Welcome to the Public Accounts Committee on Thursday 19 March 2026. Today’s session is in two parts: the first panel is on the NAO’s financial audit insights 2024-25; and the second panel will be on the restoration and renewal of the Parliament building. Therefore, the first order of business is to consider the NAO’s Report, which collates insights from its audits of about 400 Government accounts, covering £1.1 trillion in public spending each year. We will look to understand the Government’s progress in achieving timely annual reporting, to examine the quality of the Government’s financial management and reporting, and at how good practice is shared. We will also explore how the Government respond to financial management risks associated with digital transformation and IT change. Today, we have a very eminent cast of witnesses, who are well familiar to the Committee but, James, will you start the brief introductions of yourselves? Also, I have introduced a new thing where I ask every witness to say when they were appointed to their post—2022, in your case.

Yes. I am James Bowler, permanent secretary at the Treasury since 2022.

I am Andrew Cartner, deputy head of the Government Finance Function, also since 2022.

I am Philip Duffy, chief executive of the Environment Agency. I was appointed in July 2023.

But you also have a Treasury background, don’t you?

I was Treasury previously, yes.

Morning, James. Do you subscribe to the view that the state should be as accountable to the taxpayer as the taxpayer is to the state? Do you start from that premise?

Yes, and I think that the state is pretty transparent in what it shows it is doing, more so in some areas than even the private sector. Absolutely, as part of my job, I am keen to work with this Committee and the NAO to make sure that we deliver transparency and value for money for every pound of taxpayers’ money spent.

Good, because I hope we can all agree that timely and accurate accounts are the genesis of any form of stability and of people being held to account for failure.

I do agree.

When we look at figure 1, we see the situation is varied: pre covid, 76% of public accounts were meeting the summer deadline, but that then dropped to 42%; currently, it sits at 64%. Is that acceptable?

We as a Government must, and plan to, do better. In the legislation, we are required to publish our accounts by 31 January at the end of each year. Everyone did that, apart from the College of Policing, which had a particular transitional accounts issue. We are meeting our legislative standard, but the aim is to do far better than that, and for public bodies to publish before the summer recess. Your stats are exactly right. In terms of the larger Government Departments, I think eight were post-summer recess this year; we think that at least four of those will move to pre-summer recess and have credible plans to do so. Perhaps the Department of Health and Social Care are the largest outlier, and I think you have questioned them on that. They have a plan to deliver pre-summer recess by ’26-27. I am happy to go into the issues around all that.

It will be helpful if you could explain the responsibilities of accounting officers, the Treasury and the Government Finance Function in this process. I think it would be useful to know what part you all play.

Okay. The Treasury sets the framework under which people must prepare their annual report and accounts. We publish a financial reporting manual by December each year, which sets out all the detail of what people are required to do. An advisory body helps us do that. The Treasury sets the framework. Accounting officers across Government are required to deliver to that framework. They are responsible for preparing and for themselves—not as their institution, but as individuals—signing off their accounts, in accordance with the manual. The NAO audits that, up to, I think, 400 accounts per year, which is a key part of the assurance that Parliament and the public can get. The Government Finance Function, which is the function that does finance across all Departments, is—with the Treasury—here to help Departments share best practice and push for improvements, so continuous improvements and learning from each other in that regard.

I think, Philip, you have been at the coalface doing this yourself, so my question—we have been over this before in a previous meeting—is because in the private sector there are quite severe penalties if people do not file their accounts within a statutory period. I think we have identified that that is not the case for Departments and Government accounts. If you miss the timescales, yes, you may get a bit of criticism, but in the end you do not suffer the same statutory penalties that people do in the private sector, such as, in the end, being banned from being a director, or having HMRC come and lean heavily on them and start to assess them for taxes it thinks they owe. That does not happen in the state sector, as we have identified, so there is a schism between the private sector and the public sector in terms of penalties for not adhering to a statutory timetable. Why is that not the case with the state?

I hear what you say. We have discussed this before.

Shouldn’t it be the same?

As I said, they are legislatively required to lay their accounts by 31 January, not by summer, but they are encouraged to do it before summer. Everyone is meeting the legislative standard apart from the College of Policing. It is very serious if your accounts are late or qualified. I have signed the Treasury’s accounts personally a number of times, and I am personally responsible to Parliament, this Committee and other Committees for doing so. To your point about getting a whole host more scrutiny in the private sector, that is the model we have set up. We call it project reset. If you are a highly competent finance function delivering on time and to a high quality, the Treasury, which can be even worse than HMRC, will mainly leave you to your own devices, will have higher delegated limits for you and will trust you to deliver. If you are not doing that, we will micromanage you more, and we are stepping that up. So I do not think that there are no consequences.

How many public sector employees have been fired for failure?

For late accounts?

For failure to deliver accounts on time.

Quite a number of permanent secretaries have moved recently. I do not know whether you would link it to lateness of accounts—probably not. I do not think there is a total lack of accountability in the system, but I take your point. I would not be able to list the numbers of people who have been fired just because of accounting issues, but I am trying to reassure you that there is not just a free lunch and that it does matter. It really is very material whether you get your accounts done before the summer—that is a huge element; you have to get your accounts ready and then have them audited—and whether you have a clean set of accounts. That is not a low priority for me or fellow permanent secretaries. I would go back to the fact that people are delivering to the statutory requirements.

James, there is another reason it is important to get accounts in on time: the timeliness of the whole of Government accounts, with which you and Andrew are very familiar. You said that half of the eight Departments that were late last time—beyond the summer—will now come in before the summer. Will that affect the timeliness of the whole of Government accounts?

No, not those large Departments. The biggest drag on the whole of Government accounts when we set out our timescales is the poor and unacceptable quality of local audit at the moment, which you see as a reason for both lateness and qualification in accounts directly related to local authorities and in accounts with local government pension issues. The NHS has similar local issues. That is the drag on both the disclaimer and the lateness. We are due to get the whole of Government accounts back to the timeliness we had before. I will ask Andrew to say when that is. The other thing that came out very powerfully in the Report, and which I subscribe to, is that if you are always doing last year’s accounts right into next year, you are looking backwards, not forwards. That is very true.

I think Sarah Olney will come on to that in a minute.

Well, there is my answer.

Andrew, briefly, is the timeliness issue improving, particularly in the local government sector, but also in health and related pension schemes that are causing problems?

On WGA timescales, as James mentioned and as the Committee knows, we have a multi-year recovery plan. Last year’s WGA was laid in July. The aim for this year is June. That would complete the recovery plan and get us back to the normal timescales for producing a WGA. James is exactly right: none of the timelines of the Departments we are talking about will delay WGA. We made a decision, as we said to the Committee before, to go ahead and publish, despite the missing draft data for local government, to get the accounts there and to take the disclaimed qualification. I make the general point that the quicker the whole system can produce and report accounts—working through the local audit issues and getting all the main Departments back to the pre-recess laying—the quicker we can continue to produce WGA. That is an important point. In terms of local audit and how things are progressing—and we talked about this briefly at the last Committee—the backstops are working in terms of issuing opinions and getting published draft accounts. The last backstop was for the ’23-24 financial year. Around 94% of local authorities were issued with an opinion. For ’24-25—the February backstop this year—98% had already published draft accounts before that backstop date. Publishing draft accounts is still really important for us for WGA, because we can still take that and use that information, and obviously draft information is better than missing information. Progress continues to be made, but there is still a long road ahead to get that fully back on track.

Are pensions in the Department of Health accounts still a problem?

Yes, that is still a live issue that is attached to this. It impacts MOJ and DCMS. They are currently laying post recess due to the time delay of receiving that local government pension information that is important in their accounts. By waiting, they do not get a qualification. We expect that to continue for the ’25-26 annual report accounts as well.

Sarah Olney is next. I am sorry; we have slightly cut into your questions but ask what you would like to ask.

Don’t worry; I will speed up. Mr Cartner, can you give us a bit more detail about the knock-on impact of Departments publishing their accounts late or getting late audits on Government operations as a whole?

I am happy to answer that one. James already touched on this. Your annual report and accounts is a backward-looking document that is obviously really important for transparency over what Departments have delivered and what they have spent public money on. As we have talked about, the Government Finance Function and the Treasury want Departments and all bodies to lay them as soon as possible. However, I would say that the annual report and accounts is not critical for forward-looking budget setting and financial management within Government. It complements that, but we have other processes in place, too.

Can I ask you a bit about that? Wouldn’t the completed accounts flag up areas for concern that need particular action in various areas?

It would, but we have mechanisms in place within Departments. Before a set of accounts is audited and published, we have draft accounts and systems of internal control and management accounts within each Department. A Department would usually be aware of those issues that are then transparent in accounts. A lot of them are multi-year issues, so we have talked about DHSC and some of its qualifications, and then the Environment Agency and DEFRA. They can be multi-year qualifications that organisations will know about.

So internally, you will already know the issue that a set of accounts might point to, but what about your external stakeholders? Will their understanding of the operation of the Department not be delayed?

Yes. That goes back to what I was saying. In terms of public transparency and external stakeholders, it is true that the later we lay accounts, the later people need to wait for an update, either on new qualifications, new issues or existing ones. As I said, it is really important, and that is why it is a priority for timeliness. The question was around the internal budget setting and the internal management of issues—

I understand that. Andrew Cartner I was just trying to demonstrate that we have other processes in place. Last July, the Treasury set all Departments’ and all bodies’ multi-year spending review up to ’28/29. For three years, Departments have those budgets set, and then have their own internal planning rounds, budget-setting exercises and management accounts to manage how areas within that Department are managing that Budget in that forecasting. There is a lot of activity happening within Departments so that good financial management can continue.

Presumably that activity is taking place now for the accounts that you will produce next year. You are still doing some of the activity that relates to last year. Are you not finding that there is a—

Yes. This is the point. The longer a set of annual report and accounts continues, the more resource that takes; people will have to continue servicing that and the audit remains open. It is better all round for public transparency and internal capacity for accounts to be laid quickly. Then we can turn our attention to forward-looking matters.

What are the reasons for the delays? I am thinking of some of those that have been delayed recently. What specifically is holding those up? Is it lack of resource, or is it something else?

I wouldn’t say it is lack of resource. For two Departments, it is the local government pension scheme. That has impacted MOJ and DCMS. In a couple of other Departments, there are some complex provision estimates, so for the Horizon compensation scheme and the infected blood compensation scheme, because it is really difficult to estimate for new schemes. A lot of audit work has gone into them, trying to get assurance over those balances. That has led to some delays as well. The other category you would have for late accounts is the long-standing qualification—DWP and HMRC fraud and error qualifications that have lasted a long time. Sorry, I talked about qualifications there—I meant to talk about timeliness. The other thing on timeliness is that you can then still get unexpected things that may crop up in an audit as well. You could have a plan, but something unexpected could crop up late in the audit, which could then push a Department to lay late. They tend to be one-off issues, and then the Department would be back to pre-recess laying the following year.

Can you give us a bit more detail about what the Treasury and the Government Finance Function are doing to support those bodies that are routinely late, or late several years in a row?

Yes, I am happy to talk about that. Within Treasury, we do a lot of joint work with the NAO. We publish best practice guides. We publish a review into Department annual report and accounts. Where Departments are doing things well and they have good ideas on why things can be laid pre recess, we look to share that through our various communities, like our finance foundation group. That could be specific project management initiatives or it could be specific things that some Departments do. For example, in DWP, they have a launch event for their annual report and accounts process and then a finish event, so the whole organisation is behind that. We do a lot of that sharing of best practice. Within the Government Finance Function, we have quite clearly set in our new strategy that the finance foundations continue to be a priority, and that includes timely and accurate annual report and accounts. We are sending the signal that that is a priority for Government. Then we have other help available to people. We have technical help. We have a technical accounting centre of excellence, which can support organisations with complex accounting issues. Every Department and body has a point of contact within Treasury who they can go to to discuss things with. UKGI, which is one of the arm’s length bodies of Treasury, has a specialist contingent liability and provision centre of expertise as well, to help with some of those really complicated areas. The faster we can work through them, the faster those accounts can be laid. We offer a lot of support across Treasury and GFF.

Mr Bowler, can you give us examples of where the various measures that Mr Cartner outlined have been successful in bringing accounts to a more timely conclusion?

The most obvious one, when lots of Departments are doing the same thing, is sharing best practice. For example, when a new accounting standard comes up—the one that would have been in front of this Committee for a couple of decades would have been the leases changes, which caused the MOD headaches—it is about making sure that everyone is doing the same thing, training everyone up to a standard and bringing people forward. It is also about shared training and understanding. The one thing I would like to promote in this regard is the Government Internal Audit Agency. You don’t see it so often—you see the NAO—but it is a large agency that independently does all our internal audit. That exists to help to have better controls and governance, in particular, in Departments, which will lead to a better quality of accounts and shared best practice. It is a really quite big organisation that is much less visible to you, but it helps with that. I am biased, but I would say that the Government Finance Function is one of the more impressive cross-departmental functions. It is very much a community where everyone is trying to get through the issues together. Our relations with the NAO are good: we are on the same side as this Committee and the NAO in wanting value for money. I am sure that Gareth would ape this: best practice for Departments would be to go to the NAO and say, “We’ve got this issue and this problem. How are you going to deliver that in audit? How does that work? How can we do better than last year?” I think quite a bit of that does go on—

You think.

I know—quite—but Gareth will be able to tell us more. Obviously the clock is ticking, and you want to be producing your accounts in spring—in May—so that the NAO, which has a very large bottleneck, can audit off all 400 of them at the same time of year. You do not want to go to the NAO very late in the day with some massive new issue.

Mr Duffy, can you talk about the steps the Environment Agency are taking to bring forward the publication of their accounts?

Absolutely. We are aiming to do that by July this year. The history of the Environment Agency accounts is somewhat chequered. In 2022-23, which was the first year I was signing the accounts, we had four separate qualifications on them: irregular spending, valuation of land and property, failure to account for capital and revenue split in major projects and a depreciated replacement cost of the asset base. We cleared three of those up in the first year I was there, but that took quite a lot of time and effort. It is worth stressing the amount of work that the NAO, the Environment Agency and the Government Finance Function have to do on our accounts. There are 255,700 separate assets that provide flood and water management across the UK, varying from a penstock, gate or pipe to the Thames barrier. To get the account lifted, we had to use 10 million data points across those assets—the height, the width, the size and the condition of all those areas—and a lot of that data just wasn’t there. It took two years to get the methodology right, so that the NAO were happy with the methodology, and another two years to get the actual data to a quality at which the NAO felt comfortable to lift the qualification. That is a huge amount of work. This can be seen as a process thing—“Oh yeah, you haven’t filled this form in properly”—but it is not really. If we do not know the condition of our assets, you cannot scrutinise the state of the nation’s flood defences, you cannot give advice to your constituents about the risks they face, we cannot debate with Treasury colleagues what spending is needed on those assets, and we cannot even forecast risk of failure. That is really serious.

We know.

Any conversation about the state of our flood assets or water management needs this data, not only for us but for everyone else involved in the system—local government, farmers and others too. Now that we have the data, we have been able to do quite clever things with it. For example, previously we couldn’t group it. Often, flood defence assets work together so that the water does not flow around the back of one thing. Now we are able to say, “Can you group the assets in the database to say which work together? If this one fails, will that mean that one will fail as well? Where shall we prioritise our operational effect?” To make that work, we had to get the NAO embedded and we had to have governance. To make sure we had perspective on how to do this, we had the Government Finance Function, MOD, Network Rail and Highways England, because we all do similar kinds of work. We had to get quite a lot of IT spend to build an effective database for 10 million datapoints. We also had to get the operational people to buy into it. Operational staff like to deliver effect—they want to make sure that the floods are fixed. If you ring them up in the middle of a flood to say, “Stop what you’re doing and go and measure the height of this embankment and fill in this form,” they will think you are crackers. But if you explain to them, “Actually, this is really important, because this will help us to work out what we need to maintain, give you more credible plans for your work in the year ahead, work out how much money we need to spend and where, and work out risks and dangers across the hydrology you’re working on,” people get the memo. We saw the finance function, the NAO and my operational staff working out when they would do the surveys—not in winter; you do them in summer—how to prioritise and how to communicate the benefits. Now, we are using this dataset to do operational planning. We had a very, very wet winter, and it was the first where we were able to say, “That embankment has declined. Fix that one in spring.” That has been really powerful. That should give you a sense of the work going on. It did take time, and we did not file on time the last year. The NAO had to take more samples, and those samples are difficult. We are very hopeful this year that we will lay—

Hopeful?

No, more than hopeful. We have a clear project plan, and we are aiming to do that.

Finally, Mr Bowler, will the target of 70% of accounts being laid before the summer recess be met this year?

Yes, I think it will—I look forward to you playing that back to me.

We’ve got you on the record.

We do want to recover. I am particularly focused on the big Departments because, you know—

There is a bigger impact.

There is a bigger impact on bigger Departments.

Before I come to Rupert, there is a really important point here, James. That was a really helpful explanation, Philip, if I may say so, of why the valuation of leases, equipment and assets is really important. Equally, there is a balance to be had here, isn’t there? IFRS 16, for valuing leases, does not necessarily need to be done every year. You have introduced this thing; are you now going to come up with a streamlined way of valuing leases so that they do not need to be done every year?

Is that for me?

Yes.

I was wondering whether it was how often Philip was going to measure his—

It applies right across Government, and it is a problem.

Sorry—Andrew might know better on IFRS 16. I think it is the case that once you have done the enormous amount of work, you are in a much better place going forward. To give you the stat that you will perhaps know only too well, it took the MOD two decades to be compliant with IFRS 16—is that right, Gareth? Gareth Davies indicated assent.

Once you have revalued everything, you are in a much better place. It is the process, which I agree Phil has just described really eloquently, of having to do it that is a block. But the MoD has done it, and we are at the other side of that, so we are now into use. First, on the NAO Report—Mr Duffy just gave a really good example—it is really important that people understand that they are doing this for a reason. There are really good case studies in the analysis as to why knowing more about your assets, and about your contingent liabilities, is very important. Secondly—

We have to speed up.

I was also going to say that there is a proportionality issue here, and we do have this discussion with the NAO. You can polish something and polish something and polish something; there is a proportionality issue.

We are going to have to speed up. One simple question please, Rupert.

I do not know if you saw that Anna wanted to ask a question. I do not want to take over the Chair, Geoffrey; it just seemed germane.

Anna, there was confusion as to whether you wanted to ask a question. Do you?

If there is an opportunity to ask Philip a very short question that follows on from his previous response, then yes.

Very quickly, because we are well behind time.

Philip, you have obviously had to prioritise asset mapping. The docudrama “Dirty Business” has exposed that your predecessor cut costs, taking inspectors away from some of the frontline work to clean up our rivers. Has the work to meet the audit requirements in terms of asset mapping required you or your predecessor to divert staff away from some of your essential work on inspecting pollution incidents?

No. Look, it is very difficult for me to comment on things that happened a long time before I took up post. Many of the events in that docudrama were 25 or more years ago. I have said publicly that when I arrived at the Environment Agency in July 2023, I asked to see the water teams, given that that was the main issue I thought I was coming in to deal with. There were only 91 staff working full-time on water at that point, and there are 16,500 waste water assets, so whether those assets were getting adequate attention is a difficult question. Since then, with support from successive Environment Secretaries, we have increased our staffing to nearly 500 people working on that. This year, we will do 10,000 inspections of waste water around the country. I think I said to the Committee previously that we found a disturbing picture in those inspections. In the first year, 25% of the inspections found non-compliance, and 6% of those non-compliances were really serious—they could have caused really serious sewage spills. The good news, from the latest data I have, is that the level of non-compliance has reduced. Clearly, turning up and inspecting works. We have got that down to about 3% non-compliance, but that is still extremely high for a regulated industry, so we still have a lot of work to do. There is a principle that links this work to water, and it is this: if you try to follow the Nolan principles, as I do, of transparency and accountability—the points you were making, Mr Lowe, are very important to those principles—we actually want, in environmental terms, everyone to be an auditor. The agency is not just doing 10,000 inspections; we are publishing them all on an interactive website where you can look at them yourselves. You do not have to take my word for it, Ms Dixon, when it comes to what we found; you can look it up yourself, in your own constituency. That is how you gear environmental change in this country: you make visible to everyone what is going on in their own water environment, so that they can take action on it there and hold water companies to account. Sadly, we cannot enforce 2,500 non-compliances a year, but we can shine a light on that work, and that is what we are doing. That is very much in the spirit of aligning the operational data and financial data that we see in the Report.

I am going to call a halt to that line of inquiry. It is extremely important stuff, but it is outwith the scope of this hearing. Rupert, one short question, please.

My question is for Andrew, I think. I see that you are a passionate Newcastle supporter. It is a great, passionate football club—love it. I also see that you have spent 23 years in the civil service, and you have worked in two of the worst-performing Departments: the DWP and HMRC. I think I am right in saying that the DWP’s accounts have been qualified for 37 consecutive years. How are you actually helping these bodies to improve? They have become used to having their accounts qualified. In the private sector, your bankers and suppliers start to get worried, and your credit lines start to dry up, but that does not happen in Government. What are you doing to help these people who seem to have their accounts regularly qualified? What are you doing to help them to avoid that?

Lots of stuff, using the experience of my time in the civil service. It is actually 27 years that I have worked in the civil service, and I have done a variety of roles in the Government Finance Function—not just financial accounting. I have done management accounting and finance business partnering as well. My career spans different roles and has not all been involved in the qualifications attached to those different organisations. From that experience—

Should I have asked James, maybe?

I was just going to finish answering. Basically, my role is to work with the finance community across Government to prioritise removing qualifications and ensuring timely accounts, as well as having effective finance functions. I have mentioned that we do a lot of training, networking and best practice. We do a lot of recruitment to bring qualified people into key roles within Government—that is both private sector and internal candidates. Lots happens to do that. The other thing we do is an end-of-year finance assessment. We have a set of criteria that, within Treasury, we assess and rate Departments against. That will be things like, “What is the quality of your business cases like?” and “How good is your financial forecasting?” We do all of that activity and then, when we are doing those assessments, we know what the best practice is and about issues that other Departments have encountered. We try to share that and give recommendations for Departments to improve their finance function.

But 37 consecutive years suggests that that is not working. Do you need to do something different?

You make a really important point. DWP and HMRC are qualified on the basis of fraud and error—something that you have pushed on specifically and we have given evidence about before. Although some of the qualification is almost inherent in the systems that they run, there is never going to be any complacency about accepting fraud and error for year after year, and there have been significant pushes to reduce that. The one I am closest to is in HMRC, where we have put an awful lot more people into the compliance side of HMRC to push down fraud and error and to close the tax gap. I think the stat is that they have reduced that by over £850 million compared with where it was in 2021. This is money that the taxpayer otherwise doesn’t have to provide Government, and its reduction is deeply important to us. We work really hard on reducing fraud and error, even if that qualification persists. The NAO are the experts, but I think it is almost inherent in some of what DWP and HMRC do that their systems will have some forms of fraud and error. It is about trying to reduce that to the absolute minimum, and we have very tangible progress to show there.

Finally—before the Chairman cuts me off—given my knowledge of the DWP and HMRC, I think they are now both structurally flawed. You’ve got to do a lot from a governance point of view and from a cultural point of view to change the way they operate. Figure 4 in the Report deals with DWP, the Department for Education and HMRC. I think you have a cultural problem in these organisations, and the key thing the Committee would like to understand is how you are going to change that culture. I think DWP and HMRC are deeply flawed culturally. How are you going to change that? You are not going to change the outcome after 37 consecutive years. No, it has become ingrained and—

I have no doubt we will cover both those things in many future hearings, but will you give a brief-ish answer? The question about culture is really important.

I would point to the very aggressive and important measures that both Departments have taken to reduce fraud and error, which must be part of a cultural change that never accepts an ingrained level of fraud and error as okay or good enough. HMRC is the one I know best. We can see, in higher tax revenues, what you get from really bearing down on that fraud and error. There is £850 million more in tax because of less fraud and error, thanks to the actions HMRC has taken since 2021. The cultural point is absolutely important, and they, and I think I, would push back against the idea that there is any form of complacency in that regard.

You have indicated that you see the assistance given to Departments to get this right as an increased priority compared with where you have been in the past. Is that fair?

Yes, I think the Government Finance Function is much more interventionist and helpful than it has been in the past.

We had a very helpful explanation from Philip Duffy about how the problems in the Environment Agency have been addressed. There is clearly a big spend there, and important issues, but there are a lot of small bodies that have perhaps been asked to do too much of late. We have looked at this before in the Committee; I think that we are waiting for a response. Could the requirements on some of the smaller bodies be reduced, so that they do not really have to do as much work on this and can get their accounts in in a more timely way, without as much effort being put in?

Yes, I think proportionality comes to the fore there. We welcome the Committee’s push on not having overbearing requirements on small bodies, so that they are not spending all their time delivering the requirements rather than doing what they are there to do. We agree with the standards there. Yes, I think that we can reduce the requirements on them. Then, of course, the larger Government Finance Function and the larger best practice can come in and help them even more than the very large Departments, which have really rather advanced internal capability and capacity.

So are we going to see that?

This links to the hearing in January on the small bodies. We are doing a piece of work to think about what the options could look like, as per the Committee’s recommendation. To go back to Sir Geoffrey’s point earlier around IFRS 16 and how we do accountancy within Government, we are going to think about what other opportunities there may be for proportionality in terms of financial accounting across Government. We are going to look to do that review and report back in the timescale agreed with the Committee. We are looking for specific things to do there.

That is helpful. Paragraph 2.9 of the Report drew to my attention the £452 million in DBT for “unrecoverable redundancy…payments”.

I did know what that was, for a minute.

It seems to imply that the payments should not have been made, if they should be recovered but cannot be.

I did know about this. Special payments and losses are something that the public sector does better than the private sector, as it is more transparent. In the private sector, it has to be material to the accounts, but in the public sector, you have to declare them more often. I am trying to get to the details of—

Perhaps you will have to write to us.

What you see in special payments and losses is a change of Government policy—such as the Rwanda scheme, retiring assets or the MOD deciding that it is not going to go ahead with sets of things that it was going to do, as well as some of the things on the covid side.

This is about—

It is a very specialist point, James, so I am happy for you to write to us.

Chairman, Elaine has given us an answer on the line.

Can I ask one specific question? I am trying to get you on board, Permanent Secretary—this is a bit of a hobby horse for me, as I have raised it with other Departments. Fraud and error is a serious matter, and it is important that we get people to identify and report it, but we need some information and feedback. What has happened to the concerns they have reported? When you deal with people in HMRC, they do not want to talk, and it is always confidential. They say, “We can’t discuss it, as it is third-party information.” This really is an issue. If you are encouraging people to come forward and report, you probably cannot give them a blow-by-blow commentary on what is being done, but there is a secrecy agenda across Government that does not help encourage people to report. Will you take that on board as an issue across Government that might be worth looking at?

I will very happily do that.

I have had a bit of inspiration from the NAO team on these redundancy payments. On the Department for Business and Trade, I am told it is where it has paid redundancy payments to employees of insolvent companies.

Oh, this is the Insolvency Service.

Yes.

Sorry, yes, I better write to you on the Insolvency Service. The Insolvency Service exists to cover situations where companies go insolvent, and it is often the case that they do not get their money, although they try to.

That is really helpful. Unless anybody wants to come in on anything else, I have three entirely different but important questions to ask. First, James, can we go to page 15 and talk about the auditing requirements for small bodies? You will be aware that we had a hearing the other day, and we are in the process of producing a report. Since then, we have had a really good real-life case study of a small body that faces a disproportionate cost on a very small budget. You may want to comment on that now, but auditing requirements for small bodies is something that we need to look at really seriously.

I agree, and I think the proportionality point is important. The only thing I would add to what Andrew said earlier is that we also need to work with our audit colleagues. If the bar is set at the same height for small bodies, they have to respond to that, and not just our side of things. I think there is willingness in that regard. Elaine Lewis indicated assent.

I am sure discussions have already been had on that. Next, I want to go to page 16, which is a subject that you have touched on: losses and special payments. Paragraph 2.8 states, “the 17 departmental groups reported more than 2.7 million losses totalling £6.6 billion and special payments totalling £293.5 million.” These are big figures. Again—this is rather like going back to the beginning—the MOD incurred losses of £1.6 billion simply for cancelling projects. In the public sector, we should surely be able to get better at deciding whether we want to spend money on a project and see it go through. To actually cancel it, and incur these huge public losses, seems unacceptable to me.

I would say that an advantage of the accounts is to absolutely bring all of this out. While it is painful, we encourage that, and transparency is helpful there. You are right that the Department for Transport cancelled eight road schemes, and that was linked to the change of Government. The new Government came in and said that they did not want to go ahead with those—the A303 tunnel under Stonehenge is the one that jumps out at me. There is a value for money trade-off in this, because it is not necessarily the right answer that once you have said “Go” you must always complete it—as we see with heavy rail. I agree that you do not want to be incurring large costs in things that you do not then proceed with—although obviously people have the right to change their minds. This comes to the challenge whereby Departments over programme against their budgets and essentially look to do more things than they might be capable of doing. That particularly comes to pass if there is a change of Government and people have slightly differing objectives. However, I agree that you do not want to incur large costs on things that you then do not proceed with. This transparency is a way of trying to push back at doing that.

Yes. Looking at page 24 of the Report and these compensation schemes, there is the alarming sentence right at the very bottom, “in terms of combined compensation paid to date and future provisions—at £102.8 billion as of October 2025.” That is the total payments on compensation schemes. It seems that the compensation culture is getting much bigger in this country and that is a huge figure. I am wondering if you should do some work and look internationally to see how we compare as a percentage of GDP on paying out compensation?

I am happy to look at that. What to say? First, on clinical negligence—

Indeed, both are over the page.

Yes, both the Department of Health and Social Care’s accounts, but also the whole of Government accounts. To Ms Olney’s point, the advantage of timely accounts is seeing the balance sheet and the contingent liabilities and provisions. On things such as clinical negligence, we see this very high amount paid out every year. However, there will always be clinical negligence in a health system. The question is: what do you do about that? That has led to quite a lot of action, and the 10-year plan represents a desire to do something. A barrister called Mr Lock is doing a review of it. At the moment, the large percentage increases in the last year or so are because of those two large compensation schemes—the infected blood and Post Office compensation schemes—coming on train.

I accept that there are certain costs in compensation schemes baked in already. However, in future we need to design compensation schemes that are less litigious and can therefore be settled quicker and more cheaply—and the Committee will be doing some work on that.

Yes. They have made some progress on clinical negligence, where there is a large number of low-value claims and the legal fees can exceed the payment. There is definitely some work to do there. To your point, this all comes out of the money that you could spend on other things, and the larger the compensation, the smaller the remainder is. You want to be designing into any scheme upfront assurance and a requirement that you are not going to do something that is going to lead to large payouts. A large part of what the Government legal service does is advise Ministers and officials about ensuring policy developments avoid successful legal challenge.

The Committee will be doing more work on that. Finally from me, the Committee is very keen on digital transformation. That is very important if we are to meet the Government’s efficiency targets. On the other hand, looking at page 30 of the Report and the £10 billion that is going into the NHS, this is a huge job of work. The Government, as we know, are not, overall, very good at doing these large transformation programmes. They tend not to be well designed and are over budget and over time. What is the Treasury doing to make sure that the £10 billion is being spent on a value for money basis and achieving what it is supposed to? I talk to all my GPs, and they cannot contact specialists digitally because they run different systems. Yet here we are, with bits of the NHS purchasing digital systems that are not compatible with one another on an almost a daily basis.

You make a very good point. It is even more important, with the advent of AI and the changes we can expect, that we as a Government do AI and digital transformation better than we have in the past. The NAO’s conclusions here—that day‑to‑day IT is generally done well, but larger projects less so—is correct. I will ask Andrew to say a little about what we are doing, but the shared services clusters are all about getting people on to the same IT and using the same financial management. There is a role for the Department, DSIT and for the Cabinet Office’s digital functions to drive up best practice and performance. DSIT did not exist so many years ago; that Department now has lessons it can use to help. Then there is the day‑to‑day learning from the realities of where we have been in the past, and from the value for money reports that you have done, to try to push this better. The issues that come out here—people not getting their governance right, poor testing and go‑live decisions, and the challenges of moving from one provider to multiple providers—remain, and they are a concern.

I will mention very briefly three things within the Treasury. NISTA sits within the Treasury, with the project delivery authority function as part of that, working with major programmes and projects to help them improve best practice—all that type of stuff—and ensuring that Departments have capable people to deliver those projects. James mentioned GIAA earlier—the Government Internal Audit Agency. That is an ALB of the Treasury, and it does specific internal reviews, recommendations and controls for major programmes, as well as general control environments within Government. The third thing I would mention, as part of the Government Finance Function, is the effort around skills that finance professionals will need in future. It is worth thinking about digital, data and change management skills, so that finance professionals working with these big change programmes are fairer with that, and know the issues and the pitfalls to look out for. Those would be three specific things as well.

I am sorry that this has been a rather quick canter through a very important subject. I thank all three of you today—three extremely busy people—who have given us some very useful evidence to think about. There will be elements of what we have and have not covered today that we will want to come back to. An uncorrected version of the transcript will be available in the coming days. We will also decide how we are going to handle the evidence we have heard today and how we can progress it.