Public Accounts Commission
Oral evidence: NAO Annual Report 2024-25 and Supplementary Estimate 2025-26, HC 1345
Wednesday 19 November 2025
Ordered by the House of Commons to be published on 19 November 2025.
Members present: Clive Efford (Chair); Phil Brickell; Sir Geoffrey Clifton-Brown; Sarah Olney; Chris Vince.
Questions 1-31
Witnesses
I: Dame Fiona Reynolds DBE, Chair, the National Audit Office; Gareth Davies, Comptroller and Auditor General, the National Audit Office; and Rebecca Sheeran, Chief Operating Officer, the National Audit Office.
Witnesses: Dame Fiona Reynolds, Gareth Davies and Rebecca Sheeran.
Chair: Welcome to today’s meeting of the Public Accounts Commission on the NAO’s annual report and accounts, and the request for a supplementary estimate. Can the witnesses introduce themselves, please?
Dame Fiona Reynolds: I am Fiona Reynolds, chair of the NAO board.
Gareth Davies: I am Gareth Davies, the Comptroller and Auditor General.
Rebecca Sheeran: I am Rebecca Sheeran, the chief operating officer.
Q1 Chair: Thank you, and welcome. Why was it not clear in the main estimates that the NAO would be taking responsibility for the new train operating company audits?
Gareth Davies: At the time of the discussion of the main estimates with the Commission—you may remember—we were clear that the expectation was that we would become the auditor of the train operating companies as they transferred into the public sector, but the pace of change was uncertain at that time. As we explained, at the beginning of this calendar year, the expectation from the Department was that we would be taking on two additional audits for 2025-26—that was reflected in our main estimate—but with the caveat that, as we discussed with you back in March, that was uncertain. It was possible that there would be more than two for 2025-26.
The Department had assumed that the incumbent auditors would stay in place for an additional year; that is what changed. In May, the incumbent auditors let the Department know that they were having to rotate off the audits for 2025-26, because of commercial conflicts that they had built up in other parts of the rail group. I think that pushed the decision forward, as far as the DFT operator was concerned, so that number of two became eight—the operator itself and seven of the train operating companies.
That is what we have now had to gear up for in the 2025-26 audits, which, of course, have to start in this financial year, because we have to do the planning and the interim work before the end of March. That is a substantial commitment. These are large organisations. Our rough rule of thumb in terms of audit time is 2,500 hours for each of the train operating companies, and there are eight, so it is a big commitment. Around about May or June, we had to put in place firm plans for gearing up for that volume of work, and obviously that was too late to be in time for the main estimate.
Q2 Chair: Since you got late notice of the change in the number of contracts that were coming back in-house, could the Government have given you earlier notice?
Gareth Davies: I think it was because of the decision of the auditors to resign a year earlier than the Department had expected. I am not sure it would have been possible for the Department to have known that earlier than it did. That is obviously a question you would have to ask the Department directly. As far as we are concerned, we have had good communication with the Department throughout all of this, and with the train operating companies we are taking on this year. The process of getting to know the audit risks and understanding what we are taking on in technical terms has already started and is going well.
Q3 Chair: You have calculated that the cost is £1.2 million. How did you come to that figure?
Gareth Davies: I have given you an idea of the size of each of the audits. That is an important assumption that was built into it. That is quite a well-informed estimate on our part, because we have very good dialogue with the incumbent auditors from whom we are taking over the audits, as well as with the organisations themselves.
We are drawing on the NAO’s experience of taking on new types of work in recent years. Nine years ago, we took on the audit of the BBC and its group, which was very similar in scale to what we are describing for the train operating companies. That was very good preparation, because it taught us how much you needed to do up front to understand the risks of a new organisation and make sure that any specialist technical requirements of a new sector were properly understood.
In the case of the rail companies, how the revenue system works—how the allocation of ticket revenue between the train operating companies works—is obviously a key technical issue for the accounts. There are many others—the leasing of rolling stock and so on—that our teams are having to understand. That process is going well, but we also learned that we have to treat this as a major change project.
This is not just taking on a new audit task; we need to make sure that our workforce plans are updated in a way that means we have the resources to do the work. We have to make sure that our IT capabilities are able to deal with the systems in use at these organisations. Part of our gearing up for this work—part of that £1.2 million—involves hiring two IT audit staff with the relevant specialist expertise that we will need. We have hired a new director with commercial audit experience and a senior manager who has worked on train operating company audits in the past. That is why we have needed to invest in line with the project plan, which we think will allow us to deliver good-quality audits from the first year.
Q4 Chair: As you just said, you have brought in people with knowledge of auditing TOCs. Have you had to upskill your existing staff as well? Are there specialisms that the train operating companies bring with them?
Gareth Davies: Yes. Commercial-type work has become a bigger part of the NAO’s portfolio over the years. We are approaching 100 Companies Act audits that we sign off each year. That is a significant increase compared with, say, 10 years ago. We now have more audit staff who understand the requirements of commercial audits—not just the technical requirements, but the risks that exist in those kinds of organisations in a way that they do not exist in Government bodies. That general expertise has been very helpful in preparing for the train operating companies, but we have also needed some very specific things, of which I just gave you examples.
Q5 Sir Geoffrey Clifton-Brown: Good morning, C&AG. We have discussed this a lot, but how much of the additional expertise will you not be bringing in-house? I am thinking about people like specialist valuers, because you will need to have values for a lot of rolling stock and all sorts of specialist equipment. How much of that will you need to buy in?
Gareth Davies: This is a live issue for us, as we have discussed previously. The accounting standards and associated audit work have become not only more demanding, but clearer about exactly what is required from the people preparing financial statements—in relation to the basis of their estimates and then what the auditor has to do—so this is a substantial piece of our work nowadays. We have had to outsource expertise in valuations, as you say, as well as in pensions and financial instruments.
We have built centres of expertise in the NAO to act as the intelligent client for that work—knowing who to commission the work from and how to make sure we get what we need. Increasingly, those teams are becoming experts themselves. We are looking very carefully at this “make or buy” decision for specialist work. At what point would it be right for the NAO to have actuarial or valuation expertise in-house, given that this is increasingly a predictable volume of work for us, rather than a one-off or something we need occasionally? We have not yet committed on any of those, but it is a big project for this year and will inform what we do next year.
Q6 Chair: Your estimate highlights £400,000 in efficiency savings from investments in audit methodology and tools like DataSnipper. Can you provide detail on how these tools have changed audit delivery and whether further savings are expected?
Gareth Davies: We have been talking about this in the audit profession for a long time, and it feels as though it is now happening in practice. Those are examples of where commercially available tools have proved to be very effective in increasing the efficiency of the audit process. The one you mentioned, DataSnipper, is a tool that allows what previously were routine tasks carried out by junior members of the audit team—matching numbers in the accounts to invoices, confirming that those details are corroborated and transferring all that to the audit file—to be automated. The tools use optical character recognition and can match numbers and flag anomalies. Of course, they still need expert auditor oversight, but we reckon they have increased the efficiency of that part of our audit by three times—it is three times faster than the way we have done it for many years.
We are also piloting the use of other commercially available software, like Copilot, which many people will be familiar with. Copilot turns out to be very useful for things like scanning the board minutes of an organisation and extracting relevant matters for audit consideration. Again, it needs human oversight to make sure you are not missing anything, but it can really speed up those administrative tasks.
We are in the foothills of a big change in how audits on large organisations are carried out. We rely heavily on IT controls and extracting entire populations of data, rather than just taking audit samples, and we are then able to identify anomalies and errors much more directly using data analytics. Rather than science fiction, this is now happening on our audits, as it is in the private sector. A big challenge for us is how we target our investment in these tools accurately so that we are using things that are proven to be effective, but not falling behind either. We do not want to be too cautious and then take too long to deliver the available efficiencies.
Q7 Chris Vince: You have described this year as a “pivotal year”, based on the modern methodology and the software you have touched on. Could you add a little bit more to that? You have talked about some of the technology and the methodology, which is obviously really important, but how effective has that been? How do you see it moving forward?
Gareth Davies: Yes, I did call it a pivotal year in the annual report. Dame Fiona will want to come in and share the board’s perspective on this. In the first five-year strategy of my term of office, we obviously had a lot to do, with covid and many other external challenges, but we concentrated on getting better at what we do by investing in our methodology and our tools in the way I have just discussed. I think that has been effective. Our new software platform really does set us up for exploiting the technology that I have described. A really encouraging sign is that other audit offices around the world are showing an interest in using the platform we have developed. Audit Scotland is already formally doing so, and others are in discussions with us about how that might prove useful. That is a good indication that there is something valuable there.
It is also pivotal in a deeper sense: having increased the NAO’s capability and effectiveness, it is now about the purpose of that. We are keen to maximise our impact on better financial management and reporting in government, but also the productivity and resilience of public services. That is what our new strategy prioritises. We do not want just to produce well-researched, hard-hitting reports; we want change. We want to see more value for money and we want better services for the available resources. Everything we are doing is designed to use the efficiency gains that we are making in our core work to be more influential in change in the civil service and in government.
Dame Fiona Reynolds: I am sure you have seen our strategy document, which highlights trust, value and impact. You have heard me say many times before that, since I have come to the NAO, actually making a difference has been a hugely motivating factor. As Gareth said, we have spent the last five years building our capability internally, and building our confidence internally that our systems, processes and the outputs we produce are really high quality, and we are very proud of that. The strategy, on which we spent a lot of time as a board discussing, debating, thinking about, getting external perspectives on, getting internal perspectives on and consulting our staff on, is all about making that difference. We are really proud of the document and the ambition that we have set out.
What I am particularly proud of, though, is the way that staff internally have responded to the strategy. It has been hugely motivating to think that we are aligning every single thing we do around how to make a difference, not only to the resilience and productivity of public services but to financial management in government. This group of MPs, and Sir Geoffrey in particular, know very well that there are challenges there that need to be addressed. We are feeling really proud, which I think is why the “pivotal year” remark really resonated throughout the organisation.
Q8 Chris Vince: Let us move on to spending. You exceeded your budget for audit and assurance last year. I appreciate that you managed to come in under budget because of higher income than forecast, which has balanced it out, but it is important to look at why you exceeded your budget for audit and assurance work. Could you give us some idea of why that was?
Rebecca Sheeran: You are quite right: we spent around £1.8 million more in support of our audit and assurance work than we had planned to at the start of the year. In essence, that reflected the work that we needed to do, across the more than 400 financial statements that we certify every year, to ensure that we had a fully standards-compliant audit for each of them. It is not always possible to predict with absolute precision exactly what work you will need to do before you have carried out the detailed planning work for each one of those engagements.
That £1.8 million broadly breaks down to around £1 million more that we spent on temporary staff to support our audit teams during our very busiest period, particularly in the run-up to the parliamentary summer recess, which is the real peak in our workload, and around about £0.7 million to engage specialist skills and independent expertise in the areas that the C&AG just spoke about—things like complex pension liabilities and valuations, or complex financial instruments, for which we need to bring in external expertise to comply with standards. As you say, we more than offset that with the increased income we were able to recover from our fee-paying audits.
Q9 Chris Vince: On temporary staff—we have discussed this previously—I appreciate the offset bit, but we want to bring the overspend down if we can. Is that a challenge when it comes to temporary staff? I recognise that there is an element of seasonality to your audit work, but are you able to limit the additional cost in any way, or is it just part and parcel of the nature of your work?
Rebecca Sheeran: We are always trying to look at the balance and to establish the right balance and the right value for money in terms of having the right size of permanent workforce and bringing in some temporary resource. Because we aim to complete 70% of our audits before the summer recess, we are always going to have a peak, so I do not think we will ever be in a world where we have no temporary support, because that probably would not represent the best value for money overall in delivery of those engagements to that profile.
Gareth Davies: Quite a chunk of our temporary resource is people who come back every summer to work with us, so they are people familiar with our systems and able to do a good-quality job without having to learn everything from scratch. That is an important point.
Chris Vince: There is obviously a saving inherent in that fact; you are not just getting someone in off the street.
Gareth Davies: Yes.
Q10 Chris Vince: We talked about the additional income, which is very welcome. I wonder whether that is that something that you sought to gain, or something that happened—not by chance, but you know what I mean.
Gareth Davies: Some of it was just because more audit work was required and we therefore had to charge additional fees. That is a well understood process in audits. It is a guarantee of quality, really. If the auditor feels too constrained by the fee available to cover the cost, they may not do the work necessary, so that is an important principle.
As well as additional work being necessary, the other element of the additional income in the year was that we did slightly better than we had expected in closing the gap between what some of the fee-paying audits cost and what we have been able to charge them. A table in our accounts demonstrates that there is still a gap between those two numbers—we are not fully recovering all the costs from the fee-paying audits—and we have a programme of steadily reducing that gap year after year. Some of that progress is reflected in the additional income.
Chris Vince: So you are shifting cost from the NAO books to the bodies that you are auditing. That is a basic way of looking at it—
Gareth Davies: Yes. If that is the cost of doing the audit, that is what they should be paying—that is the principle. Of course, we are all keen to keep the cost down as far as possible, as these are all public bodies, which is why the efficiency measures I was describing earlier are so important, because they will feed through into benefits in fee reductions for those organisations. Collectively, we are trying to make sure that the fees are no higher than they have to be, but consistent with us doing a good job.
Q11 Chris Vince: There has been a decrease in the proportion of MPs and senior officials who agree that your insights are accessible, clear, easy to read and understandable. How can we address that issue, and what does that say about the effectiveness of the new insights team?
Gareth Davies: We get good feedback on them. The figures are, as you say, in a couple of cases slightly down on the previous year. I think that in the case of MPs, that is largely explained by the fact that there was such turnover at the election. We worked very hard to introduce the NAO to new MPs after the election, and that has gone pretty well, but this survey was carried out about four months after the election. Actually, we were pleased how high the figures remained in the light of that scale of change.
Dame Fiona Reynolds: I definitely agree with that. We were expecting that there would be a significant fall, given a 50% turnover in MPs, so we were delighted that the figure remained so high. That does not mean that we are not working very hard to build relationships with new MPs, but we were very pleased.
Gareth Davies: Of course, the civil service point is not explained by the election. That one we are determined to turn around. I talked earlier about being more ambitious about the influence we were having within the civil service, on applying the lessons from our work, so a big focus of what we are doing now is designed to improve those figures that you mentioned on the civil service.
Q12 Chris Vince: That is reassuring. As a new MP, I had a slight advantage because, obviously, I am on this Commission. Finally, what is your measure of success for the insights team?
Gareth Davies: It is some of the measures you have just mentioned, because the whole point of them is that practitioners in government find our work useful in improving what they do. That is why we ask that question every year, and that is our key measure of success. It will also be reflected in the financial impacts and other impacts of our work. Our annual report has lots of examples of those.
Clearly, with a more ambitious strategy, we think it is logical that we should be more ambitious about the financial impacts of our work in government—how much does our work help Government save and do more efficiently? We are increasing the 10:1 target. We have had a target for a long time for our work to provide financial benefits to the taxpayer that are 10 times the cost of the NAO each year. We are increasing that to 15:1, to reflect the greater ambition in our new strategy. We would hope that the insights teams are making a substantial contribution to that figure as well.
Chris Vince: You have led nicely on to the next question.
Q13 Sarah Olney: I want to go back to the question of staffing. In previous years, the NAO has struggled to retain staff due to a competitive market for qualified auditors, but now your turnover is significantly below target in every grade. Is that a concern?
Gareth Davies: Well, it is a much better problem to have than the previous one. We are really pleased that some of the action we took to deal with high levels of turnover has been effective. Some of this is down to the fact that we restructured our grades after qualification. It is more attractive, I think, for our staff to stay with us now on qualification than it had been, because progression is quicker.
But it is not all down to that. It is also down to the external market. There is no doubt that the firms who are one of the main competitors for our talent have been less aggressive in their recruitment in the last 12 months. That changes quickly, in our experience, so we are not at all complacent about that. That is why we are working so hard on developing our staff and increasing their engagement with the organisation, so that they want to build their careers with us. As I say, it is a much bigger problem to deal with if your best people are leaving with the skills that you have spent years training them in. We are very pleased about the improvement there.
Of course, that leads to two pressures. One is budget pressure, because we have to assume a level of turnover in setting our budget. That is manageable, we think. The other one is progression through the organisation. We do need a level of turnover to free up manager vacancies, director vacancies and so on, so that ambitious people can move on and progress in the way their skills are allowing them to. We do not want blockages in the organisation. We obviously do not want to encourage people to leave, but we do want to make sure they feel well equipped with skills that allow them to take jobs in other organisations at higher levels. We have a big emphasis on our staff development programme for that purpose.
We are also revitalising our secondment programme. We have had a very successful long-standing secondment programme with Parliament to Select Committee teams and so on, but we are now pushing that and extending it to other organisations—crucially, where we are not the auditor. There are ethical constraints on us seconding auditors to our audited bodies, but there are obviously a lot of other places where people can be seconded to great benefit to their career development and so on, so we are pushing that, too. It is something we are focused on, but as I say, I would rather have this challenge than the shortage of staff problem.
Q14 Sarah Olney: In addressing that previous problem, do you think you overcompensated in your pay offers to NAO staff?
Gareth Davies: I don’t think so, no. I think we have judged that about right. Clearly, turnover is one of the things we consider when we look at each year’s pay settlement, but there are other factors as well—for example, the cost pressures on our staff, which have been substantial. If you had the staff representatives here, they would quite strongly make the other case. They compare our pay rises to the civil service, and sometimes we match them and sometimes we do not. That is always an issue. But I think we have got that balance about right in recent years. We are obviously having to think about next year now, so this is a big consideration in our minds for the coming year, too.
Q15 Sarah Olney: What is your expectation for the next year? Will it be a more competitive market?
Gareth Davies: It turns very quickly, in our experience, so we are cautious about that. We must not be complacent. We need to make sure our staff feel well treated, engaged and properly rewarded. Also, as we have just been saying, with the train operating companies and so on, we have an increasing workload coming up. Our programme of insourcing some of the previously outsourced audits is continuing as well, because we can do them for better value for money. All of that means we need our qualified, engaged staff in high volumes. To be honest, our main consideration in thinking about next year’s pay award is what will retain our best people and make sure we can do a good job.
Q16 Sarah Olney: In the last financial year you agreed eight exit packages totalling £236,000. Seven of those were disclosed in the accounts as special payments and not agreed within the usual provisions of the civil service compensation scheme. Can you explain that?
Gareth Davies: Yes. One of them was compulsory and carried out in line with the scheme, as you would expect. The others were not compulsory redundancies. Every case is individually different. It could be that somebody’s skills are no longer what we require and, despite a lot of work on development and so on, that has not proved effective. Often there are cases of long-term ill health that fall short of the standard required for ill health retirement under the civil service scheme—our staff are not civil servants, but they are members of the civil service pension scheme. Those are the two criteria, really. We take a business case decision in each case. Clearly, we cannot allow long-standing ill health problems to continue, because that means the team is short of the resources it needs, and it is not good for the individual either, so in those cases we will come to an agreement with the individual for one of these voluntary departures.
We use the civil service scheme as a guide to what is appropriate, because it is based on years of service and so on, but we are not always completely bound by it. We may be able to come to a resolution with an individual that is slightly different from the terms of the scheme. That is what we mean by departing from the scheme. There is a business case for each one of these, proposed by the chief people officer. I review each business case as the accounting officer and sign them off where I think they are justified. The payback on these is very good. Usually it means we are simply saving cost or we are replacing somebody who does not have the skills we need with somebody who does, and obviously that is a big payback for us. For the size of organisation we are, this is the kind of level I would expect each year.
Q17 Sarah Olney: Can I quickly ask about that? It was eight last year but only two the year before. Were you keener to sign off these business cases because of the pressures you already identified?
Gareth Davies: Yes, and because of sickness absence, actually. The number after the pandemic had crept up, as it has in lots of organisations in all sectors. There are lots of reasons for that, but one of them was that we needed to deal with some of these long-standing cases a bit more efficiently and expeditiously. I would say the level from last year is more like the number I would expect now.
Sarah Olney: So you think two is the outlier rather than eight.
Gareth Davies: Yes.
Q18 Sarah Olney: Interesting. You also disclosed an extra contractual payment to a supplier of £165,072 as a special payment. What are the details of this payment, and why were the terms of the contract not adhered to?
Rebecca Sheeran: Yes, there was a payment of £165,000, as you say. This was to an IT service provider who provided end-user computing and infrastructure services for us; essentially, they looked after all our devices—our laptops and things like our mobile phones. We were not far into the contract and it became clear that we needed to exit the contract in order to secure a service that was going to better deliver what we needed for our IT estate and represent better value for money longer term. To your point about the contractual requirements, we feel that the payment we made reflected genuine services and work that we had received that was helpful in setting the new contract with a very clear specification of the service we would need going forwards.
Sarah Olney: So it was an exit payment, effectively—
Rebecca Sheeran: It was an exit payment.
Sarah Olney: To one supplier, because you found another supplier who was better able to meet your needs.
Rebecca Sheeran: Because it became clear that they could not deliver the service we were looking for at an affordable cost, and to give us the opportunity to go back to the market and find a service that was going to better meet our needs at better value for money.
Q19 Sarah Olney: Were you able to procure that more suitable service at a reasonable cost?
Rebecca Sheeran: We have done, and we have a service that is working very well for us at the moment.
Q20 Sarah Olney: In that context, you think £165,000 represents value for money.
Rebecca Sheeran: Absolutely. Gareth did a thorough accounting officer assessment and considered the value for money. As I said, the amount that we paid genuinely reflected the services and work that had been done that was genuinely valuable to us, to that sum.
Q21 Chris Vince: So there was no additional cost to exit the contract early, in effect, because you were paying for what had already been provided.
Rebecca Sheeran: No, this was a negotiated settlement to exit the contract.
Q22 Sir Geoffrey Clifton-Brown: C&AG, your financial impact has jumped from £17 per £1 spent last year to £53 per £1 spent. You have also introduced a new methodology to calculate and attribute financial impact, which no longer relies on Government agreement. Is it fair to say that you have watered down your methodology for calculating financial impact?
Gareth Davies: No.
Sir Geoffrey Clifton-Brown: How surprising. [Laughter.]
Gareth Davies: I don’t think that would be an NAO thing to do. The background to that change is that we had a review of our approach from our auditors, who queried whether it was ethically robust essentially to be negotiating with the audited body on the attribution of the savings from a particular piece of audit work from us. There was just a slightly uncomfortable feeling that we are independent, we are their auditors, and we want that to be a very clear distinction, so wouldn’t it be better if we just calculated, on a proper basis, the financial impact of implementing the recommendations from that piece of work and reported it, without the awkward point of needing the agreement of the audited body on the amount?
In practice, of course, we still discuss that; there is no dispute that what we are setting out here is accurately measured. It is just a cleaner process that, as the external auditor, we are giving our professional view on the results of implementing those recommendations, and reporting that clearly to you in public. We thought it was important to do this because it is crucial that Parliament understands what benefit this audit work is having, and it is important for us to understand where we are having the most financial benefit from our work as well, so that if there is an opportunity to do more of that kind of thing, we should be taking it. That is the background to the change.
Had we not made the change, we would still have had a much bigger number this year. It would have been 40:1 instead of 53:1, which is clearly still a huge leap from the previous year, so it is not as if we have suddenly distorted the position by this change of methodology. The other point to make is that we unfortunately cannot bank on this level every year. I mentioned earlier that we have increased our target from 10:1 to 15:1. I think that is more realistic and a good measure of our ambition to do a better job on this, so we will hold ourselves to that for the current financial year. There were a couple of very large financial impacts in the 53:1 figure, which are not likely to happen every year.
Q23 Sir Geoffrey Clifton-Brown: Thank you. We had a briefing earlier from your external auditor, with which you have a contractual arrangement. There was quite a lot of discussion about whether five Companies Act and two non-Companies Act audits were a sufficient sample size. Do you have a view on that?
Gareth Davies: This is the FRC’s inspections of our audits?
Sir Geoffrey Clifton-Brown: Yes.
Gareth Davies: We discuss this every year with the FRC. The context here is really important. If you look at the large firms—the biggest firms, in tier 1 of the FRC system—around 10% to 15% of their audits fall within the FRC’s remit, so only a minority of their audits are eligible for inspection by the FRC. In our case, it is pretty much 100%, so all 420 of our audits are eligible for inspection by the FRC. That is a really important factor to bear in mind when you compare sample sizes between us and the firms.
Every inspection is an intensive process that demands a lot of time both from the inspection team at the FRC and from our audit team, and we have to manage the total impact of that on the organisation. Seven is a big commitment for both of us and gives us valuable feedback every year. We always bear in mind, though, that it is a small sample, so we do not over-interpret any single year’s results. As I am sure we will talk about in a few weeks’ time with the Commission, our results have improved significantly in the last year from a base that was too low.
I think both of those are probably exaggerated. I do not think we were quite as bad as the previous figures suggested, but I do not think we are quite as good as the latest figure suggests, and that is a function of the sample size. As long as we are aware that seven audits do not tell you everything about a population of 420, the results are still incredibly useful, because they give us thematic areas for improvement, which have proved to be effective for us in previous results. It is a massive driver of quality improvement in the organisation. One year we tried to do nine. We found that quite burdensome, and we did not learn enough of additional value to justify the cost and time taken for that larger sample. I am comfortable with seven as a good indicator of where we need to improve.
Q24 Sir Geoffrey Clifton-Brown: Can you assure us that they can do exactly what they want in respect of Companies Act and non-Companies Act audits? Do they choose completely independently?
Gareth Davies: Well, it is a principle of the regime. It is entirely down to the FRC which audits they pick to inspect.
Q25 Phil Brickell: Your financial audit performance target for 2024-25 was 70% and you hit 55%, so you missed the target. Your annual report attributed that to the general election, because the audits had to be completed before the parliamentary summer recess. If the general election had not been called when it was, would you have met your 70% target?
Gareth Davies: I think we would have fallen a bit short, but we would have been much closer—somewhere in the 60s, I think. You will remember the timing of the election: if you wanted to disrupt the audit timetable, that is when you would have an election. It was in early July, which is the peak period for signing accounts. Clearly, for Departments and principal public bodies, you need ministerial sign-off. That was the problem.
Q26 Phil Brickell: Were there other contributory factors that were not attributable to the election and would have meant you did not hit your 70% target? In the preceding reporting period, you did not meet it either—you were at 60%.
Gareth Davies: Since the year we are talking about, we have obviously had another summer recess—summer 2025—and we hit 64% this summer. It is still not quite there, but it is obviously better than the previous year. I would attribute some of the improvement on the previous year to the fact there was no election, but some of it is due to the work we are doing with Departments to clear the qualifications to their accounts, which in many cases arose during the pandemic. That has been a long job, which we are still doing. We have had fewer qualifications this year, which is a good sign of progress, and that also speeds up the process. We are planning to build on the momentum that improved it for this summer so that next summer we hit our 70% target. It has been a long haul from the pandemic to get back to where we should be on this.
A crucial point to make, though, is that when we do get back to that level, which is comparable to before covid, we are talking about quite different audits now. These are not the same audits we were doing in 2018; they are much more sophisticated. The auditing, financial reporting and audit standards have changed significantly in that time—we talked earlier about the work we are doing on valuations and other areas of expertise. They are kind of unrecognisable.
It is a bigger achievement to deliver 70% of our audits by the summer recess now than it was in 2018, because the work between us, the Departments and the arm’s length bodies is significantly more complex. I want to recognise that and the work that our teams put into that. They have worked really hard to help Departments to, first, clear up some of the accounting residue of the pandemic and, crucially, adapt to the new standards that have been a challenge for the preparers as well as auditors of the accounts.
Q27 Phil Brickell: On the topic of the NAO operating segment expenditure, there seems to be a shift towards more expenditure on financial audit. Does that reflect a change in NAO strategy? If so, what does it tell us?
Gareth Davies: It is largely about volume. We have no choice about whether to audit the accounts of the bodies we are responsible for, and there has been a volume increase in recent years. We have talked about the train operating companies being the latest main example of that.
On the value-for-money side, we have stuck to our objective of producing just over 60 value-for-money reports for the Public Accounts Committee, and Parliament generally, every year. The PAC has a finite capacity, and at the moment Sir Geoffrey and the PAC are not saying, “Could you produce more work?” The challenge is in following up the existing inquiries, as well as accommodating the new reports we produce.
We have had to accommodate more financial audit work and we haven’t increased the volume of our value-for-money work, so the balance has shifted in the way you described.
Q28 Phil Brickell: To drill into that a bit more, you mentioned earlier more support for Parliament being part of your strategy. According to our briefing—correct me if I have this wrong—expenditure on support for Parliament was £5.8 million in 2023-24, and that reduced to £5.5 million in 2024-25. The figures seem slightly to contradict what you are saying, unless what you are outlining is forward-looking for the next year?
Gareth Davies: I think there was election impact in 2024-25: the PAC held something like 37 sessions, compared with the normal 60, because of that. Very unusually, there was no PAC meeting between May and October, because of the election. That change is because of the reduced activity in Parliament, rather than any change on our part.
Q29 Phil Brickell: Thank you; that is very helpful. On governance and risk management, your audit and risk committee considered the lessons learned from the British Library cyber-attack and the implications for data loss across the NAO. What can you share with us about lessons learned from that?
Gareth Davies: This is one of the biggest risks we manage day to day. Rebecca can explain.
Rebecca Sheeran: We are always keen to learn from the experiences of others who have been so unfortunate as to experience such an incident. As we set out in our January ’25 report on cyber-resilience, the cyber-threat that we all face is severe and advancing quickly. We take cyber-resilience and information security incredibly seriously at the NAO, and opportunities to reflect on things like the British Library incident have helped us to think about where we need our investment and attention to be in the organisation.
In terms of giving some assurance, we have in place the things you would expect of an organisation of our scale. We have Cyber Essentials and Cyber Essentials Plus accreditation, which has been recently recertified. We hold ISO 27001, which is the international standard for information security management. We now have all our key systems in the cloud and multi-factor authentication in place on all our key systems. We are currently in the process of rolling out biometric identity access for our laptops and devices, and we have implemented unphishable identity controls for our critical accounts and services.
Those are some of the things that we have increased our attention on in response to those learning exercises, as well as doing quite a lot with our people. Everyone have to undertake mandatory information-security training at least once every year. We carry out regular phishing simulations and use them as an opportunity for organisational learning and reflection. Reflecting partly on that British Library experience, we have also been working on improving our disaster-recovery protocols and systems, should we be so unfortunate as to have a cyber-attack.
None the less, this is a constant area of attention. The complexity and sophistication of the threat continue to increase. Earlier this year, we decided we needed to reprioritise some of our resource to increase the pace of some of the work we are doing on our information-security maturity, such as increasing our capacity to proactively detect and respond to any unusual activity we notice within our estate, and starting to build our approach to secure development in the cloud so that we can safely harness some of the opportunities from things like generative AI.
Q30 Phil Brickell: That is very helpful. On another governance and risk management topic, Gareth Davies mentioned long-term absence because of sickness. You have highlighted in your report that sickness absence, mental health and wellbeing are among the highest concerns in your risk register. You touched on the pandemic, but have you identified what the contributory factors to some of those risk-register items are?
Gareth Davies: Yes. We survey our staff three times a year now, and this is always something we ask about to make sure that we understand the developing picture. Obviously, our HR team also builds up a good understanding of individual cases and helps to feed that into our discussions.
First, we think this is economy-wide—these are general issues that we are seeing in organisations across the piece. For us specifically, as well as the general factors, workload is obviously a significant challenge, particularly in a seasonal business like audit, with an annual peak, as we have already discussed. We take that very seriously.
Some of the responses in the last 12 months are that we overhauled our system of time off in lieu, which is quite common for organisations with a peak in work like this. Where we need some extra effort at the busy period before the summer recess, people then get that time back when it is quieter, and we do not need quite so much of their time. We improved the terms of that system for last summer’s peak, and that has gone down well with the staff and has been better used.
The main effect has been to spread the additional effort across a wider group of staff. More people have given us their discretionary extra effort as a result of that, but fewer people have worked worryingly long hours. We obviously keep a careful eye on our obligations under the working time arrangements, and we always intervene if it is shown that anybody is overworking for an extended period. Workload is a key factor in that.
We have also modernised our offices; we had a two-year programme for overhauling our offices so that they are better environments in which to work. Again, that is going down well and making a difference to staff wellbeing. Our chief people officer is also leading a new wellbeing strategy. For example, we have always had a gym in the basement of the NAO, and we are making that more accessible to staff than it had been before. There are lots of practical measures to try to help tackle that. As we say in the annual Report, it is good to see that our sickness absence came down significantly between 2023-24 and 2024-2025. It has continued on that path during this year, which I think is a good sign.
Phil Brickell: Thank you.
Q31 Sir Geoffrey Clifton-Brown: Dame Fiona, sadly your term as chair comes to an end next year. May I thank you very much for the service you have given to the NAO? I know it is hugely appreciated by those of us in Parliament and no doubt those in the NAO itself. I want to give you an opportunity to reflect on your term as chair, and how the NAO has changed, with particular emphasis on staff. We have discussed the new AI tools and others that are going to require better, experienced oversight, but I am concerned about what happens to the more junior graduates coming in. Will there be jobs for those people? Particularly with your other hats on, how does this affect youngsters coming into the professions? What are your reflections on the overall changes in the NAO?
Dame Fiona Reynolds: Thank you very much for your kind remarks. It is a bit of a shock to realise that I am nearly five years through my six-year term, but I still have a year to go—I am not anticipating my demise by even a day. It is a privilege to serve this organisation; it is a bit like having a bird’s eye view of the whole of central Government expenditure, which is a unique perspective. Even more motivating is the opportunity to try to make that better—to see public resources spent more effectively and programmes being delivered with more effective value for money. That is an incredibly motivating cause.
For me, the staff have been a joy to work with, and they will be right through to the end. They are highly intelligent and curious, and they have energy and perspicacity. Watching them become engaged in this energy around better impact and more purposeful application of our work, whether that is in financial audit or value for money, has really been an incredible change, and also a success. It builds on many years of preparation, as you have heard, and I think it is reaching a point where there is an enormous sense of opportunity in the organisation that we can really fulfil our purpose in ever more effective ways.
I emphasise that the hubs and the insights work have added enormously to that. The old division of FA over here and VFM over there is now less categorised, and people are working and using their skills collectively to those ends. Financial audit itself is becoming more purposeful in that we realise that the skills and the insights gained from an audit are going to help better resource allocation and better programme planning, and enhance the effectiveness of how public money is spent.
We are seeing change; it is evolutionary rather than revolutionary, but it is very exciting. I would add that we have a fantastic board, and our non-execs are really brilliant and great to work with. Making sure that that remains the case is also an important lesson for the future.
Sir Geoffrey Clifton-Brown: Thank you very much. You and I share that bird’s eye view of the whole of Government expenditure.
Dame Fiona Reynolds: On your point about the next generation, this is something we talk about. I do not believe the NAO will ever not have a cohort of trainees coming through. We are passionate about our trainees, and they are increasingly motivated by the new strategy ambition. They may be doing different things, and I think that they are probably rather pleased that DataSnipper and other tools are making their lives slightly more interesting. Many of them have always had the opportunity to work on VFM as well as on audit during their training periods with us.
Obviously, there may be a question about the volume, which changes slightly every year anyway, but it is far too early to say that there will not need to be a cohort of trainees. In fact, it is rather the reverse: we know we will be training the auditors for the future as long as the NAO exists.
Chair: Your term comes to an end in January 2027, so we look forward to seeing you over the next year. I understand that you are looking for your replacement now; there is nothing like starting early.
Dame Fiona Reynolds: It is a sensible process, but I am determined to serve.
Chair: We look forward to working with you for the time you have left, and I suppose we will be seeing you in two weeks’ time. That concludes the evidence session. Thank you very much for your attendance today. The Commission will now deliberate in private.