mess.

Wait a minute. Of the 10,000 bodies, local authorities—about 200—are the ones with missing data. No one would ever try to do it, but I am sure if we tried to amalgamate the entirety of the private sector into one account, which would be the equivalent, there would be all sorts of issues.

At least you would have accurate data, because most of them do have audited accounts—in fact, all of them do. They have to.

Indeed. I am not disagreeing with you—I am enjoying the debate—but you can see the point.

Let me make two points, one about WGA and one about your excellent point on how you can be reassured that we are making progress. On WGA and the value of producing it, we have six or seven people working in the Treasury consolidating this account. If we did not produce WGA, we would still want all those 10,000 bodies to produce accounts, right? It would not be the right decision in local audit to say, “Okay, let’s not bother.” It is absolutely an obligation to produce audited accounts. There is a serious problem that we are trying to fix.

But 4% have audited accounts, Conrad.

I totally agree, and that is why we need to do something about it. Once you are requiring everyone in the public sector to produce accounts, it seems to me that it is good value to consolidate that, particularly if you have a system that allows you to do that with six or seven people and allows you to put all that data together. If you did not have it, I think you would invent it. That is my first point. Secondly, you are right to challenge us on how we can be sure we are making progress on this stuff. We really are making progress on the missing data. You are right to pick on missing data, as that is the most serious problem with the account. The improvement from no data to disclaimed data is a much bigger improvement than going from disclaimed data to fully audited data, although fully audited data is what you want. We know that the number of missing English local authorities was 187 in the ’22-23 account. It is going to be 107 in ’24-25, and it is going to keep falling, because Will’s backstop plan is getting those accounts produced. There is a long way to go to get the disclaimers off local authority accounts and then WGA, but we really are making progress.

Thank you. While we have got the supremo of OSCAR II here—the man who drives the national assets and liabilities—I have a bugbear, and this “rubbish in, rubbish out” point plays into it. As I understand it, this feeds up into OSCAR II, so the Whole of Government Accounts will be a subset of what goes into OSCAR II.

OSCAR II is the system that we use to collect the data from all the entities to consolidate WGA.

My opinion is that OSCAR II basically understates our liabilities and overstates our assets. I think to some extent it is delusional. If you put nonsense data in, you get nonsense data out. If we are feeding all this nonsense data into OSCAR II, how confident are you that we actually do know the financial position of UK plc?

Going back to the challenge about the value of WGA, I will make two points in addition to everything James and Conrad said. We know that for the assets, the liabilities, the income and the expenditure that is there, over 93% of the value for UK Government is reflected in the WGA. The vast majority of those large balances are in there, and they are audited by the National Audit Office. That is one thing. The other thing is that, in the same NAO Report, for the two years that we have been disclaimed in WGA, the NAO still made the judgment that WGA is still a valuable product for Government and for the public. On how we use OSCAR to collect the data for WGA, we go through several layers of assurance. Departmental accounts or entity accounts are produced. They are independently audited by the NAO. We have teams of professional accountants ethically preparing those accounts and uploading them. When they are submitted to OSCAR, if the value of the entity is over £2 billion, in terms of assets or liabilities, that needs to be audited again by the NAO before it is submitted. Within Treasury, we have the OSCAR system, which does validation checks, the teams do checks as well, and the whole WGA product is then audited again by NAO, so it goes through three layers of assurance to ensure that it is as accurate as possible.

How does that square with a briefing we had this week that said that the Government do not know how much they are spending on external consultants? They do not have a washboard of who they spend the money with, and they do not know how much they are spending. How does that feed into OSCAR II correctly?

Obviously that is a known, reported issue. The NAO has qualified DWP for a number of years, so that is common knowledge. That will be consolidated into WGA, and obviously that is referenced in there.

The thrust of your question, which is correct—how can we know that we have a reasonably accurate picture of the overall state of the public finances?—points to one of the values of WGA. It is very easy to count the number of gilts outstanding from the DMO; it is much harder to put values on clinical negligence liabilities, the pension liability and all those things. That is what WGA does for us. Is it perfect? It is never going to be perfect, and you can look at the sustainability of the public finances in lots of different ways. This is, I would suggest, a better way of looking at the overall balance sheet than you would get in, say, France, where they look at the fiscal statistics—the kind of statistics that Eurostat produces—so I think it is really valuable.

I have just a few sweeping questions, if I may.

Are we allowed to give sweeping answers?

No, you are not allowed to give sweeping answers, and if you do, there will be follow-up correspondence, so if you want to save time, give us the proper answer now. The first is on a relatively small thing. You refer to European Community contributions twice in the report as what it is, but you do not put any numbers on it.

It is in the Treasury’s accounts; it is probably a materiality issue, as to the extent to which it is in—

I am slightly stumped; I am sure we could give you a note on it.

I am starting at page 165. The next issue is compulsory redundancies and public sector employees. It looks from page 183 as though the number of compulsory redundancies is going down from 2022-23 to 2023-24, from 3,000 to 1,900. Is that correct?

I am struggling to find it, but I am sure it is correct.

Actually, I draw the wrong conclusion in my own mind, I think, but I think that is correct: it is going down, because you had 1,934 in these accounts—

And 3,100 in 2022-23.

Then I want to turn to the vexed subject of academies’ accounts. You wrote to us about this. It still seems fairly weak that we cannot get them into these accounts at all, because we cannot get the year-end sorted. Returning to Rupert’s private example, I don’t care what you say, Permanent Secretary, the Inland Revenue is trying to force private sector businesses into producing a year-end to coincide with the fiscal year-end, because if you do not, they ask all sorts of queries. What we do in my business is produce a year-end in line with what the Inland Revenue wants at the end of the fiscal year, but we have management accounts to the end of September.[1] Why can the academies not do the same?

I think the position on this is the same as when we discussed it in January. The conclusion that we and the DFE have reached is that forcing them to produce those two full sets of books is not value for money for the additional benefit you would get.

Conrad, come on. You are forcing me as a business to have value for money—in fact, to do this at considerable cost. What is sauce for the goose is sauce for the gander. We want these academies in the Whole of Government Accounts because we want to see what is going on in that sector, so why can’t we have a little bit more stick from the Treasury on this occasion to make it happen?

We do have a good idea of what is going on in that sector, because of the sector annual report and accounts—the SARA—that they publish. I am happy to go away and talk again to the DFE about whether—

If you are making the private sector do it—at a cost to ourselves—why can’t the public sector do it?

I am happy to take it away and look at it.

It’s two-tiered, Geoffrey.

It goes both ways, though; either it is value for money to be doing this or it is not. There’s some “having your cake and eating it” going on here.

My answer to that is that if the job is worth doing—as you so eloquently adduced, Permanent Secretary—let’s have it complete, not partially complete.

From an Abingdon boy.

Oh no! I won’t ask you where you went to school.

Radley.

I have one last but important set of questions, then Chris is going to have the last word. It is on the decision to unwind the quantitative easing purchase of gilts. That is being unwound now, with considerable sales of those gilts. First, why are they being sold before they mature? Secondly, on what basis is the judgment made as to how quickly this whole position is to be unwound? I can’t find the page in here quickly, but I think that will cost the country between £50 billion and £60 billion this year.

That is a perfectly reasonable set of questions. On the accounting and where you can find it, the OBR does a forecast of future projections, the Treasury accounts has the impact on spending, and the Bank of England publishes a quarterly view of what it is doing. That is where you can find it. The situation is that, up until October 2022, cash was being transferred to the Treasury in the order of £124 billion from the quantitative easing decisions; after that period, that has now turned the other way, as we thought it would. The exact answer to your question is that the decision on pace and maturity versus active sales is a monetary policy decision. It therefore rests with the Bank of England and is an explicit decision taken by the Monetary Policy Committee. They take a view every September and publish the answer in their monetary policy decision, that September, of what they are going to do in each forward year. That includes a quantum and therefore how much is active sale and how much is just managed by reaching maturity. They publish that decision, and that is reflected in all our accounts. So the answer to your question is that it is their decision. The last thing to say is that lots of people have examined this in detail; the Treasury Committee did a fairly lengthy inquiry into it just last year or so. In that inquiry, the focus was on value for money, which I am sure is also your focus, and it was essentially to ask the Bank, with the support of the Treasury, how we can be assured that the choices they are making are value for money. I hope that is a very full description of the answer.

That is very helpful, Permanent Secretary; thank you.

If I can just very quickly come in on that subject, I understood that the Treasury has given a guarantee to the Bank of England so that any losses that the Bank of England suffers are paid by the Treasury. Is that right?

Yes, as an indemnity.

I think that is a very good question, Chair, because I do not understand why we do not just let these gilts run off. If the Bank of England is getting a payment from the Treasury for making the decision that you say is theirs, surely, if the Treasury is paying the losses, it should be the Treasury’s decision, not the Bank of England’s decision? You just told us that the MPC are basically the Bank of England, effectively.

There are two points here. They are taking a decision on the pace as part of their monetary policy decision. They take a decision on interest rates and on the pace thing as part of how much they are affecting demand in the economy. So point one is that it is part of their monetary policy decision. Point two, of course, is that with letting them run off versus active sale, there is potentially an indemnity and a cost to both options, because it obviously depends on what you paid for it versus what the maturity is. It is not as if it would be cost-free, if you let them—

It is all a function of interest rates, James. But at the end of the day, if the Bank of England is being covered for any losses by the Treasury, surely he who says, pays? I mean, it should be the Treasury’s decision and not the MPC’s decision.

I take your point and it is a point you can make. However, where it sits is as part of a monetary policy decision, which is independent in law and for the Bank of England. If you wanted to change that, you would have to change that.

Good question.

And a good answer, I hope.

Er—[Laughter.]

In terms of production of the WGA, you are right; we are consolidating Scottish and Welsh public bodies, like all UK public bodies, into the WGA. Regarding the performance of public services in devolved Administrations for the kind of money that is provided, which tends to be—

The different devolved Governments are obviously accountable for spending their money and for the value for money of what they get from it. They have their own audit regimes and other mechanisms for publishing the performance they get from different sectors. Those arrangements are probably appropriate for those devolved Governments to be held to account. I think it might be difficult to combine all that into WGA, which is trying to do quite a lot of stuff already. A point we have made to this Committee before is that, although the WGA is a very valuable consolidating document, you need a suite of other ways of looking at both the finances and the performance of public services that you have for—

I would make the point that, although we have talked a lot about missing local authority data, underneath local authority data, Scottish data is the second biggest area that is not submitting entities into the Whole of Government Accounts. Again, there is an issue about the extent to which we can compel them to do so. Actually, both central Government and Scottish local Government are not submitting entities, and that is on an increasing path, not a decreasing one. There is a challenge for us to get a really good UK picture, if that path continues.

I think we do know. I think the answer is that we are asking rather than compelling.

Well, I would say that the situation in Scotland is quite different. In England, Wales and Northern Ireland, they have a legal requirement to submit a WGA return. Scottish entities do not have that, but we work really closely with counterparts in the Scottish Government. There had been an increasing improving trend; it worsened slightly, but we are confident that it will get back on track, and we have arrangements with them. On your question about devolution and the back end of the accounts, we need to produce them in compliance with IFRS accounting regulations. We would not look to split that by devolved Administrations. If that is an area of interest for the Committee, we could look to put a spotlight in a future version of the performance report, just to give bit more information on the spend of money and outcomes in the different countries of the UK.

If we can have the Committee’s help to ask local authorities and Scottish entities to contribute their reports, that would be really helpful.

Permanent Secretary, your wish will be granted; I think we should shine a spotlight on this. The whole purpose of this debate is to shine a spotlight on bodies that have not submitted, so that we are getting some the missing data that we were talking about. We have come to end. I think my summary of today’s hearing would be that we are making progress, but there is still a long way to go. Just as we had DWP in here the other day and berated them about their 37 years of qualifications—we will be discussing with them how they might reduce the fraud so that they might get into a situation where, in time, their accounts will not be qualified—we would like really to see these accounts not disclaimed. That means a lot of input from these witnesses to try to achieve that. Thank you very much indeed. It is a big, complicated field, and we appreciate your evidence this morning. We will be producing an uncorrected version of the transcript in the coming days. Following that, there will be a report with recommendations. Thank you very much indeed for your time today.   [1] See Register of Members' Financial Interests for further information.