Welcome to the Treasury Committee on Wednesday 25 February 2026. We are here to start our inquiry into the future of the Office for Budget Responsibility. It is 15 years since it was created by the then Chancellor, George Osborne, and since then it has had two chairs. We are delighted to welcome Sir Robert Chote, who was the first chair of the Office for Budget Responsibility and did that job for 10 years—welcome back to this Committee. It is a while since you have been here; none of the Committee members were here when you were coming regularly. We are also pleased to welcome back Richard Hughes, who was a regular visitor to this Committee and stood down as the OBR’s chair at the end of last year. Before we go into the main session, I thought it might be a good opportunity for you to speak, Richard. You have been in front of a Lords Committee, but this is the first time you have been in front of the House of Commons since your resignation, and I wondered whether you wanted to say anything about that.

Thank you, Chair. I should start by saying that I am now a senior economic adviser at Taula Capital. Thank you for the invitation; it is a real pleasure to be before the Committee again. As this is my first appearance before the Committee since I stood down as chair, I would like to start by reiterating my apology to the Committee for the inadvertent early disclosure of the OBR’s EFO on the morning of 26 November, and the disruption that caused to the business of the House. It was a technical but serious error for which I took full personal responsibility. I hope you will agree with me that the OBR’s actions since then—including commissioning two independent investigations, one led by its non-executive directors and one led by the National Cyber Security Centre, publishing the full results of both in a timely manner and implementing their recommendations—reflects the organisation’s commitment to recognising its mistakes, transparently investigating them and taking the action necessary to fix them. I have full confidence that the changes the OBR is introducing around the publication of the next EFO on 3 March will ensure that such an error will not recur.

Thank you very much for that. Your resignation speaks volumes about taking responsibility for something that obviously did go badly wrong. I should stress that we heard evidence from the Treasury permanent secretary at a recent meeting that the OBR papers for the spring statement next week will be published via the Treasury, because the OBR has not yet got its new arrangements in place. They are still independent, but they are published through the offices of the Treasury without any interference. I want to make that clear. You have both had this big job. Fifteen years on, are there any changes that you think should be made to the Office for Budget Responsibility? Sir Robert, from your perspective five years out, what is your take?

I think the rationale for creating the OBR is as robust now as it was back in the day when it was being considered—various options were considered before the final model was come up with. That rationale was about providing rigorous, professionally based analysis of the public finances to inform both policymakers’ decisions and the public, and it remains a robust one. It is important to be clear about the task that the OBR was originally given by Parliament. Two key components of it were to produce forecasts looking ahead five years and to use those for the evaluation of the Government’s progress against its fiscal rules. As I am sure we will get on to, there is a debate about whether there is too much focus on estimates of how close the Government is to achieving those fiscal rules, but that was literally the job that Parliament gave the institution at the outset. The other thing to bear in mind—and it comes into the questions that I know you want to raise around forecast performance—is that, again, Parliament has specifically tasked the OBR with producing forecasts. They are not unconditional forecasts of what you think the most likely outcome for the economy and the public finances is, but what you think the outlook is based on current Government policy. If, for example, you look at the performance of forecasts for Government borrowing over a five-year period, you see that the OBR under-predicts Government borrowing. That is a feature and not a bug. It is a reflection of the fact that Governments have chosen to increase the spending plans that were originally in place when those forecasts were made. In terms of the OBR’s role, it has, I think in good faith and pretty professionally, done the tasks that Parliament has set for it. There has then been an interesting set of issues, over time, as to how Government have dealt with the task of putting together a fiscal framework, developing policy and, essentially, deciding how responsive policy ought to be to what are inevitable moves up and down in forecasts. Obviously, there is a good deal of work ever to be done at the OBR on looking at its forecast performance and its analytical work, and on ensuring that it can do that in the best possible way. The OBR produces the single most detailed and comprehensive evaluation of its own forecasts—although I see, interestingly, that the Bank of England has popped up with a new one in January this year, so maybe it is joined there. In terms of the way in which the broader framework has been operating, the issues of how Ministers and the Government use the forecasts, how they talk about those and how they set policy in the face of an outlook that inevitably moves is something that it is well worth the Committee looking at.

That is a good start. We will have detailed questions about some of the points you raised, but basically you think that the OBR is still the right thing to have and is broadly doing a good job.

Yes, I think so. Others have thought about various models—for example, as some other countries do, having the fiscal watchdog not producing the forecasts itself, but scrutinising and judging the reasonableness of forecasts that come out of the Government. That is a perfectly feasible thing to do, but it is important to bear in mind that particular practical and process difficulties would arise with it. In particular, the evidence from other countries is that the access to information that the body needs to reach those sorts of judgments could be much more contested and much more difficult—much more frictional in the relationship between the institutions and Ministers—than with the model we have at the moment in which, like the Netherlands equivalent, the OBR is a monopoly producer of the official forecast and has a private interaction with the Government as that forecast is being developed with the policy package in parallel.

I agree with all of what Robert said. I reiterate the point that the legal institutional architecture on which the OBR rests is very strong and sound. I think both Robert and I, during our time, leant on it from time to time when we needed to, in order to deliver the mandate of the OBR, especially around access to information and the right to produce that information. I think it is important that the organisation continues to adapt to the changing landscape for fiscal policymaking. During my time, a lot of that was just about reacting to changes in the outside environment. I arrived in the middle of the pandemic, and we then had the energy crisis and the mini-Budget. I was important to understand what all those events were doing to the economy and the public finances. It was not so much about changing the institutional arrangements for the OBR as making sure that it was flexible in, basically, understanding novel shocks, talking to different kinds of experts to understand what pandemics or energy shocks do to the economy, and making sure that it was an outward-facing organisation, with access to expertise throughout the UK’s very rich economic community and the wider community of experts, including in public health, energy markets and those sorts of areas. The OBR has to stay flexible and adaptable to the rapidly changing environment in which fiscal policy is made. During my time, a fair amount of our work on trying to progress the institutional architecture for the OBR was to do with addressing gaps and lacunae in the institutional framework, one of which was the OBR producing this report on fiscal risks and sustainability. That is a very important report. It gets a lot of attention in this Committee and it gets discussed in the Lords Economic Affairs Committee, but it does not get very much attention from Government in terms of a policy response in some sort of formalised way. EFOs do, because they are accompanied by a Budget. The FRS report is basically an EFO for the next 50 years, and it tends to highlight some much more fundamental questions about the trajectory of the public finances, but it does not get anything like the kind of response that the EFO gets. It started out with a response from Government of over 100 pages, but it now gets a cursory 10 to 20 pages in a written statement, and there is no parliamentary debate about it. I think that the lack of focus on risks and sustainability, which in many ways are much more important to determining the long-term trajectory of the public finances, is an area that does need more attention within the framework. The second area in which this Committee has been very much involved is trying to improve the way in which the OBR forecasts departmental spending—departmental expenditure limits. We had an investigation and a report to this Committee during my tenure about whether there was adequate exchange of information between the Treasury and the OBR about risks to departmental spending, and whether the Government’s spending plans are fully explained and fully justified. More recently, in the past few years, the OBR has had an explicit legal right basically not to believe the Government’s departmental spending plans and to put in something that it thinks is a more realistic reflection of the cost of Government policy in areas like health, education, transport, asylum and other areas. That is a very important to having a credible forecast. Our own exploration of the OBR’s forecasting record demonstrated that one of the biggest misses in forecast accuracy was around with departmental spending consistently ending up higher than what the Government said their plans were, oftentimes for reasons that were pretty well understood and people were well aware of. Thirdly, in the past few years the OBR has taken a more transparent approach to the scoring of the supply-side effect of policy measures, which is an area the organisation has kept under review. You will have seen that, alongside the November EFO, the OBR published a paper drawing some lessons from the initial outing of that new, more transparent approach to so-called dynamic scoring. That is an area that the organisation needs to keep under review, because it is relatively novel, not that many countries do it—the US does it and the Dutch do it—and it creates different kinds of incentives for policymakers. It is very important that those are borne in mind.

Yes, we will move on to who is wagging the dog. Before I hand over to Mr Murphy, when you were at the OBR, Sir Robert, the EFO had grown quite a lot—useful information was kept—but when you arrived at the OBR, Mr Hughes, I think you stripped some of that out. Will you explain why you took the decision to make those changes from your predecessor?

There was a constant struggle over how long the EFO should be. It tended to balloon and then people would complain. We would do a survey of our readership and they would say, “On the one hand, we’d like it to be shorter, but on the other hand, we’d like to see all these extra things that you do not currently provide.” Over time, I suspect that my average EFO was about as long as Robert’s average one, although it grew and shrank over time. On one occasion, after the mini-Budget, we had to produce it so quickly that it ended up being only 60 pages. Some people loved that and some people hated it. One of the things that we tried to do is to have the EFO be a comprehensive summary. It is a Command Paper to Parliament, so it is important that it is a comprehensive and transparent summary of the public finances. But that has to be complemented with working papers, briefing papers, and lots and lots of disclosure on our website, with ready reckoners that people can use, so that we are also providing people with the tools to develop their own understanding of the public finances, and not just having that delivered by us in words and verbiage.

For the record, Sir Robert was nodding in agreement. Do you briefly want to add to that, Sir Robert?

Yes. We were always conscious of the different audiences with different interests and expertise who were consuming the report. My sense of it, as a former journalist, was that there are always people who basically want the one-page to four-page summary, and people who want the 20 pages. One of the reasons why the report grew is that there are a whole lot of areas of the public finances, and people might have a very legitimate interest in one small part—for example, how the student loan system works, or what the trends are in welfare spending. The temptation is always that, on the one hand, we need a product that provides something to a journalist who is not an expert on any of these areas but needs to get a grip of them quickly, while on the other hand, we also need to provide that level of valuable information on key areas of the public finances that people are not getting from anywhere else. Also, the other thing I always had in the back of my mind is that the Treasury needed to know that no part of the public finances was so boring and incomprehensible that if they tried to do anything dodgy in it, the OBR would not devote three pages or a box to explain what was going on.

Okay. On that note, let’s hear from Luke Murphy MP.

I want to pick up on the point Sir Robert made a moment ago about the comparison with how this is done elsewhere and who does the analysis. I think you were suggesting that we should be wary of the other model, because of the friction it could create and the relational problems, but what about the policy outcomes? Your point was about the process that occurs as a result of the different relationship, but what is your view of the actual policy outcomes that that different kind of relationship delivers in other countries?

In terms of the background on who does what in the other countries, there are roughly 55 fiscal councils. The International Monetary Fund regularly updates a database, and I think that at the end of the last calendar year there were 55. Of those, 18 prepare forecasts like the OBR. In only two of those cases—the OBR is not one of them—is the forecast binding, so it is perfectly feasible for a Chancellor to say, “I don’t agree with this forecast,” or, “I think things are going to turn out differently,” explain why that is the case and present an alternative view. I think that 44 out of the 55 monitor fiscal rules, so they have to reach a judgment about whether the Government is on course to achieve what it says it is going to do, and how far away from that it is. Interestingly, of the 55, 42 make policy recommendations, so they say what the Government should do and not just, “This is some background analysis to inform Parliament and the Government.” The OBR is not one of those, so we are in the relatively small minority that is actually relatively distant from the policy decision process. We do not tell the Government, “Our forecasts show that the public finances are not consistent with the rules. We think you ought to change the policy.” It is rightly for elected Ministers to decide in that scenario whether they change the rules or the policy, or whether they say, “I’m willing to live with this apparent discontinuity.” To the extent that the relationship affects policy outcomes, it is because Ministers choose to let it be the case. There is one fiscal council of the 55 that can actually stall the Budget process, so these are not like independent central banks with policymaking scope. As I say, the OBR does not even make recommendations, so I am very wary of language like, “The OBR has forced us to do this or that.” That is not the case. It is the combination of the targets that the Government has set itself and its decisions on how to respond to forecasts that show them on or off course.

Sir Robert, you made a comparison between the friction introduced by models like the Netherlands model and models in other countries where the equivalent of the OBR only scrutinises and checks the Treasury forecast. That comparison is really interesting, because the Netherlands has about three or four times the staff to do the forecasts as the OBR here in the UK, and comparable departments at the IMF, the World Bank or the Bank of England certainly have vastly more macroeconomists involved. Do you feel like you risk falling into the gap between two scenarios? You have been asked to do a really big task with much less of the resource. Does that lead to some of the volatility and unpredictability of the forecast process itself?

I don’t think it leads to volatility in the forecast process. It is certainly the case that the OBR has fewer macroeconomists to play with than, for example, the Bank of England. I remember that when Charlie Bean’s first appeared before this Committee, he was asked whether the OBR had a sufficiently large staff to produce its macro forecast. He said, “Well, of course I had many people to do this at the Bank of England, but you can produce a perfectly good macro forecast with 12 people.” Little did he realise that we actually only had six. It is unusual, because the OBR needs to produce a different sort of macro forecast from most of the ones that are produced outside—much more focused on nominal variables, or in other words cash amounts, because that is what drives tax receipts and so on. I do not think the OBR is particularly understaffed in that respect. Clearly, anybody in the chair would always like more resources to be able to do more and different interesting things, but I am not sure that doubling the size of the OBR’s macro staff would make a great deal of difference to the performance or volatility of the forecasts that are produced. You make a comparison with the Netherlands and the size of the institution, but the key difference is that when the OBR produces its very detailed, bottom-up forecast of revenue, spending and financial transactions, all of which then meshes into what you say about budget balances, debt and so on, it draws on what we counted up in my day to be 125 full-time equivalent people in places like HMRC, DWP and so on to feed in the material. At the end of the day, the OBR always made the judgments and was deciding what was put out, but you were relying on the kindness not of strangers but of friends to pull that together, whereas the larger fiscal councils, of which the Netherlands, the US and the Koreans would be the big ones, are just doing much more work in-house. The fact that you have the model of relying on the experts in the tax field and in welfare, for example, is an artefact of the way in which the OBR was created and originally took some functions out of the Treasury. I do not think the size affects the volatility very much.

Sure, but in those cases you are still relying on HMRC and DWP experts and so on to input into your forecasting. There is still that shared task.

Exactly. As I say, at the end of the day the OBR owns the forecast, and if the OBR takes a view on what explains the trends in a particular part of a disability budget, the Department is perfectly entitled, if asked, to say that it takes a different view. It is a very good system. If you are looking at what you can learn from the latest output data on North sea oil production or capital gains tax, making use of that expertise and the people who are involved in the maintenance of the tax system, as well as the design of it, is a strength of the system, not a weakness of it.

This is a question either to Sir Robert or to Mr Hughes. One of the initial intentions for the OBR was to produce more accurate and credible forecasts independently than would be produced under the Treasury. Looking back over the 15 years since its creation, has that happened?

Just to reiterate Robert’s point, from day one 15 years ago the OBR produced what I think is still one of the world’s best forecast evaluation reports, which transparently reports how its forecasters perform from year to year, both economic and fiscal. It has been copied by lots of other fiscal councils around the world and now emulated by the Bank. When I arrived, I commissioned a backward look at the OBR’s forecasting records since 2010 and how we compared with the Treasury, internationally, and with other domestic forecasters. What we found was that looking from 2010—this was up until 2023, when we did the report, but we now keep it up to date—in all successive forecast evaluation reports, our forecasting record compares favourably with other forecasters in the UK and other official forecasters around the world. We are beaten only by Belgium in our year-ahead forecasting accuracy for GDP among European official forecasters. We were both more accurate and less biased than Treasury forecasts in the first three years of our forecast horizon.

But that reduces over the longer period. You are better in the first three years, and then you are similar to the Treasury over a longer period.

Better in the first three years. We were wider of the mark in years four and five, which drove further work on our forecasting infrastructure during the rest of my tenure. The reasons were reasonably well understood by the time we got to the end of the process, and to some extent reasonably well understood at the time. On our economic forecast, it was because we were overestimating productivity. There has been a consistent downward adjustment to our productivity forecast. The OBR has been around during much bigger shocks to the economy than during the Treasury’s forecasting record. You had the financial crisis, covid and the energy crisis—a succession of very large shocks to the economy and public finances. If you do not adjust for those, it looks like your forecasting record is worse, but in fact you have just had a tougher time trying to forecast the future. On the fiscal side, you had the obvious consequences of the fact that if you overestimate productivity, you are going to overestimate revenues. That will mean that you have a larger deficit because revenues are higher, but it also revealed a structural problem with departmental spending. Especially in years four and five of the forecast, after a spending review period has come to an end, Governments tend to be far too over-optimistic about their public spending plans, which always get revised up when the time comes to do a spending review and an allocation. That is what has informed our discussions about strengthening the OBR’s role in forecasting departmental spending.

Over the first three years of the forecast, you would be more accurate than the Treasury was previously. But over year four and year five, at the end of the forecast period, you would be less accurate than the Treasury was previously. Some of that is structural, as you said, and about the changing era that you were in.

I would expect that to improve now that the OBR has a more central forecast—what we thought was a more central forecast—for productivity, as well as being able to take its own view on the path of departmental spending. This gives it the option of putting in something that it thinks is a better reflection of the actual cost of Government policy.

But overall, does that mean that when assessing a five-year forecast period in the 15 years of the OBR, compared with, say, the previous 15 years of Treasury forecasts, the OBR forecasts have been more inaccurate at the end of that five-year period, compared with the Treasury previously?

You have to adjust for the environment in which you are operating. The OBR was just facing much more significant shocks than the Treasury was. I was in the Treasury in the 2000s, and forecasting the macro economy was easy. It was always between 2.5% and 2.75%, up until 2007. It did not take very much dexterity to think about what GDP growth was going to be on average over that period, and inflation was low and under control. We then had periods when GDP went down by 25% and inflation went up to double digits. So you have to bear it in mind that life since 2010 has been rather different from life in the 2000s.

That sounds as if the OBR was created to solve a problem that did not really exist in the 2000s, because, as you say, those forecasts were so stable and credible.

I was not in the country at the time of the creation of the OBR, but its creation was designed for us to have more transparent and credible forecasts. Even when the economy was growing relatively strongly, the Treasury had relatively over-optimistic forecasts for trend growth. It also tended to have over-optimistic forecasts about where the public finances were going to head, and insufficient analysis of the risks to the outlook, which turned out to materialise in spades in 2008. So part of setting up the OBR was an attempt to get more central and credible forecasts, but also much more transparency around what the actual ingredients of putting the forecast together were.

On the problem you were trying to solve when the OBR was created, I think the view then was that when Gordon Brown was Chancellor you had a succession of over-optimistic forecasts for the budget deficit, and that was more because of judgments that went into the fiscal forecast than what you thought about the macro forecast. On the macro forecasts, in a sense there is no reason to believe that the OBR will be any better or worse than anybody else; they are all based, essentially, on the same publicly available information. The main reason why there was over-optimistic forecasting in the period prior to the financial crisis was the judgments that went into the financial sector corporation tax forecast, which—as the IFS, where I was at the time, pointed out—was persistently over-optimistic and hard to justify on the basis of where you thought things were moving in the medium term. The disease that you were trying to cure was more in the less visible, less comparable judgments that went into the fiscal forecast than in the view you took of the macro forecast, where, yes, you might say that trend growth was pencilled in at slightly too high a rate, but that is an issue on which reasonable people can differ. On your point about the three to five years, I come back to the fact that I am very wary of the idea that the OBR should be put in there just because Governments change their mind and push up public spending in years four and five before you get there. The OBR is there to say, “What are the consequences for the economy and the public finances of the current Government’s current policy?” You can certainly point out, particularly when you are talking about risks, that if you look at the past, the numbers in years four and five always tend to be changed. Those are policy changes. When you are thinking about public expenditure or DEL expenditure in public services, it does become more of a forecast when you are in the very near term. Once you get to this stage of the financial year, the out-turn for DEL is essentially a forecast, not a policy choice, but in years four and five it is. So it would have been a problem if the OBR had plugged in those sorts of spending numbers ex ante, because that was not the policy ex ante and that is not the job that Parliament gave the OBR to do.

Sir Robert, you have mentioned that the underprediction of borrowing and the overconfidence in departmental spending limits being stuck to is a feature of the system, not a bug. But do you not think that there is a role for more scenario planning, so that the OBR is able to produce credible forecasts of departmental spending, not just the forecasts that the Departments give you?

There have been various suggestions around that. For example, should you assume that DEL spending evolves in line with what would be necessary to maintain a constant quality and quantity of public services, or what would you need to do to deliver a particular set of policy priorities? Doing it in that fashion is an interesting exercise, and probably an exercise that is as well done outside the OBR, for example with the public services tracker material that is produced by the Institute for Government. I do not think that you want to go down the route of essentially saying that the job of the OBR is to predict what future Governments will do in terms of policy, or to say what future Governments should do in terms of policy. I think you can say, “Confronted with a situation like this, what do we learn from how Governments have behaved in the past?”, and that can provide you with a scenario. A classic case, away from spending, would be that it is Government policy to raise fuel duty in line with inflation, but it is never Government action to raise fuel duty in line with inflation—just explaining what that difference is. But I would be very wary of going down the route of the OBR saying what spending should be, or saying, “This is how Governments are going to behave, so that is our central forecast.” The OBR and others could say things in that space, but I think it would drag the OBR into a very politicised space, rather than saying that, at the end of the day, DEL levels are a policy choice that the Government can take; they historically change them once they have made an initial judgment, but you carry that and explain it, rather than trying to get ahead and say, “Well, we think that in five years’ time the Treasury’s in-year DEL target will be x.”

I am generally in favour of scenario planning when it is within the OBR’s mandate. Highlighting sensitivities and trade-offs is, I think, what we have always tried to do in EFOs. Where we think that there is particularly uncertainty around a particular item of Government spending or revenues, it is always useful to illustrate that, and also to illustrate patterns of behaviour. The classic one has been on fuel duty. The Government always say they are going to index fuel duty. We always have to put that in our central forecast, but after advice from this Committee, we have also shown what would happen if that did not materialise. On departmental spending, one point to underscore is that, if you look at the report we did on our forecasting record, in recent years 100% of the borrowing miss in years four and five has been departmental spending. All the other changes are basically offsetting in other items of spending and revenue. It really was what was explaining a lot of the error in the OBR’s forecast. That is an artefact of what Robert was saying, which is that we are a conditional forecaster. We have to take Government policy and put that into our forecast. Other outside forecasters can say, “I don’t believe what the Government are saying. I’m going to put something else in this forecast, because I believe that that is what year five will actually look like.” The OBR does not have that choice, so it has to rely on scenarios, sensitivities and patterns of behaviour as a way of saying, “Look, this is what is in our central forecast, but these are the risks around it.”

I want to ask about measuring the growth impact of supply-side measures. As you alluded to in the previous discussion, Andy King said that the forecasting of the OBR acts as a guard against “politically motivated wishful thinking”. I wonder whether you think we have a new problem of process-driven, short-term thinking, in which a Chancellor, rather than being incentivised to do what they believe will drive growth, looks instead to what the OBR would assume will drive growth. What do you think of that?

I think you would have a bigger problem if the OBR did not pay any attention to the growth effect of policies. That is the attitude that some forecasters take—not most, but some. They just say, “Look, we are not going to take account of the economic effect of any Government policy. We have a macro forecast. We are going to layer tax and spending decisions on top,” and they just cost or yield what they cost or yield: “We are not going to think about the interactions. We are not going to revisit our macro-economic projections in the light of that.” That has never been the OBR’s approach. From the very beginning, when Robert set the OBR up, the OBR has always very transparently accounted for the demand-side effect of policies through the use of its fiscal multipliers. Also, the OBR did from time to time make adjustments to the supply side to reflect the impact of policy.

I am not suggesting that the OBR should not do it. Reflecting on the period in which you were in charge of the OBR, do you think that that has happened? Have you observed that Chancellors have been incentivised to do something different that is not good for long-term policymaking? This is not about picking on the OBR, but about a relationship and dynamic that has been created. Do you have reflections on the dynamic that has happened? Do you think that that is a fair assessment, rather than it being a question of whether OBR should be scoring it?

I think I had sort of the opposite experience. In 2022, we decided that we would be more transparent about how we accounted for the supply-side effects of policies. That was partly because Governments were talking a lot about the growth effects of policies, and they wanted to understand the implications—partly because the OBR just believes in being more transparent about the way it comes to judgments, and partly because we felt as though the OBR had an obligation to policymakers to be predictable about how it treats things, because Chancellors do not like being surprised when they suddenly discover that we think differently about a particular employment policy or tax measure. We set out a framework that said, “Look, we are going to be more transparent about how we think about the effect of tax policy and spending policy on employment, investment and productivity.” Our experience in the first few forecasts was that we were overwhelmed by policy initiatives from the Treasury and the rest of Government, as a way of trying to get things through the system. To be honest, when we looked at them, the growth effect of most of those policies was very small: it was far less than 0.1% of GDP on the level of output.

In the forecast period? What about in the longer-term period?

It was 0.1% of GDP by the fifth year of the forecast—on the level of GDP, and not on the growth rate. On the growth rate, it is not just a tenth of a percentage point, but something even smaller than that. When we look back, in the end we had taken account of 19 policies that we thought would have some kind of effect on the supply side, where, according to the criteria we had established, there was a reasonable basis for reflecting those in our forecast, based on the evidence that you could find for their effects, and the fact that they looked additional to the forecast that we had. But then we thought about the materiality of those. We ended up putting in a threshold of 0.1%, as a way of saying, “Look, we are not interested in being the arbiters of Government growth policy. What we want to do is get our forecast right. If there are some measures that are of sufficient magnitude that might move our forecast, we will take explicit account of them. If not, we just assume that they are in the baseline outlook for the economy and public finances.” Once we put that significance threshold in there, the number of measures that actually mattered to a significant degree went from 19 down to six. The vast majority of the measures that came through were very small labour market measures, and a few measures on the tax and welfare side, that were making a difference of a few thousand people to a workforce of 30 million. In that sense, disproportionate energy was being put into feeding into an OBR process that could actually have been better devoted to trying to improve the growth outlook of the country, which would show up in the data and in our forecast when it materialised. For us, as macro forecasters at the macroeconomic level, it was really about trying to make sure, when you saw very big changes such as the big increase in public investment, planning reform, the freezing of thresholds or the increase in employer NICs, that those measures that have material impacts in the near term and medium term on labour supply decisions and business investment decisions were captured and reflected in our forecast, because otherwise it would have been wrong. Many of the other things were too small for us to even know whether they worked in out-turn, because you would not know whether you could observe them in the data or not.

Sir Robert, have you anything to add?

The only thing I will add is that there is inevitably an asymmetry here. Certainly in my period—I suspect that this may have been even more the case in Richard’s period—you had, during the course of each forecast process, something called the indirect effects meeting, which is when the Treasury would bring a series of papers saying, “One of the measures that we are considering in this budget is x, and we think that this will have the following beneficial effects to the economy.” I do not think that in my 10 years a paper ever came saying, “A past policy that we implemented is not working out quite as we anticipated,” or, “We have done something that has negative effects here as well.” So there is an asymmetry built in. The more you take on what in the end, as Richard says, can be relatively small changes to an underlying number, say for productivity growth, around which there is an enormous amount of uncertainty anyway, the more you will see the things that the Government hope will push it in one direction. They will chuck as much as they can at the wall and hope that it will stick, whereas if you were being fully transparent and symmetric about it you would want to go into a lot of detail around things that have previously been announced. That would mean looking at things like where planning reform has actually delivered and whether the change in welfare reform actually affected labour supply in the way that was originally anticipated. That could result in an enormous amount of work that ends up not moving the dial that much on your inevitably uncertain forecast for productivity growth.

On how you assess the impact of public investment, there has been criticism by bodies such as the IPPR that you do not capture second-round dynamic impacts in the way their modelling does, regarding how public capital crowds in private investment. I should declare that I used to work for the IPPR. How do you respond to those criticisms? Do you think that they are fair? Do you have the appropriate resources and capacity to do accurate enough forecasting?

When we scored the impact of the increase in public investment in the 2024 Budget, we assumed that £1 of public investment catalyses 30p of private investment. That was drawn from the literature—both in the UK and in other jurisdictions—and was a central estimate based on work by people who have done empirical studies of these things. However, there is uncertainty about these numbers, because you never quite know whether the present looks like the past—where these studies inevitably draw their data from—so we also looked at alternative scenarios. One was where every £1 of public investment catalysed £2 of private investment, as well as one where public investment was a substitute for private investment—so every £1 of public investment meant 50p less private investment. That was because there are sectors where Government investment is just a substitute for private investment. If the Government build a house, that might be one less house that private builders need to build to accommodate people living in the country. Part of the challenge we have as forecasters is that the Government just announce one big number for public investment, and at the time they did not say where they were going to spend it. You have to assume a blended average of all public investment. Some of that will be good for growth and some of it will be good for delivering public services but be fairly indifferent to growth. We thought that it was important to do scenarios around that. We ended up having 30p for every £1 as a central estimate. Looking at what has happened to public investment in the UK over the last 10 years, it has grown from 1.5% to 2.8% of GDP, so it has nearly doubled as a share of GDP. Business investment was flat during much of that period, so I do not see a lot of evidence in the data for a catalytic effect from public investment driving several multiples of what it is investing in the private sector. Maybe that will be different in future; that is why we did alternative scenarios. We thought 30p for every £1 was a reasonable central estimate.

In a lot of these areas in which policies are aimed at driving supply-side change, they actually take quite a long time to deliver that. The amount that will show up in a five-year forecast is relatively limited. Part of the reason is that even if you do things that change the flow of investment in the public sector—the public capital stock—or the flow of labour into or out of the country, what in some senses matters to the underlying productivity position is the stocks of those things. You might not be making that big a difference, even with a relatively successful policy on flows to your stock of public sector capital, housing or the size of the working population. You would expect some of those things to come in relatively slowly. Any Minister or Chancellor with confidence will say, “Well, this is the right thing to do and the fact that not much is showing up over a five-year period is no reason not to proceed.” I certainly hope that that would be the approach they take.

Going back to where we started, to summarise what you were saying earlier, are Governments too focused on these tiny policies that they think might move your assessment, rather than the more substantial policies that will move the dial on growth? Is that your assessment?

I do not want to criticise Governments. They should do whatever they can to get growth going because it is definitely a problem for the country. On where they focus their energy, we try to ensure that in the design of the system, when creating a forecast, we are only focusing on things that are material and are not just looking five years ahead, but longer term. Wherever we had a policy that we think would have a bigger impact after 10 or 30 years, you could see that in the five-year forecast, but we also always provide an additional projection—be it for public investment, planning reform or other areas. As Robert said, if you want to affect the growth trajectory of this country, you have to move some very big stocks. That is 30 million people in the workforce and 200% of GDP in the capital stock. Typically, Government policy operates on flows. To see the effect of just adding a few tens of thousands of workers back into the workforce from a welfare-to-work scheme or just investing an extra 0.5% or 1% of GDP in public investment, you need to be patient. You either need to do something very big to affect that stock, or you need to do something small, but be patient and wait while that flow makes some kind of meaningful difference to a large stock of workers and a large stock of capital, in order to make a meaningful difference to the growth potential of the economy. We tried to give the Government the best sense of what we thought, from a forecaster’s point of view, was material in the near term, but also what could have an effect if you looked beyond and were patient enough to get to the long term.

Sir Robert, I will just come back to you on a couple of points. I completely agree with you that a lot of policy effects on growth take longer than five years to materialise, but I do not share your optimism on the political ease to make this change, if those effects are not showing up in an official forecast. For example, Draghi has just called for the EU to borrow to invest in defence R&D, arguing that that will actually lower debt-to-GDP ratios over the long run, because growth will rise faster than the debt. How do you solve this five-year issue? Do you think it would be a lot easier for the Government to adopt long-term economic policies if the OBR were putting out both a five-year and a 10-year forecast, where a lot of these effects could materialise?

Looking at some of the impacts of, for example, a relatively substantive change in the share of GDP spent in capital stock, the OBR is already producing relatively stylised 50-year projections in which you can get a relatively informed view. If you were to move the five-year horizon to 10 years, you would obviously get some more of the effects of policies that you need to be patient about showing up. In the US, the Congressional Budget Office does projections over a 10-year period. There was a period in which a Chancellor under my watch did think about and warned us that we might need to be ready to move to 10-year forecasts, but in the end they decided not to. One of the challenges with moving to a 10-year forecast is that the OBR would then have to require the Government to be explicit about their policies on a whole range of payments, tax thresholds and upratings over that 10-year period. My sense was that one of the things that got in the way was the reluctance of a Government to set out all those details over the 10 years, so as to provide the underlying basis of a forecast that might then show the greater impact of a particular growth measure. In terms of major changes and substantive persistent changes in, say, the amount you invest in public sector net investment, what you can learn from the 50-year projections is probably the best bit of information to get. That underlines the importance of the Committee and others focusing on those projections, and not just the five-year forecast.

I will just come back on your point about forecasts being completely fine, small teams, resources and top-down forecasts. To be clear, I was staggered in previous Treasury Committee sessions with the OBR that the basis, as I understood it, for the productivity forecasts over the past 14 years or so has essentially been that at some point productivity will return to the pre-2008 level, or a 2% haircut of that. It seemed pretty finger-in-the-air to me, to be frank, for a G7 economy. If you look at the last Budget, so much of the political debate around that was driven by the £16 billion productivity downgrade. Again, the basis for that seemed to be fairly finger-in-the-air. I struggle to understand why having more resources and doing bottom-up forecasts is important, when there are stacks of economic research over decades about how productivity is generated. Would that not be a better way of doing it?

With producing a medium-term productivity forecast—we are not talking about investment or the flow of labour supply—there is, to be frank, a limited amount of science that you can bring to these sorts of things. The challenge that we had—certainly during most of the period for which I was in the role—is that, prior to the financial crisis, we had just become used to productivity or the economy growing by 2%, and it had done so for decades. That was followed by a dramatically weaker performance. You should not mischaracterise the decisions that we had to make through a lot of that, which were essentially about how much weight we placed on the relatively recent weak period versus the evidence that, over a much longer period, these numbers had come out relatively strongly. You are right: I think I must have presided over four reductions in the underlying productivity forecast during my time, and most of them were based on the view that, “Now we’re getting more evidence that this is not directly related to the financial crisis. It looks like it’s consistent with global development, and it looks like it’s not financial sector or crisis-specific.” We were making adjustments on that basis. That, I am afraid, is the territory for that sort of area. One of the things you are looking at is how your forecasts compare with others. More recently, the OBR moved its forecast to be more consistent with those taking a different view, and that is reasonable.

Do you think I am being harsh or unfair in describing the OBR’s productivity forecast as a finger in the air?

Yes, I think you are being unfair. Uncertainties around those forecasts confront any forecaster that is producing a productivity forecast, particularly one over five years.

Mr Hughes, you were not here when David Miles, a current member of the Budget Responsibility Committee, described what happened, but what is your perspective?

Productivity assumptions always tend to be the least understood and modelled aspect of any macroeconomic forecast. One positive development during my tenure was something that happened at the ONS. When we put together the supply side of our forecast, you have a stock of labour, a stock of capital and then total factor productivity, which is the efficiency with which you combine those two things. It is only recently that the ONS has produced an estimate for the stock of capital, and that has allowed us to isolate that third residual term, which is total factor productivity. In everybody’s forecasts, that involves a certain amount of future-gazing and trying to think about how innovative the future is going to be compared with the recent past. It is a residual in all the growth accounting equations after you have accounted for growth in labour supply and the stock of capital that builds up over time from investment, and it is very susceptible to shocks. During Robert’s time, the big question was, “Is TFP ever going to recover from the shock of the financial crisis?” During my tenure, we had been hit by the pandemic and then by the energy crisis, and we knew that both those things were going to be bad for TFP, but that at some point it might recover. We ended up splitting the difference between the past and future. In the most recent downgrade, the conclusion that we came to was that it did look as though the impairment of TFP in the UK economy was not due primarily to the effect of that series of unfortunate events, but more to some underlying structural shifts that had been happening. That includes the UK economy being more closed off to trading with Europe and more recently the US, due to tariffs being introduced by the US Administration, and demographic changes which look to draw more employment and labour supply away from higher productivity employment into lower productivity employment, and as a result, change the composition of the workforce over time. For those two reasons, we made a decision to further downgrade our TFP assumption. That now brings us back into line with the Bank and some other forecasters, but it is something that everybody keeps under review, because it is the least understood aspect of what drives growth these days.

Finally, it seems very restrictive for policymakers that so much political debate is around a single point forecast that is five years out, with these sorts of crazy, tiny variations driving so many things. How would either of you reform the OBR to have less emphasis on a single number being put out by it?

Fiscal rules in the UK and the concept of headroom have been around since the late ’90s. When Gordon Brown had fiscal rules, there was talk about whether he was on track to meet them or not, and the same was true of George Osborne’s fiscal rules. The big change that has happened in recent years, and why we are talking about headroom so much, is that there has been so little of it. The reason why it has driven changes in policy is that very small perturbations in the OBR’s three to five-year forecast meant the difference between meeting or missing fiscal rules—and not big shocks, just small changes in gilt yields and the take-up of welfare benefits. The reason why it became a thing that drove policy was not that there was a rules-based fiscal framework. Every country in Europe has a rules-based fiscal framework; we have had one since 1997. It is entirely an artefact of leaving yourself a tiny margin against whatever rule you set for yourself and the Chancellor’s unwillingness to see that rule broken. If they were tolerant of a near miss and said, “Look, it’s a near miss, I’ll make it up later,” people would not focus on it so much. But because Ministers treat it as a binary pass or fail, it also raises the political stakes around having that fiscal rule.

I think that is true. If you go around saying things that are iron-clad and non-negotiable, it is not surprising that people look very closely at the degree to which you think you are likely to achieve that. Again, the fact that the OBR produces forecasts and uses them to assess how close the Government are to achieving the fiscal objectives they set out is literally the task that Parliament gave the OBR. Right from the very start, despite the fact that that was the job that we had been given, we put an enormous amount of effort, which Richard then took on afterwards, of explaining the uncertainty around the central forecast. We explained the uncertainty about the likelihood that a central estimate of a headroom of 20 actually meant that you could make judgements about the likelihood or probability that you would hit the target or not. We used past forecasting error-scenario analysis, looking at the sensitivity to key parameters to try and take people away from the idea that, “This is how things are going to work out and that’s the difference between that and the target.” Richard is right: the reason it has become a live issue, in the sense that the headroom is driving policy decisions, is that at the end of the last Conservative Government the headroom was reduced to a considerably lower level than it was through all the previous periods when the OBR and fiscal rules regime had been there. Then, when Rachel Reeves came in as Chancellor, she did not use the first Budget to rebuild that margin. As a result, you are in a situation where the forecast does not need to move around very much before you hit the borderline of the rules. That is a recipe for policy volatility.

You have both given a very reasonable assessment of the theory of the OBR and how it should work. You have pointed to the medium-term and long-term assessments that you make and the fact that you are just trying to do what you would do in most jurisdictions, which is give a dispassionate assessment of whether the Government’s policy is working. The reality is that if you look at any investment in education, infrastructure, energy policy, or planning reforms, those cycles of delivery are much longer. In a world where we have the level of debt that we have now, the volatility around the cost of the debt, a legacy of the state growing quite significantly with expectations of what the state will do, and demographic pressures, the sensitivity of the political cycle does not mesh well with that calm, dispassionate assessment that you are making. I have heard what you have said about the five and 10-year horizons, but as we are now—with the level of debt we have got, the size of the state, the growth trajectory we have had over the last 15 years—how could we design the OBR differently from a starting point of today so that it would be a more effective, dispassionate commentator on Government policy, rather than an absolute arbiter of it with the authority that it seems to have? In the run up to the last Budget, we saw an extremely tense apparent stand-off, with alleged briefing between one entity and the other. Perhaps Sir Robert could answer first because this is an opportunity where you are not talking to George Osborne and his chief of staff Rupert Harrison. If you were starting now and talking to the Opposition or Government about what they can do in the future, how would you design it differently?

Implicit in your question is the issue that if you have a situation where the way in which political decision-making is taking place interacts with the provision of, as you have described it, dispassionate analysis and assessment of what the outlook is, the solution is not to get rid of the dispassionate analysis and assessment.

I am not saying that we should get rid of it. I am just asking whether you can refine it so that the interaction is better—because it is not now, is it?

Sir Robert, the point is that you were there right at the beginning, designing it and setting it up. If you were there, as Sir John Glen— sorry the right hon. John Glen—although I am sure you deserve the title.

I don’t mind the slip.

It’s only a matter of time.

As John Glen says, you were there at the beginning. If you had a free hand now, and you were sitting down with a Chancellor and saying, “Right, this is the Office for Budget Responsibility,” how you would design it? We touched on this at the beginning; you said not much change, but there is an open opportunity for a new blueprint.

I do think, fundamentally, that the model of having a professionally based, dispassionate analysis of how the public finances are likely to evolve under current Government policy is fundamentally the role of the OBR. We are not a commentator or a giver of views as to whether policy is right or not. The aim is for it to be a technocratic exercise in producing the best analysis you can, on the basis of all the available information and what the Government tells you about what their policy decisions are. The framework and the application of the framework are what need to be changed, not the value of having somebody providing that sort of dispassionate analysis.

My purpose in asking that question is not to dismiss the OBR. I have deep respect for you, for your successor and for the theory of this. I am just saying that it is quite obvious that the way politics works, the way the OBR’s documents are pored over to look for any discrepancy and the way those numbers have to sync up over the short term rather than the long term will militate against any Government having the strength to take the long-term decisions that are likely to elevate the level of growth. I am desperately trying to find a way to improve this. Nobody is disputing the theory of the OBR, but it is not working in the way that was probably envisaged, and as it probably did work in earlier seasons, because the public finances have been so different over the past five years.

That is primarily because the Chancellor had put in a much higher level of room for manoeuvre against the targets they had set, so that there was less political pressure to change policy when, as is inevitably the case, forecasts moved up and down because of new data, new financial market movements and so on.

The average headroom from 2010 to 2024 was, I think, £25 billion. I totally acknowledge what you said about the last Conservative Budget, and that was not replenished. But we are also seeing a situation in which the level of debt, post covid and the energy crisis, and the cost of that debt mean that the public finances—notwithstanding that the state does more than it did a generation ago—create that problem whereby there is little room. Public expectation and political cycles require Governments to do things. In practice, there is very little room to do much at all, because you are so hemmed in by that short cycle. Your answer seems to be, “Have a high level of headroom,” but I am saying that the lack of emphasis on that 10-year cycle surely militates against effective interventions and continuity of policymaking that is probably in the national interest.

With a 10-year horizon, if the Government are willing to articulate what current policy looks like over it, you could certainly have a richer discussion, as in the US case. As you say, it is not about what will be done in the course of the current Parliament, but about what this looks like as a medium-term programme and so on. But the fact is that we still have a budget deficit at the level we do. In terms of the reaction of financial markets, the underlying concern is that we are in a dynamic now—and people probably suspect this across the political spectrum—that when there is bad economic news or bad news for the fiscal position, the goalposts get moved, and when there is good news, it is spent. That results in an upward ratchet in the level of the debt to GDP ratio, and probably in greater concerns that you are hitting ill-defined inflection points where the market will take a dimmer view. If you want an environment that is more conducive to taking and sticking to long-term decisions, persuading people that you are not just going to spend any good fortune that comes down the way and ignore any weaknesses is probably a good place to start.

Throwing it back to the politicians!

If you want to get policy-focused on the long term, you have to look to the long term. The OBR produces a document that allows precisely that kind of discussion and debate: it is the risk and sustainability report. It does not just look 10 years ahead—

It is not the one we have, though, is it?

We don’t have it, for several very good reasons. We have one hearing, in this room, once a year on it; there is no Government policy response and no parliamentary debate. It gets nothing like the focus of the EFO. However, if you want to show a difference to the growth trajectory of the country or to the fiscal trajectory of the country from different changes, that provides the mechanism to do it. It is the only way to capture things like changes in the state pension age, changes in the productivity of the health service, the impact of demographic changes on the productivity of the economy and on the tax take, and the impact of things like falling fuel duty from the electrification of vehicles. If you want to get ahead of those long‑term trends, it provides you with a basis for looking at the impact of those choices and assessing trade‑offs. It is particularly important at a time early in a Parliament, when Governments are setting fiscal rules, to think about whether those rules are consistent with sustainable public finances over the longer term. It is a report not just about sustainability but about risks; it also highlights the risks that might hit the fiscal framework over the course of a Parliament or several Parliaments. It is a really important piece of work. I know that it gets a lot of attention from this Committee, but it gets nothing like that attention from Government or from the rest of Parliament. If you want that perspective on fiscal policymaking, you must draw much more attention to that perspective in the way in which discussions had around it.

Would you welcome a parliamentary debate on it?

Absolutely.

Sir Robert, would you welcome that?

Yes, indeed.

It is a bit like the Whole of Government Accounts, which are exciting to a small group of people in Parliament—I am one of the few—but which are never debated in the Chamber.

Could I go back to the question that I asked Sir Robert about the design of the OBR? At the point we are now at with the public finances and the situation in which we find ourselves, as if the OBR had not existed hitherto, is there anything you would suggest, Mr Hughes, that would be helpful?

Much of my time at the OBR was basically watching the crystallisation of a set of risks and sustainability problems that had been highlighted in previous risk and sustainability reports.

And not acted upon? Richard Hughes And not acted upon—and then observing the consequences for the fiscal situation in which we find ourselves. The more we can do not just to focus on the central forecast over the medium term, but on the pressures on the public finances of the long term and the risks around those scenarios, the better fiscal policymaking will be. You may get people from other countries talking about how fiscal policy is made. It is a huge focus of how they do things in the Netherlands when they set fiscal rules and produce fiscal reports: they very much focus not just on the central forecast, but on the risks around it. In the end, most surprises that Governments face tend to be bad ones, especially these days. If you do not take account of them, you will always end up with this upward drift that Robert talked about, of deficits and debt going up. As Robert says, when good things happen, Governments have no problems spending the money, but when bad things happen, they struggle to offset them.

Thank you. I think I have taken enough of your time.

Can I push further on Mr Glen’s point?

I think we have done it to death.

It is about the 10-year horizon. Sir Robert, you mentioned that one of your Chancellors considered asking you to make this change, but the political concern was that they would have to reveal their policy preferences over 10 years. Given that the OBR has a 50-year projection, which involves keeping the current policy direction constant, could that not be done in a similar way over 10 years?

You could do it on that basis. As you point out, in the absence of the Government saying what their policy is on lots of things over a 50-year period, you have to make stylised assumptions. One key assumption is that tax thresholds, for example, rise in line with earnings over time, so that you do not have, essentially, an explosive or deeply shrinking path for receipts as a share of GDP built into that. You could make those sorts of assumptions over the second half of a 10-year period. I think the US Congressional Budget Office has something that is relatively more stylised, but its projections are not quite so rooted in a bottom-up forecast of things like tax receipts, which require you to reach judgments around the levels and rates and so on. The danger of doing a 10-year forecast with five years of concrete policy and 10 years of stylised things is that it will end up being neither one thing nor the other and therefore not producing a particularly good guide. If you went to 10 years, you would need the Government to be explicit about most of the things that they already have to be explicit about over five years for it to be meaningful and useful, particularly if you are trying to send any messages about credibility in how the Government will behave beyond the next election.

Can I chip in and get a quick answer from both of you? Presumably that could be quite political, because it cuts across or over an electoral cycle, so there could be attacks on the Government of the day for saying something that will happen in 10 years’ time, on that basis, that other parties could then criticise and run against. Would you be worried that that would lead the OBR to be dragged into the politics, more than it already is?

I would not disagree. I think it also explains why that has not happened: it may be that reluctance to have a need to set that out with that degree of specificity.

I think it is a matter of degree more than kind. If you are in the fourth year of a Parliament, you are forecasting mostly for after the next general election anyway, so I did not find it a particular problem during my tenure.

I appreciate that you both said, in response to Mr Glen’s question about the framework, that as technocrats you operate within the framework set by Government, but neither of you currently runs the OBR and neither of you is bound by civil service impartiality. You are free to have political opinions before this Committee, as much as any of our other non-civil service witnesses. Have you any comment, for example, on the IFS’s traffic light system proposals? It suggests that, rather than having specific numbers that come out in the headroom discussion and a pass/fail, there be a broader assessment about sustainability in a red-amber-green fashion. It is much more a narrative format than a pass/fail format. What do you make of that suggestion? If you dislike that, are there other suggestions that you have seen from other fiscal frameworks that you think would make the whole system healthier?

I have been working in fiscal policy in the UK and around the world for 25 years. The one thing I would point out is that the UK is not unusual at all in its fiscal rules, which are very similar to those you see in almost every other advanced economy. It has some target for the balance and some target for debt, and they are typically expressed over a three to five-year horizon. Sometimes there are carve-outs for some types of spending. At the moment, capital spending is carved out of one of the rules. In Europe, there is discussion of carving out defence. What is really unique about the UK’s fiscal rules is how frequently we abandon them. The average shelf life of a UK fiscal rule in the last 10 years has been 1.8 years. No single fiscal mandate has been around long enough to actually be measured in outturn. Our biggest single problem in this country is that we do not stick to any of the rules that we set for ourselves. We should not be surprised that we have one of the highest deficits in Europe and a rising level of debt, because we are constantly changing our fiscal rules rather than trying to meet them. My plea is to set fiscal rules, try to deliver them and have them be around long enough to be there at the target dates that you set for yourself. This is my experience of looking at more qualitative or colourful assessments of fiscal performance: since 2021, we started publishing the EFO, which, exactly as you describe, is a traffic-light assessment of the sustainability and vulnerability of the public finances. It is mostly amber and red, I should say, because our public finances are not in a healthy position. I never saw any journalist, any member of the public, any Member of Parliament or anyone on this Committee pay any attention to it, because there are a whole bunch of different numbers and it takes a very long time to explain them. In the end, the question you ask yourself is “Which one of these things is the binding constraint on fiscal policy?” Once you figure that out, it is pretty much the number that you focus on, because you know that it is what will drive the next round of decisions. Even our own fiscal framework has three fiscal rules in it. I am not sure that anybody outside this Committee could name all three. There is one for the current balance, there is one for net financial liabilities, and there is a welfare cap. Nobody focuses on the welfare cap. It is important in relation to sustainability, but it is not the binding constraint, so it does not get a lot of attention. My worry about having traffic lights with colours and lots of indicators is that it just becomes an excuse for the Government to target-shop and say, “We’re not meeting that rule, but look at this one—it’s gone light green. It turns out that is the one that we are now much more in favour of.” My fear is that it would be a mechanism to make it more difficult for this Committee and Parliament to hold the Government to account for its actual performance, because the Government would pick and choose what exactly they want to be accountable for. From my perspective, simple rules that you stick to are more important than complicated rules that you pick and choose among. More generally, my heart sinks every time that there is a discussion about setting a new set of fiscal rules for this country.

Your heart must have sunk many times.

It has sunk many times. I would just like to see some of them being met at some point.

Sir Robert, have you anything to add?

Not really. I will just say that a rich set of data and a rich set of things to look at, to analyse the underlying fiscal position, is very desirable. But, as Richard says, a multi-component traffic-light model is simply an opportunity for people to pick the bit of the traffic light that is green and ignore the bit that is red.

I have a couple of questions about the speculation prior to the November Budget. I think most observers saw it as being more speculated about than many past Budgets. Indeed, you wrote a letter to Dame Meg just before the Budget in which you talked about the “unusual volume of speculation” around the pre-measures process. Do you have any thoughts about why that unusual volume of speculation was abroad, as it were?

I do not and cannot interrogate the motives of others. What I would say is that it is not unusual—in some cases, it is desirable—for a Government to have public discussions and public consultations about its proposed policy measures. You design better policy by having public consultations about policy. That used to be the norm in the run-up to Budgets, but it was done in a very formal way. Consultation documents were published and industry was consulted on tax measures, for example. That then fed into the development of policy. What seems to have happened more recently is that this stuff happens through anonymous briefings from sources allegedly close to Government about what policies are under development. That level of informality is unfair to the people who are potentially affected by the changes. It also feeds into the general sense that the process is not orderly or transparent. What was particularly notable about the run-up to the November forecast was just how much speculation there was about the OBR’s pre-measures forecast. The reason that I wrote the letter is that I was concerned about the OBR’s reputation for professionalism and integrity, on several scores. One is that the OBR had somehow been fickle and changeable in the views that it was taking about the overall economic judgments around productivity, earnings and other things, when in fact we had made all those judgments at the outset and were at pains to make sure that we had made all the major economic judgments at the outset of the forecast, so the Government had a reasonable basis on which to plan—including about the downgrade to productivity, but also about the higher earnings forecast, which was a substantial offset, as well as on inflation. There was also reporting in the media that somehow our forecast had improved at the last minute and that it offered some relief to the Chancellor after our pre-measures forecast had closed. That was not remotely true either. I thought it important to set that out on the record and give the figures for the evolution of the forecast. That was an unusual step, but we took it because of the circumstances at the time. Finally, there was also the suggestion that we had somehow colluded with the Treasury in choosing the window for interest rates that we had chosen for setting the debt interest forecast. That was also not remotely true. The decision that we took about the time window was taken back in the summer, long before we would have known whether or not it was advantageous to the public finances, once it was actually utilised. For all those reasons, to make sure that the OBR’s reputation for integrity and fair dealing remained intact, I wrote the letter that I did. I did not end up sending it on the day, because there were other priorities in our communication with this Committee around our own mistakes regarding the pre-publication of the EFO, but at the Committee’s request I shared that letter in its original form.

Do you think that speculation generally is a symptom of the way in which the pre-measures forecast process works? Are there some lessons that you would look for us to learn about the interaction that the pre-measures forecast represents, and how to do it better so that we do not get some of this speculation, which is clearly damaging?

I don’t think it is helpful. Organisations such as the CBI said that it was having an impact on business confidence, and I do not doubt that that was the case. The Government seem to have learned lessons, from the run-up to the forecast that we are expecting next week. I have not seen very much speculation at all about what is in the OBR’s forecast or indeed around policy measures. It seems as though lessons have been learned from that exercise, and I think that is very welcome. In the end, you can have a very formalised process of disclosing information. That is how things work in some countries, where they actually publish the pre-measures forecast well in advance of the Budget, and that provides a framework within which policy is developed and discussed in Government, Parliament and more collectively. That is not the way we run things in this country. Instead, we publish the pre-measures forecast and the post-measures forecast all on the same day. That has great advantages in not dragging out the Budget process and the debate, and allowing people to see the complete picture all at once, but other countries do it differently.

Sir Robert, there was speculation back in the day when you were chairing the OBR, though perhaps not to the same extent as we saw in November. What was your sense of what happened with speculation prior to the Budget, and how did you manage it? Do you have any suggestions for how we can make sure it is reduced and managed better?

Through my period, there was a lot less than one saw in the run-up to the most recent Budget. I would concur with Richard’s view that anonymous kite-flying is no substitute for consultation, if you are trying to think about whether a particular policy idea is a good idea or not. On the sharing of bits of the forecast in advance, there has always been a desire to, at the margin, roll the pitch. If the Treasury or the Chancellor think that the numbers are going to look very different from those that the markets are expecting, then occasionally—George Osborne has spoken about this publicly—you can use methods to try to ensure that those sorts of things do not come as a complete surprise without leaking. When sharing components of the forecast, particularly in a way that does not actually, when the forecast comes out, correspond with the overall picture, the danger is that you are not rolling the pitch but ploughing a field. The main learning from this is that it really did not work very well in the run-up to the last episode. As Richard said, there does not seem to have been a great deal of speculation ahead of this one. I do not think there is anything that the OBR can particularly do to prevent this, but it has been an opportunity for people to recognise that that sort of combination of kite-flying and rather cack-handed pitch-rolling does not do anybody any good in the long term. The other thing is that it sets up a culture. If stuff is brought into the public domain, whether semi-officially or not—I do not know what was official, what was not, who was doing it, etc—other people will think, “If that sort of information is being shared, why can’t I share some as well?” You are setting up unfortunate internal dynamics and incentives as well. I think a return to a presumption of a relatively taciturn, quiet run-up to these fiscal events would be in everybody’s interests.

It would certainly have meant that the OBR produced a forecast in the run-up to the so-called mini-Budget, or at least in the immediate aftermath. Once the OBR had become aware of a policy change of that magnitude—tax cuts of more than 1% of GDP—that would have triggered an OBR forecast. At the time, we would have tried to accompany those policy decisions, or at the very least, do it briefly after. The amount of uncertainty that was prevailing in the aftermath of the mini-Budget would not have existed or persisted to the extent that it did.

It has strengthened the independence, and also the transparency around the public finances. Again, it is not unusual in other countries to have that kind of provision that big fiscal decisions need to be accompanied by an assessment of their impact on the public finances. I think it is definitely an improvement to the fiscal framework, and one that plugs what turned out to be a very big source of vulnerability.

Yes, although I do not have strong views on the fiscal lock. In a sense, with the OBR being required to produce two forecasts a year, and the Treasury and the rest of the Government machine required to support that, you can only run and hide for so long before the forecast process catches up in the end. The idea of not taking big decisions without thinking through their impact is clearly a desirable one. There is also the practical issue that the idea that the OBR can decide, off its own bat, “We need to do a forecast, and we need to do it now,” is not uncomplicated. This is a very interactive process involving a lot of Departments that have their own work schedules of the things they need to be doing in managing and administering different bits of the welfare and tax system. Introducing a system in which you can simply drive an eight-week forecast process to get under way, straight off, is not as straightforward as it might sound to some. But the notion that it is a good idea, when big decisions are taken, to think carefully about what their consequences are is right.

It is not an exact science.

I would never begin to claim that the OBR is a perfect institution, or that perfection is possible in the world of forecasting. In the end, forecasting is about the future, and it is uncertain. One of the things that we have always underscored in our work—this was also true in Robert’s day—is really emphasising the uncertainty of forecasts and the risks around them through any number of tools.

Absolutely. As I said at the outset, and on several occasions, the lack of attention that the risks and sustainability report gets is a serious shortcoming in the fiscal debate in the UK, and the raw material is there to have it. The Committee has the ability to draw attention to these issues, as it has done in important ways in the past when discussing the work of the OBR and highlighting risks, weaknesses and shortcomings that need to be addressed. The OBR jointly serves Government and Parliament, and it is dually accountable to both. It is really important that, where the Committee thinks that there are issues or analysis that needs more attention, you bring that out and provide those changes to the framework and your own mandates to the organisation to help fill them.

The other thing missing is that there is remarkably little debate about the fiscal rules at the time when they are adopted. The charter for Budget responsibility, when it is adopted by Parliament, is one of the most fiscal decisions that any Government take. It tends to not be supported by that much analysis, and it does not get a lot of attention or discussion in Parliament, but as you see, it is an incredibly binding and consequential choice that Governments make, because they then try to live within those rules. We need to have more context and more analysis—from the Treasury, the OBR, or somebody else—on what the implications are of the set of rules that we are adopting, how much confidence we have that they will be met in the medium term, and how consistent they are in the long term. It is important that whenever those rules get introduced, one kicks the tyres as much as one can before going snap on them, because we want them to be around for quite a long time and actually shape decision making. Otherwise, they are pointless.

Not specifically on the interaction with the OBR. As Richard has referred to, it is worth looking periodically at the whole structure of the Budget process and how participative it is—for example, the points at which material is brought and the time that Parliament is able to devote to discussing key issues, particularly spending reviews, which are an important part of the discussion of what the role and size of the state should be over a medium-term period. The OECD has produced various reports. I remember one on the Irish budget system, which said that its strengths were the opportunities for stakeholders to participate and for Parliament to have detailed consideration, draft plans and so on. I remember talking to one Treasury person who said, “Well, they think they’re all the strengths of the Irish system; we think they’re all the weaknesses. We just want a much more Treasury-centralised process.” At the end of the day, for Parliament to decide whether it feels that it has the appropriate amount of traction in the Budget process—in a context in which it is relatively Treasury-centric in this country and has been for many years—is not necessarily the right answer, but probably one that should be thought about periodically.

One thing I can say from my tenure, and I learned this from Robert, is that the OBR works very hard to stay carefully within its mandate and colour within the lines set out in legislation and given to it by Parliament. That mandate is very clear. The OBR’s job is to forecast the public finances, say whether the Government are on track to meet their fiscal rules, and highlight the risks around the forecast. At no point during my tenure or Robert’s tenure did we ever give policy advice to Government or express our own views on policy. We understand that we are given a job as technocrats, and we are meant to inform the policy debate but not be part of it. We were always at pains to make sure that was the case and to not be drawn on policy questions, even when we were invited to Committees like this. I think the OBR does its best job to inform the policy debate, even when that debate does not end up in the place that people would like. In the end we are just messengers who are trying to make sure the debate is best informed.

Finally, a quick one from me, because you have covered this obliquely. We talked a bit about resourcing along the way. Do you think there is enough staff? Obviously there is the IT system; we won’t go over that again. Are there any resources—IT modelling software—that the OBR should have in the modern world? I will start with Richard because he was most recently there.

The OBR has grown to about 54 people. The Treasury has 200 economists and 2,000 people, so it will never be able to man-mark the Treasury as an organisation. In my experience, whenever the OBR was given new responsibilities, it was given resources commensurate with the delivery of those responsibilities by the Treasury. It is unusual that the OBR relies on the Treasury for its budget. There is an important check on that in the form of this Committee. If at any point I felt, as chair, that I was not being given the resources to deliver the tasks I was given or that the Treasury was trying to squeeze the organisation to make it harder to do its core job, I could come to the Committee and say, “Look, we have a problem. We want to provide you with information, but we do not have the resources to do it.” This Committee is an important source of oversight of the OBR’s finances to make sure that it is resourced to do its job. That has always happened, but it is partly because I always knew that I could have come here and it would have helped my case.

That is interesting. The National Audit Office, with 950-something people, is not quite an equivalent, but there is a special body called the Public Accounts Commission, made up of MPs, which determines its budget. It has that budget power and then the Treasury has to cough up. Do you think there could be a benefit to having a further independent step about how the Office for Budget Responsibility budget is set?

I think there is definite benefit in this Committee taking an interest in whether the OBR is sufficiently financed to provide you with the information that you want to hold the Government to account. I should say that, to date, that has always been the case, but it may not be the case in the future.

I would have to go back through the annals of our Committee’s history over the last 15 years, but I am not sure there has ever been the need for the Committee to ask the Treasury for more money, so it has not been tested. From what you are saying, if there was a test with a mad, bad or dangerous Government slashing the budget of the OBR, is there no power held by this Committee or anywhere else to reverse that?

Not other than the power of my successor to raise it with you, and you to draw attention to it, but formally it is a decision made by the Treasury, and the OBR’s budget is part of the Treasury.

In summary, you do not think you are short of resources, but there is no mechanism to absolutely guarantee that you would get resources if they were cut.

I agree. For my time, the OBR was adequately resourced to do the job that it was given. Among other fiscal councils—the other 54 that I spoke to—the two things that they would complain about most frequently were access to information and the adequacy of the budget to the job they were given. So that has been a positive over time. I remember looking at the Public Accounts Commission model when people were talking about what the status of the body should be—whether it should be a non-ministerial department—and my feeling at the time was that the incentives for the Treasury and the OBR were sufficiently well-aligned that most of the time, if the Treasury squeezed the resources going to the OBR, it would make their lives more difficult, as well as the OBR’s life more difficult.

A Faustian pact.

It was one of the reasons why I never had to come to the Committee to make those points. One point on the role of this Committee in addition to funding is that you are one of three key lines of defence against political interference and inappropriate behaviour. You have officials at the Treasury, who are obviously conflicted because they are serving Ministers as well, you have the oversight committee, and you have this Committee. At a time when any institution goes through difficulty, and there is the potential to use that as a pretext for moving things in an unsatisfactory direction, your role is absolutely key.

We always do prize our independence, but if there was a scenario with a mad, bad or dangerous Government with a large majority, this Committee would have a majority of that party on it.

Well!

I won’t be tempted, Mr Glen. If there was a Chancellor of that mad, bad, dangerous Government who wanted to slash the funding, there could be a nasty alignment of interests against the OBR that you would be quite powerless to defend against. Is there any need for legislation or constitutional change to guard against that doomsday scenario? Well, we hope that it is a doomsday scenario.

I am not sure what you could do, but I take your point. Interestingly, when we were set up and I explained dual accountability, people would look at the Treasury Committee and say, “But surely this is a Government-dominated body. How on earth does that provide any assurance?” I did point out that is not how the Treasury Committee has operated in practice for many years, and that the sense of independence is well found there. It is an interesting question, but I literally do not know the answer. Could the Public Accounts Commission literally stop any squeeze on the resources of the National Audit Office? There is a parallel: I now work in Northern Ireland where you have a set of institutions that are separately financed through the Assembly rather than through the process from the Department of Finance. That may be a potential, but I suspect that once you got to a situation where things were as mad and bad as that, the money might be only one of the problems that you were confronting.

Indeed. Richard Hughes, have you a final comment on that point?

This Committee has a duty over the appointment of the members of the BRC. It would not be unusual for it to have a duty over, or at least some right of consultation on, the resources that the OBR is given to ensure that it is resourced adequately. The incentives are asymmetric on Government, and as Robert said, in some cases the way in which fiscal councils have basically been muzzled has not been by removing the personnel but by cutting the budget, which is undoubtedly something that stops them producing meaningful analysis. That has happened in European countries, so it is not science fiction.

I thank our witnesses very much for their time. It is a privilege to have both the former chairs of the Office for Budget Responsibility—not quite in one place, but we nearly got there, with Sir Robert online. Thank you for the work that you have both done leading the Office for Budget Responsibility. This is just the beginning of our inquiry into this topic. For the record, I should say that there is now an advert out for Mr Hughes’s successor as chair, and whoever is recommended for that job will come to this Committee for a pre-commencement hearing. I thank our colleagues at Bow Tie and Hansard for their work. The uncorrected transcript will be available on the website in the next couple of days.