Laura Gardiner, Tom Josephs and Professor David Miles.

Q76 The Chair: Welcome to the Lords Economic Affairs Committee’s sixth evidence session of our inquiry into the UK’s fiscal framework. We are delighted to have with us three senior members of the Office for Budget Responsibility team. We have Laura Gardiner, chief of staff at the OBR; Tom Josephs, member of the Budget Responsibility Committee; and Professor David Miles, also a member of the Budget Responsibility Committee. We are broadcasting this session on parliamentlive.tv. We will give you a full transcript after the meeting, so that you can make any corrections of fact. Thank you again for your time; we really appreciate it.

I will start with a general question about how you, as the people running the OBR, would assess the contribution of the OBR. What should the metrics be to assess the performance of the OBR over time? How do we know whether the OBR has made a difference compared with the absence of a rule or fiscal council? I will start with you, Tom, and then we can see whether your colleagues want to add anything.

Tom Josephs: I will start and I am sure my colleagues will want to come in. First off, thank you very much for the opportunity to give evidence at this hearing. The starting point in assessing the performance of the OBR is what we are here to do. The legislation on that is pretty clear: our primary role is to examine and report on the sustainability of the public finances. The legislation also says that we should do that objectively, transparently and impartially.

We can talk a bit about how you might assess our performance against all those criteria, but I would particularly focus on transparency, which is the area where the OBR has really made the biggest difference and added real value. I know you have heard that from previous witnesses, but I do think it is important.

To me, it is really about how we provide a deeper understanding of the public finances and better analysis for policymakers to make decisions. We also provide more and better information that allows Parliament, the public and the markets to scrutinise the decisions that policymakers choose. There are three particular ways in which we have done that. One is our core role of producing medium-term forecasts and assessing the cost of government policy. We do that in a way that provides a much deeper and broader information set on the public finances than was the case before the OBR was in existence, with in-depth information on all streams of tax, all streams of spending, the public sector balance sheet and the impact of policy choices on those going forward.

As importantly, we provide a much more transparent assessment of the risks and uncertainties around our central view of the prospects of the public finances. We do that through a number of different techniques and methodologies, which we can talk about in more detail. That includes scenario and sensitivity analysis, but also identifying particular areas of risk and pressure on the public finances and being transparent about them in a way that, quite often, has been uncomfortable for Governments—the previous Government and the current Government. That therefore shows that, without the OBR, there would not have been so much transparency around those kinds of pressures.

The final analysis we do is of longer-term sustainability and broader risks. The Treasury did a bit of that before the OBR was in existence, but we have again been able to provide much more detail and a broader set of analysis than was the case in the past.

One final thing I would say on this is about why the OBR is able to provide that transparency. A key thing there is our role in producing the numbers and the analysis that the Government actually use to produce the Budget, which means that we need access to government information. In contrast, other fiscal councils around the world have a different role, in which they are commenting from the sidelines on government policy. They therefore do not have that access to government information and can bring less transparency to the public finances.

Professor David Miles: I guess that one way of assessing whether the OBR has fulfilled a useful role is to ask whether people, on the whole, now think that the assessment that is made is unbiased, is not subject to wishful thinking and does not try to favour one position over another for reasons that are not backed by evidence; and that, in presenting forecasts, we are transparent both about the assumptions that we make about things such as the path of interest rates, oil prices, the stock market or foreign exchange rates and in explaining the logic behind the outcomes that we have used in our central forecasts, in terms of GDP, inflation or tax revenues, both in aggregate and for individual measures.

Being unbiased is probably the single most important thing. I hope that we have established some credibility on that front. Unbiased, of course, is not the same as forecasts that are always very closely aligned with outcomes, which is impossible, frankly, given the horizons over which we forecast and the kind of shocks that have come along even in the relatively short period that the OBR has been there.

Laura Gardiner: Building on what David said, I guess there are inbuilt ways in which we are required to assess our performance. David and Tom have outlined that the transparency of how we work rather than the numbers in the forecast itself might be the best thing to focus on but, each year, we do a forecast evaluation report that looks specifically at the accuracy and bias of our forecasts. Our assessment of those over the years has shown that, abstracting the very large shock of the pandemic, they have tended to be more accurate and less biased than the Treasury’s prior to that.

We are pretty much in line with external forecasters when you focus on the key metrics such as GDP and borrowing. I do not think there is any expectation that we should be any better than the external average; that is not a good yardstick by which to measure the OBR. On the economy side, we have access to the same information as other external forecasters and the Bank of England. On the fiscal side, we have access to much more detailed information and models from government departments, but we are also required to forecast conditional on government policy, especially on things such as public spending, which does not constrain external forecasters. We use our forecast evaluation process to shine a hard light every year on how well we are doing on the forecasting side, but the yardstick should not be that we are doing better than the average of externals; the law of large numbers says the average is quite hard to beat. It should be that we are doing it as transparently and informatively to you and government as possible.

The Chair: Thank you very much. I have one follow-up, given that we are talking about OBR performance and given the moment we are in: obviously I would like to ask you about the Budget process that you have just gone through, which I am sure has not been comfortable for you in the public eye to an extent that is not normal. We are going to look at different aspects of the Budget process in some detail, but overall I just wonder whether you wanted to say anything about the challenges that you think you faced with the Budget process that has just happened. In particular, we are interested in the letter to the Treasury Select Committee that you sent in the immediate aftermath of the Budget, examining some process issues in your relationship with the Treasury. Maybe you could explain the logic behind why you felt the necessity to send that letter.

Professor David Miles: Maybe I can say a few things about that. I think the main purpose of the letter was to try to clear up some misconceptions that were doing the rounds in the media. One was that the OBR had rather helpfully seemed to have come up with some extra money—some bit of good news, some money lost down the back of the sofa that we found—that was useful to the Government at the last minute. That was not the case. After the 31 October so-called final pre-measures forecast, in fact there was not any good news. What there was was an assessment of policies that the Government announced at the Budget, but there was not any good news that came along after that.

I think there was also a perception that maybe the OBR had come under some sort of pressure to fix the window over which we measure the likely interest-rate path into the future in a way that might be helpful. That was not true either. We explained in the letter that we decided on the windows to be used to calculate expected paths of interest rates right at the beginning of the process, and well before it was clear where yields would be at the time of those windows.

I think there was also a bit of a story doing the rounds in the media that there had been enormous volatility in the assessments coming out of the OBR leading up to the final pre-measures forecast on 31 October, and that had somehow added to the chaos and made things even more difficult. That was not true either. We showed in that letter that actually the central forecast for the margin against the target had not moved very materially in the weeks leading up to that.

It was really about trying to remove some misconceptions. It seemed helpful also to get the agreement of the Treasury that we would send such a letter—an agreement in a sense, both to the facts in the letter, but also to the point that this was an unusual thing and we were not setting any kind of precedent that in future we would show, possibly even in real time as we went along, what interim forecasts looked like. We were not setting a precedent for that.

Q77 Lord Lamont of Lerwick: What is your assessment of the effectiveness of the UK fiscal rules? David, you said a moment ago that you thought one of the values of the OBR was that it ensured the forecasts were impartial, not based on optimistic assumptions. But many people might think that the fiscal rules, as they currently stand, are somewhat Augustinian. They are easily gamed, with Governments writing in unrealistic expectations on either taxation or expenditure, and of course, you have the fact that several years down the line it may show a reduction in the final year, even though that final-year estimate of the debt-to-GDP ratio is higher than at the beginning. Surely this is rather meaningless.

Professor David Miles: As you know, Lord Lamont, the targets are set by the Government. The OBR role is to take those targets as given and form an assessment of what the probability is—the chances that they will be on the right side of it—as you say, quite a long way down the road. It is not our role to push back against the targets and say, “They are a bit too lax. They’re too easy to game”. Just as an observation, I can see the argument that because the target is set several years into the future, it can lead to a situation where a Government can be appearing to say, “We will do something difficult right at the end” and actually have a lax policy in between. It is inevitable, though, that there is something of a trade-off with having a target that is very close—in front of one’s nose, almost—that forces you to do something almost immediately in order to bring things into line with the target. The disadvantage of that is that you might have a juddering crash—foot down on the brake and hitting something that is looming up in front of you. It is actually quite a difficult judgment to make about what is the optimal horizon over which to be trying to hit a target?

Lord Lamont of Lerwick: Could you not have an absolute figure at the end date?

Professor David Miles: You could, and you could set a target for the stock of debt. In the past, though, if there had consistently been a target for the stock, then at least a few events—Covid being the most obvious—would have meant that you would have blown through that stock target. That is almost certain in the aftermath of what happened in 2020. You might end up actually just resetting the stock target several times, which probably does not enhance credibility.

But there is something to be said for having a stock target. There is also something to be said for having the primary target as being a deficit target. If you have a target for where the deficit will be, you could set it in such a way that it was likely to mean that the stock of debt would then not be continuing on an upward path. Where we are at the minute with a target for current balance some years down the road—the number of years down the road is coming toward us on the government policy at the moment—is that it is set in such a way that it probably coincides with a setting of policy that would stop the stock of debt rising. In other words, it is quite close to what some people might define as a sustainable position of stabilising the debt-to-GDP ratio.

Tom Josephs: Could I add one point? On the time horizon, as David says, the current set of rules will eventually be three-year targets. That does strike a reasonable balance between the need, as David says, to keep some flexibility in the short term to deal with shocks, but also to be on a horizon which is reasonably biting in the policy constraints it creates. Three years will typically encompass the years in which the Government are setting their spending plans—their spending reviews in particular—so there will have to be a firm set of spending rules covering those three years. A particular problem with five-year rules was that you often had two or three years at the end where there were no spending plans, and it was therefore quite easy for Governments to essentially pencil in a spending assumption in order to hit a rule. As I say, with a three-year horizon that is not possible.

Lord Lamont of Lerwick: Could I just ask another question? David, is there any problem with a set of fiscal rules that might require a tightening of fiscal policy, even when economic activity is low? I think the rules do allow for exceptional events, and I think they do allow for the automatic stabilisers. Personally, I am not sure I see this as a problem, but I am interested in your view.

Professor David Miles: Because the target has in the past typically been four or sometimes even five years down the road, it is now coming toward us. As Tom just said, it is going to become three years. That is usually enough time, for whatever the cyclical position of the economy might be right now, to have at least an expectation to move back toward a position where the economy is perhaps closer to its natural potential—something more like an equilibrium in the economy. Therefore, this potential tension between the cyclical position of the economy and the need to perhaps tighten policy because the target date is close to you has not actually been much of a problem pretty much in the whole time the OBR has been there, because the target date has been far enough ahead that it is more likely that you are in a more neutral cyclical position at that point in the future.

Laura Gardiner: Just as a quick follow-up on that, there have been eight different sets of fiscal rules since the OBR was established in 2010. In a few of those—I think three—the deficit was a cyclically adjusted one, designed to adjust the problem that you are talking about. As David said, because it was a cyclically adjusted measure, with the deficit targeted usually three years out, it was typically very similar to the kind of non-cyclically adjusted measure at that point in our forecast. It is certainly possible to have a fiscal rule set with some cyclical adjustment in it and, for a lot of the 2010s, we had that for the deficit target in the UK. In practice, it did not make too much difference to the set that we have now, because you are typically looking a few years hence.

The Chair: We have a couple of quick follow-up questions on that topic.

Lord Blackwell: Can I just come back to your option of a stock target, Professor Miles? The disadvantage of the flow target, as a number of witnesses pointed out, is the difference between two very large numbers, both of which are difficult to forecast. It is therefore incredibly volatile, and it is a one-year measure at the end of five years, which means that there are lots of options to game in the meantime.

The advantage of a stock target is that it is relatively stable. It is the ultimate objective and it is cumulative, so the Government cannot mess around in the first few years without affecting that. Your reservation was about shocks, but surely that is true of any measure. If there is a major shock, such as a pandemic, the Government will always have the option of explaining to the markets why special measures are needed and adjusting the target. I do not see that peculiar to a stock target. Ought we to be giving more serious consideration to that?

Professor David Miles: Yes. I think there are pros and cons; it is not obvious that a stock target is either better or worse than what we have currently and have had for quite a long time, which is primarily now a target on the current deficit. There is a link between the two because, if you have a target as exists at the minute for current balance, given that the stock of debt is not very far off 100% of GDP, the sustainable rule—defined as being what position on the current balance will stabilise the debt at that level—is not very far off from what we have with the current balance rule. The sustainable rule for having the debt-to-GDP ratio not rise but flatten off depends on the difference between interest rates and the growth of the economy. At the moment, given the configuration of interest rates and the likely growth of the economy, that is not far off from looking for balance on the current deficit.

In a sense, therefore, there is a link between the two. You can always work out from a target for the current deficit in the future what is likely to be consistent with the stock of debt. My main point is that there are differences between the two, and it is not obvious which is better. There are pros and cons, I think.

Q78 Lord Agnew of Oulton: I just feel that you are rather dancing on the head of a pin here. If we zoom out, the big picture is one of huge concern for our economy. It is all very well talking about stock of debt, but we are borrowing money to pay interest, and that amount is compounding on an annual basis. There is not a correlation between the base rate set by the Bank of England and the bond rate that we are having to pay for in the international markets.

You seem to be condoning the, I would say, extraordinarily complacent approach of the Government to let this all keep drifting forward in the hope that something might happen in three or four years’ time. All the bad and tough decisions are being pushed forward, and the idea that we are going to get to a year before the election and suddenly they are going to wake up and confront reality is simply not realistic, in my opinion, but you seem to be condoning that.

Professor David Miles: I do not think it is our role to condone. We take the target that the Government has set and our primary responsibility is to make an assessment of the chances of staying on the right side of that target. Our publication, which usually comes out in the middle of the year, on long-term sustainability issues, shows pictures that are very worrying indeed about long-term fiscal sustainability. We have argued for a few years now that, on current tax and spending settings, the UK is almost certainly on an unsustainable path. Indeed, we show projections of the stock of debt to GDP rising, if you look several decades down the road, to something approaching 300% of GDP. That is not a forecast, because that would not happen; there would be some juddering explosion some way down the road if you were on that unsustainable path. I do not think that we are blasé about the situation.

Tom Josephs: I have one thing to add on this. As David said, our role is not to comment on the merits of particular choices of rules, but we do comment on the risks to the public finances very extensively. We do that in the long-term analysis that David was talking about, and we also do that in our medium-term reports. One thing that we have pointed out in the last couple of medium-term reports is that, despite successive Governments having some kind of target for debt to be falling, what we have actually seen in outturn is debt ratcheting up. We have made that point.

We have also made the point that successive Governments, particularly since the end of the Covid pandemic, have also had plans for borrowing to fall pretty gradually over the next five years. In fact, what you have seen is borrowing stay pretty stubbornly flat at around 5% of GDP since the pandemic, despite those plans. We do point out the risks and pressures on the public finances alongside the assessment of the rules.

Lord Turnbull: I rather share Lord Agnew’s dissatisfaction. Two words came out of your opening presentations: transparency and unbiased. They seem to paint a world in which, if you provide really good information to the Government, they will take good and sustainable decisions. I do not think that is the safe assumption at all. There are very strong pressures, on both the Prime Minister and the Chancellor—they are hopelessly outgunned in terms of the numbers—to want things but not be prepared to pay for them or pay for the interest on them.

I would have thought that the whole purpose of financial councils—indeed, the whole apparatus: the rules plus the people to enforce them—is to lean against this complacency, as I think Lord Agnew would call it, and help not just to set the target but to help them achieve that target.

On the monetary side, we have a monetary target and the MPC and, on the fiscal side, we have rules and the OBR. They exist because, if left to itself, I do not think the system corrects itself or naturally tends toward sustainable outcomes. Therefore, underlying your position is something that is not unbiased; you are actually trying to help the Government, or provide the Government with the ammunition so that they can take and enforce more sustainable decisions than would otherwise be the case.

Professor David Miles: It is a political issue for the Government, I think, to decide whether they want to take actions that then create a very high probability that they are on the right side of their targets. What we have seen is Government after Government, almost since the OBR was established, and certainly in more recent years, pretty consistently being on the wrong side of targets, with debt having gone up and up. That reflects a number of things. First, there have been some shocks, the most important of which is Covid. One can argue about the fiscal policy response to that, although at the time it seemed that there was broad agreement that it made sense to have what was an extraordinarily lax fiscal policy, certainly in 2020. However, in the aftermath of that, the issue has been a difficult political one for Governments, which the OBR cannot sort out.

Lord Turnbull: The point I am making is that the Prime Minister and the Chancellor have created these institutions because they need help; on their own, we do not get the right decisions. I think the evidence seems to be that on the monetary side, that seems to be working a bit better—gradually we are getting within touching distance of a target. On the fiscal side, I am not sure that it is getting better.

Professor David Miles: If I may make one observation, which in a way is an obvious one, the Monetary Policy Committee sets the policy and the OBR does not.

Lord Turnbull: The target is set by the Government.

Professor David Miles: We could have a system in which the Government set the target and the OBR set fiscal policy. But that seems a bit of a conflict between that and—

Lord Turnbull: I am not saying that they should set it; the system has been created in order to reinforce the position of those who are in a weaker position—in Trump’s words, they do not have the cards. They are trying to create some cards by using institutions like yours to—

Tom Josephs: One point on this is the model of the OBR compared with the model of other fiscal councils around the world. The role of other fiscal councils around the world is to comment on government policy and say whether they think it is consistent with sustainable public finances, and I think in some cases even to advise the Government on what a sustainable fiscal policy setting would be. To go back to why the OBR was set up initially in the way it is, which is to not have that role, I think the view was that there were already institutions in the UK which essentially play that role of assessing the merits or otherwise of the Government’s fiscal policy choices, such as the IFS and other institutions, and that that role is performed fairly strongly in the UK compared with other countries. So, in a sense, there was not a need to set up an independent fiscal body to do that. But the issue in the UK was more one around the perceived bias in the forecasts and the lack of transparency, and therefore the OBR was set up in the way we have been to address that problem.

Q79 Lord Razzall: We have had a number of witnesses talking about the role of the OBR vis-à-vis the Government. As you will be aware, a number of commentators have said that because you basically produce the forecasts, the fundamental role of the OBR now is policing government policy. One of the other issues that has been raised is the role of the Government in disagreeing with what you have said. I do not think that, since you have been founded, the Government have ever put out a statement saying they disagree—I may be wrong but I cannot remember one. Do you want to comment on that? First of all, do you feel that your role is primarily, as you produced the forecasting, to comment and police government policy, and what do you think about the possibility of the Government disagreeing with you?

Tom Josephs: Our role is very clear in the legislation, and it is that we produce the forecasts that the Government use to set policy. We assess the cost of government policy choices and whether the Government are on track to meet the fiscal rules, and then we assess the risks around all that analysis. We have absolutely no role in setting policy or commenting or advising on it.

Lord Razzall: But you monitor against government policy.

Tom Josephs: Exactly—we monitor against government policy, and policy decisions are rightly for the Government to take. The sort of perception that we have seen build up in more recent times around the role of the OBR in constraining government policy, if you like, just reflects the fact that Governments have chosen to run their fiscal rules with very narrow margins, and that means that judgments we take in the forecast can be important in terms of the sort of space that the Government have when taking their policy decisions. But, again, it is absolutely up to the Government. They set the fiscal rules in the first place; they then choose whether or not to meet fiscal rules; and, if they are going to meet the rules, they choose the policy settings they are going to use to do that.

Lord Razzall: I do not think they have ever disagreed with anything you have said, have they? Well, in public.

Tom Josephs: There is nothing to stop the Government saying that they—

Lord Razzall: But they never have.

Tom Josephs: Well, the Chancellor has said that she wants to beat our forecast, which in a way is saying, I think, that she thinks that the economy can grow more quickly than we are saying, and that is perfectly fine.

Lord Razzall: Quite. David, do you want to add to that?

Professor David Miles: In many ways it would be not just acceptable but absolutely normal for a Government—a Chancellor—from time to time to say, “I think the OBR is being much too pessimistic on this or that. We’re going to base policy on a more optimistic outlook. It might mean that, on the OBR’s projections, we are perilously close to missing the target, or even slightly the wrong side of it. But we are confident we have a very good evidence base for thinking that the OBR is being too pessimistic, and that’s what we’re going to do”. That would be a perfectly reasonable position to be in. You are right, though: that has not happened recently.

Lord Razzall: Except that she has put it rather more subtly on growth, I suppose, and productivity.

Professor David Miles: I suppose so, and that is fine. I hope they are right.

Laura Gardiner: I was going to add a slightly different point, but there was at least one instance since we were created where we were forecasting that one of the fiscal rules was going to be missed. I think maybe Chancellor George Osborne in the mid-2010s was content to have a Budget or fiscal statement where that was the case. That underscores the point that we provide forecasts and judge the degree by which Chancellors are set to meet their fiscal rules, but they have choices to react to that, to not react to that, to change the rules, to explain to the public why they are behaving in the way they do. Different Chancellors have done all those things since we have been around, including missing the rules.

Lord Razzall: I am glad that you talk about Governments rather than government.

Q80 Lord Londesborough: Can we come back to the issue raised in the first question by our Chair in relation to the challenges around the Budget process? How would you characterise the OBR’s relationship with the Government and Treasury in particular? What, if anything, will be reviewed in light of recent events?

Professor David Miles: There is actually an agreement between the Treasury, the OBR and the Chancellor that this was not a very helpful process in many ways. There was lots of speculation about measures, there were leaks, there was briefing; I do not think it did anybody any good. I am pretty confident that there were no leaks or briefing in the lead-up to the Budget from the OBR, but there was plenty of such from other places. As I say, I do not think that helped anybody, really. I think I am right in saying that there is widespread agreement—from the Chancellor, from the Treasury officials, certainly from us in the OBR—that it would be better if that did not happen again.

Laura Gardiner: In terms of the relationship with the Government—particularly at the official level, which is what I oversee—as other witnesses have told you, it is a highly collaborative process with a huge amount of co-working between the OBR and the Treasury in particular, but also other departments, particularly HMRC and DWP, which have hundreds of staff working with us on the forecast. There is a necessary degree of creative tension in that—disagreeing on analytical judgments, looking at the evidence for income tax and what have you. But that is a feature, not a bug. The process is set up to get the best out of analysis on both sides via that tension.

There has been some characterisation that maybe the relationship has deteriorated a bit in recent events, at the official level. I do not recognise that at all. It was a busy, long Budget process, with a lot going on in the forecast and a lot going on in the policy package, but it was approached in the normal, collaborative creative tension way by officials from all the main departments, who we work very closely with and get along with as well as we always have. So I do not see any issues there and I do not think it has got any worse. I do not recognise that at all.

The final point I would make is about a problem that I have not had at all—but which I think is the biggest problem faced by fiscal counsellors in other countries—and that is around access to information. One of the real assets of the process that we have in this country—which is partly because we have that unusual role in providing the Government’s forecast—is just how well understood and adhered to that is by everyone.

I do not have to spend any of my time writing to my counterparts in other departments complaining that we are not getting access to the things we need, because everybody just provides that. I think that is something that trips up fiscal councils hugely in other countries.

So, from an official and an information-sharing perspective, our work is highly collaborative and things have not got worse, despite what you might have read.

Lord Londesborough: It is interesting, Laura, that you use the term “creative tension” because for those of us outside the OBR and the Treasury, that feels as though it has spilled over into a lack of trust.

David, when you were giving evidence in front of the Treasury Select Committee in relation to the unusual, and I think you said unprecedented, letter in respect of the OBR setting the timeline of communications with the Treasury, you said it was important to “set the record straight”. I am quoting directly. You said that this was because of “misconceptions … circulating in the media” about OBR forecasts, and that the OBR had raised concerns with the Treasury about leaks.

You added: “I think it was clear that we did not find this helpful”. You made that clear.

That suggests to me that, perhaps, there has not been in this challenging set of circumstances a level of trust and respect, taking on board that there will always be a level of tension, particularly in relation to your forecasts.

That brings me on to the second part of the question. We would like to understand how much freedom the OBR has in communicating with Parliament. As regards the letter that you sent to the Treasury, for instance, could that have been sent without the Treasury’s agreement? What is the process there?

Professor David Miles: It could have. We thought it was sensible to get Treasury agreement to the facts as they were in the letter, but also to publicly show that there was an agreement between the OBR and the Treasury that we were not setting a precedent that we were now routinely going to do this. I do not think it would be particularly helpful in future to be giving a step-by-step running commentary on the many intervening steps in producing a final forecast.

Tom Josephs: I would like to add to that. I agree with David. There are no constraints on our communication with Parliament, other than some issues around confidentiality of the policy-making process and confidentiality of certain taxpayer information and that kind of thing. Otherwise, there is no constraint on us in the legislation.

I would underline Laura’s point, which I think is important. The reason that we sent the letter, as David has said, was due to what was happening in the media, in speculation externally. However, the internal relationship between us, the Treasury and government departments, throughout the whole Budget process, was very positive and strong, as it has been. As Laura said, that has certainly not deteriorated in recent times.

It is important to understand that it is a complicated process that we have to undergo. It is pretty unique in the sense that we are one of the few independent fiscal bodies that works with the Government to produce forecasts. So there is a huge amount of interaction—a huge amount of back and forth—that has to take place.

We have to work with hundreds of officials across government, and we have to do that in a way that maintains our independence and our right to take independent judgments on their analysis. Quite often, we tell officials or Ministers that we do not agree with their analysis.

So, that does mean that there is a fair amount of disagreement. But we have developed a process by which we can do that in a positive and productive way.

Lord Londesborough: Is there nothing you would like to see improved in light of recent events? I appreciate the inquiry into the leaks, but what lessons have been learned?

Tom Josephs: As David says, the issue is around media speculation and briefing, and the Treasury has a review into the leaks and Budget security. That will play out and that will be important. When it comes to the internal process, we are always open to reviewing and developing that, and we have done over the course of the years that the OBR has been in existence.

It is always the case that different Governments and Ministers have different ways that they want to run Budgets. We are as flexible as we can be on that, and we have adapted the process. We will definitely continue to do that.

The Chair: We have three follow-up questions, so if colleagues could be brief that would be great. First, Lord Lamont.

Lord Lamont of Lerwick: I have just a brief point for David on the difference between public statements and leaks. The Chancellor of the Exchequer repeatedly said publicly before the Budget that the OBR review of productivity had downgraded productivity, and that meant that tax receipts were lower. Is that not a breach of confidentiality? Is that not meant to be private information?

Professor David Miles: I do not think it was much of a secret; it was no secret at all that we were looking very carefully at our productivity judgments, or that they had turned out to be on the optimistic side of outcomes. So I suppose it was not saying anything particularly surprising that the OBR was likely to be coming up with a new assessment of productivity growth and it was going to be lower; nobody thought it was going to be higher.

Inevitably, if it was the case that it was a lower projection for productivity growth, that would obviously be bad for tax revenues and would create a fiscal issue for the Government, all of which was true.

I would add that we were keen, though, not to say anything about our assessment, which all came out at the time of the Budget. So it certainly was not coming from the OBR.

Lord Burns: I get the impression that Treasury Ministers find the process of dealing with the OBR a bit frustrating at times, mainly because of the response lags as you exchange messages. For example, they might find it frustrating that they do not have more conversations with the OBR themselves. Are there ways of making the process work more smoothly? Does it require legislation, or is it just a question of working out the working practices, particularly for new Ministers who have to become comfortable and have to learn more about just how this process works?

Tom Josephs: Yes, and Laura might want to add to this. As I have said, we very much recognise that new Ministers, new Governments, will have different ways that they want to run the Budget process. We have always been flexible and responsive to that and we will definitely continue to be so. We are very happy to continue to look at ways to get the process to run more smoothly.

When it comes to the legislation, one point is that the UK is quite unusual in the sense that the Budget date is not fixed in legislation. The process more broadly is not legislated for in much detail at all compared with some countries where it is set out in much more detail. That is potentially one area that could help with creating a bit more stability around the process.

Laura Gardiner: I was going to make the same point. I think the goal of concentrating fiscal policy-making in a single Budget event per year might particularly merit consideration of some more clear, open and visible—to this committee and others—fixed processes, as are more common in other countries. Not least, if Budget events are as large in relation to the size of policy as the one that we have been through, that is quite a big challenge, particularly for the forecasting departments such as HMRC and DWP.

It might be worth thinking about how that can be a bit more sequenced, which is more common in other countries. The big challenge of the process is a tension between two goals. One is having a good, high-quality and up-to-date forecast at the time it is published, particularly for things such as new data on tax receipts and market interest rates. That all goes into our pre-measures forecast. The second is time for policy development against a stable forecast base that does not subsequently move. When we come to agree the timetable for producing each Budget or fiscal event with the Treasury and other departments, we have some parameters for that set out in a memorandum of understanding. But a lot of it we come back to every time. That is the tension—the Budget Responsibility Committee’s desire for the forecast it publishes to be up to date and not months out of date when it publishes it, and the Government’s desire to have quite a bit more time with that forecast fixed in order to make policy. I point out that other countries where the fiscal council plays a leading role in the Budget process have different ways of getting around that—for example, a clearer distinction between the pre and post-measures phase. The Netherlands publishes the pre-measures forecast. Everybody knows that and moves on to the next phase. That is the tension we come to. A stable base for policy-making is one thing, but we also want our forecasts to be up to date and credible at the time they are published. That is why there is a lot of time pressure in this process and why there are some of the frustrations you have heard, and why we might push back against some of that to ensure the forecast is up to date.

Lord Liddle: Can I go back to what Lord Lamont was talking about in terms of the review of the productivity numbers? I have no inside knowledge but, in the Financial Times and other newspapers, it sounded as though Ministers and their advisers were cross about the timing of this productivity review because they felt that it should have taken place immediately after the general election when they were thinking about their inheritance and whether they should raise taxes straightaway in order to deal with the inheritance. To my mind, that is a reasonable point of view for advisers to take. What are your comments on that?

Professor David Miles: One of the difficulties in changing one’s view about the trajectory of underlying productivity growth, and in working out when to do it, is that you want to abstract from the temporary effect of big shocks. One is trying to forecast what the underlying trajectory might be far into the future, rather than what the measures will show for a particular quarter or month. In 2023 and even into 2024, we were still living with, to some extent, the aftermath of Covid but particularly the huge increase in energy prices after the Russian invasion of Ukraine, plus the fact that there were severe problems in measuring labour input in the UK, given the ONS difficulties with the Labour Force Survey. The productivity numbers have been gyrating all over the shop since 2020. More recently, two things have changed. One is obviously that much more time has passed since Covid, and a lot of the big increase in energy prices have more or less reversed themselves. Those effects, which clearly were affecting productivity in 2020 and probably through into 2023 and beyond, have largely gone away and we have some alternative measures of labour force input, which gives you a more robust estimate of what is actually happening to productivity.

We waited for the dust to settle from those big earlier shocks and for better measures of the labour input behind productivity. There were good reasons to do that, rather than get a bit trigger-happy, look at some bouncy quarters back in 2023 or 2024 and then make a judgment that was clearly going to have a material impact on the fiscal outlook. I completely understand that any Government, particularly when the news on the productivity assumption we were making was bad news fiscally, are never going to like it. It is a bit of a tricky judgment as to the right time to do it but it was something not to rush into when the data were very difficult to read.

The Chair: We are going to go to Baroness Wolf. We may well have a vote in the next few minutes. If so, we will suspend for 10 or 12 minutes and then resume. Thanks for your understanding.

Q81 Baroness Wolf of Dulwich: I should like to ask you about what the OBR says about longer-term trends and the fact that so much of the attention is focused at the moment on the Budget process and the changing fiscal rules. This picks up a little on what Lord Agnew was saying. What more could be done to get the media, the political classes and the general public to focus on what you are discussing in the Fiscal Risks and Sustainability report? Is there anything that can be done to shift that, or are we just doomed to have a yearly bout of excitement around the Budget and not enough attention paid to the longer-term analyses?

Tom Josephs: A couple of things on that—first, we have very much tried to do what we can to raise the profile of the longer-term analysis. It attracts a reasonable amount of attention. It gets a lot of that in the media on the day that it is published. On an ongoing basis, you see the analysis being used in a lot of wider forums and analysis. For example, the analysis we did this time on pensions and long-term pressures on them was picked up pretty widely, and we have seen it used on an ongoing basis. So it gets traction but, we agree, not as much as we would like. We are trying to do more in terms of bringing some of that analysis into the economic and fiscal outlook that we publish on Budget day and at fiscal events in order to give it more profile at that time. We did more of that in the Budget this year and we intend to do more going forward. The fact that the Chancellor has now said that the spring event will not be a policy event and that we will not make an assessment of the fiscal rules at that event gives us space to present more of that analysis in the spring. We will be thinking of ways in which we can do that. I guess the other way in which this could get more traction is if there were potentially more of a requirement for a fuller government response at the time. The Government respond to the Fiscal Risks and Sustainability report but tend to do so later, with a bit of a time lag, and quite often alongside the Budget or another fiscal event, whereby, again, it does not get as much profile as maybe it deserves. That is one potential route through which it could get more attention.

Baroness Wolf of Dulwich: Can I follow up on that? One of the things that you get everywhere, but particularly in the longer term, is that you are dealing with probabilistic estimates and estimates of risk. A number of people have said to us, “Well, there should be more presentation of ranges, confidence intervals and more emphasis on that”. I have never had any success in getting anybody to be interested in confidence intervals or ranges rather than a particular point. Do you have any comment on that?

Professor David Miles: Your experience exactly matches ours. People focus laser-like, particularly around the Budget, on the pass or fail. Is the so-called headroom a positive number or the wrong side of zero? It should not be seen as a pass/fail, zero/one thing. I completely agree with you. It is a probability assessment.

On this Budget, for example, we made an assessment that, on the policies that the Government said they are going to follow, given the shocks that might be expected to come along, there was something like a 58% or 59% probability that it would be on the right side. That is intrinsically not a pass/fail statement. It is what it is. It is better than tossing a coin, but it is not a whole lot better than that. Then we read in the press the next day that the headroom is £21.7 billion. Maybe we should not put numbers after the decimal place. It is annoying that there is this laser-like focus on a single number when we are really trying to paint a picture of a probability distribution, a range of things that could happen. The middle of the distribution happens to be that number, but the chance that it would be very close to that number is negligible.

In a way, I can understand why the media do that. The way the OBR was set up in the legislation forces us a bit into a pass/fail statement—if you look at the middle of the probability distribution, is it on the right side of the target or the wrong side? I should not just blame the media, because we slightly get pushed into this pass/fail thing. There are pros and cons of it, but one of the advantages of making an explicit assessment only once a year alongside the single Budget is that we will be able to move away from this pass/fail, zero/one thing in the spring.

Baroness Wolf of Dulwich: I am interested to hear you say that, because obviously the other problem is that, once you have a combination of not much headroom and a precise number, everybody starts talking about this score, that score or something else. Do you think having it once a year really will help, or will it just mean that everybody is running around like crazy as the Budget advances and the rule is not being met?

Professor David Miles: It is an opportunity for us in the spring, as it will now be, to move away from whether the number is £21.7 billion, £22.4 billion or £17.6 billion and paint a richer picture of the fiscal outlook. It has the potential to be useful in a way.

Lord Davies of Brixton: On that point, is it not an unavoidable truth that normal people want a single figure, not a narrative?

Laura Gardiner: David made the point about decimal places, rounding and recognising some of the challenges that the legislation sets for us in terms of producing a central point forecast, which we are required to do. We have tried to take steps in our communications on recent events not to overly specify numbers in our paragraphs, particularly on the first page of the executive summary. I do not think we said £21.7 billion there; I think we said £22 billion—maybe even that was too precise. It was similar for other numbers. But other people went straight to the table.

We are trying to evolve our communications to provide the specificity and transparency that people want but also to be clear when the differences between two numbers are non-significant. It is not just us communicating these things. We took some steps away from being overly specific, but I did not necessarily see that feed through.

Q82 Lord Petitgas: You spoke about a 58% chance on the outcome. Do you have a rule about where you want to land? Is it below 60% or above 55%? Or do you section it—for example, as long as it is about 55%? How does it work? Secondly, I now understand the headroom better from these conversations; you have a central number and then you compute it by multiplying by the probability of outcome. Is that where you get the decision? Why do you pick 58% and should it really be 63%?

Professor David Miles: Here is how it goes: we form an assessment of what would happen in the absence of positive or negative shocks. That is the number that, when we turned the handle on the various machines we use, gave us £21.7 billion or £22 billion. I would prefer that people call it £22 billion, but we did put £21.7 billion in the book. There it is. The chances of that happening are negligible because it relies on there being no shocks, either positive or negative, on balance between now and five years down the road. There is zero probability that that will happen.

We then look at the kind of shocks that have happened in the past and the impacts they have had on revenues and spending. That then gives you a degree of volatility that you can anticipate, symmetric to some extent to your central forecast. When you have roughly £22 billion of headroom, you can work out the chances that the shocks will be less bad than wiping out £22 billion—either good shocks or negative ones that are not enough to wipe out £22 billion. That then throws out a number of the chance of that happening—being on the right side of the target. That is where the number came from. I think it was 59% actually, rather than 58%. It is not in any sense a target. It is what is implied by the statistics on the past levels of shocks that can come along and hit you, the Government’s choices on where to set the policies and our central forecast of what would happen in the absence of shocks.

Lord Petitgas: It is obviously not a normal distribution.

Professor David Miles: You are right: you would not want to impose a normal distribution. It is asymmetric in a sense, because when there are really big shocks they are nearly always negative, and the good ones tend to be smaller, so you would not want to impose a normal distribution.

Lord Petitgas: The problem is that nobody outside understands the 58% and people focus on the decimal. People do not really understand that there is a less than two-thirds chance of getting there; it is not something you hear.

Professor David Miles: No. Maybe we could do a better job. For what it is worth, I think most people understand what a 60% chance is.

Lord Petitgas: Better than £21.7 billion.

Professor David Miles: Indeed. It would be a triumph if people were quoting 59% rather than £21.7 billion.

The Chair: Lord Blackwell, you may be competing with a vote in a second, but we will see.

Q83 Lord Blackwell: Can I move on to the way you score changes in government policy, the so-called dynamic scoring? I can see that there are some things where you can use normal econometric tools—multiple regression analysis, et cetera. There is a lot of data, but a lot of policy must be quite subjective. The debate on which we may be voting is on the Employment Rights Bill, which you have not yet assessed. When you have something like that, what process do you go through to try to incorporate that into your forecasts?

Professor David Miles: It is to try to gather as much evidence on the issue as is available. Sometimes it is relatively straightforward. One can rely on econometric estimates that have appeared in learned academic journals which focus exactly on the issue in front of us. But as you rightly say, something like the Employment Rights Bill is almost unique in what it is trying to do. It is difficult to look back in British history and see what happened last time we had something like that. There is some statistical evidence from labour economists who have looked at different countries and what different kinds of regulation did to employment when they were brought in. Sometimes there are natural experiments in the US; there may be a rule that exists in a particular state and a bordering state has a rather different one, so one can look at things such as whether jobs moved across borders.

The employment rights measures are a good example because there is not a straightforward piece of empirical statistical analysis which says that we can be pretty sure what it is going to do. It is particularly difficult in this case because the ground keeps moving as to what exactly the legislation will be. We will wait and see. I think it is a case of saying—this is true of a lot of things we are asked to make a judgment on—what this will do, for good or ill, to productive potential. They are all different. Sometimes there is a very good, firm body of evidence. Sometimes it is more a case of finding some experts who know a lot about the issue and getting a judgment from them. On the Employment Rights Bill, I am sure we will talk again to lots of the interested parties. Some of the views are well known. The CBI has well-known views on it and we have certainly spent time talking to it. So it is drawing on a broad range of evidence, which is sometimes much less reliable and more shaky than it is for other things.

Tom Josephs: We published a paper alongside the Budget on our approach to looking at the economic effects of policy. One thing we pointed out there is that the experience of having done that process over the past few years has made us realise that this is a more difficult exercise for policies such as the Employment Rights Bill, which are essentially regulatory changes, compared with looking at changes in tax rates or benefit rates, where there is more often a more standard economic model that you can use to assess the effect of a change in the tax rate on something like labour supply. That is quite often not the case when you are looking at regulatory changes, which by their nature have more complex and wider-ranging sets of effects.

For those types of policies, it is more difficult for us to establish the policy baseline against which we would assess the difference that a change in policy will make. So for a tax change, for example, we know what the tax rate has been over the past and we know what employment has been over the past, and when you change the tax rate you can make an assessment of the difference that that will make against that baseline. But with regulation, the policy baseline is much more difficult to pin down.

For that reason, we have set out some criteria that essentially will mean that it is less likely that we would assess that those sorts of policies have a material economic impact that we should incorporate into our forecasts, just because those effects are more uncertain and more difficult to assess. They are therefore less likely to meet our threshold for incorporating that sort of policy into the forecast.

Lord Blackwell: Even for conventional tax policies, such as a cut in tax, you can model the normal multiplier effects on demand, et cetera. But there is room for debate about the supply-side effects of tax cuts—how far you believe in the Laffer curve, et cetera. When you are making those kinds of judgments and feeding them into the results, should there be, or could there be, more transparency about what the range of outcomes might be? Is the process sufficiently clear?

Professor David Miles: When we have made judgments like that in the past—for example, on the changes in national insurance contributions and some impacts of freezing thresholds, which obviously pushes more people into higher tax rates at the margin—we have been explicit about assumptions we made about key parameters that affect the outcome on labour supply and employment, or so-called elasticities of demand and supply. We have tended to try to look at a broad range of the literature, primarily focused on estimates from the UK labour market, and we have used a central estimate from a range of different studies on those elasticities.

We also draw on models of labour supply developed over time by the Treasury—HMRC has been involved in this as well. We can draw upon those models and feed in our own assessment of an elasticity if we think that the default numbers used by the Treasury and HMRC might be a bit off from what we think is a central estimate.

Laura Gardiner: On the transparency question, in recent years we have done a bit more of this dynamic scoring, or indirect effects, and it is an area where we try to be really transparent. On your question about tax changes, when we made these kinds of judgments on the cost-of-capital framework, which was used to judge the super-deduction in corporation tax or the labour supply model that was used for various national insurance changes, we published papers for all to see that set out the exact elasticities we used, the assumptions we made and the things we plugged into the model. Those are on our website. Via our advisory council and our forthcoming areas of research interest that we plan to publish, we are very happy to engage with the academic community on whether those are right. In this area in particular, particularly because we have done quite a bit more of it in recent years, we have sought to be really transparent and put all those elasticities out there.

You were talking about tax but, when we did some work on public investment just over a year ago, we published a 50-page paper. We had a kind of submission process and we published a response to those submissions and questions that people asked us. We are always keen to do more of that kind of thing and bring in academic expertise on the judgments that we are making. We think that being really clear about what we are plugging into the models and what elasticities we are using is the best way to invite people to tell us where they think we are getting it a bit wrong.

Lord Blackwell: You have to have a number at the end of it to put into the central forecast. Where there is a range of views about what the impact would be, could you do more to set out what those different views might be and therefore the uncertainty?

Laura Gardiner: There are some examples when we have, but I am sure we could do more. One I can point to is when we incorporated the supply-side effects of increased departmental capital spending and public investment a year ago. In the relevant bit of the economic and fiscal outlook, we presented a range—in essence, upside and downside scenarios—not just over five years but on what happens if this carries on over 10, 20, 30 or 40 years. David might need to remind me, but I think that range was based on different assumptions of the extent to which the public investment is complementary to or substitutes for investment in other areas of the economy.

So when there are quite uncertain parts of our judgments on policy and its effects on the economy like that, we try to show ranges and some of the effects of using different parameters. It is certainly something we could consider doing more of, if it is helpful.

The Chair: We are free; the threat of a vote has gone.

Tom Josephs: I will make one point on this: we very much recognise the uncertainty around these assessments. It is part of the reason why, in the document I was describing earlier, we set out a new, more transparent threshold for policies to incorporate the impact on the supply side of the economy into the forecast of 0.1% of GDP by the fifth year of the forecast. Part of that reflects the fact that we had found that we were incorporating into the forecast some fairly small impacts of policies—on your point, we recognised that they were very uncertain—on to an already uncertain view of potential output in five years’ time. That led us to decide that we should have this threshold. So, in a sense, we are only going to incorporate policies into the forecast that we think will have a reasonably material effect.

Q84 Lord Liddle: Just on this modelling of impacts, one thing that struck me from looking at the Budget was how, at the end of the period, young graduates will be facing a very high marginal rate of tax—40% once they have crossed the threshold, plus their 9% university loan repayment, plus 2% national insurance. That is a 51% tax on a 50-grand-a-year graduate. Are there any precedents for trying to model the impacts of such a change?

Tom Josephs: We do not model the distributional impact of policy. We do not go down to the level of looking at the impact of policy on different groups, such as graduates. It is outside our remit generally, because we focus on the macro side, but we have a lot of analysis and illustration in the EFO of the impact of the personal tax threshold freezes, in particular on the number of taxpayers who will be pulled into the tax system over the period of the freezes and the number who will be pulled into the higher and additional rates.

Lord Liddle: Yes, but do you have a model of how that will affect their behaviour?

Tom Josephs: Something that we have incorporated into previous forecasts is the impact that that will have on labour supply in particular. There is also a feed-through from higher marginal tax rates into household income and consumption. That is in the modelling as well.

Professor David Miles: The answer is that it is negative and substantial because, on current policies, the thresholds will be there for almost 10 years. If you think of the last 10-year period, it has had years in which inflation was close to 10%. So it is a huge change in the value of the thresholds and more people will be driven into being higher-rate taxpayers.

Lord Liddle: There is very little public discussion of it, because it was introduced by the Conservatives and sustained by Labour.

Professor David Miles: And it happens slowly, in a way that people do not notice. Well, eventually they notice.

Lord Davies of Brixton: They will also get hit by the removal of salary sacrifice if they are academics.

The Chair: Lord Petitgas has a follow-up and also the next scheduled question, so he is going to do a smooth amalgam of the two.

Q85 Lord Petitgas: I will not overdo it, because I actually have three questions within my question and my question was quite complicated, so I have reset it into three points. The first one is addressing the OBR’s, I dare say, constant overoptimism such that we—I see the Government and OBR as one package—never seem to meet our forecasts. I would like to understand that better.

The second is a productivity point, which is a key factor in the core model. I know that we have talked about this before, but I would like to understand a little of what goes into that. Perhaps we should take it from the market instead of doing it ourselves.

The third is the question that we have talked about, which is whether the OBR or the Government should own the forecast and the model. I will take the points one by one.

On the first, there was an article in the Daily Telegraph by Neil Record, formerly of the Bank of England—I am sure you have read it—who went through a list of the errors, as it were, from 2010 to 2020. There was only one year, 2014, when we somehow outperformed the forecast. So it is greater than a 50:50 chance; basically, we never really meet it. We also know that debt grew in that period from £1 trillion to over £2 trillion, so it doubled. Certainly it was reducing in the fifth or fourth year, but it grew overall, even as a proportion of GDP. So the question is: is it because of the fiscal rules? You abide by the fiscal rules; maybe the fiscal rules are too lax and do not work. Secondly, is there too much bias in the system, exogenous factors and lots of excuses? Or is it perhaps something to do with the headroom being too tight? Can you maybe explain why, in only one out of 10 years, we did what we thought we would do?

Tom Josephs: I could start on that question and then maybe David can address the question on productivity. We have done a comprehensive assessment of our forecast performance since the OBR was set up and that finds three main reasons why our forecasts have been too optimistic. The first is that, over that period, the economy has been hit by a succession of big shocks, which obviously had very significant negative effects on the economy and public finances and which we and every other forecaster did not see coming.

The second reason is that we were overoptimistic on productivity, and David can talk a bit more about that. In response to that, we have made a succession of downgrades to our productivity forecast.

The third is more on the policy side. In particular on public spending, we take, in essence, the Government’s plans for departmental spending and use that as the basis for our forecast. Over much of that period—in particular, the latter half of that period—what actually happened was that, when it came to setting firmer plans in spending reviews, Governments increased spending compared with the plans that they had set previously. That was the third main reason for overoptimism on the borrowing forecast.

Going back to the point that we made right at the start about what the OBR adds in terms of transparency, that is another area where we are very transparent, in the sense that we do our forecasts, but then we go back and do a very thorough assessment of the reasons why our forecasts do not match the outcomes—which, as David said, is always going to be the case. We set that out very transparently and try to learn from it.

In particular, on the public spending side, that finding led us to be more transparent about the fact that Governments have tended to increase spending levels when it comes to spending reviews and to try to point that out more clearly and transparently so that Parliament and the public can understand the scale of that risk. David, do you want to talk a bit about productivity, specifically?

Professor David Miles: Yes, but I just start with why the stock of debt has gone up so much and which was not in OBR central forecasts, going right back. What has happened—we all probably understand this—is that there have been three very big shocks, going right back to even before the OBR was there. There was the global financial crisis, a huge fiscal negative shock. Then things calmed down somewhat but, although there was an attempt, it did not bring the stock of debt back down anywhere close to where it had been before the financial crisis. The same thing happened after Covid. There was a huge shock and a massive loosening of fiscal policy, GDP fell 10% in 2020 and the stock of debt ratcheted up very substantially further. Then things became somewhat calmer. However, before that happened, there was the energy price shock, with huge fiscal loosening again to protect households from the impact of the shock.

When you then get a run of years in which there is not a big negative shock, one might hope that, before you get to the next negative shock, there is an attempt to tighten policy on a scale that would bring the debt back down so that you are in a better position to face the next storm. That did not happen, and one can speculate about why. I think that part of the reason—something I perhaps have said before—was that precisely because successive Governments have protected people from the immediate impact of these shocks, people’s expectations about what the Government will then do, and the support they are able to give, have not adjusted to the reality of an economy that has shrunk relative to what people had expected. Governments have found it very difficult, in a sense, to bring the public with them in understanding just how diminished the ability of the state is to deal with negative shocks. That is why the debt has gone up, then levelled off a bit, gone up, levelled off for a bit and gone up again.

On the productivity issue, it is possible that the OBR could canvass forecasts from private sector forecasters, think tanks and academics, but they would probably be slightly all over the shop. The last two Nobel Prize winners who focused on productivity have been at two ends of the spectrum about what the impact of artificial intelligence might be on productivity growth. Looking forward over the period that we are interested in—the next five or 10 years—one of them, Daron Acemoglu from MIT, said that it would do virtually nothing. Then the most recent Nobel Prize winner has said that it will increase productivity levels by something like 7% or 8%—almost 1% a year faster growth—over the next 10 years. So they are all over the shop.

Part of the difficulty in just saying that we should take the consensus view is that different people are focused on slightly different things.. Part of the difficulty in just saying, “Well, let’s just take the consensus view”, is that different people are forecasting slightly different things. We are trying to forecast the underlying trend of productivity growth into the future. Some of the forecasters out there are looking at shorter horizons and saying, “What do we think that measured productivity is going to do over the next couple of years?”

It sounds as if I am trying to dodge the issue and say that it is all too difficult and we should make our own judgment. Some people will look at what we have done and say, “You were a bit slow in reducing your productivity assessment”. It was not something that we wanted to rush into and be trigger-happy about—particularly, as I was saying earlier, because of the real statistical fog that existed until recently, which is still there to some extent but has lifted a bit. Even then, it could be because no one is confident about what the size of labour input has been in the UK. Even if you believe the GDP numbers, you have to then look at GDP relative to labour input. Who knows-in three or four years’ time, they may reveal that productivity growth has been rather strong over the past three or four years. That is not our central guess but it is not inconceivable. There are smart people who think that is true. Productivity is really difficult to get right. We set out the evidence we looked at in a long document that we produced and that came out on the day of the Budget. It runs to about 60 pages. If you have a sleepless night, it might put you to sleep.

Tom Josephs: Another issue on the question of why we do not just use the consensus economic forecast is that we are here primarily to produce a fiscal forecast for the public finances. In order to do so, we need a detailed forecast of the economy that covers all those economic determinants that drive tax revenues and public spending. No one else externally produces that detailed economic forecast over the five years that we need. Productivity is obviously an important judgment and driver of the economy but, as we saw at the last forecast in November, although we downgraded productivity, we actually had a higher tax revenue forecast because of other moving parts in the economy forecast, in particular higher earnings and inflation, which meant that the tax revenues were forecast to be higher than before. That is just an example of why it is not as simple as taking the consensus view on productivity.

Lord Petitgas: The third question is something that we have talked about before, which is: should the OBR carry on owning the model or should you be more the auditor of the model? The model would be owned by the Treasury and the Government, and you guys would be auditing what you see. You could still do the dynamic scoring but the model will be owned by the Treasury.

Professor David Miles: The difficulty with that—this is my personal view—is that the forecast comes out on the day of the Budget and everybody is focusing on it. The incentives of officials within the Treasury is, in a sense, to serve Ministers. I will choose my words carefully here. Officials will be seen to be doing their job better in the eyes of their masters if the forecast is more favourable to the Government. It is a bit of a recipe for producing something that might not be a central forecast. Now, you might say, “Well, okay, but the OBR can come along a bit later and say, ‘We do not think this is a central forecast’”. But one could imagine the way in which that might happen. There would be a Budget, let us say, on 26 November. Then the OBR would see it and have four weeks. It would put out its assessment on Christmas Eve. Nobody would read it and the Budget would have gone. It is just a way of making the OBR considerably less relevant. That might suit some people, but you run straight into the issue of—

Lord Petitgas: Impartiality.

Professor David Miles: Exactly. Bias and impartiality.

The Chair: We have three quick follow-ups.

Lord Verjee: We hear how difficult it is to forecast and we take it that forecasts will not be right. I do not believe we have it, but is there a case for putting in place a contingency reserve? As a businessman, that is what I think I would be doing. I would say, “I cannot forecast this accurately. We’re going to have black swan events but do not know what they are. In any case, as a matter of prudence, let me put in a healthy contingency reserve”. If we outperform on the good side, then great, but would it not be more prudent to run our finances by putting in a healthy contingency reserve?

Professor David Miles: In some ways that is the so-called headroom that we forecast, the £22 billion or thereabouts at the November forecast. If that number had been considerably larger, £40 billion-odd instead of £20 billion—which would not have been enormously large relative to the fact that you are looking at a target five years down the road with lots of uncertainty even six months ahead—the probability of being on the right side of it, which we thought was about 59%, would have been materially higher. It could have been 65%, or even 70% or 75%.

What you are saying makes lots of sense. Why would one not have a bigger buffer? The difficulty for the Government is that, in order to generate the buffer, you have to make some choices. You either increase taxes even more or cut spending even more. There is not much else you can do. It was hard enough for the Government to reach a set of policies that took the headroom from £10 billion to £20 billion. If you said, “Why not get to £40 billion?”, I think that would have been extremely difficult. But it is a good place to get to if you can get there, for exactly the reasons you said.

Lord Verjee: Is it not your job to try to bring reality rather than let the Government game the system?

Professor David Miles: In a sense, if you had to summarise it in one number, our reality is the 59% chance of being on the right side of the target. If you thought that was good enough, then fine. If you thought a 40% chance of missing the target was too high, then there is the difficult decision of how you make it not 59% but 75% or 80%.

Tom Josephs: The other way of thinking about it is more on a long-term horizon. As David was describing earlier, debt in the UK has ratcheted up because we have been hit by a series of very big shocks. From the point of view of sustainability of the public finances, between those shocks you would want to try to reduce debt so that you are in a more sustainable position when the next shock hits. This is similar to the point you are making around having a contingency reserve. It is not our place to advise the Government on the appropriate level of debt or the pace of cuts in debt, but in our longer-term reports in particular we try to point out these risks and the risks to debt sustainability if Governments just allow debt to continue to ratchet up. I think those reports do that pretty effectively, but it is up to policymakers to decide what they do in response to them.

Q86 Lord Agnew of Oulton: I want to pick up on your explanation, Professor Miles, of the debt journey of the last 15 years. You have really only answered half of the question for me. We started in 2008-09—these are very rough numbers—with a debt of about £800 billion, give or take. A couple of hundred billion was used as interventions for the global financial crash. About £350 billion was used for Covid interventions and maybe another £75 billion for the energy crisis with Ukraine. If you add those up, it is nothing like the journey from £800 billion to £2.8 trillion that we sit at today.

It worries me that august institutions such as yours seem to be putting a blind eye to the telescope on the extent of the problem that we face as a country. I feel like we are in the 1930s, with the general failure to recognise the seriousness of the build-up of Germany as it was then and what we ended up with. I see you dancing around with these very elegant answers, but the elephant in the room is not addressed. You really are the thin blue line, because we have a huge Labour majority in the Commons who appear to be economically illiterate. We are about to be utterly emasculated up here. We are really relying on organisations such as yours to explain this narrative to the wider electorate.

Professor David Miles: Our reports that have come out every year for at least the last four years paint a grim picture of fiscal sustainability. As I mentioned earlier, when you look a little further down the road, given the demographics and the sorts of shocks that come along and hit an economy such as the UK’s, on the current policy settings on the tax side and the spending side, we are on a trajectory which is clearly unsustainable. It would take debt to 300% of GDP, except you would not get there because the Government almost certainly would not be able to sell enough debt to finance what they were doing. At some point, there would be a juddering halt and a financial crisis.

That has been the message of our longer-term assessments for certainly as long as I have been at the Office for Budget Responsibility. In some ways, coming back to an earlier line of questioning, it is a bit of a shame that these reports do not get the same focus as our much more short-term things which focus more on the question of, whatever the fiscal rules are, whether there is a reasonable chance of getting there just a few years down the road. If you look further down the road, the picture is, frankly, very grim, at least on unchanged policies. In many ways I am not pushing back against what you are saying.

Laura Gardiner: I think we have tried to bring more of that characterisation to our medium-term forecasts—our economic and fiscal outlooks. There is a chart that I really like, which we have included in the last two or three; it looks a bit like a claw. It makes the point that, in successive forecasts that we have done since 2010, we have forecast that debt is going to rise for a couple of years and then stabilise or maybe fall a bit. As you move on, that gets pushed out and you never quite get to the bit where it is stabilising and falling. We have taken the time to describe the factors that have driven that. Some of it is due to shocks, but some of it is upside news and changes in our forecasts being spent, policy reactions being asymmetric, planned consolidation via spending cuts or tax rises being delayed or reversed by successive Governments over the years, and other things.

It is not our role to make the policy choices to get debt on a sustainable path, but we try to bring increasing focus on the factors that have driven the Augustinian nature of our forecasts, where debts are always going to fall in the future but it has not happened. That has had increasing prominence in our major reports, our economic and fiscal outlooks, as well as the long-term ones.

Lord Burns: I read over the weekend that the Bank of England has published a database of its forecasts against actuals, which someone was examining in the article I read. Is there a published database of OBR forecasts against actuals, unfiltered by OBR text, so that other researchers can examine it?

Laura Gardiner: It is dismaying that you think OBR text is a problem, but, yes, there is a spreadsheet containing only numbers that does exactly that.

Tom Josephs: It has been there for quite some time. On this one I think we are ahead of the pack.

Q87 Lord Davies of Brixton: This is going back a couple of iterations, but Lord Petitgas mentioned that the forecast is right one time out of 10. That sounds pretty good to me, albeit something that could well happen just by random luck. The real issue is not that figure but whether you are going in the right direction. I end up slightly agreeing with Lord Agnew. There are different views, but the real decision is about where we are going on public expenditure and where we are going on raising revenue.

Focusing on this figure, which is the difference between two highly uncertain figures, does it really work? The more we have discussed this whole thing, the more I have lost faith in the idea that that is a sensible way of proceeding. Let us talk about expenditure and income, and not this sort of proxy that mashes the two together.

Tom Josephs: We very much try to bring out and explain what is driving borrowing on both the tax side of that and the spending side of that. In the report we have a lot of analysis on the question: in this medium-term forecast, what is driving the reduction in borrowing? It is primarily the fact that, as a share of GDP, the Government are planning to increase taxes by a significant amount while keeping spending as a share of GDP broadly flat.

We also try to explain the underlying drivers in terms of which of the tax policies are driving the increase in the tax-GDP ratio, and what the risks are around that. There are considerable risks given that a lot of it is driven by personal tax threshold freezes. The revenue from that is very dependent on what happens to inflation and earnings over the next five years.

Quite a considerable amount of that revenue is coming from capital taxes. Again, that is very uncertain. It is dependent on what happens on asset prices but also on the behaviour of taxpayers in response to those tax policy changes.

Then, on the spending side, there is a fairly flat projection in relation to spending as a share of GDP. It relies essentially on a fall, or a slight decline, in departmental spending as a share of GDP.

Going back to the point I made earlier, what we have found in the past is that, when it comes to it, Governments have tended to increase departmental spending compared with the assumptions that they have pencilled in. I am not saying that that is going to happen again, but that is what has happened in the past.

So we try to give a pretty broad picture of the risks around both the tax and the spending side of the forecast. As regards the sustainability of fiscal policy, it is about the difference between those two things and, essentially, about whether those levels of borrowing are sustainable. Can the Government find people who will lend them that money at a reasonable price going forward? That sort of drives sustainability, so it is important that we focus on that as well.

Lord Davies of Brixton: It is more a consequential rather than a substantive issue. It is an important factor, but it is affecting the real decisions, which are about what taxes you want to levy and what public services you want to provide.

Tom Josephs: I guess that is why we are trying to be very transparent about the choices. This Government’s choice in their plan for reducing borrowing and getting debt stable is to do that primarily through these increases in tax. We try to be very transparent on that and, as I said, to explain the risks.

The previous Government were planning to do their reduction in borrowing much more through tightening spending rather than tax increases, although they also had some tax increases pencilled in. Again, we explained the risks around that and the pressures on those spending assumptions in particular.

The Chair: Lord Verjee has the last question.

Q88 Lord Verjee: We have touched on this a little bit already, but some people have called for the OBR to produce only one forecast a year, and market participants giving evidence have told us that this would probably constitute an improvement. What is your view of that? How many forecasts a year would be optimal?

Tom Josephs: The UK has been producing two forecasts per year for at least 50 years, if not longer. It is standard practice in pretty much every advanced economy to produce two forecasts a year. I do not think there is a right number of forecasts, but it is very standard in respect of international fiscal frameworks and fiscal policies.

I think to go down to one forecast a year would be a very significant reduction in transparency. As we have seen over the past few years, a lot happens in the course of a year. Updating the view and understanding of the public finances and the economy every six months feels reasonable.

What the Chancellor has announced is that the Budget is not going down to two forecasts a year. There will still be two forecasts a year, but there will be only one assessment of the fiscal rules and only one policy event. Again, that is quite standard compared with how many countries around the world run fiscal policy, where the first forecast in the year is more of a preliminary forecast and sets the ball rolling for policy development ahead of the main Budget, which is when you have your second forecast and when the Government announce policy. This change moves in that direction.

As we have discussed a bit in our answers to other questions, there is an opportunity for us to use that spring forecast—where there is no policy and no assessment of the rules—to provide more analysis of the wider fiscal position, trying to address some of the points that you have been making around sustainability in a broader context. We will definitely be looking at ways in which we can do that.

One point to make here is that many countries have this model where you have two forecasts a year but one is more of a preliminary forecast; it is not a policy event, but it provides the context for the Government to start thinking about policy ahead of the Budget. However, there is typically a smaller time gap between those two events. So one potential issue with doing this in the UK set-up is that there will be quite a long gap between the spring forecast and the final Budget in the autumn; quite a lot could happen between those two points in time. As I say, other countries have a smaller gap between those two things.

Q89 Lord Blackwell: Can I come back on something? Professor Miles, you mentioned AI at one point. As we reflect on the uncertainties in your £21.7 billion number five years out, as you said, AI and its impact on the economy could be huge; it could be a hugely positive shock for productivity, depending on the extent to which that is offset by a reduction in the number of people who are employed. In looking at longer-term fiscal risks and sustainability, how are you going to cope with that huge uncertainty? What work are you doing on that?

Professor David Miles: On AI in particular, we will learn quite a lot in a relatively short period about how it is rolling out, how people are using it and whether it is more of a complement to labour input—that is, does it improve the productivity of labour and not affect employment negatively, or is it a substitute for people? We will learn quite a lot about that over not the next few years but the next few months.

I am relatively optimistic on the employment front because, if one looks back at the history of technological improvements, one sees that there has always been a worry that such improvements will generate huge unemployment and that we will have to get used to armies of people who do not have work. That has not been the story of the UK or, indeed, of most countries. Certain jobs disappear, but people find other things to do. So I am less worried than some about an army of unemployed people coming down the road because they are going to be replaced by machines.

There are always great sources of uncertainty. This is a good one, in a sense, because AI is more of an upside risk. The worst thing that can happen with AI is that it is a lot of hot air and does not actually change much or boost productivity much. Relative to that, though, there is definitely an upside. So in a sense, it is good uncertainty. It is asymmetric, too; there is no huge negative downside, apart from people who may have invested in AI companies that were overvalued and who, as a result, may have lost some money.

Laura Gardiner: On your question about how we are thinking about this in our forecasts, a substantive bit—probably one of the most innovative bits—of the productivity paper that we published did some of our own modelling, bringing together others’ estimates on the impact of AI. We provided some ranges. We used AI to do the analysis of AI; I was very impressed by the staff’s work there. We have built some frameworks there, which we can continue to look at as we see how this thing unfolds. We are developing our modelling tools to account for different estimates of how AI will affect employment and productivity; we put quite a lot of investment into that this summer.

Lord Blackwell: I love the idea of using AI to predict the impact of AI.

Laura Gardiner: We were very pleased with the staff’s effort on that.

The Chair: Thank you so much. We have taken you across a pretty wide range of topics; we are really grateful for both your time and your candour in replying to all of that. With that, this meeting is now closed.