Good morning, colleagues, and welcome to this session on the ONS and the UKSA. We are delighted to be joined by Professor Sir Ian Diamond, a past leader of that organisation. Welcome, Sir Ian, and thank you for joining us. We look forward to hearing what you have to say. Let me start with a bullish question. You may well have followed the Committee’s previous hearings into this issue and will doubtless be aware of the enormous concern that we have on behalf of Parliament with regard to our nation’s ability to collect national statistics in a meaningful way for use across the whole range of policy formulation. I know that you will need no lessons on the importance of robust statistics. To colloquially summarise a fair assessment of the immediate past situation: the ONS had gone to hell in a handcart. First, did it really? Secondly, what is your assessment as to why? If you think it did not, and your period at the helm has been either misrepresented or misunderstood, now is your opportunity to put your take on the record, and colleagues will look forward to questioning you on it.

I fully recognise that the ONS’s reputation has taken a major hit, and I feel very sorry for the thousands of dedicated public servants who work there and who are doing great things. I do not believe that the organisation has gone “to hell in a handcart”. There are many serious challenges which have been there for the last decade. When I first joined the organisation in autumn 2019, I had been there perhaps two to three months when I had a briefing on research and development and the survey on business research and development. That was in the context of a request from the chief scientific adviser for Wales. During that meeting, I thought to myself, “Hold it, hold it, hold. This is an area in which the numbers have not changed for a long time.” Right the way through the decade beginning 2010, the percentage of our GDP in R&D has largely stayed constant, despite enormous efforts from the Government to do something about that. In my very quick view, the survey had not changed. You asked me, “Why?” The first “why” is that, in that particular incident, the survey had stayed constant throughout its time, and therefore missed the new parts of the population it was trying to estimate. I immediately said to the then director general, who is an outstanding civil servant, “I’m worried.” To his credit, he went away and worked with his team. As a result, over the next couple of years, using both administrative data and a new and extremely good survey, with a new design—to my first point—the numbers have changed. There had been an error for a period of time, and the design, frankly, needed updating. One of the mantras I used all the time when I was at ONS was: the economy and society are changing, and therefore we need to change all the time. This case was very similar to—I am sure you will remember it—the 1996 average weekly earnings issue. It was a very similar situation, which is one of the reasons I picked it up quite quickly. The first issue was: were the surveys fit for purpose? Over time, my former colleagues have input a lot of work into really understanding the designs and thinking through the surveys, because in some situations, the designs were not right. I was assured, in a meeting probably six to nine months ago, that we feel now the designs are about right, but alongside other national statistics institutes across the world, there really needs to be a thinking through of the role of social and business surveys in national statistics. Let me be very clear, Chair: that does not mean to say, for a minute, that we stopped doing surveys. There are things you can only get by asking people or by asking companies. I said that right from the beginning, but we also need to recognise that people are busy, and people want to answer questions, if they are persuaded it is good, when they want to do it, in a way that they wish to do it. That means you need to think through—I think the future has to be the linking of administrative data and survey data. Problem number one was that, in some cases, the designs needed thinking through. I am sure we will come back to other examples. The second problem, which became more apparent to me as we were going through, was—again, largely during the last decade—the strategy. You had the statisticians, but not everyone, writing the programmes, and the consultants coming in to write some of the really complex programmes, and you had a large number of programmes that were written largely—not always—in a statistical package called SAS. They had been added to over time, often by consultants without the proper documentation. What needed to happen was to move from there to open source software, and to do so, really understanding the documentation. When that has happened—there have been a couple of examples this calendar year—that has led to us finding some errors in the codes and therefore making small changes that were needed. That was the case, for example, with the producer price index, which has been referred to. Those are two of the main reasons why there have been errors. There have also been one or two other areas where there have been magnificent moves forward. The measurement of migration, in my opinion, is much better than it has ever been in this country. It is now done, largely, using administrative data. I am in awe of the work that the team on that have done, but to use an analogy I cannot quite remember, it has almost been like building an aeroplane while you are flying it. By not being able to use the previous survey during the pandemic and needing results, you could not actually do what you would like to do, which is to dual-run over a period time. They have done magnificently, but there has been a need, as they do that, to revise some of the statistics. The other area that clearly has caused an enormous amount of reputational damage has been labour force statistics. I really regret that in every way, but at the same time recognise that the teams working in those areas have been working incredibly hard and, in my opinion, incredibly effectively. If you ask why that has happened, it was for a number of reasons. First, historically the labour force survey, which has been going since the 1980s, had been added to and added to, and often other Departments would come in and say, “Do you mind asking an extra question?”, and it would be added. What I think happened, again before my time, was a decision that we would move away from this long questionnaire, where the interviewer comes and sits with people—it is very difficult to get really good samples, in particular at the two ends of the income distribution, and to get people to sit down with you for 45 minutes and have a conversation. I think the field work at the ONS is outstanding, but it is just hard, and that is the same everywhere. I know one colleague said, “Well, I could have told you this,” and when I came to look at it, I agreed with this person completely. What I think was tried initially was to take the questionnaire that people had for face to face and to try to mirror that in an online questionnaire. What that meant was that it was too long, people did not finish it and you had problems. I think they are going in the right direction now, with a very short questionnaire. I am very hopeful, although I clearly have no insight to give you, that over the next short while, the trial—the design is very good—will enable that survey to move forward and there will be very good employment statistics. The final point I would make in that area is that because things are done in quarters, you cannot simply say, “We are going to change,” and it will all be right tomorrow. You have to have four quarters at least of dual running of the labour force survey and the new labour force survey, for the obvious reason that employment is seasonal, and you need to understand whether you are measuring seasonality. That is why it has taken rather longer than anyone would wish, particularly working with stakeholders. I met once with the Bank of England, and the chief economist was very clear that the Bank would only accept four quarters of dual running. That meant that it would take some time.

So in summary, your assessment is that you have identified some backlog problems and are seeking to address them. In shorthand, while understanding there are difficulties, the organisation has not gone—to use my phrase from earlier—to hell in a handcart. The Bank of England clearly lost faith and confidence. Commentators lost faith and confidence. The Government were in the process of losing faith and confident in the robustness and reliability of the ONS to meet the challenges that were facing it. Why do you think they had arrived at that conclusion?

It’s very hard. Let me come back to the labour force statistics, a key and important statistic. That was what the Bank’s Governor was referring to when he spoke to the Select Committee. I regret deeply that that happened in the way that it happened and that we in the ONS did not try to shore things up as quickly as we could. That, I think, was a major problem. Please, please, please, I recognise the problems; I am not hiding from them. But at the same time the Bank and the Government were using GDP and inflation data, and there were no challenges. The major challenge that we did have on GDP, let us be clear, was a revision. You always revise GDP because you get more information, and there was a revision that came about in 2023, which was a result of the rapid adjustments that had to be made during the pandemic. Then when you got the data, there were some estimates that, frankly, had not gone as well as one would have wanted. So I think that, because the labour force statistics and, as I have tried to indicate, one or two others that happened came at the same time, a snowball rolling of errors, which we have been very honest and very clear about, led to that reduction in belief in the ability of the ONS to measure the economy.

Colleagues will come on to a variety of questions with regards to culture, governance and the like, but let me pause there for a moment and put this to you. Since this inquiry began, I have received quite a postbag from current staff at the ONS, all completely demoralised and browbeaten. Most say they had tried to raise issues and concerns that the culture of the organisation neither encouraged nor allowed. On the hiring and firing of colleagues—we will come on to questions about that within the organisation—could it best be described as haphazard, if not chaotic? You are clearly an individual who cares about and understands the importance of valid, robust statistics and how they are gathered. On a scale of one to 10, one being not very good and 10 being brilliant, how would you describe yourself as a leader of people and in fostering a culture of engagement, inclusivity, the picking of brains and the learning from others?

First, I do not think your question can be answered in one word, but on many, many occasions I started meetings by saying, “It is a speak-up day.” On a speak-up day, you have to be prepared to speak. On another occasion, it was pretty clear to me that colleagues at the most senior level outside of the senior civil service, the grade 6 level, felt disenfranchised, so I held 11 open meetings over a period of three months, just to sit down and say, “What are the issues?” One thing that was pretty clear to me was that, as you have said, people did not feel that if they said anything, it would necessarily be listened to. I have already given you the example of someone who said, “I could have told you this before.” So what I tried to do was to really improve the cascading that was going down, and when I found out—

But it didn’t happen, Sir Ian, did it?

I don’t think it did, which was what I was just going to say. So what I did was go round again. I thought, “This isn’t working as well as I would like it to do,” so I changed things about a year ago and I invited all these colleagues at grade 6—that means over 1,000 people typically—to come to the first Monday morning meeting of the month. I found that very, very helpful. People were, I found, prepared to speak up. We had a very good conversation once about problems that colleagues felt they had with HR if they raised issues of harassment, for example. So I have tried very much to build much more of an inclusive culture. I have worked very hard personally with the leaders of the protected characteristics networks. I would say—I don’t hide from it—that I have been critical in meetings at times. It has proved difficult—do I regret? Yes, I do sometimes, but at other times—if the statistics are wrong, you cannot say anything other than “I think these statistics are wrong,” I would suggest. Okay, I am not going to give myself a particularly high mark, but at the same time all I can say is that I do feel that while one person can set the culture for an institution and try very hard to make that culture happen, it is for everyone in the institution to take that forward, and I am not entirely clear that we always had the cascade through that I would have liked.

I am sure colleagues will come back to that. Lauren Edwards.

We have obviously heard a lot about leadership style and priorities throughout this inquiry. Could you briefly describe what you hoped to achieve in your time as National Statistician?

Two—[Interruption.] I am sorry; I am just trying to turn my phone off because it seems to be going off.

Sir Ian, can I just point out that the microphone in front of you is just for recording purposes rather than amplification? There is a bit of air conditioning whirring around as well, so if everybody could speak up, that would be helpful.

I will try to speak up. What I tried to achieve, very simply—I am passionate and always have been about official statistics and the role. My guiding light for the whole of my career—well, since before I had a career—has been that statistics, properly calculated and properly used, can shed light on inequalities in our society and do something about this. That is where I come from. When I was asked if I would come and take this role, my motivation was very much—I did feel before I arrived that things were not great, and my motivation was to improve the quality of statistics. I remember a very senior member of the board saying to me once when I said to him that we had a problem with a statistic, for the reasons that I just told the Chair, “Well, you could have said absolutely nothing and no one would be any the wiser.” That is not where I come from, I’m afraid—it is not where I come from. So I said many, many times, “Our No. 1 priority is quality statistics.” And I am proud of the way in which we have improved, in many areas, the quality. That has been at the price of some of the reputation, because it has come across as errors. I am very proud of what the organisation did during the pandemic. When I joined, I did not expect that we would have a pandemic—that came as a surprise. I was also very clear in my mind that there was going to be a census, and I wanted that to be really good—and it was. I am very proud of that. That was not down to me but a huge team of people, every one of whom deserves an enormous pat on the back. If you look at what has happened in censuses in other countries, the census that was undertaken in England and Wales had a very high response. I came to improve the quality of statistics. I also tried to de-factionalise the ONS a little because I felt it was siloed. Did I succeed? I tried; maybe I failed. I had a mantra that I used a lot, which was “One ONS”, and people did start working together better which is important.

The Committee heard criticism that you focused funds and resources on transformation programmes—some of which were described as a bit unrealistic—at the expense of the UK’s core economic data. How do you respond?

I have heard that, and I largely disagree. When I was there, the ONS had a funding base that came from the Treasury, and which was in many ways hypothecated. There was some money to do everything, but then there was a big project that we were asked to do on administrative data and another on the future of population data. The ONS was asked to do that in 2013, but it was to be delivered during my tenure. Then there was a programme on economic statistics, which included the R&D survey that I mentioned previously, as well as a transformation of inflation calculations away from using people going into supermarkets with clipboards to using much more automated data, for example, scanner data from supermarkets or automated rail price data. When you take that hypothecated money out, you have a relatively small amount left. You could not prioritise. I would have liked to have taken a large chunk of money from some of those programmes to prioritise other things, but that was not possible with the way the funding in the spending review came about. What we did was go back to the Treasury when there was clearly a problem with response rates and say, “Can we have some more money, because globally response rates are not great.” To its credit, the Treasury did come with some extra money, and we were then able to put that into more interviewers and greater knocking on doors. My reading is that response rates in some of those surveys, while not yet great, are improving. I know people will say, “Well, you should have done something else.” At the same time, in one financial year we had to take cuts in funding because of the third year of the spending review, pay an allocated for pay rise—which I was very pleased to do—and pay all colleagues below senior civil service a £1,500 bonus. When you add all those together, it makes things very tight. You could not just say, “Well, we will not do something else over here.” It comes down to money and allocation. People say, “You didn’t allocate enough money to economic statistics,” but much economic statistics depends on those surveys and a lot of extra money is allocated to surveys. The other thing that is important for this year, 2025-26, is that although we allocated the same money to the economic statistics team—and I have been told many times that that was a real cut; it is a real cut—at the same time less needed to be done and so there was money to reallocate. I think that has been missed. The criticism that I went chasing after shiny new things rather than, if you like, fixing the foundations I personally do not accept because I have been fighting since I got there to fix the foundations, particularly by pushing to undertake a real restructure of the coding that we use, as I described to the Chair in response to his first question. Do I believe that, put pretty simply, there need to be greater skills in coding in the organisation? Yes, massively. Do I believe that for a national statistics institute trying to deliver what it is trying to deliver, it is underfunded? Yes, 100%. I completely accept, and I have been quite open that I regret and I am very sorry about, the issues around the labour force, but this is driving an enormous economy and the amounts of money are really quite small. We need and should have a really strong field force. We should have the ability to get the right people into post and I do not think that the funding allows that to happen.

Do you think that the Government were asking you to do too much with the resources that they allocated to the ONS?

Yes, no question. Just to be clear, I think there was enough money in each of those hypothecated pots that I described to you at the beginning of this conversation. I do not think that there is enough money in, if you like, the remainder to be able to do everything properly. There is a philosophical argument here, I suppose. There is what the Government are asking us to do, but that comes down to asking who the Government are. The Treasury is doing the funding, and on many occasions the Treasury said, “All we want is economic statistics.” Actually, if you are in the Home Office, you want really, really good crime statistics. I believe those statistics need to be done, that survey needs to be transformed, there needs to be much better data on violence against women and girls. Much needs to be done. We were being asked by the Department for Work and Pensions to do extra work. The Department of Health was asking us to do more. You have to be able to say no, and we were able to say no—do not get me wrong—but we are a national statistics institute and in my opinion that is different from an economic statistics institute. Although there is much discussion about getting economic statistics right, which is incredibly important, I think that a national statistics institute should be doing more. It should not do everything, as there are an unlimited number of things that people would like to get statistics on, but it should be able to do more and in order to do that it needs to be properly funded and not tied down all the time, so that it does less.

Would you say, then, that the Government constrained your ability to prioritise the things that you wanted to focus on?

The Government constrained the funding, and certainly that hypothecation was not helpful at the time—no question. Just to be clear, the hypothecation has been, for this year, removed to a large extent. If you want a really strong, broad range, I would argue that a national statistics institute should be doing social as well as economic statistics. The two are, in my opinion, equally important, so then you need to have the funding to do that.

How, and to whom, did you make representations regarding funding?

There are very many meetings. The Treasury has a team that works with our finance director. I certainly met with them at times, but they are basically who one speaks to about funding. There are other ways, though. For example, there was a real need identified very early in this decade for much better small area statistics, and I think—

I—

Can I just explain?

No, let me recast my question. The finance director talks to the Treasury about financial issues. Did you, the finance director or the broader senior management of the ONS, raise the issue with this Committee or with Parliament?

I am not clear that we raised it with this Committee, although it is fantastic having this conversation—it is a really good one. I am not clear if we raised the issue.

You didn’t?

I didn’t. The point I wanted to make, if I may finish very quickly—

No, Sir Ian. Thank you. That answers that question, which is helpful. You have set out, in answer to my initial questions and to Miss Edwards, that you came into post, identified and assessed deficiencies, and started to craft a roadmap to move the organisation to a more “contemporary needs meeting place”—my description, not yours. Again, how, and to whom, did you set out that assessment, diagnosis and prescription?

The governance is pretty clear; you have a board, it agrees a strategy and what goes on at the same time. Under the statistics Acts, we are rightly civil servants and therefore also report to the Cabinet Office. I had periodic meetings—sometimes with the chair, sometimes without—with the Minister for the Cabinet Office, or with the civil servant in the Cabinet Office charged with working with the ONS.

Could you define “periodic”?

It depended. I would like to say quarterly, but sometimes it was more than that. Certainly, for example, since the last election, I only had one meeting that I am aware of with the Minister, but much more regular meetings with the civil servant in charge. For the last couple of years, for example, there were two directors—the director for data science and the director for statistics in the Cabinet Office—who took a view on the ONS, and I spoke with them often on a fortnightly basis.

Would you have taken senior personnel at the Bank of England into your confidence to say, “Look, there are going to be some bumps in the road, from what I have identified. We have a plan to resolve and rectify. Bear with us; this is the timeline expectation”?

I do think it is incredibly important to say that the work on the transformation of the labour force survey, and addressing the problems that were had, has been done in very close collaboration with the Bank of England and the Treasury and with external academics—the point being that we are saying that we have a problem here and it would be great if it were a problem that we could all own. One of my former directors would meet with them on a very regular basis. There was a committee that I chaired, which only met once, that involved the chief economist of the Bank of England, the chief economist of the Treasury and the Office for Budget Responsibility. It was just to say, “Here we are, these are the problems, can we make sure that we are all owning them and that we are all agreed that we are going forward together?”

Let me now take you to my confusion. I am sure that it is my fault, for which I apologise in advance.

I’m sure it’s mine.

You set out to us this morning a calm rationale of your assessment of baseline operations when you arrived, as well as your assessment of where things were going well and where things were going wrong in the gathering of and interpretation of statistics. You had identified some issues with regards to the people management of the organisation and had sought to address those. Proper representations were made to Treasury opposite numbers on funding and the challenge, and key stakeholders such as the chief economist of the Bank of England and others were sighted on your analysis and the road map to reform and change. So far, so good. My confusion is why suddenly, just a few short months ago, the whole thing implodes, politically and reputationally. The Bank of England appears, in our understanding, to be effectively setting up or trying to set up its own in-house shadow ONS, such is its loss of confidence in the Office for National Statistics and the reliability of the data it provided. Everybody—stakeholders, paymasters and the teams—is brought along on the grand Sir Ian Diamond vision for ONS today, ONS tomorrow and how we get there; I am not putting words in your mouth or being facetious when I say that. Nobody rings an alarm bell, apparently. Everybody presumes that things are just moving along at pace and in a reliable way, and then suddenly the proverbial hits the fan. My confusion is why? What is your assessment of the trigger or the straw that broke the camel’s back?

The initial trigger, frankly, was not producing labour force statistics in one month. It is deeply embarrassing and I have not hidden from that at any time in this conversation. I also made clear—well, probably not very clear, but I tried —the problems involved in turning that particular tanker around. Because of the need for analysis, you have a period of time when the statistics are not as good as you would want. While you are putting out monthly labour force statistics—

We have all heard that and we understand it. So you believe it was just the absence of that data?

I think that is the major catalyst. That, alongside a number of other smaller issues—not changing economic history, not changing anything—they also did not help. The revisions to migration data in November 2024 also did not help. So, while I would say that the labour force statistics were the initial major catalyst—

It brought to the fore other things.

—there were other things going along. I feel that some of them were blown out a little bit, but I am not hiding from any of them, because I think it is much better to be absolutely open and say, “Look, we’ve got this problem; let’s solve it.” Some of those issues—again I don’t want to go on at great length—were other parts of Government giving us data, which they then had to say, “I’m really sorry.”

The cars issue, wasn’t it, from the Department for Transport?

That was one, but let me be clear: I am not trying to pass the buck.

No, no, I get that. Notwithstanding that, let me ask you this question before I bring in Mr Lamont. In the still watch of the night, do you ever just feel that you have been scapegoated on all of this?

I don’t sleep a lot at night—I haven’t slept a lot for a long time, I tell you that. Do I believe that all on my own I have caused the, if you like, demise of the reputation of the ONS? No, I don’t; let me be absolutely clear about that. I am owning it; I am not in any way trying to hide from it.

I understand that.

But is it just down to me? No.

Is there anybody else’s collar we should be feeling?

Look, I think there is a much broader spectrum here, as I have been trying to say, of many stakeholders, and over a very long period of time. Some of those errors that we have been finding could have been there for 10 years, and we are now finding them. They haven’t changed economic history, but at the same time I would much rather look you in the eye and say, “Here is a problem; we have found it and it has been sorted,” than brush it under the carpet. That is not going to happen on my watch.

Thank you.

Before I ask about the integrated data service, I want to come back to the previous point about funding. You said that funding had been an issue previously. In the discussions you had with the board and your ministerial masters in the Cabinet Office, were you raising concerns about funding?

Being careful about what I am saying, the board were always aware that we were having to find money—well over £20 million extra in one year, just like that, to break even.

But did you ask Government? Did you ask Ministers for more money?

On the surveys, for example, as soon as we knew, we went to Government and said, “Look, we need more money for surveys to get the response rates up, because we are going to need to run these surveys for longer,” and the Government gave us that money. They didn’t give us the money previously, but that they certainly did, and—

So you were able to get extra money in response to your call for help.

Yes, 100%.

Correct. So funding then wasn’t an issue.

It was.

Sorry; I am not clear. You earlier said that funding was part of the problem. I have now asked, did you request more funding? You were then given more funding, so did that resolve the questions?

Initially no, but when we needed it—when we said, “We can’t do this without more money”—more money was forthcoming.

Right. So funding wasn’t an issue after that.

Well, in response to Lauren Edwards’ question, very clearly, we still haven’t got enough money to be a really, truly wide-ranging national statistics institution.

That is my point. Were you making that point to Ministers?

We were making the point, but we were being told—for example, I was explicitly told in April this year, although I am paraphrasing—“If anyone is asking you to do anything other than economic statistics, come back and see us again.” We were being told to, if you like, rein in and only produce economic statistics.

Told by whom?

The Treasury.

So they were trying to restrict requests?

Yes. Their view was, “Do not try to do too much.”

Ministers or officials?

Officials. I never met a Minister.

At ministerial instruction, or at their own behest?

I couldn’t say, because I—

Did you not enquire?

No, I did not.

Did nobody from the ONS ask from whom the instruction was coming?

I couldn’t tell you whether anyone asked. There were regular conversations between my economic statistics colleagues and Treasury colleagues; very regular conversations. But I could not tell you whether that was a ministerial decision.

Were these discussions all informal or would there be a formal note of these discussions?

Mostly informal conversations, but I would imagine, and hope, that there was a formal note of all the meetings between the finance director and the Treasury spending team.

You would hope?

I would hope.

It was not normal practice to record and note something about—?

I was not at those meetings, so I couldn’t say.

Okay. Can you explain to the Committee what the integrated data service was, and whose idea it was?

Sure. I couldn’t tell you exactly—

And in the interests of time, could I add “briefly”?

I can give you 50 minutes on this without even trying. Without hesitation or repetition— [Laughter.]

Or deviation, yes, exactly.

To go back to what I think was the first authoritative piece—the 2013 report by Professor Sir Alan Langlands on using administrative data to improve our understanding of society and the economy—many people, including myself, have believed for a long time that if you can take the large amount of administrative data that exists, bring them together, link them together, you can answer the sorts of questions that Ministers, Governments, citizens want answers to, and you can do so effectively, authoritatively and relatively quickly. An example would be the wonderful work that has been done in education, where schools outcomes data at the individual level can be linked to subsequent employment data so that you can look at the payoff of different types of education. I think that is brilliant. The view that came from a number of people post 2020 was that we needed to turbocharge this. Already the ONS was running a smallish project called the secure research service, widely used by economic statisticians, which enabled people to access administrative data to do analysis. Given that existed, it was suggested that there should be a real turbocharge of this: we should get together all Government data and link them, and the sorts of work that could be done would be mind-blowing, wonderful. We had some conversations with people in the Cabinet Office at the time about whether we would take this on. We agreed to take it on. The idea, as I said, was to get very large amounts of data, link them together and enable great work to happen. I do believe that this project has delivered a good product. Now—

Has it been a success, do you think?

I am not going to say it is a success. I will say that it works, and the advice I have been given by the two directors general who work in that area, Pete Benton and Nigel Green, is that it is good. Indeed, there is an excellent piece of work that links surgery to subsequent employment outcomes. In other words, it looks at health and employment data, and looks at the pay-off from various types of surgery. It shows that employment goes down, then you have that surgery and employment goes up. Given the current discussions around getting people who are long-term sick back to work, I think that kind of work is incredibly important. Has it been a success? I would not say that it has been a success, for the simple reason that, first, it was very difficult to get access to as much data as one wanted, and, secondly, it has not been used as much as I would have liked. That could partly be an ONS problem, in getting it out there, and partly be that people were so busy that others did not come to use it. There is actually a good piece of work by Richard Price, a former Ministry of Justice director general, arguing that, rather than the integrated data service—the “build it and they will come” approach, if you like—you are much better going small and then upwards. Justice has a programme called BOLD, which stands for better outcomes through linked data. In other words, I do not think our vision was to have this very big project with everybody using it—and it is a Government project, not an ONS project—but I do not think it has got as far as I would have liked.

Other witnesses identified the issue with data access. Are you able to tell us which Departments, or individuals within Government, were blocking access to data, or not sharing data in the way that you envisaged?

I think it would be quite inappropriate to give you a rogues’ gallery—

Why? Tell us.

We like a rogues’ gallery.

We like honesty and transparency in this Committee. Professor Sir Ian Diamond—but there has been a lot of effort on data sharing, and it has not been successful.

But you are not going to tell us who has blocked it?

Well, I am going to come to that. I just want to say that I do think it is quite successful in Scotland, in Wales and in Northern Ireland. Part of the answer there is that they are smaller jurisdictions where people know everybody and where the data are part of that in the same way.

How can we make this project in the UK—or here in Westminster—better if you are not prepared to identify the blocks in the process, in terms of the people who are not sharing data in the way that we want them to?

First, there is an inherent conservatism. If I am the data owner of a particular dataset in Department X, what is in it for me to let somebody else have those data when, if something goes wrong, I am going to get blamed? That inherent conservatism is a real issue. The second problem is that different Departments have different systems, so people say, “Well, it will take an awful lot of effort for me to get my data into a system that you can use.” Even when the Cabinet Office said, “Well, we’ll send a team in to help,” that did not work as well as it should have done, in my opinion. Thirdly, people are very busy getting the things out that they have to with those data for their own purposes. Then you come along and say, “Actually, I need three months of your time to get these data flowing,” so people say, “Oh, I haven’t got time to do that.” All in all, I think there are many reasons, so there is not any one person, or any one place, where things are slow. For example, I had a meeting to talk about the future of population statistics with all permanent secretaries, and everybody was saying, “This is great!” So I do think the will is there, particularly at the high level, but—and I am not hiding ONS away from this—it has not worked, as yet. I would really urge this Committee to continue to press on this, because I think the benefits of improved data sharing, for all of our fellow citizens, are immense.

It would help us greatly if you told us which Departments were not complying, but anyway. Was Nigel Green appointed as director general to try to resolve some of these challenges?

Yes, that came about when I met with the then Cabinet Office Minister, Baroness Neville-Rolfe, together with the two directors—one was the director of data science and the other was the director of statistics in the Cabinet Office. Their view, which I fully shared at that time, was that the project was not progressing at the speed at which we wanted it to. Their suggestion was that Nigel Green would be able to help, and he was someone whom they really respected. At that time, it was clear in my mind that the project needed a bit more of a push, so I was happy to bring in Nigel Green to start off by saying, “Where are we? What needs to be done? Could it work?” He turned out to be an absolutely excellent colleague.

Was the funding for his post part of the additional resource that came from the Cabinet Office?

It was not additional resource from the Cabinet Office; it was funded from within the hypothecated money for the integrated data service.

So it was within your funding.

Sam, do you want to come in?

Lovely to see you again; we met at the Treasury Committee. I was not instructed because, as I said quite clearly, I was totally agreed that we needed to bring someone in. It was someone who came in as a consultant to help us, and he worked a couple of days a week. He gave us some really excellent advice and worked very closely, following the 2024 election, with the new Department of Science, Innovation and Technology to make sure that what we were doing linked in with what they wanted and were trying to do.

I am going to be very careful what I say. Alison was appointed in 2020—

By?

By a committee that included a commissioner, myself and an external. She was appointed as director general for data capability. Her portfolio was basically all the IT, and she also initially had surveys and data science. With the integrated data service, there is only so much one person can do, so we moved one or two things around. She was an excellent colleague. Sadly, at the beginning of ‘24—maybe late ‘23—she has had to have a period of very long-term sick leave. While she is still there, we are still working through that sick leave. I could not tell you anything more.

Before I bring in Mr Quigley, I want to come back to something that I have been trying to compute—some of the threads of the answers that you have given to me and to colleagues. The ONS has perfectly properly championed its independence, its ability to speak truth unto power, and the fact that the statistics it produces cannot be manipulated by Ministers who want statistics to say, “Black is white”, or the other way around. In response to Mr Lamont’s questions, you have spoken with a great deal of experience and clear commitment about the merits to society, and therefore the better administration of governance in this country, of maximising data sharing, collaboration between Departments, harvesting as much information as possible and distilling it to create as meaningful a picture as one can. That is all to the good and all to be applauded. I want to take us back to an answer that you gave a moment or so ago, in regard to the anonymous instruction from Treasury, communicated by an official to your finance director, you believe, at a meeting—which was either an official or an unofficial meeting, and which may or may not have a minute, and we will look into that in due course—where, in essence, you were told, “The Treasury just wants you to concentrate on economic data gathering and nothing else.”

Let me just be clear: that was an informal conversation. I would be surprised if that was not said elsewhere.

Okay. Let me ask you this question. Given that that approach would fundamentally undermine the bigger picture of the merits of data sharing and gathering to amplify statistical assessment, and that it would be a clear instruction in terms of how you would gather data—so the operation of the organisation—the fact that no alarm bell was rung by ONS to this Committee or to Parliament, and the fact that nobody inquired as to whether or not it was a direct ministerial instruction, passed through a senior official, does suggest a lack of inquisitiveness, does it not?

I wouldn’t say that.

You would or you wouldn’t?

I wouldn’t say that.

In which case, why wouldn’t you? You are conscious and very persuasive of the merits of the broadest approach to data harvesting and sharing, which requires more than just economic data. Somebody from the Treasury is, in essence, strangling this initiative at birth by saying, “All the Treasury wants is economic data.” However, nobody in the organisation saw fit to ask, “Is this coming directly from the Chancellor, the Chief Secretary or the permanent secretary to the Treasury?” It was just, “Okay!”

I would have done, had I still been in the organisation, but the conversation happened pretty soon before I left. I did feel, and I do feel, that there is a real need. I am not going to hide from what I have told you, but there needs to be a much broader spectrum. But equally, very clearly, the purse strings are held in one place, and that is no different from other Government Departments. But I would have raised it—

I will go back to my scapegoat question. Do you think you might have been eased out? The presumption was that with this new Treasury-led directive, your magnum opus, as it were, was not going to come to fruition or progress.

I left for a reason—a prime one. You clearly have experience of surgery. You are very welcome to have the pain that I have been through over the past few months, and the only way I have been able to manage it is by not working.

Yes, but without necessarily going into the weeds of all of that, it is a very unusual position for somebody who requires orthopaedic surgery to resign. I was a Member of Parliament when I had both my hips replaced. We are now going into very personal detail.

You are a tougher person than I am.

No, but it is very unusual for somebody requiring what is actually basic orthopaedic carpentry to resign from a position merely to have the job done. As you already answered in a question to Mr Carling, one of your colleagues is on very long-term sick leave.

But I was not prepared to do that. I think there comes a time, when you are 71 years old, when you have taken a view that you have given this your best shot and now is the time to move on.

In September 2020, Sam Beckett was appointed as the second permanent secretary to the ONS and was made responsible for economic statistics. Can you tell us how and why that appointment came about?

Sure. We were approached—I was approached—by the then permanent secretary of the Cabinet Office, who said that they wondered if we might like an extra post because they were looking for a role for Sam Beckett, who they held in very high regard. I met with the board and we had a very long conversation about whether we should accept that invitation, which was initially for a chief operating officer. That was how it was posed to me. On balance, we thought it was a good move. Sam Beckett was then interviewed by the first civil service commissioner and by the permanent secretary of the Treasury, and joined us. She was very clear that she wanted to do more. She was a fabulous colleague. She took over economic statistics.

Did you and the board not question what problem somebody outside the organisation was trying to solve when they came to you? You have expressed at some length your frustration with the lack of resource and funding. If somebody came to my organisation and said, “We’re not going to fund you for the things you want, but we want you to take this person”—did anyone question what problem you thought they were trying to solve?

Not really. The argument at that time, which I think is a good one, was that we had just taken over—well, we had not taken over anything. We invented the covid infection survey, which was an enormous undertaking and was hugely successful, I believe. The suggestion was, “Would it be helpful?”, given that we had the covid infection survey and a census coming up.

I do not blame you for taking the help, but if somebody offers you a chief operating officer without you asking for it, they are actually saying that somebody is not in charge of the operation. How did you divide your responsibilities between yourself and Ms Beckett?

Sam—who is great, by the way—wanted to lead economic statistics. That is entirely appropriate. So Sam led economic statistics. She also took on the leadership of things like communications and surveys—after a while; there were new surveys in after a while—and security. In many ways, she took on the operations role. If you are going to say, “Where is the algorithm that said what she did?”, there isn’t an algorithm. Sam and I met very regularly. We got on very well. We worked out who was going to do what together.

Let’s move on to your relationship with the Cabinet Office. How would you define that relationship? How interested were they in what you were doing? Did they attempt to hold you to account in any way? When you run a business, you walk in in the morning and say, “This is the cash position. What’s happening on operations? What’s our stock level?” I know you are not producing widgets, but did you have that sort of relationship with them?

That is a really interesting point. There is a point I would like to make at some stage. The point that I would start with is that we are civil servants. I thought the Chair was going that way when he talked about independence. To me, independence is not irrelevance. The independence comes from the way in which one calculates the statistics, and then the statistic is the statistic. It comes back to what you said, Chair. The relationship with the Cabinet Office is incredibly important, because as civil servants we report into the Cabinet Office. I found that the Cabinet Office was always extremely interested in what we were up to. It certainly wanted to keep a handle on progress on various issues. Minister Neville-Rolfe, for example, came down with briefings and gave some pretty insightful work—as did this Committee, I have to say, and I do hope that this Committee takes the time to come down. I felt that we were being held to account, but not with as tight a rein as you would expect from the board. The point that I wanted to make—I have suggested this in the past, but I still believe it—is that I believe strongly that there should be a Cabinet Office observer on the board. It works elsewhere. In a previous life, I was chief executive of the Economic and Social Research Council, and at that time we reported into various places; for a while it was DTI, then BIS and then BEIS. There was always a person from the parent Department as an observer on the board. That seems to me to be entirely proper and right, and it enables some joined-upness. Otherwise, you can end up with, “Who is talking to who? How is everyone getting the same story?” If everyone has the same story, you are being held to account by the right people. That is not to say that you cannot go back, but if for example the director is an observer, and the director is not happy, the director can go back and say, “Minister, I am not happy. You need to get in the chair and the National Statistician now.” That seems to me to be entirely right.

Without putting words in your mouth, in terms of the Cabinet Office UKSA sponsorship team, would you say that that relationship was not particularly effective because you did not have that board oversight? I am starting to struggle here; you are providing essential information to Government, but at the moment I cannot see where anyone in Government is saying to you, “That is not the right information.”

Well, that is where the board has a role, and we can maybe pick up on the way the board works at some stage in this conversation.

I am quite fascinated by exactly what the board was supposed to be doing, having taken a lot of evidence as to what it could not do or should not do.

I felt that I reported to the board, but equally there is another reporting line, which is to the Cabinet Office. We had very good relationships, and they were challenging, with the Cabinet Office—no question. We met them regularly, but at the same time, they were not, if you like, at that other conversation with the board. I am not saying, in any way, that I did not feel I was being held to account. I am saying that I felt it could be much more smooth and joined up if I was held to account through them being at that meeting as well.

As you describe that situation, do you think that, had that been the case, it would have headed off the issues we are discussing today?

I think it would be helpful.

You have outlined that there is some friction between Departments in terms of what they expect from the ONS. Certainly, the Treasury’s idea might be slightly different from what you have outlined. If we were to ask the Treasury to describe their relationship with the ONS, how do you think they would describe it?

Well, I hope—I stress “I hope”—that they would say they had a very good relationship, because Treasury officials meet on a very regular basis with the economic statistics team, both formally and informally. The economic statistics team also meet regularly, I understand, with the spending team, and when they were also due to discuss population, the head of population joined that conversation. So I would hope that the Treasury were kept very well informed. I would also say that, for example around labour force issues, the Treasury were absolutely central to that, if you like, triple group of people—the OBR, the Bank and the Treasury—as stakeholders working with them. That is not to say that they all agreed all the time, but they were there and having those conversations—incredibly important conversations.

With the wonder of hindsight, if you had a magic wand to let you go back and correct one thing to do with governance and accountability, what one thing would you change?

I think, personally, I would—obviously, not just alone, but with the director general and so on—have regular meetings with Ministers and officials in the Bank and elsewhere. Also, in terms of governance, I think there are important issues with the board. May I expand on that?

We will come on to the board shortly.

Okay. I just did not want to get through this conversation without that.

Don’t worry—we will cover all bases. Before I bring in Charlotte Cane, may I ask this question? Recruitment to posts within the organisation seems to be the old boys’ network writ large. A Minister or an official says, “I know so and so; they are frightfully good. They could be very helpful to you. Why don’t you take them on?” That might be for a post you had not even been thinking about, which is not advertised, so one does not really know if one is getting the cream of the crop to do it. Are the recruitment processes of the ONS fit for purpose for 2025?

I very much hope that they are absolutely impeccable. That is what I would want them to be, and I certainly would not have had anything to do with anything that was not entirely within the civil service rules.

I am going to ask mainly about the culture of the organisation. We have heard from other witnesses, and it came out of Sir Robert Devereux’s review, that there is a culture where open discussions regarding challenges are not welcomed. Do you recognise that description?

I have heard it. Personally, I welcome robust debate. In answer to the Chair’s earlier question, one of the things I regret is that, perhaps, I have been a bit robust at times, because that is the way, in my opinion, academic discussions can go forward and we get to the right discussion. I welcome that, and I have said so many times in the organisation: “Speak up. You have to be prepared to speak up.” But I recognise that that is something that many people have said. I recognise as well that, therefore, even more needs to have been done. I am clear that I would have continued to push as much as I possibly personally could for people to feel able to speak up. I can certainly report examples of when somebody has spoken up in a meeting and I have taken it forward. So I personally welcome that. I do think that there is sometimes a hierarchy, where people might be a bit nervous, and that perhaps, as an organisation, not enough was done to encourage people to speak up.

In your answer to the Chair’s question at the beginning, you talked quite a bit about having open meetings, but you talked about those meetings being quite large—in one case, you said it could have been 1,000 people. You also talked about trying to get your managers to cascade down. How well do you feel that that worked?

The one that was 1,000 people was online. It was not always 1,000, but it could get to that number. I just felt that it was super-important to enable all those people. All those colleagues at grade 6, to me, are incredibly important leaders of the organisation, as were those in the senior civil service. Let us make sure that if I am going to give a message, everybody gets the same message, because then there is no question about cascading. I think things got better, but I do not feel that, initially, cascading was done as well as it should have been, because, as I said, I spent a lot of time talking to people, and I did not feel that people knew what I expected them to know. Maybe that is my fault for not having been clear in the first place. Again, I am not trying to hand this over—I own the problem—but I certainly do not think the message was getting through. We held monthly meetings for a very long time, when people could ask me any question they wanted, and they did. They were robust, I can tell you. I felt that it was important that anyone felt comfortable to ask me any question. Those questions that were being asked initially were always anonymous. There was not an issue with people feeling, “Hmm,” and I feel that that was the right thing to do. I am not hiding from the fact that I have heard the point about people not feeling comfortable about speaking up. I do not know why, personally, that is. I do accept that I would have been robust in conversations sometimes.

You have talked about systems for getting your message out and for cascading messages down—systems that may not, in themselves, have worked as well as you liked. What systems were there in place, other than the questions, for cascading things back up again?

I did often say, “Look, if you don’t think you’re getting listened to, come to me.” Back in about 2021, a few people did. My private office at that time did a really good job at mediating and sorting things out, but I could not answer the question, “If people were not raising things, where were they stopping, and were people taking the decision that they did not want to share them at the top level?” Certainly, people did want to share things at the top level, and they were quite clear.

You said that quite a few people came to you in 2021.

A few.

Did that continue? Did people continue coming to you? Do you have any idea why that might be?

I don’t know. Maybe, because I was serving for the first time, people felt—I don’t know. I have no idea.

This Committee is clearly of the view that the HR function was not working particularly well. You have said that colleagues of yours raised concerns with you about HR and how things were operating. You were aware of the problems, and yet you effectively promoted the leader of the HR operation to be a director of operations, rather than focusing on HR. Why did you do that?

At that time, I was not aware of the extent, as perhaps I was subsequently, of the, if you like, dissatisfaction with HR. Certainly, I have to say very clearly that I worked with a number of deputies, and I thought they were all extremely professional and extremely high class. We have talked with Mr Quigley about the appointment of Sam Beckett. In 2023, when Sam Beckett moved, it was at a time, as I have already pointed out to you, of really difficult financial times in the organisation. I thought it was wrong. My ideal would have been to have had a new director general or a new CFO, but I felt that that was the wrong thing to do at that time, given that there were cutbacks right across the organisation. One wants to be able to look people in the eye and say there are cutbacks everywhere. I was clear in my mind. We had one director general who was off pregnant and another who was on long-term sick leave. We had another director general who was being asked to take on more responsibilities on economic statistics. We had another director general who was relatively new—they were brilliant, but they had an awfully big portfolio. So I said, “Well, I will take it on.” That is appropriate because I think I should, if it is with another permanent secretary, be meeting regularly with the director of communications, the director of security and the director of business surveys. Those were the three. It was argued to me that we should do that, but that many of the day-to-day issues, often around personnel, would be best handled by going through the director of HR. So it was suggested that we should try, for a short period of time, the operations issue being managed by the director of HR in a kind of expanded role, while at the same time I met regularly with them to deal with strategic issues. We reviewed that after three months. We reviewed it again after six months, and it was going pretty well. After six months, that was at the height of when the labour force issues had come up. I took a judgment. I was advised that it was entirely within civil service regulations, so there was absolutely no issue. I took a view at that time that stability was better than anything else, because we were going flat out, and I felt that there was support being given to the surveys team in the recruitment of new people. I am conscious that you are hearing negative things about HR, I am conscious that I heard those increasingly as time went on and I am conscious that, therefore, one needs to look very carefully at that function to see what are the issues.

When did colleagues first raise concerns with you about HR?

The first time? Well, in every organisation I have ever been, you will find people who say, “I am not keen on HR.” The real issues, I think, came in late 2024, into the beginning of this year.

So not until late 2024?

Yes. I mean, there are always issues—people will say, “I didn’t get treated properly,” or something like that. There were, I think, two catalysts. I do feel one was completely unjustifiable. One catalyst was when we were required to attend the office three days a week, and I felt that the HR team put an enormous amount of effort into meeting individually with colleagues to look at their personal circumstances and work out the best solution. In that instance, I felt they got, quite frankly, with the organisation, an unfair press. The other catalyst I would point to is, with any national statistics institute, in the period after a census, you will have a lot of staff leaving. You build your staff up, and then it goes down. That was exacerbated at the ONS by the fact that we had also undertaken the covid infection survey, which you have not asked about, so there was a period of time in ‘24, in particular, when large numbers of people were rolling off their contracts. I thought that an enormously successful job was done around redeployment, but again the HR function got, I felt, a lot of negative press. I think perhaps we were not as good in the communications of that redeployment process. We never went to redundancy, for example. We simply managed to redeploy people in different ways. Certainly, there were some voluntary—a lot of people voluntarily leaving for different areas, and that is fine; we were managing that by moving people around. I feel that the HR function did not get the greatest press. I actually think it managed that well, but I am pretty conscious that it was in that window of ‘24 that I felt increasing negativity.

You seem to be saying that it was a function that was actually doing a pretty good job, but there were mitigating reasons why people felt unhappy with it. In the time that you were at the ONS, did you really feel that the staff culture and staff morale were good?

I have to say that, in the first two or three years that I was there, the people survey results were at a level they had never been before—positive, I should say—and that even in the last year, when I expected them to go down, given the issues around hybrid working, they stayed relatively high. I am not, for a moment, going to sit here and say that everybody thought that things were brilliant. I am saying that a very large number of people at the organisation massively enjoyed working there and were very dedicated. I feel privileged to have called them colleagues. I feel that there are always issues and—to raise something that you have raised—a rethink through how messages scale up is important. I am not trying to tell you that everything in the garden was rosy, but when I was there I felt that the culture was good. The evidence base from the people survey and other surveys was pretty positive.

Did none of your direct reports report any concerns or issues about staff morale, either that they had or that had been raised with them by their direct reports?

No. Not about staff morale. There were clear concerns about budget allocations and issues like that. We had conversations about how we could improve morale, about speaking up and about psychological security, so we were not hiding from any of those things. But, certainly, people were not saying to me, “Ian, we’ve got a real culture problem,” or anything like that.

You were trying to deliver quite a lot of change in the organisation, and you were under quite a lot of financial pressure. How much messaging did you get back about how the change was being delivered, how people felt part of the process, and how well they thought it was going?

I think there were times, when I was initially there, when people felt that in some areas we were overwhelmed. One of the reasons for that was that when I arrived, the same people were trying to deliver readiness statistics and to deliver the transformation. Certainly, it was Sam Beckett who first said, “We’ve got to do something about this.” That is why we put together the programme—the hypothecating programme—that had the new projects, while maintaining the others. For example, we were putting out monthly inflation figures, while another team was building the new automated methods. Then over time, we were splicing them in. I think that was an improvement. Enabling that to happen removed a lot of the feeling of being overwhelmed that I think was definitely there. I am not going to try and say it is the hardest job in the world, but when you are producing monthly statistics, it is an unbelievable grind. People really do work incredibly hard and in an incredibly dedicated way. Sometimes it must feel that you have scarcely put out a set of wonderful inflation figures and then, “Oh my God, we have to start again on GDP.” I mean, I was told—

I am sorry, but in the interest of time, Mr Campbell-Savours, Mr Taylor and Mr Quigley want to say something, if you have finished, Charlotte?

I just want to pick up on the way that the director of operations post was appointed. You said it was within the civil service rules. Your chair seemed to be concerned about the way it was done, yet you did it anyway. Why was that?

The chair was written to by—I assume; they never told me who—a member of staff saying what is going on here. I asked the deputy director, who had given the advice on this, to write to the chair with chapter and verse on why everything that was done was within the civil service rules. That was done, and he never mentioned it to me again.

Okay, thank you.

I wonder if we can be explicit on something. Do you not recognise the criticisms in the Robert Devereux review at all?

I recognise—look, I think I'd say—

I am going to make an observation, if it is okay. It might help you to answer the question differently than you did previously. When we were talking about culture, you talked about some of your own interactions, but this is an organisation of over 4,000 people. This is a report about culture; this is about that wider thing. Yes, as a leader within an organisation, you have a role there. I want to see that wider analysis and how you helped enable it. Charlotte has tried to get that. Show us that you had a broader understanding of the culture of this organisation, so that we can really understand what your issue is with the Robert Devereux review.

Robert Devereux’s report—faithfully, I’m sure—reports the conversations he had. I am surprised that he chose not to triangulate them with the conversation with me, but that is for him. I recognise—I am not hiding from the fact—that people have said there are cultural issues. That is where I was just going with Charlotte Cane at the end when I said people are working incredibly hard. When you feel perhaps that things are not going very well, it must not be good for morale. I believe that I recognised the fact that there are issues that have been raised around culture and the environment that they are trying to address.

Your workforce and workplace priorities were listed as build a “brilliant place to work, creating an inclusive and collaborative working environment based on flexibility and trust.” There is nothing wrong with that at all, but you need a large HR buy-in to get anywhere near that in an organisation so big. Is the real problem that, post the excellent work you did in covid—I think everyone would agree it was an excellent response—the culture became “Anything is possible”? During covid, you were allowed to jettison a lot of your priorities, so when you reformed, you say “Anything is possible”, but you now have a broad portfolio you cannot keep hold of. You were trying to do a transformational programme based on “Anything is possible” when really the Government wanted “Let’s just have what we need.”

That is a really interesting point. I tried to say from my position, “You do what you can do with the resources you’ve got. Then let’s have a conversation about what’s not going to happen.” The thing that I said more than anything else is that our No.1 priority is quality. There is still a huge job to do. I know that in evidence to you, Emma Rourke also said that it is difficult sometimes to stop things in the ONS. That is a real issue, because I was very confident and comfortable about stopping things. Some things can be improved by being more efficient. For example, the team in environmental statistics did a wonderful job by becoming much more efficient in the way they did their programmes, and, with much less resource, produced the same statistics. That is fantastic. The opportunity to do that across the piece still exists, but it requires—it comes back to what I said to a colleague who is no longer with us—improved coding right across the office and improving the coding we have to make things more efficient. I do believe that can happen. I do not have a problem with “Anything is possible”, but “Anything is possible” cannot happen without it being subject to a budgetary constraint. What is possible? What do we do, subject to the constraints? For example, in January this year, colleagues in economic statistics said it will cost them about £30 million more than the budget will allow to do everything that they wanted to do. I said, “We don’t have that money, so let’s have a conversation about what will be lost.” The director general and director came to a meeting of the executive and said, “These are the things that won’t happen.” That was very sad, but those are things that had to stop, because you cannot do everything. What you do, you must do because—I have said it so many times; everyone must have heard me say it—quality is our No. 1 priority.

The Labyrinth report was, I think, helpful. I asked for an action plan, and I hoped that that action plan was being taken forward. On Lievesley, which was a really excellent and helpful review, there was a progress report that went—I would like to say every board meeting, but it was probably every other one—regularly to the board. It received a progress report on each of the recommendations in the Lievesley review, to be able to monitor progress against it.

Can I just ask you this? You are clearly committed to the quality and reliability of statistics.

Quality was my No. 1 priority.

Right—you are a statistician. The casual observer might say, “Sir Ian is very interested in that, but in terms of the operational mechanics of an organisation, he was less engaged or less inquisitive.” Maybe, to add to that perception—because one cannot play all instruments well; we all have strengths, and we all have things we are less interested in or less good at—you have just said, in answer to a question from Mr Carling, that the Labyrinth report was produced, it caused you concern, you asked for an action plan and you hoped it would be implemented. That does not suggest to me, with the greatest respect, any sort of hands-on energy, drive or “I want a weekly update.”

Anything but—when I say we have got an action plan, the question is how we are going to implement it. It comes back to your question, which implied that you cannot do everything yourself; you have to rely on people to deliver.

But you are the head of the organisation, and there is a report produced that adds to the canon of concerns—Mr Carling has referenced Lievesley and others. One accepts one cannot do everything—

But you try very much to give a strong message.

And one can give the message in the strongest and most robust of terms, but surely as the head of an organisation, the expectation would be of regular progress reports, the identification of inhibitors and so on. Did you expect, ask for and receive those, or was that a matter for the board?

I am not clear that we did get regular updates on the progress.

You are not clear?

I can’t remember.

Was it that you asked for them and they just did not give them to you?

To be honest, I really cannot remember. I mean, it is three or four years ago, and I was very clear that we needed to drive things through and that there was a—

Sir Ian, with the greatest respect, you have said in answer to several questions that the culture and morale of an organisation play a fundamental part in driving its outputs. The reports are published, you take them seriously, and then we are left with—perfectly sincerely, I am not questioning that—a rather lackadaisical, “Well, I just hoped things might happen.”

Okay, I apologise for the use of that one. I wanted things to happen, but I was relying on other people to deliver. But I must have said a thousand times, culture is not for me; culture is for everybody.

Yes, but not everyone is the head of the organisation.

No, I take that point. But that is why one of the key things with every organisation is to have a constant and clear message, and I tried to do that.

Yes, but a message is something one transmits; it is also what one receives. If one is not receiving regular progress reports on how the organisation is addressing, in this instance, the deficiencies identified in the Labyrinth work, that surely is a negation of responsibility.

What I have said, as I have said, is that Lievesley was followed up on a very regular basis.

Yes, but we were talking about Labyrinth.

Yes, I know; Labyrinth was earlier. I am just saying very clearly that I cannot remember how we monitored it.

Well, what we did was restructure. One of the problems that was being addressed was that people felt—this was said explicitly to me—where is the decision being made? I felt that the decision was being made at something called the investment committee, but we decided, for absolute clarity, to introduce an executive committee, which gave real clarity on where decisions were being made.

No, not as far as I am aware. It was an internal report that we then tried to act on. I am sure that it would have been raised periodically—perhaps not as often as it should have been, but periodically the director of HR would come to the board.

That is because it was before his time. Labyrinth was ’22.

Sorry, this is a report on culture, HR and so on. When the chairman of the board began is, frankly, an irrelevance. The question has to be, was Sir Robert’s predecessor sighted on it? The ancillary question to that is—because this is a work in progress—was Sir Robert Chote briefed on the Labyrinth report and its main findings, and how things were being addressed, as a work in progress?

I would have expected Sir Robert to have been briefed on that report as part of his introductory meetings with all directors-general and directors. I am surprised that he was not.

It was not discussed at the board, no.

It was not a conscious decision—anything but. I cannot remember whether it was in the report of the director of HR, who periodically did come—not as often as I think they should have done—to the board.

I don’t regret saying, “We’re moving forward.” I do regret not just saying, “Here it is for you to take it forward.”

I am going to ask a question that Mr Taylor was going to ask, as he has left to ask a Treasury question—he was not being disrespectful to the Committee. You have said in answer to a number of questions, “I would have hoped”, “I would have expected” and “I would have thought”. You have made the unarguable point that not one person can expect to do everything supremely well. My take, as I have said, is that Sir Ian Diamond is a very good statistician—tick—but a very good manager of an organisation? Cross. The Government is right, is it not, to recognise the huge skills challenge of meeting both requirements brilliantly, but is the Government right to have split the post of National Statistician?

It is very generous of you to give me the tick for the first one.

Nobody, Sir Ian, would question your interest in, devotion to and expertise in trying to drive statistical excellence.

I am not looking for an equal tick on the second one. Were we to have a much longer time, there are a number of mitigating factors that I might share. Do I think it is important to have the role as it was undertaken by the last couple of people? Yes, I do. The National Statistician should ideally be the permanent secretary. I also believe there should be a director general of operations. I also believe that there should be, at all times, a fully functioning senior leadership team, and I think there needs to be some thinking about what happens if you do, sadly, have long periods of sick leave for a number of people. I accept that the decision has been made. I think Robert Devereux says “at this time” and it may go back. My own view is that you should try to get a National Statistician who can do both—and I would not give myself a complete cross on the second one—but if you cannot get one, you need to do something of the sort that is proposed at the moment.

Let me be clear: my view has changed. My starting point on just about everything is, “Let’s not change things just for the sake of changing things; let’s make things work in the way that they are.” Having said that, personally, I think the structure of the board now is not ideal, because you have the regulator and the producer in the same place. I cannot see that it is a good idea that the chair sits on the regulation committee and chairs the board. What happens if the producers disagree with the regulator? It is clear that the regulator feels—and it is absolutely right that it should—that it can be unbelievably critical, but if you disagree with some of those criticisms and they have already been through the regulation committee a couple of times, you are on the back foot in trying to make your arguments. Where I think this whole point was exacerbated in the last few years was that there was a period of time when the view of the Cabinet Office, broadly defined—not the ONS team in the Cabinet Office, but the Cabinet Office in Government as a whole—was that non-executive directors should not be renewed. In other words, you have one period. I disagree with that profoundly. I do not think they should go on forever. My own view is that two periods of three years is, in governance terms, ideal. Some people would argue that, in exceptional circumstances, a third period of three years is fine. Fine. In other areas, some people say two periods of four. But you certainly need a period of time to get up to speed and then a period of time when you are really able to provide the scrutiny that is needed. Certainly when I joined the board—I was on the board before I became the National Statistician—I felt initially that there were people there then, who had been on the board for some time, who really had a grasp of things and who I was learning from. In the last couple of years, two or three really good—in my opinion—non-execs were stopped after one period and not renewed, simply because that was the policy. I do not think that is right. I think that had they been there, there could have been greater scrutiny. That is the first point. The second point is that, in order to have a fully functioning regulation committee, you ended up with a pretty tiny cohort of people who were not on the regulation committee. In summary, there are two key points. First, I think it is time really to think—and I hope that your Committee does so—very deeply and profoundly about whether it is time to split the two. I argued this to the chair, and the chair’s view was, “Oh, they wouldn’t pay for us to split it up.” Well, I think it is money worth spending, frankly. Secondly, you should really make sure, not that people automatically get two periods of, shall we say, three years, but that, on the basis that they are good members of the board, they are continued for a second.

I should be clear: had you asked me four years ago, I would have probably been in the same place as Ed Humpherson, but given my experience over the last couple of years, my view has changed.

The board has a role in being a central part of setting strategy. It also has a role in monitoring strategy and a role in the delivery of that strategy. The board also has a role in being a critical friend to the organisation. That is why I was really, really clear that I wanted the board to have absolute sight of everything. I did that in three ways—well, more than three. First, I met with the chair on a weekly basis and went through everything that I was thinking about. Secondly, I met with the deputy chair to make sure that was happening. Thirdly, I am not sure that the chair really thought it was a good idea, but I asked for a private meeting with the board ahead of the formal board meeting, so that I could discuss personnel issues and capacity and capability, for example, in a way that we could not in the meeting. Then the board received an up-to-date briefing on everything. In answer to your question, it is setting strategy, monitoring strategy and being a critical friend of the organisation.

The board in 2019 and 2020 was deeply involved with setting a strategy, which I think was a good one. The board was involved. You would have to ask them how they felt about the setting of the new strategy when that comes. There were meetings with them without the executive as well, as we discussed. So I would hope that the board would say that they were—well, I know that they were—

Yes—absolutely.

The board.

There was until, I think, this year. When I say this year, I mean ‘24. It was decided that we would not have one, because of the way that it fitted; it was two weeks on from the previous board meeting and there was only so much you could do.

That would be my expectation and that is certainly what I believe happened, or I wouldn’t have said it.

Modernising the website is something that I tried so hard with. I hope that it is going to get there pretty quickly, because, honestly, it does need doing.

They were aware that the numbers were going down, but they did not feel, as I did—let’s be honest. At that time, response rates were going down, and at that time, the policy on the labour force survey was to have two quarters of dual running, so we were trying very hard to get through that period and move on, so to speak. It was then, sadly, in August ‘23 that there was a problem, particularly with—well, the real problem was young men, men aged 18 to 24. That caused the reasons why—I am really embarrassed about this; I am not hiding from it—we had to delay. Alongside that were the conversations with the Bank of England, who said, “Actually, we want four quarters of dual running,” which pushed things further back. That is when, instead of just trying to keep things going with reduced resources, we looked back and said to the Treasury, “Look, we need money, and we need money quickly, to be able to get the response rates into a stable place,” and then Treasury gave us the money.

It has been going on for decades.

It did not collapse. It had been going down steadily. This was not just an ONS issue; it was an issue with all social surveys, so one of the things that I had done previously was talk to the other social surveys organisations—the three big ones. I also talked to the Economic and Social Research Council, and I persuaded it to put in a programme aimed at increasing survey response rates, which we and the national centre and what is now Verian and Ipsos MORI would all be prepared to contribute in kind to, as a real national effort. Let me be clear: we knew this was happening. We didn’t expect it to get to the situation where there was a crisis, which is what hit us in October ‘23.

I think it was probably a mix. Perhaps the board were aware, but, if you like, the team were not putting their hands up and saying, “We are lying awake at night worrying about this.” That may speak to a lack of survey—I would have a survey statistician, as I was, on the board at all times.

Good afternoon, Sir Ian. As a Scottish MP, I am particularly interested in the issue of harmonising the collection and production of data across the UK with the devolved Administrations. As a former principal of the great University of Aberdeen, you will have a particular perspective on this from your experience over that time. What was your experience of trying to harmonise statistics across the four Administrations of the UK?

Difficult. Let me say that I believe that our country has missed a real opportunity to analyse the national experiment that was devolution, and having different social policies in different areas. We have missed it, and the reason is that it was never a priority. I will come in a second to when it became a bit of a priority, but it had not been a priority. Of course, in different jurisdictions, you are producing statistics to measure the policies in those jurisdictions. Therefore, issues in Scotland would be measured against Scottish policies, and the same in Wales against Welsh policies. It is difficult, and I have to be honest: not every jurisdiction wants harmonised data initially, because you can then make real comparisons. Around 2022, 2023 and 2024, there was much more of an interest in harmonised data, led by the Union committee, which was chaired at that time by Secretary of State Gove, and there was a group led by now Baroness Gray who were very keen on moving harmonisation. That was great. Some of the thrust came, I think, from Welsh MPs. It costs money—let’s be absolutely honest—and the Lievesley review makes a really clear point that the funding of the statistical offices of Northern Ireland, Wales and Scotland is not great. We did manage to find some money, we prioritised some areas and we put a lot of work into some harmonised statistics. I think that work is great; it is not as good or as long term as I would like, but it is really important. My own view, to go back to where I started, is that we should be doing this much more to be able to study what is going on in different places, but it requires proper funding and commitment. We talked about that funding as a potential in this spending review, but I think the impetus for it had gone away. That was my view.

Were there any Administrations that you found to be particularly resistant to harmonisation?

No. Everybody was quite happy to get involved when they understood that things were going to be done in a properly harmonised way, but the other key point was resources.

Were there any particular issues caused by the Scottish census taking place in a different year from the English census?

Let’s be honest: the Scottish census taking place in a different year was and is not ideal, nor were the challenges that that census faced in terms of response. I personally put a lot of work into getting together a strategy for, first, adjusting that census for under-enumeration in a really pertinent way. Some wonderful stuff has been done by colleagues in Scotland, and I think the results are really good. Secondly, you then have to make some adjustments so that you can harmonise population data across the country. That clearly has to be done by either rolling forward the England and Wales data a year, or by rolling back the Scottish data. Personally, I prefer the former, and we can do it, but it is not ideal. It is better to try to do that over one year than it is over many years. I hope that in 2031 everything will work together, and I know there is a great commitment to that. Certainly, in some of the conversations I have about 2031, we hope the platforms will be the same for the first time, rather than the Scots having a different platform to the others.

Professor Lievesley recommended that, in a future spending round, the Treasury should ensure that funding is available to support harmonisation, but this money does not appear to be made available. We heard in a recent session that the ONS has withdrawn funding previously used to harmonise UK data in favour of addressing quality issues in economic data. Presumably, you feel that this is a worrying development?

I think the funding had to come. Let me just be clear; Lievesley was proposing that that should come through a spending review—that was my understanding. The Treasury was not really interested, and I think we just have to be clear about that. It was probably getting pushed from elsewhere in Government in the way that was happening in the previous round. It is then a decision of prioritisation, and I think it is the right decision for the ONS to take to focus on the quality of our statistics, particularly if you do have not the money. I think it is very sad.

It is very sad. In the longer term, if we want to fix this issue, it will take funding. Are there any other further steps that should be taken?

It is really about funding. The commitment, ability and teams are there. Sometimes, when you then have conversations with somebody, you need to have conversations about definitions to ensure that you have the same definitions of, for example, social care in different places. Ultimately, you can do what you want with the numbers, but if you are actually comparing apples and oranges, it is not worth doing.

The Office for National Statistics is an unusual one, with lots of different accountabilities. Who would you say that you thought you were ultimately accountable to?

Government.

To Government?

Yes, I think that is right. You are independent, and you have the board. However, there is no question at all that, ultimately, you are accountable to the people of this country, and the democratically elected Government.

Did your feeling that you were reporting to Government, as opposed to other institutions, drive the way in which you undertook the role?

I am reporting to the board because it is an independent body. For the decisions that I take, I am reporting to the board, but I have a dual responsibility to report to the Cabinet Office.

The Government have split the post of National Statistician. Could you give us your view of that and what the positives or negatives might be?

I have already said that, in my own view, in an ideal world the National Statistician should also be a permanent secretary. They have tried it before. If you really cannot find someone to do the job, I think you have to move to a different model, and I think that different model is the one that is there.

There are other chief executive positions in the country that require a combination of leadership, managerial and technical skills. Do you see the role as being similar to them, or do you view it as having significant differences?

You certainly need the executive leadership for the role. On top of that, you clearly need to be able to deal with some pretty hard statistical questions. To return to the Chair’s very kind tick for me, before he gave me a cross, I have needed that tick on many occasions when you get a phone call asking, “What do you think about this?” You need to have a pretty strong and broad understanding of statistics and not to be in a position where you say, “I will come back in three days and give you an answer.” There is an important and strong case for the national statistician to exist and be able answer those questions. It is also easier and better if that person is also running the Office for National Statistics.

Mr Campbell-Savours, were you trying to catch my eye?

Yes. I was hopeful that the question on accountability would come back to the push from the Treasury for you to concentrate on economic statistics. There is that wider set of Departments that want more. You have mentioned hypothecation and how that is pushing you to concentrate on certain things. I also got a sense from you that you wanted to prioritise looking at a wider set of statistics. Where is the real direction coming from in terms of that purpose?

The real direction is coming from the ONS, which has said, “This is what we would like to do, and it is going to cost x”. Effectively—and this is my experience—you are having a conversation with the Treasury about which set of things it is going to fund. You try to express your priorities, but there has to be a conversation. The point that the Chair did not allow me to answer earlier is that there is a second phase you can play, which is that if something is really, super important to a Department, and you believe you can deliver it without impacting on quality or anything else, then you can ask for the money to do it. That happened with regard to small area statistics post 2021, when the Department for Levelling Up was prepared to provide some funding, which was then used to deliver an excellent programme of small area statistics—which is very widely used—and to have ONS staff in different parts of the country working with local, regional and combined mayoral authorities so that we can have local data impacting on local decision making.

So if another Department came forward and pushed you for something, would you always have to go back to the Treasury to ask for more money to deliver it?

No, only if the Department had the money in its budget. You are certainly not going to do things without the full economic costing. On my point about small area statistics, that was a conversation that we had been having. We wanted to take that forward and there was a cross-Government view and a cross-Government high-level committee on small area statistics, which was really committed to it. We were not funded to do it. The Department for Levelling Up wanted to make it happen and was prepared to pay the full economic costing, so we were able to make it happen in that way.

I have struggled a bit—

It fitted within our priorities; it did not fit within the Treasury’s, but somebody else had some money that they wanted us to use for that, and we were prepared to do it.

You spoke earlier about the importance of quality. This is about quality. Ultimately, as we have seen with the economic statistics, we did not get that quality. We have seen a shift in purpose, and we have seen a report that describes how the culture may have led to that. Does there need to be much clearer direction on what you should and should not do? Was your criticism of hypothecation at the beginning of the session consistent with trying to be clear as to the purpose?

I think that there should be very clear priorities, and those priorities should be published very clearly and at the same time.

Are they not published very clearly?

They are. There should be clear priorities, and those priorities should be published clearly. I also think that the funding should be in one pot. You can then prioritise as you move forward if things need to be prioritised, rather than being hamstrung from day one.

But if it is not hypothecated, how do we protect the most important statistics—the ones that are market sensitive and could cause big problems?

Nobody is ever going to take them away. The commitment has to be, “These are the highest priorities.” I am embarrassed and upset about what happened with the labour force statistics—I don’t hide away from that—but at the same time, they were a priority. At that time, there was a feeling that we were about to move forward into a new online survey, and perhaps the drop in response rates was higher than expected in that one month. The future has to be that we agree on what we are going to fund and what it will cost. There has to be that awareness of a proper budget to deliver what the nation needs.

So would the labour force survey issues have been fixed by more budget?

Yes. If we had had more budget in the first place, we would not have had to cut back on surveys as we cut back on everything else. The number of interviewers and callbacks would have been greater, and the response rates would have been higher.

But considering the importance of the labour force survey, why was resource not shifted into that area to protect it?

There were cuts everywhere. That was my point about hypothecation. There were big cuts being taken everywhere in the organisation, for the reasons that I have already indicated.

But they would have been discussed.

Sir Ian, let me ask you this final question. Do you feel you have been treated fairly by the Government?

I think I have been treated very fairly by this Committee.

That is kind of you, but could you answer the question?

I started off by saying very clearly that I feel privileged to have worked with many dedicated civil servants who, I feel, have not always had the support or redress that they deserve. I hope that I have not tried to hide from anything—certainly my intention was not to hide from anything. I also hope that I have given you some views on things that will impact the future in a way that will enable my successors to have a supportive Government. Are there more errors in coding? If the director of methods were here, I think that person would say that there are probably some more errors. I am not ashamed of the fact that the organisation I led tried to put quality at the heart and to find all those errors.

Having said that was the final question, let me ask you the final, final question. A number of colleagues have asked a similar question. Ultimately, the Treasury has the pen over the chequebook. The Bank of England was making demands on the service, as were individual Government Departments, including the Cabinet Office. The Treasury’s instruction to focus on the economic data and not to do any of the wider stuff seems to me almost fundamentally to hole below the waterline the Government’s stated ambition, which was to get more joined-up government and better outcomes by breaking down silos through data sharing. We will undoubtedly take that up with the Cabinet Office in due course. Do you think the ONS suffers by having, effectively, too many “masters”? You have the board as well. Did you need to have somebody—whether it was the Cabinet Secretary, the Chancellor of the Duchy of Lancaster or the Chancellor of the Exchequer—who was effectively the person to whom you would have been directly accountable, and to whom you would have been able to say, “Look, I’m being pulled in this direction here and that direction there. We want to try to meet everybody’s needs, but it is impossible so to do within either the funding envelope or the staff envelope,” and so on? In essence, were there too many chiefs?

There is always going to be a lot of chiefs, because you have to work with multiple stakeholders.

Yes, but stakeholders are not people to whom one is accountable.

I take that point. My view is that you should report to the chair, and the chair needs to be very strong. I come back to the point I made, I think to Mr Quigley, that the Cabinet Office ought to be at the board table, because they need to be going straight back and saying, “This was the discussion.”

If that is your preferred organisational structure, would we be right to think that the responsibilities and duties of the board need to be far more explicitly set out and implemented?

One hundred per cent.

Sir Ian, we are grateful to you for your time. We wish you a continued, full and speedy post-operative surgery—we can swap orthopaedic notes at some point or another. We are very grateful to you for coming before the Committee. There was no obligation for you necessarily so to do, but you have aided our understanding of the situation immeasurably and the Committee is grateful to you for your time.

I was told that even if I was in my hospital bed, I should be here.

We have powers of summons, which we use lightly, but we did not have to, so we are very grateful.