My core role is to make sure that Parliament and the public get high-quality assurance on the Government’s accounts, but also an insightful assessment of value for money in public spending. At the moment, we are dealing with three big risks to our ability to do that. One risk relates to ensuring that our quality—particularly our financial audit quality—matches the best standards of the profession, which have ratcheted up significantly in recent years, after the various scandals in the audit world a few years ago. We have been on a long journey to make sure that our quality is at least as good as the best of the private sector. We have had to work hard on that. In the past year, there has been good evidence that we are closing that gap and, very encouragingly, are achieving what we need on audit quality, but we still have more to do. Some of the cost here is an investment in completing that journey on audit quality. Secondly, we are keeping up with the growth in the number of public bodies that need auditing. As we explain in the document, quite a lot of new bodies have been created by the previous Government and the new Government, which adds up to a fairly significant increase in our workload. That is not a voluntary decision on our part; it is us doing the job that I am required to do by statute. Not only are there more public bodies, but in lots of them there is increasing complexity of financial arrangements. The final thing is about making sure we remain a competitive employer of auditors and performance experts. All our staff are highly marketable in the private sector as well as the public sector, so a continual battle in running the NAO is making sure that our employment offer is sufficient to retain and recruit the talent that we need. We are always trying to strike a balance, as a public body, in doing that in a way that represents value for money for the taxpayer. Those three things together—better quality, more auditing and a competitive offer to our staff—have put this cost pressure into our budget for next year.

We do, yes. We have a really detailed, bottom-up operational planning process involving our teams. The information that has driven this is not a top-down set of guesses; it is detailed information about what it is taking to audit the organisations for which we are responsible. I am confident that with this budget we can come to Parliament proud of the quality that we are delivering in our audits, which is at least as good as that in the private sector, if not better; that we are continuing to give you the insightful value for money reports of which this Committee is the main user; and that our staff retention remains within the targets that we set for ourselves. In the past year and a half, we have kept our staff turnover within the targets we set for ourselves, which is a big improvement on the situation two or three years ago, when we were seeing quite high levels of turnover to a booming private sector.

I have shared with Committee members our new five-year strategy, which takes effect from April. If I could characterise the last five years, they were essentially about getting better at what we do. That has been our focus. Clearly, while doing that, we have had to contend with pretty extraordinary events like the covid pandemic, the energy crisis and so on, but I think we have successfully improved the quality of what we do. The next strategy focuses more on asking, “With that, what does a highly effective NAO allow us to achieve, working with Parliament in the interests of taxpayers?” We have been more ambitious around productivity and resilience in public services. They are two big value for money themes for the next few years. Clearly, demands on public sector organisations—health demands, social demands and so on—keep going up, but the resources cannot, so the only answer to that is productive and resilient public services. There is a big theme in our new strategy about auditing in a way that encourages positive innovation in Government rather than encouraging risk-averse behaviour or avoiding innovative experiments and so on. That will be a very important area for us in the next few years. The other big chunk that we are more ambitious around is better financial management and reporting in Government. The nuts and bolts of good financial management can never be taken for granted. In our audit work, we keep finding evidence of scope for significant improvement in that. We think that those are the foundations that will allow you to tackle productivity and resilience in a more successful way.

Some things are very measurable and important, such as timely, completed audits of Government accounts and good-quality results from our regulator. Our audits, as you know, are inspected by the same regulator as for the big firms, so there are measurable things to show whether we are doing our job well. In some ways, however, what is more interesting to us all is what difference all this has made to the productivity and resilience of public services. We are doing a lot of work to better measure the impact of our work in those areas. Of course, we cannot be confident that it is always our work that has made a difference in a particular area, but we are committing ourselves to taking far more interest in what the measures of public sector productivity are telling us; whether there is evidence that what we are doing is contributing positively to that; and if not, why not. If what we are doing is not making a difference there, that is something we should challenge ourselves over—the topics we are selecting, the way we do our work, the way we communicate or the incentives for people in the system who could be doing more with their resources but are not. What is it that is getting in the way of our effectiveness? By being more ambitious about outcomes, we are holding ourselves to a higher standard of results.

Yes, there will. Of course, there will be big debates about attribution: “This may have improved, but was it anything to do with the NAO?” That is a perfectly fair question, but without the data you cannot even start to have that discussion.

That is why it has been our top priority for three years. As I was saying at the start, the huge agenda for quality improvement in the audit industry over the last few years, driven by the regulator in response to the scandals in Carillion and other audited entities, has completely transformed the way these audits are done. It has also driven up the cost, as we were discussing at the start. Essentially, a gap opened up between the big four firms, auditing the largest companies, and the rest—including us, sadly—in the last three or four years, so we have had a quality-first plan, which is our internal plan for learning quite how far the bar has moved on these things. We are talking not about the basics of auditing, but about, for example, the standard of evidence required on valuations and on asset impairments. All that has transformed in the last few years. Not only have we had to respond to that, but the bodies that we are auditing have had to respond to it. We have had to explain to them what the new standards are and what those require in documentation as part of the audit preparation of their accounts. We have been on a big journey. In the last three years, we have replaced our audit software platform with a brand-new one. We developed that ourselves with a technology partner, and it came in on budget and on time. We have just finished the first round of audits using that system, and it was very successful. It sets us up for using automated audit tools and AI tools in future, and we have started to do that this year. That is one area of big focus for us. Another is learning and development for our auditors. There is no point in having new tools if we have not trained people to use them, so there has been a big focus on our new methodologies and a lot of learning for our teams to do. A huge effort has been under way in the NAO over that time. I am pleased to say that the programme of inspections by the FRC on the results of the ’23-24 audits is coming to an end at the moment. I cannot give you the final results, but we already know that they are significantly better than last year. That 43% figure you cited will be significantly better in the year that will soon be reporting. Now, you do not declare victory after one good set of results. There is a long way to keep going on this, and it is only after we have seen repeated results that we will be confident about the progress we have made, but that is more in tune with my assessment. Obviously, I see the inside of our audits, and I think that is a fair reflection of the effort that the teams have put into improving things over the last few years. This is fundamental to us. That is why it has been at the top of our risk register for three or four years. It is essentially our licence to operate—so that when you see an audited set of accounts for any public body, you can completely rely on the accuracy of that audit.

We are quite unusual in doing the full range of financial statements audits that we do. Quite a lot of our equivalents in other organisations contract out all their financial audit work to the private sector and do not do any. We think we are unique in being externally regulated by the same regulator as the private sector; if that is not right, I will write to you to correct it. We are very unusual in doing the range of audits that we do and having the full degree of scrutiny from the private sector regulator. I genuinely think that, provided that we maintain the level we have attained in the last 12 months, you can be confident that the NAO is the leading practitioner of financial statements audit in the public sector among any of our equivalents. In recent years, the then Chair of the Public Accounts Commission used to say, “Is it realistic to aim to match the standards of the big four?” My answer was always, “The audits that we do require us to do that, because we audit the BBC and its group, the Ministry of Defence and the Department of Health and Social Care. These are some of the largest, most important and complex sets of accounts in the country, and it is crucial that they are audited to the same standard as a FTSE 250 company.” I think that was the right answer. It has been very tough for our staff—it has been a long slog to do that, and I think it will continue to be challenging for our people—but it is the right standard for us to aspire to, and I think it is realistic as well.

We are getting more information all the time about the creation of new bodies. Often we know that they are coming, but it is a question of which year there will be a first set of accounts that require audit. It is also the ones that already exist that are coming from the private sector inside the public sector boundary. By far the biggest group that we are working very closely on at the moment is the train operating companies. They will become NAO audits over the next two or three years. Whether it is two or three is a really big piece of information for us: it is a bit like us taking on the BBC and its group as a new set of audits. We are already in quite a detailed conversation with the current auditors of those organisations on the timetable that the Government want to see for the transfer over to us. Hence, we are building up our staffing now, even though some of those audits are two or three years away. We cannot create an instant audit team from scratch; we have to build the layers over two or three years, so we are now starting to do that. That kind of information becomes clearer. In fact, even since we wrote this document, it looks like the strong requirement on us is to take those rail audits on in two years rather than three, which is quite a significant bit of information. We still think that this is the right figure for the coming year, but it will be a large exercise for us. The other information that we update all the time is the risk assessments of individual audits and the level of complexity we are expecting. Is there a new financial instrument, for example, that will require a significant amount of audit effort? Our resourcing system picks that up every month, essentially, as our teams update our risk assessments.

At the moment, we think this is the best we can estimate. We are not doing this knowing full well that we will have to come back later in the year. It is always a possibility, though, so we can never rule it out. I guess that is what the process is for—things that you could not have predicted at this time of the year, but that become a requirement later. We have done it before, so I am prepared to go to Parliament for a supplementary estimate if absolutely necessary, but I see it as my job to avoid that if I can. For example, this year—the year that is about to end—we have required every penny of our budget. We had a debate in the autumn about whether it was going to be enough, but with reprioritizing a few things, cutting our cloth accordingly and one or two tricky decisions, we are now very confident that we can land this year within our estimate. We will do the same with this one.

Well, we are doing. This is not more staff to do the same work; this is more staff because, crucially, we are having to take on a higher workload. The question is: what are we doing to improve our productivity almost regardless of the volume of that work? That is where the investment I was talking about in our IT capability is really important. The way that audit work is developing in the industry generally, there is going to be a reduction in the amount of junior-level work in a typical audit because the automated tools that we are starting to apply for the first time this year take out some of that junior-level time in reasonably manual tasks—for instance, agreeing a figure from the accounts to a supporting document. That can all be automated now in a way that has not been possible before. We reckon that just that one tool speeds up that part of the audit by three times, and so it is starting to take out appreciable amounts of time at the junior end of the skill mix. Obviously, it takes time to work that through into our staffing requirements. But by the end of this five-year strategy that we are discussing, I would expect our skill mix to look a bit different on the financial audit side and to have a lower volume of resource at the junior end of the skill mix, because we are using more automation and more intelligent tools to deal with that part of the audit. Our auditors will then be using their professional judgment for the things that they are trained to do and that require human judgment. I think that that is coming and will result in more productive audits in due course, but it will take some time to work through.

One of the areas we have not talked about yet is that we are insourcing a degree of the work that we have previously outsourced to the firms. That will mean a headcount increase for us, but lower costs than would otherwise have been the case. That is particularly the case for whole audits where there is not a high level of specialised complexity. The firms’ fees have essentially priced them out of that work for us, and we know we can do it at the same level of quality but more cost-effectively. We have estimated that by next year, this modest but meaningful amount of insourcing will save us about £1.5 million compared to the cost if we had not done that. That will require us to increase the headcount again to continue to insource that work, but we will spend less than we would have done if we had have left that work in the private sector. I would rather measure cost as our productivity measure here, rather than just headcount.

Yes. I might bring in my colleagues on this, but this is why we do relatively little of the in-house employment of specialists who are not auditors. The areas where we need this kind of advice are valuations, for example, and surveyor expertise. We also need actuarial advice from qualified actuaries, and we need financial instruments advice from people who are professionally trained to value financial instruments. We do not have any of those teams in house yet, but the case is growing for it because we are doing so much of that work, and paying so much to the specialist consultancies who provide that advice to us, that the make-or-buy decision that we have for each of those specialist areas is getting more marginal. We are having to look quite closely at that. I am worried about the point you mentioned—that we will still have to buy in some level of expertise. I am also worried about professional isolation. I would not want to be, for example, the only actuary at the NAO, because the point about being a professional is that you have peer review, you learn from other people’s workloads and so on. We need to be confident that there is a sizeable critical mass of that work before it is viable for us to have our own in-house expertise.

We used a lot of benchmarking of turnover levels and pay levels compared with our competitors. A very relevant factor this year is that, as you can see from the timing of our estimate, we always have to estimate and agree the pay award earlier than the civil service. Last year our pay award to our staff was 3% and then, a few months later, the civil service settled at 5% for the same year, which obviously caused a frisson in our organisation. We have borne that in mind when setting this year’s award, so our 3.5% has to be seen in the context of the 3% last year. That is 6.5% over two years, but the civil service had 5% in one year, so I suspect we will still be behind its two-year increase. Pay pressure in the private sector is constant, so we keep a very close eye on benchmark pay there. We think that the figure we have included in the estimate is the minimum necessary to responsibly look after our turnover and make sure that we are not faced with a shortage of key people when we have such a big workload.

No, it is our proposal. It is subject to approval by you and the Public Accounts Commission, but it is our recommendation, based on the benchmarking that I was describing.

Yes. The intake last September was our largest ever—just over a hundred graduates and school leavers—and we are looking for a similar volume in this intake. Clearly, we have to make sure that we keep our quality standards very high in that recruitment process. We are doing very well at the moment, at a time when the firms are reporting quite hard going in the graduate and school leaver markets. There is a general concern in the audit industry about the attractiveness of audit as a career. There used to be no shortage of graduates queuing up for jobs in the big firms, but those firms are now finding that more challenging. We have not found that, partly because of the unique status of the NAO. We can offer unique access to the audit of organisations that nobody else can audit. Because that appeals to a sufficiently large number of high-quality graduates and school leavers, we have always been able to fill our places. This year, so far, we are doing well on our recruitment for next September—it is so far, so good but that is an absolutely crucial bit of our intake. As I said earlier, we think that the skill mix will change, and not just in numbers. We think that we will need to start recruiting larger numbers of data scientists alongside trainee auditors, given that a large part of our audit process will be data analytics. I am sure that if we have this discussion in a couple of years’ time, I will be talking about two streams of recruitment—with accountants still being the largest—and a very significant data scientist stream, which is already starting with us.

There are two large areas where we spend money on the audit firms. One is outsourcing whole audits, and that is the bit we are reducing over the next two or three years. That process, as I said, would deliver about £1.4 million-worth of savings, compared with what it would cost if we carried on using them.

Million. We are not that big. The other part where we have seen cost pressure is the specialist advice that we have to buy—what I was just describing was the valuation advice and financial instruments advice—and the audits we are doing right now require significant amounts of that specialist advice—for example, when we audit specialist insurance entities like Pool Re, and other organisations. The cost has been going up, and that is the area where we are looking at the make-or-buy decision: would it be more cost-effective for us to employ our own experts in this area, rather than continue to pay these high fees to the private sector?

Yes, we have certainly picked up the Committee’s interest in us doing more of that. We have been working on it ourselves. We are going to make more intensive use of the OECD data, always checking the data quality as we do that, of course. Recently, we used some OECD data, and we spotted that the UK data had been submitted incorrectly. We must not just take on face value some of these international comparisons without digging a little bit under the surface.

We are certainly going to bear data quality in mind—but, absolutely. We have good relationships with our equivalents in other countries, particularly other developed nations with similar challenges around their public services. We are going to use those relationships to maximise the quality of the comparisons we can bring to you. Of course, it is not always easy, particularly in sensitive areas such as defence and so on, but where we can, we will make sure we maximise that evidence base.