Welcome back to the Treasury Committee on Tuesday 15 July 2025. We are meeting to discuss the reappointment of Richard Hughes as chair of the Office for Budget Responsibility. Mr Hughes was first appointed during covid in 2020. He has held the post for five years, and this is a five-year reappointment. He has indicated in his questionnaire that he intends to serve the full term. The OBR has had a bit of a buffeting. On 30 July last year, the new Government introduced the Budget Responsibility Act, which gave you even more powers. In particular, you will provide a fiscal statement if there is a measure that is fiscally significant, or where any combination of two or more measures is fiscally significant. That is quite a lot of power. We have heard a lot of discussion, with the IMF talking about you doing only one forecast a year and some MPs saying that you need to get back in your box or do only one forecast a year. Some people have been critical of what you have chosen to score. How have you found the last year, and do you recognise that characterisation?

It has been a pretty busy five years, I would say. I started in the midst of the pandemic and spent a lot of time talking to epidemiologists about where they thought the pandemic was going to go, vaccine specialists about when the vaccine might roll out and the Government about where the furlough scheme and the business support schemes were going to go. I had to try to understand what that meant for the economy and the public finances. Then I spent a few years dealing with the energy crisis and how that filtered through to inflation, disposable incomes and consumption. More recently, I have been having to contend with what tariffs might do to the UK economy. I guess in some ways it has felt fairly relentless. I am not sure the last year has felt that much different from the first four.

I think the other things were “events, dear boy, events”—covid, the energy shocks and the war in Ukraine—but the Budget Responsibility Act was passed by this Government as a political move to counterbalance what happened in the Budget of September 2022. That gave you more powers yet, within a year, there has been quite a lot of criticism. How has that felt for you, personally?

We do a job of highlighting the state of the public finances and their trajectory. The reality is that they have been under pressure and stress ever since I have been in this job, and I think what that does is transfer those pressures and stresses on to policymakers, who then have to deal with the consequences. I do not take any of the criticisms of the OBR personally, nor do any of the staff. To some extent, because what we do is portray quite a difficult and challenging set of circumstances, which both the previous Government and this Government have had to contend with, and quite a volatile external environment that rapidly translates, for all the reasons we have discussed in this Committee, into a revised outlook for the public finances, it creates a lot of volatility and puts a lot of pressure on the policymaking process. I think that frustration with both the constraints on the public finances and the volatile nature of the external environment sometimes translates into basically blaming the messenger, rather than not being happy with the message. I think it comes with the territory, to some extent. Where people criticise our methodologies, we try to address and respond to those criticisms where we think they are reasonable and where we think that we can do something different. I highlighted the work we have done to try to be more transparent on how we think about the growth impacts of policy measures. Where we have identified gaps in the way we do things—we have had discussions with this Committee on the forecasting of departmental expenditure and the shortcomings we have seen in that system, which we have looked to improve in more recent forecasts. We try to address it where we think there are legitimate gaps in our methodologies, but where we are just being criticised for telling it like it is, well, that is fine, but it is not going to stop us telling it like it is.

Earlier, Mr Glen asked how you make decisions about what to include and what not to include in your forecasts, and obviously there are sometimes timetabling issues in how things come in. Have you had any discussions with the Treasury about changing the memorandum of understanding on how you deal with that, or is it just an iterative process that you go through with the Treasury each time you do a forecast?

It is codified in the memorandum of understanding, and we discuss and agree the timetable for forecasting with the Treasury and our other key forecasting partners—HMRC, DWP and other Departments—and try to strike the right balance between providing a sufficient number of pre-measures rounds, so that the Government have a sense of where the pre-measures position is heading, and providing sufficient numbers of post-measures rounds so that the Government get a look—but usually two looks—at how things stack up after their policy measures have been fully incorporated into the forecast. Inevitably, there is a trade-off in when you close the pre-measures forecast and what data has gone in. Then, ultimately, there is the day of the Budget, or other fiscal event. Because we are operating in a very volatile environment, you can get changes in interest rates, energy prices or Donald Trump’s tariffs in the middle of all that, so you have to decide at what point you stop the music on changing the pre-measures forecast and just reflect the effect of Government policy. In that way, you have a transparent picture of “What are the changes we have made to our forecast, and what are the changes the Government have made to policy?”, rather than mixing up those two things. That is the discussion we have with the Treasury on the timetables for the two forecasts we do every year. We always come to some kind of agreement at the beginning, and we reflect on the lessons afterwards.

You talked a lot in the previous session about not scoring a policy until it is a clear policy, and not scoring something until a plan for delivery is properly in place. On the welfare reforms and scoring people going into work as a result of those changes, while I think you were right that there is not detail on some of those work programmes, there is quite a lot of history of work programmes, and if you get 5% of people in a work programme into work, it is a good, successful one. Did you ever consider such an approximation? And what pressure did you put on the Government to come back with proper plans to be able to score it? How does that process work? Do you just say no, or do you say, “Hang on, if you give us a bit more, we can think about scoring this”?

It is never just a straight no. It is always, “If you can tell us more about the policy, and if that is the policy you announce”—equally importantly—“on Budget day, we can try to think about what economic effects it might have.” Certainly, under the previous Government, there were employment programmes where we scored employment effects from welfare-to-work measures backed up by employment support. That is where we were given sufficient detail about the nature of the interventions, the number of people affected and the timetable over which it was going to unfold. I should not understate the point about additionality—“How can we make sure that Jobcentre Plus is doing extra stuff, rather than just replacing an existing policy that will mean that other people are not supported into work?”—or whether there is a very clear evidence base for the effectiveness of the policy. As you say, depending on the group you are talking about, you can see success rates as low as 10% or 15% in the proportion of people who are still in employment in five years’ time. Those measures of effectiveness need to come from empirical studies from somewhere. We like to see them from the UK, because we like to understand how policy works in our own jurisdiction. Where there is a reasonably analogous intervention in another country with a sufficiently robust evaluation, we try to draw on that kind of evidence. In the past, we have drawn on evaluations of other countries’ employment programmes that we think look quite a lot like what the Government seem to be announcing here. In the case of the welfare reforms back in March, as I said, there were two problems with the employment support. One was that the entire policy package—even the benefit changes—came to us very late in the day, and there was just not enough specificity about the employment policy for us to do anything other than reflect its costs and say, “Tell us what you think will be the employment impact of this thing.”

Thank you very much indeed.

The OBR has to make judgment calls in forecasting, and that is really difficult. In particular, productivity is a key forecast. Obviously, it very much hits the fiscal forecast as well. I am going to present some of the autumn Budget EFO forecasts to see what the out-turns for the years have been. In 2021, you predicted 0.5% productivity growth; out-turn was 1.5%. In 2022, you predicted 0.3%; out-turn was 0.9%. In ’23 and ’24, productivity was missed on the low side by almost a percentage point. Clearly, there are big sensitivities, but these are big misses within year. What has been the miss, and why do you think it was missed?

For productivity, you are dividing the entire output of the economy by the entire hours worked in the economy, so it is the difference between two numbers that over that period were changing enormously. The pandemic affected both output and hours. Then you had the recovery from the pandemic, and all of us were trying to predict how quickly that was going to happen. Then you had the energy crisis, which hit energy-intensive businesses, some of which are the most productive enterprises in the economy, quite hard. The national health service also goes into productivity calculations, and its productivity is measured directly from the Office for National Statistics, so the performance of the public services also feeds into those calculations. On the misses you have described, we are symmetric either way. That would suggest that, at least in recent years, there has not been a bias one way or the other. But how productive the economy was going to be after you had sent about half the people out of the workforce during furlough and then brought them back, and massively disrupted the operations of the NHS and other public services and then tried to turn them back on, was a very hard thing to predict. You saw a very dramatic fall in productivity and then you had to ask yourself how quickly it was going to recover to its pre-pandemic trend, and then what kind of trend it was going to be on after that. And we were then hit by an energy crisis along the way. Those are tricky judgments. It is also worth bearing in mind that, over this period, we had a big influx of migrants into the workforce and a whole bunch of people falling out of the workforce, so thinking about the average productivity of the people coming into the workforce versus the productivity of those leaving made a difference. And last, but not least, our various data collection mechanisms were breaking down over this period. The labour force survey started to become unreliable. As you might recall, the level of GDP was being massively revised by the ONS, coming out of the pandemic—as it was in lots of other countries. That is not a specific criticism of the ONS, but it was hard to take definitive reads from any one bit of data, and lots of bits of data were changing quite quickly.

Let us look at the period before the pandemic. It would be fair to say that there were post-pandemic issues but also, before the pandemic, the OBR did keep getting it wrong. My concern is that it kept getting it wrong in the same direction. During the period of fiscal consolidation of 2010, we had large negative output gaps and productivity consistently over-forecast. In 2021-22, high public spending and a positive output gap. Fiscal consolidation back in 2023-24, productivity then missing on the downside. If you change your fiscal demand multiplier, would your productivity and GDP forecasts be more accurate?

We keep our multipliers under review; they are based on the ones that have been empirically observed between different kinds of public spending, tax measures and output. The last time we did a review of them, and talked to people who estimate these models, they turned out to be roughly in the right place. More generally, looking back at our forecasting record for both the economy and public finances, one of the things I asked for, and we published a few years ago, was a comprehensive assessment of how our forecasting record compares with those of other forecasters, both UK forecasters, other official forecasters such as the Bank of England, and external official forecasters elsewhere in Europe. In the end, our forecasting record compares pretty favourably with all those comparators. That is especially when you bear in mind the legal constraints we are under. We have to forecast Government policy as it is given. We know that there are already lots of things in there that are probably not going to happen, such as indexation of fuel duty. That is already going to make bits of our fiscal forecast wrong. Other forecasters can take that into account, which we cannot. A comprehensive evaluation of our forecasting record against others shows that we do a pretty decent job. Everybody has been too optimistic about productivity since the financial crisis, because everybody assumed a stronger rebound in productivity after 2008 to 2010, which just did not materialise. The issue is, at what point people recognise that and just revise down their medium-term forecasts on productivity, which we did in the middle of that period. As I said, optimism was something that affected all forecasters during that period, because we did not realise what a legacy the financial crisis and subsequent shocks had for productivity.

That is great, but my question was about changing the fiscal demand multiplier. My contention is that there were cuts in public spending that hit productivity in a number of different ways that cannot be accounted for in a very hard-to-quantify model, as we have come to several times today. I appreciate that everyone got it wrong, but your forecasts are far more important that anyone else’s because they have a real-world impact on people’s lives, how much we are spending and fundamentally on growth. If the fiscal demand multiplier were different, would your forecasts have been more accurate?

Not necessarily, no, I don’t think so. They are calibrated on the best data we have got, on the relationship between different forms of public spending and growth. For direct public spending, our demand multiplier is one. If you think it is bigger than that, that means that the bigger your estate, the bigger your economy for ever, and that cannot be right. More generally, you saw a relatively big increase in public spending in the autumn Budget, raising the level of public spending, big increases in public sector pay. It is not evidently driving much more dynamic economic growth in the past few months or indeed in the second half of last year. We take a long view of these things. We try to estimate them off long-term relationships between different forms of public spending and tax, output and other measures. If that relationship is changed, we would change our multipliers. We have got demand multipliers and supply multipliers, and they are based on the best evidence we can find.

Have you tested it?

We test it on the whole history of UK economic and fiscal policy, so yes.

Have you tested what would happen if you changed the demand multiplier? Would it be more accurate?

They are already estimated to be as accurate as they can be. I am not sure how else I could answer that.

You mentioned a few times that your model is only as accurate as the data you put into it. The Committee has spent quite a lot of time on the labour force survey and the ONS data. How much is the reliability of that data keeping you up at night?

Our forecasts are as only as good as the out-turn data on which they are based. When that out-turn data changes, we have to change our outlook, and that can create instability for people who have to make policy on the basis of those forecasts. I have sympathy for all national statisticians, in that there seems to be, across advanced economies, less participation in surveys and so it is getting harder and harder to collect information in what has conventionally been the most reliable way to collect economic information from households and businesses, which is to ask them a load of questions. People don’t answer their phone any more. People don’t answer the door any more. But we seem to have more of a problem with that in the UK than in other places. We have seen a bigger fall in response rates and we have struggled to get them back up again. When you are making big judgments about how big the labour force is, how many hours people are working or what they are earning—big ingredients to our forecast—that is a problem. It is good and encouraging that the ONS is looking at the compliance burden of collecting data from the LFS—it seems to take a very long time and takes a lot of people’s time to collect that data—and thinking about other means of collecting data that are quicker and put less of a compliance burden on the respondent. Also, the general message coming out of the ONS is about focusing on the fundamentals around the production of good macroeconomic statistics, possibly at the expense of some of the more innovative things that they have done. I should say those are very valuable, but you have got to get the bread and butter right, especially the things on which the Bank of England is making decisions about interest rates and the Government are making decisions about tax and spending policy. You really want to make sure that those are based on the best possible data that you have, and that those are gold standard. Other surveys about how people are spending their time, which are interesting for social researchers, also need to be done and it is really important that they respond to people interested in those things, but it is possible that too much attention got diverted on to those latter things and not enough was being focused on the bread and butter.

How much would real-time PAYE, HMRC and DWP data support the work that you are doing to get a more accurate forecast? Do you think you are overweighted on the LFS?

For years now, we have looked at a variety of indicators. Those things have the big advantage of being very timely. You understand the population, because they are on someone’s payroll. The issue is that what you do not know is what is going on with everybody else, so you have to make guesses that infer how big the self-employed population is, whether people are working informally and who is just not being picked up by high-frequency data that comes out from very formal systems and direct from employers. The nice thing about the labour force survey is that it was a random sample of the whole population—we thought—and so it was just like stopping the average person on the street and asking them about their employment status. You felt more comfortable that you were picking up everybody who might possibly be in the population, rather than just the people who are on a formal payroll in a company. If you just rely on that, you have to infer what is happening in the self-employed population.

Would you have to only infer if you had the HMRC data annually? HMRC will have the tax returns.

With a lag, because they will be filling out self-assessment returns. Much of what we have been trying to figure out, especially in recent years, after we had the pandemic and loads of people fell out of the labour force is, were they coming back? When? How quickly? At what level of earnings? What you really want from these things is a sense of the dynamics. If you are waiting a full year to get self-assessment tax returns and you are only doing a five-year forecast, that is a long time to wait to get a signal about what is going on right now, especially when we have seen big jumps in the participation rate and trying to figure out why that is going on. In the end, we have looked at a variety of sources, including the LFS, but appreciating its limitations and picking up on these more real-time indicators.

We have heard a number of times about when it was flagged to the Government that the labour force survey was unreliable. At what point did you start noticing and did you flag to the Government that the data was not telling us a proper picture?

Quite early on. Let me give you a very precise answer to that—that will involve me going back and checking my records. Let me come back to you. To my recollection, we were still recovering from the pandemic when it was obvious that response rates to the LFS were falling. Let me not rely on my imperfect memory—let me go back and give you a specific answer.

We have just published our implementation plan in response to that external review. It has always been a valuable exercise for us—this is our third one—but this one was especially valuable because it came from a practitioner. The Dutch CPB has been around for a lot longer than we have been and it has a lot more experience in doing the kind of work that we do, so it was particularly useful to get insights as well as notes of caution from her about where to head next. I very much accept her recommendation on doing as much as we can to highlight uncertainty around our forecasts. I think this Committee has seen how much we are at pains to produce fan charts, do scenarios and highlight areas. We have published this whole green report on risks and uncertainty, looking not just five years ahead but 50 years ahead, and at where the world might go and what might happen on a different trajectory. We try to use as many different tools as we can to highlight uncertainty. In the end, the Government have a fiscal rule that is either met or not met on the basis of our central forecast, which the law requires us to produce, and it is interesting question to answer, is the Government on track to meet its fiscal rules or not? We always try to highlight whatever margin there is against that achievement, whether it is a big number or a small number, and how it compares to the average level of uncertainty around any forecast that we produce, for all the reasons that we have discussed, as well as what are the known unknowns in our forecast. In the last forecast we did back in March, for example, we knew interest rates had been highly volatile and we have got lots of debt; we knew trade policy was rapidly changing, and a week later Donald Trump announced loads of tariffs on the world; and we knew our estimates around productivity were very uncertain, so we made sure that our forecast included scenarios on all three of those things, so that we could say, “Look, £10 billion sounds like a lot of money, but it is a very small amount of money when compared to any one of those risks, and even smaller if compared to them in combination.”

We do not just report whether the Government’s current budget is in surplus, balance or deficit in five years’ time, but give probabilities around that. The sad news is that those probabilities are never higher than the high 50s, so usually there is a roughly even chance of meeting this thing or not, because the Government run their public finances really close to the wire, and they have done for years now. In the 2010s, Governments would run very large amounts of headroom against their targets, so they could absorb lots of risk; more recently, they have not, and probabilities bring that to light. There is a huge fan of uncertainty around the forecast for the current balance. The Government are right in the middle of that band, not towards one of the safer extremes of it. Making use of those fan charts and confidence intervals is useful and relevant in studying fiscal policy, if you want to meet that target with confidence, given what we know about the distribution of risks around them. There are features in the current framework that have been built in but do not come into effect for a number of years. They are about having ranges around that number for particular forecasts, which give the Government a bit more flexibility around exact balance. They allow them to be 0.5% either side of balance, which gives the Government a bit of leeway. I still assume that this Committee would be interested in the answer to the question, “Are the Government on track to meet their fiscal rules on the central forecast or not?” rather than a bunch of handwaving from me saying, “It depends”, “Who knows?” and “The world is very uncertain.” I get paid to answer that question and I assume this Committee wants me to answer it.

We already do it; it is in the book. It is 50-something per cent, but we already answer that question in probabilistic terms.

A Government could express its rule in probabilistic terms and say, “We want to not just meet the current Budget by one penny but we want to meet it with 60% confidence,” and that would then have to be calibrated off the risks around the forecast, and the distribution of errors.

It is important to say that, as I understand it, the IMF’s recommendation was that there be one assessment of the rules in the autumn but still two official forecasts produced by the OBR, and that they continue to think that doing two forecasts a year is international good practice. I think their more detailed staff report will set that out. From my point of view, the UK has been doing two economic forecasts and fiscal forecasts a year since 1975. Ether advanced economy also does two forecasts a year, so were we to reduce the number of forecasts to one, that would make us one of the least fiscally transparent countries in Europe and of any major advanced economy. But I think what the IMF was getting at was, could you make that assessment happen in the autumn, and then, if in the spring the Government were off track for meeting their fiscal rules, let it ride, and make more policy decisions in the autumn? The Government have that option now, and they always had that option. There have been times in the past when previous Governments have been missing one of their fiscal rules in a particular forecast and didn’t take policy action to deal with it. That option is already available to Government; it just depends on how Government reacts to the piece of news. There do seem to be elements of the fiscal framework, which, as I said, do come in in a few years’ time, that give Government a bit of flexibility around the spring forecast: they can be up to 0.5% of GDP out from fiscal balance in the spring but they have to make it up in the autumn Budget, and that is a way of focusing fiscal decision making then. In principle, that option is available to them now. They can just say, “We’re meeting our fiscal rules in the spring; we’ll get back to you in the autumn with how to do it.”

How Governments react to how they are performing against their own fiscal rules is a matter for the Government. Governments have missed their rules in the past; they have changed their fiscal rules in the past; and how much leeway Governments give themselves against their fiscal rules is entirely a decision for the Government. The reason why recently policy has been very responsive to changes in our forecasts is because the world has been very volatile, and the Government have left themselves hardly any room for manoeuvre against the achievement of their rules. If you run the public finances really close to the wire, a very small change in the outlook for inflation, interest rates, growth or employment can put you the wrong side of that, and if you do want to meet your fiscal rules, which as I say is entirely the choice of the Government—previous Governments have not, or they have changed them—then Governments adjust policy. But Governments have lots of other choices about what they want to do in those circumstances.

I will say a few things. One is that we have had fiscal rules for longer than most countries—we practically invented these things in the late 1990s.

The half-life is dramatically shorter.

Exactly. We have had a fiscal rule since the late 1990s. We have had nine or 10 specific fiscal rules since then, which comes back to my point about Governments changing them all the time if that is what they want to do. To some extent, rules-based fiscal policymaking is very well established in this country. Everybody just expects a new Chancellor to have a new set of fiscal rules and articulate them very clearly. We have set out in legislation that one of the reasons why the OBR was set up was to assess whether Governments were on track to meet them. In other countries, like the US, they are not really bothered with fiscal rules. They have a debt ceiling, which everyone knows does not work and just repeatedly gets raised, because having a nominal limit on debt is meaningless unless, at some point, you decide to default on it or make dramatic changes in tax and spending to deliver it. You have other countries in the G7 where fiscal rules do not exist, so there is nothing to measure headroom against. In other countries in Europe, you have fiscal frameworks, a number of which derive from European directives and treaties. A bunch of European countries are just meeting them, with very large headroom, so this issue does not arise. They are meeting their fiscal rules by considerable measure, so they do not have to respond to the margin against the limit, because they are miles from it—their budgets are in balance, and they have limits up to 3% of GDP. In other cases, they are far above them. Greece is not getting back to 60% debt to GDP any time soon, so the discussion about fiscal policymaking is about where we are on this very long journey of trying to get our public finances back to sustainability. In the UK, because we have rules that are chosen by Chancellors, and therefore quite closely connected to the specific decisions that they make—and that Chancellors like to meet and be seen to meet—they get much more attention here than in other places.

It seems as though there is a wide range of fiscal rule possibilities that the market will swallow, but the argument put forward by this Government is that we need to have very strict adherence to relatively tight rules, at least when you are looking at the rolling side. Trying to make that work with this volatile environment is creating policy problems for the Government. I think that people would assume that the process should be that the Government create a policy, the OBR assesses it, there is an exchange and then there is a forecast. The concern now is that the OBR forecasts are driving policy creation, and the example people would give is what happened with the welfare reforms initially, which were seemingly trying to chase that £9.9 billion headroom. Do you think that, culturally—not just from a Government policy perspective but from a media narrative perspective—people’s attitudes towards fiscal rules need to make much more of an allowance in this volatile global economy?

The rules that the Government have in place at the moment are about as loose as we have had. Balancing the current budget in five years’ time already leaves out investment, so you are just balancing day-to-day spending against current spending in five years’ time, and you are trying to get a measure of the balance sheet that allows debt to keep rising, but so long as you are building up financial assets, it can still fall. That is a reasonably permissive fiscal regime compared with ones that we have had in the recent past and those that are in place in other countries. What leads policymaking to be very sensitive to changes in forecasts is what I came back to, which is that you have a relatively permissive set of fiscal rules by historical standards, and Governments are running fiscal policy all the way up to the wire—all the way up to the line that they set for themselves. I do not really accept the characterisation that our forecasts forced the Government to make any particular set of policy decisions. Back in March, the Government decided to make £5 billion-worth of welfare savings. They had £10 billion-worth of headroom against their fiscal rules. They could have settled for £5 billion.

The perception emerged, though, because the changes came quite late in the day, and presumably they were already in discussions with you at this point and so maybe had an idea of what the direction of travel was. It seems to me that they made an active decision to, again, try to meet the £9.9 billion headroom and keep that consistent. It seems quite remarkable, out of a £1 trillion budget, that they would arrive at that figure by accident. People therefore would have a legitimate concern that that is now driving policymaking, as opposed to the other way around.

As I said, Chancellors can make a point about how much headroom they want against their fiscal rules. Recently, they have left themselves very little. At the time of the March Budget, the Chancellor apparently decided that she wanted to maintain the £9.9 billion-worth of headroom that she had back in the autumn. She could have settled for less, and not done the welfare reforms. That was entirely a political choice that was available to her. She could have not done the welfare reforms and met her fiscal rules with our forecast, so I do not see how it was our forecast that somehow drove the decision on welfare. She also has £1.5 trillion-worth of choices about other spending and £1.5 trillion-worth of choices about tax. She chose the ones that she chose back in March to get to a particular configuration for the fiscal rules that, strictly speaking, overachieved on where she needed to be. She also has the choice of not meeting the rules, doing so at a later date or changing the rules.

The purpose of the OBR is to provide independent supervision of the public finances. But there are some people, including in this place, who say we need to get rid of the OBR altogether. For a reappointment hearing, you might not want to hear that. The outcome of having independent supervision is presumably confidence for people who lend to the United Kingdom. What you outlined earlier and in your report is that we are losing a big buyer of gilts over the next 50 or so years in the form of DB schemes, so we have to find some other buyers. What would you say to people who say the OBR should be abolished?

The legislation setting us up, which you passed, gave us two jobs: one is to provide the forecast the Government use to make policy; the other is to provide you with information that you want about the state of the public finance, its outlook and whether the Government are meeting their fiscal choices, as well as how those affect the forecast. At the OBR we take both those responsibilities very seriously. As you said, there is the added benefit of providing some confidence to people who buy Government debt and invest in UK gilts, which at any point in time is a choice. No one is forced to buy Government debt, not even pension funds; if they do not want to, they can find a better asset. Our forecasts also serve the role of providing information that creditors are interested in when looking at the outlook for buying Government debt. Maintaining that confidence is important. We have lived experience of a point in time, in the autumn of 2022, when the Government announced significant fiscal policy changes without an economic and fiscal forecast from us or anyone. There was a loss of confidence in that Government’s fiscal policy, and that was reflected in the cost of Government debt.

In your questionnaire, you talk about how you have had to demonstrate your independence. You are clearly demonstrating your robustness in this session. You talked about asking your non-executive members to review and report publicly on any instances in which questions are raised about the integrity of the forecast process, and you have done that a couple of times, including in March 2024. In both cases, all recommendations have been acted upon. It is notable that the Prime Minister was at the Liaison Committee and was critical of you for not scoring. You record all contacts with Ministers on both sides; did you have any contact with the Prime Minister after that?

No, I have never met the Prime Minister.

Right. You have never had a meeting with the Prime Minister; that is useful to know.

If I had, I would have logged it and informed you.

We will keep an eye on the log list from now on. Maybe this is a little prompt for the Prime Minister to have a meeting with you, Mr Hughes, if you are reappointed today. You also talk about how you would inform your non-executive members and this Committee in the event that any undue pressure is placed on you to behave in a manner that contradicts your legal duties. You have not informed us of that, but have there been any attempts or forays into trying to push you in any particular direction?

There has not, but I would emphasise that one of the reasons why I think there has not been is because of that precise proviso, which is that, if there were, I would inform you.

You would have to tell us.

I would inform you immediately, as I have in recent instances, including the issues around DEL forecasts a little while back, where there were issues that we felt we needed to look into. We immediately informed the Committee that that work was under way, told you what review was under way, and then reported back to you on how it had improved. This Committee is a really important safeguard of both a good forecasting process and also of our independence. We know that we have people who we can come to who value the role that we play, independent of the support that we provide to Government in putting together fiscal policy, and that support the case for transparency and integrity, if those instances happen. Perhaps because that backstop is there, I have never in my five years at the OBR come under pressure.

That is a nice compliment to the Committee, which is about to adjourn to consider your reappointment—over which, I would just repeat to anybody watching, we do have a veto. I thank you for your time today and for the last five years. We are about to have a conversation on the basis of the last five years, and also this last half hour or so, to discuss whether we will say to the Treasury that we think you should or should not be reappointed. If you step outside, we will be in touch momentarily.