Treasury Committee
Oral evidence: OBR: Budget 2026, HC 682
Wednesday 4 November 2026
Ordered by the House of Commons to be published on 4 November 2026.
Watch the meeting
Members present: Dame Meg Hillier (Chair); Dame Harriett Baldwin; Chris Coghlan; Bobby Dean; Jim Dickson; John Glen; Andrew Lewin; Seema Malhotra; Dame Siobhain McDonagh; Ms Julie Minns.
Questions 230 - 305
Witnesses
I: Professor Jonathan Haskel CBE, Member, Budget Responsibility Committee, Office for Budget Responsibility; Professor David Miles CBE, Member, Budget Responsibility Committee, Office for Budget Responsibility; Tom Josephs, Member, Budget Responsibility Committee, Office for Budget Responsibility.
Witnesses: Professor Jonathan Haskel, Professor David Miles and Tom Josephs.
I am delighted to welcome the new chair, Professor Jonathan Haskel, who took over only earlier this year. He is joined by the previous members of the Budget Responsibility Committee, Tom Josephs and Professor David Miles. Before we go into the main questioning, Professor Haskel, can you confirm to us that you did not receive any undue pressure from Ministers or the Treasury in general in preparing the economic and fiscal outlook?
Professor Haskel: Thank you for that welcome. I am happy to confirm that that is indeed the case.
Chair: Okay, so no undue political pressure. Brilliant.
Professor Haskel: I will start and I am sure David would want to chip in as well. You are right. We have downgraded our forecast for growth this year to around 1.25% and next year to 1%, and then it comes back to around 1.6% before going back towards potential output growth. That downgrade this year and next year is mostly, as you said in the question, because of energy prices and the negative effect that energy prices might have.
In the models that we work with, when energy prices zip up and have a utilisation effect, the capital stock of firms is less productive than it otherwise might be. Because we rely on the futures curve for energy prices, those energy prices then come back down again. It is a temporary interruption to the path of growth, so in the medium term, as I think your question indicated, John, we are back to where we were.
To unpack your question slightly, that is why, in the short term, we deviate from that medium-term trajectory. Then the question is what determines that medium-term trajectory. The medium-term trajectory is much more determined by our overall forecasts for investment, which we are a tiny bit more optimistic about, and a little extra on AI, which feeds into total factor productivity. That gives that potential productivity growth of around 1.1% and potential output growth of a little higher.
Professor Haskel: The Budget was basically fiscally neutral. It was an extra on policy measures, so £5 billion to £6 billion in the near term on average, and then £5 billion to £6 billion less in the more medium term, if you see what I mean. Relative to other Budgets, it was a small Budget in terms of those fiscal measures. That is statement 1.
Statement 2 is that, as you will know—I have it here in the Budget document itself—there is a chapter on growth. It is chapter 4, if I remember correctly: “Good growth in every postcode”. There are a series of measures set out there, but there are not a lot of details about what those various measures are. Those measures are around devolution. There is some stuff around energy and so forth. When the detail of those measures comes in, we would be in a much better position to take a view on how they would score. Does that help?
Professor Haskel: Uncertainty is bad for investment obviously, because firms wait to see how that uncertainty is going to play out. As we discuss in the document, we would want to look at various indices of uncertainty and so forth, and look at the path of investment as to whether that was held up. That would be an early warning as to whether uncertainty about the future was somehow or other impacting on current growth prospects.
Professor Haskel: I will say a word and I should bring David in as well, if I may. Let me back up a second, but tell me if this is not answering your question. GDP growth has been a little better than most commentators expected for the first couple of quarters. Our view is that that is because the lagged effects of the difficulties in the middle east around oil prices and so forth have yet to show through. That is why we are more pessimistic therefore about the prospects for next year. As far as things being priced in is concerned, as you know, in the document we have a number of different scenarios, which work through what the consequences might be if the middle east situation is more protracted or if it resolves itself.
Professor Miles: As Jonathan said, the first half of this year looked pretty strong. I think, John Glen, you mentioned this. Our forecast for this year, 2026, is marginally stronger GDP growth than we thought back in March before we had taken on board any of the energy price effects of the situation in the middle east.
What is priced in is perhaps not so much a reflection of uncertainty but more a reflection of what you might call the central expectation of the people who put money on the table in the futures and forward markets for oil and gas. That is very substantially higher levels of prices for the rest of this year and through much of next year. Relative to where we were in March, right now oil prices for immediate delivery of oil today are something like 70% higher than they were. Gas prices are probably three times as high as they were at the beginning of March.
It is very hard to imagine that unwinding very rapidly. It could. Who knows where things are going in the middle east? The market is pricing in that those prices stay at really high levels for much of the first half of next year and then start to gradually come down. In our so-called central forecast, we assume that that is a central forecast of what will happen to energy prices.
It causes a squeeze on household disposable income through particularly the first half of next year. It is going to feed in, and it is already feeding in, to the costs of almost every business in the UK, which are going to use either oil and gas directly or certainly electricity, so it also pushes up inflation in the first half of next year, we think to a level of maybe around about 4% at the peak, which would come in the first part of next year. It would be the peak, that is, if you follow the path of the market expectations of where oil and gas prices go, because they start coming down from early next year.
Two or three years down the road, they are a long way back to where they were before the middle east situation became serious back in the spring. One reason why we get a bit of a bounce back in growth further down the road is that you get this unwinding of the energy price shock.
Of course, that might not happen. It is very much a two-sided risk. Who knows where we will be even at Christmas? We could find that the oil and gas prices come down a lot more sharply. That would be very welcome, because it would mean inflation would be lower next year. If that were to happen, in the spring when we do our next forecast, I am sure we would be forecasting both lower inflation, other things equal, and stronger growth. I would put my finger on that as a major source of uncertainty at the moment. There is so much volatility in energy prices.
Chair: Yes, it is difficult to predict.
Tom Josephs: We did not include that announcement in our forecast because it came after the Budget. We will have to review it in detail, which we will do for the spring forecast. I cannot really say now, in advance of doing that, what the impact will be, I am afraid.
Tom Josephs: We have to go through these things in full detail.
Tom Josephs: That is not an estimate that we produced. That is a government estimate of the long-run fiscal costs over the whole lifetime of migrants. That is not a number that we have produced or inputted into.
Tom Josephs: We will assess the policy as a whole.
Tom Josephs: If we think it is significant enough to adjust the forecast, we will do so, yes.
Chair: I think we will see that, whether it is or not, in the next forecast. The pudding will be cooked with whatever the ingredients are. That is a very bad metaphor.
Bobby Dean: Professor Miles, can I go back to the impact of energy prices on living standards? There is a really striking graph.
Sitting suspended for a Division in the House.
On resuming—
Chair: Welcome back to the Treasury Select Committee on Wednesday 4 November 2026. We were just warming up in our session with the Office for Budget Responsibility’s Budget Responsibility Committee. I am going to hand back to Mr Bobby Dean MP, who was mid-sentence. Would you like to start the sentence again, Mr Dean?
John Glen: Which graph was it?
I want to unpick that. You have said that this takes account of some recovery in energy prices already. Can you tell us about what other factors are at play here that keep that figure so low, despite your expectation that energy prices will recover?
Professor Miles: There are two major ones that are ongoing. The first, which has been the story of the UK economy back to just after the financial crisis, is one of productivity growth holding back the growth of GDP per person to levels that have been unusually low for an extended period now. We think that there is slightly better news down the road, but no return to the kinds of productivity growth rates we saw almost every decade since the second world war. That has already had a very big effect on standards of living over the last 20 years. GDP per person may be of the order of 30% lower than you might have expected before the financial crisis if the UK, over the next couple of decades, were average relative to the last 60 or 70 years or so.
Then there are the energy prices, which we have talked a little bit about. It is a very sharp increase now. It somewhat gradually unwinds, but has a lasting effect, certainly for the next two or three years, on standards of living.
Then there is an ongoing increase in the overall effective tax rate in the UK. Much of that comes because of the freeze on thresholds where people move to higher rates of income tax. That is pushing more people into higher-rate bands. It is increasing the average tax take out of people’s income. That comes straight out of disposable income.
It is those three things, which, if you look much further down the road, may be lessened. It depends a bit on how much longer we are going to have the thresholds squeezed. They are responsible for a continuation in the short term of some declines, we think, in average household disposable income and then a return to growth. If you took, let us say, the decade from just before covid to where we think the economy might be in 2029, so over that 10-year period, on our central forecast anyway, we think that the overall growth in real household disposable income is about 1%, or maybe a little bit north of 1%.
That would be an exceptionally low growth in average household disposable income and probably unprecedented since the second world war. Wartimes may be somewhat different. In a sense it started with the financial crisis, but then we have had a series of other shocks more recently, covid being one, and two very major energy price shocks. They are the main reasons why it has happened, plus the fact that, in order to try to bring borrowing down and get back to a more sustainable position down the road, one way that is happening is through a fairly substantial ongoing increasing taxation.
Professor Miles: There are definitely risks on the upside, so on the good side, as well as there being some on the downside. The downside one would be that energy prices, for example, do not come back down as the futures and forward prices suggest and we stay at prices that, for gas, are three times as high as they were at the beginning of this year. That will have a lasting effect on people’s disposable income if that is where we stay. There is as much chance, in some ways, that things are much better than that.
One could view the futures and forwards prices as, if you will, the central expectation. There are significant probabilities that you will be on one side or the other of that. It is certainly not all potentially worse than that. There is a good chance that it will turn out to be better than that.
We also, in a way, take a relatively modest view on the impact of AI. There are plenty of optimists out there who think that, even over a period as short as the next four or five years, artificial intelligence will boost productivity and feed through to incomes, although there may be some job losses, but people who keep their jobs may find their incomes are higher, along with productivity and profits, which tend to feed back into household income indirectly as well. I would not want to paint our central forecast as one where it is going to be either that bad or worse. I definitely do not think that.
Professor Miles: That would take an awful lot of good luck, in a way. It is very difficult to move the dial that much through policy. It is not impossible. AI has done stuff that nobody would have believed possible in the space of the last five years, and energy prices are so much higher than people thought and could drop out very sharply, but that would be some going to get up to that level.
Professor Miles: It is pretty difficult in some ways to understand what has happened to productivity. A big driver over the long term of productivity is what people discover that is new that makes them produce more output more effectively. That drives wealth over time. A 10-year period, or even a 15-year period, where there are just a few fewer really transformational inventions can explain quite a lot of it.
It has also been a period where investment in the UK has been pretty weak since the financial crisis, with some bits of recovery now and again. The UK has never in modern history been a very high-investment country. We have certainly continued to be at the low levels of investment relative to most other rich countries.
It is a combination, in that there have not been many obvious transformational technologies—and I am putting AI to one side because it has not yet probably had its major impact on productivity—and investment has been relatively low. The real technology pessimists have said in the past that the most transformational technologies over the last 150 years are the invention of steam engines, electricity and huge advances in medicine. We cannot expect that to keep on happening every century. It did not happen for many centuries up until the 17th or 18 century, where, as far as the economists can make out, there was very little productivity growth for maybe 1,000 years. The really pessimistic view is—
Bobby Dean: A thousand years of low growth.
Professor Miles: The period of the last 100 years or so is the exceptional one, rather than being what you can expect for ever more into the future.
Chair: I think the OBR might be making headlines today.
Professor Miles: I am not that pessimistic.
Chair: If I could paraphrase that, “We are all doomed”.
Professor Haskel: Thanks for the question. We do not yet have a concrete quantification of that. That is the honest answer. That would have to work through the supply and demand for rental accommodation. There is elasticity of substitution between renting and owner occupation.
Professor Haskel: Yes. I am not sure that I could quantify that on the basis of the work we have done thus far.
Professor Miles: There is a lot of variation, for what it is worth, among some of the economists who have spent a lot of time in the last few years trying to work out how transformational, or not, AI might be. I think that I may have mentioned this to the Committee before, but I will be very brief about it. Two of the most recent winners of the Nobel prize are at absolutely two ends of a spectrum on how much of an impact on productivity AI will have.
Daron Acemoğlu won the Nobel prize, I think, three or four years ago. He thinks that, over the next 10 years, even in the US, where adoption of AI is probably ahead of where it is in the UK, it barely moves the dial on productivity. Another economist, Philippe Aghion, who has done a lot of work on productivity and on AI in particular, takes the view that it will be little short of transformational over a 10-year period, and that in countries that are using AI, of which the UK is one, you will get back to the kinds of productivity growth rates that were normal in much of the period since the second world war up to the financial crisis, so 2%, 2.5% or 3%.
If that really happened in the UK, it would generate changes just fiscally, holding other things constant, that could almost be described as transformational for the fiscal position. Even 0.5% higher productivity growth maintained over, say, five to 10 years, is worth something like a fiscal improvement each year by the end of that period of £50 billion to £60 billion. If Aghion is right, things are going to look an awful lot better five or 10 years down the road. If Acemoğlu is right, even our forecast with a modest assessment of what AI does is a bit too optimistic. They are smart people and both won Nobel prizes largely for work in this area.
Professor Haskel: Can I advertise table 2.2 in our report, which goes a little bit to your question, if I may say? It unpacks these various effects in different sectors. One helpful way of thinking about AI, which your question kind of indicated, is that there is going to be a sector that is producing a lot of AI-related goods. That might be computer hardware, for example, or software, which is large in the UK. Then there are sectors that are going to be using all of this. The health service will be using better software to process patients and all that sort of stuff.
In table 2.2, we look at the productivity record in the different sectors. As I say, I think it talks to your question. In the information and communication technology services sector, which is basically writing software, although it is a little broader than that, it has productivity growth in the last two years of 7.6%. That is an astonishing rate of productivity growth, given that the rest of the economy is growing at a couple of per cent if we are lucky. That seems, to me at least, to be an indication that, at least in that sector, that is a strong effect of AI.
As David has said, to work out how it goes out through the whole of the rest of the economy, it depends on how else that spreads in the different sectors. That is the thing that we are looking at very strongly. I hope that that unpacks that question a little more.
Professor Haskel: We will look at that. These are experimental numbers by the ONS, as you know. It has a different way of measuring hours. You have had a lot of discussion in front of this Committee on the labour force survey, which has been through a whole load of trials and tribulations. The ONS has a different way of doing it.
Those revised hours data, which are doing all the work in revising productivity per hour, as you know, are not official ONS statistics. I forget the exact badge. I think they are still experimental statistics. If they are badged as official statistics, we will obviously take them on board and rework in the March forecast. Tom was talking about the changes that we would make in the March forecast. That might be one of them as well. We have to wait for the ONS to move on this, rather than us second‑guessing what the ONS might do.
Tom Josephs: I will say a bit about this; my colleagues want to come in.
As you say, this is the first time that we have made a full assessment of the ERA. Over the past couple of years, it has very much been a set of legislation that has passed through a lot of consultation and a lot of changes to secondary legislation, but we felt that at this point there were enough final details confirmed that we could take a fuller view of it.
The other thing to say is that, over the course of that two-year process or whatever it has been, it is definitely the case that a number of the elements of it have changed quite significantly since the initial proposals in a way that we basically see as having less of an impact on business than the original proposals would have done.
Our assessment is that it will have a negative impact on business costs. As you know, we have a threshold for measures that we score in the forecast as having a quantifiable negative impact on productivity.
Tom Josephs: It is 0.1% of GDP. We assessed that it was too uncertain to say that the ERA would exceed that threshold, but that it would have a negative effect. Part of the reason is that it is very difficult to quantify with certainty the impact of regulatory changes such as this. By their nature, they are inherently more difficult to quantify in a very precise way. It is also the case that it is quite difficult to determine the wider regulatory framework within which changes such as this fit in and how they change that relative to the regulatory framework.
Tom Josephs: No, that would not be large enough to meet the threshold.
Tom Josephs: In this case, I do not think it would. There are certainly some policies that we have assessed where we have said that the impact on growth is likely to grow over time. When we made the assessment of the Government’s initial increase in public investment, we said that, if it was sustained over time, you would see a bigger impact on growth over a longer-term time horizon. The same applies to the assessment we made on the planning reforms, but in this case it is likely that the impacts would materialise more quickly than for those.
Tom Josephs: It is still hard, yes.
Professor Miles: If we saw, over a sustained period, that vacancies in the UK, which have fallen quite a lot in the last few years, just carried on falling, if companies were consistently reporting a reluctance to hire people and were saying, via institutions such as the CBI in surveys, that one of the reasons for that was this legislation, and if that really was emerging as a clear pattern, we would then take that into account with our projections of employment and levels of vacancies.
Where we are right now is different from perhaps where we were 18 months ago at the beginning of this journey. The initial message we were getting from employers, employer organisations and the CBI, as people will be familiar with, was that they were expecting a very negative effect of the proposed legislation.
Then what happened, as you will know better than I, was that there were various changes made. They were quite substantial changes. Now there are probationary periods; it looked initially as if companies would not have probationary periods. Zero-hours contracts initially were going to be essentially outlawed, and now that is not the case. The message we were getting was that the potential negative effects were pretty clearly diminishing through time.
Chair: Thank you. That is helpful and something for us all to keep an eye on.
Chair: Who wants to take this? Which of our professors wants to step in?
Professor Haskel: I am hesitating about this, because of course it depends upon what the headline wholesale prices are. As we have said, the wholesale prices are around 47% of the final bill, so there is not a one-to-one pass-through from the wholesale prices to the final bill and what the cap will turn out to be.
It is pretty difficult to forecast what the energy price cap is going to be. We have therefore taken the futures curve, which we have just been discussing, as our central way of forecasting what the future cap is going to be, but how that will turn out obviously depends upon the types of uncertainty that we have been discussing.
Professor Haskel: No.
Professor Haskel: Indeed, yes.
Professor Haskel: That is going to come off in July next year, if I remember the measure correctly. Sorry, you are asking about if it did not come off and if it were extended past July.
Ms Minns: Yes, if the VAT cut is extended beyond July next year.
Professor Haskel: Then we would have to go back to the drawing board and redo it, but at the moment that is what the Government’s announced policy has been, and we have done our forecasts accordingly.
Professor Haskel: Yes.
Professor Haskel: Am I concerned? Again, from a forecasting point of view, if it is a temporary measure, such as that measure we were just discussing that comes off in July, we stop it in July and then work out the rest of the year, an average over the year and so forth. That is all rather fiddly, if you see what I mean. I am concerned for our staff’s wellbeing in dealing with all these fiddly measures, but, as for whether I am concerned in some broader sense, those are the measures, and we will forecast as best we can on what the measures are.
Tom Josephs: We have a section in the report that sets out where there are temporary measures that are reoccurring temporary measures, and we flag those as risks to the forecast for the reasons you are outlining.
The biggest and most obvious one is fuel duty. Fuel duty rates have not been increased since 2011, despite the Government at every single fiscal event setting out a policy that suggests that they will increase them at some point in the future. That has happened again at this event.
On that one also, following a request from this Committee, we set out the fiscal implications if the Government were not to follow through and increase fuel duty rates. Our estimate is that that would increase borrowing by about £4 billion a year over the forecast period.
The cumulative cost now of not increasing fuel duty since 2011 is about £130 billion.
Professor Miles: In some ways, the biggest measure in terms of pounds equivalent to help households with energy bills was the £100 council tax deduction, which probably hits 80% or so of households. That too is temporary; it is a one-off, in a way.
The reason why a lot of these measures are temporary is just because of the fiscal situation we are in, which is one where the fiscal deficit is still large. Borrowing this year is still almost 4.5% of GDP; if you stay in that position, you are going to blast through the fiscal targets. The stock of debt to GDP will just carry on going up.
The Government took the view that, in order for that not to happen, you could not offer a package of relief to households that was anywhere like as big as the package in 2022, which probably, in pounds terms, ended up being about 10 times bigger than this one. I am sure the Government would have liked to have done a bigger package, but they were obviously mindful of trying to hit a target that, if they miss it, would then have debt relative to GDP continuing to go up.
Chair: I am sure we will be able to put those points to the Chancellor next week.
Professor Miles: It is partly that the labour market seems a bit softer now than it did then. We have talked a bit about vacancies already today. If you go back a few years, the number of vacancies in the UK was about the same level as the total level of unemployment; that is a pretty tight labour market. Vacancies have come down a great deal, and now the measured unemployment rate is something like four times as high as vacancies.
That slack in the labour market, if you want to call it that, makes it a little bit more difficult for workers to negotiate wage settlements that pass on most of the effects of the higher inflation. We think that wages will not, in the near term, rise by enough to match inflation. That is one of the factors behind real disposable income being a bit squeezed.
There is another factor there, which is the rate of return on corporate capital. If you go back a few years and take an average over a couple of decades, the rate of return on corporate capital, a measure of profitability, was 13% or 14% or so. That was in a world in which interest rates were lower.
There has been a bit of a double whammy there because, as far as one can make out, the measure of the profitability of the corporate sector has come down a bit, so it is more like 10% or 11% now rather than 13% or 14%, but that is in an environment where interest rates have gone up a lot.
Just to give you an indication of the scale of how much interest rates have gone up, if you go back just three or four years, the real, inflation-adjusted interest rate the UK Government were paying on inflation-proof, index-linked gilts was minus 3%; that is an extraordinarily low level. That was about three or four years ago. It is now plus 2%. That is an increase of 500 basis points, or five whole percentage points, in the real interest rate.
That is the number you might compare with the real profitability of the corporate sector. Profitability has been squeezed a bit and the cost of capital has gone up a lot. That means that many companies, not all, will have a greater determination not to see profitability squeezed further in that environment. One thing they can do is push back against wage increases of the 3% or 4% you might need just to even match inflation.
Professor Miles: It comes back to the temporary nature of the support packages for households. They have their biggest effect early next year. They then run out, and the year-on-year picture shows inflation moving back up.
Chris Coghlan: I was moderately worried by your forecast of 1,000 years of low growth.
Chair: I should stress for anyone watching that that is a slight paraphrase of Professor Miles, who was being humorous.
Professor Miles: That was a description of the past, not a confident prediction for the next 1,000 years.
Professor Miles: Yes, there are, and some of them were mentioned in the Budget documents by the Chancellor. One is the area of planning and planning reform. We have had phase 1 of that, if you like, which was directed at residential housing, back in the Budget a couple of years ago. The Government are saying that there is much more to do on that when it comes to business investment, planning and infrastructure, and the positive impact that might have on commercial investment. It is linked to investments in the grid and infrastructure. Those are factors that have potentially held back investment in the UK.
Professor Haskel: Total investment is around £600 billion in the whole economy, so £5 billion is a small number.
Chris Coghlan: Total investment in public R&D is around £20 billion.
Professor Haskel: I beg your pardon. Yes, that is total investment in both non-R&D and R&D.
Chris Coghlan: In R&D it is about £22 billion, I believe.
Professor Haskel: There is a fair amount of evidence that the rates of return to scientific R&D and basic R&D are pretty large. The nature of that in the military depends upon what it is spent on now. This is just a casual reading, but I suspect that the types of investment one does in military R&D are very different from what they might have been in the past. Presumably, in the past it was about materials, tangible things, making more bombproof tanks and so forth; I would imagine that it is probably much more oriented to software now, so one would have to have a look at that in a bit of detail.
Professor Haskel: We talked about this a little bit at our press conference. Can I pass over to Tom?
Tom Josephs: Just to give you the background on how we do the forecast, we use production investment projections that are produced by the North Sea Transition Authority. It produces a very detailed field-by-field assessment and projection. For several years, it has assumed in those projections that Jackdaw and Rosebank would go ahead. It makes various other assumptions about the number of undeveloped fields that are likely to be developed in the future. I should emphasise that that is not a new thing; it is a completely separate thing from the Government’s actual decision on that field.
The big picture, if you look at the forecast of oil and gas revenues, is that they have been for a long time and are projected to continue to be on a downward track. Currently, they are elevated because oil prices are obviously quite elevated. We are estimating this year that we will raise around £5 billion or so, but in five years’ time we assume that oil and gas prices have fallen back down and that production continues to decline. Also, a lot of companies are using past losses to offset profits in terms of their tax liabilities. We currently project that overall revenues from the whole of UK oil and gas production will fall to close to zero at just a couple of hundred million in total. That includes all fields.
I cannot give you a number specifically on Rosebank. That is confidential taxpayer information and we do not see that. HMRC sees it, but we do not.
Andrew Lewin: The trend is clear regardless. Thank you.
Tom Josephs: I am not sure whether it is the first time we have mentioned it, but you should not read anything into it. The North Sea Transition Authority has had that assumption in its forecasts for several years, and it is a forecasting assumption; it is completely separate from the government process to take a decision on it.
If you compare that with the typical revision to your forecasts between fiscal events, is the Chancellor more likely than not to use up that headroom before the next fiscal event?
Professor Haskel: I will have a go at that, and I am sure David will want to come in as well. You are right that £15 billion is small relative to the typical forecast revision. Those revisions are around £18 billion, but of course they go up and down, if you see what I mean, so we cannot say whether the Chancellor is more or less likely to do it.
Professor Miles: Yes, we do. That is the method by which we come up with a probability that, if the Government stick to the policies that they have said they are going to implement on spending and tax, what are the chances, three years down the road, that that headroom is still the right side of zero? The answer is that it has about a 55% or 56% chance of being on the right side, so it is better than tossing a coin but not all that much better than tossing a coin.
The reason is not because it is a crazy policy; it is more that the headroom, as you would appreciate, is the difference between two huge numbers. It is government spending of £1.4 trillion and tax revenue of £1.4 trillion, and it does not take an awful lot of a hit and a shock to move those two numbers apart by £20 billion or £30 billion.
Professor Miles: Yes, it is, and 56% is what it is. It is not 60%, 70% or 80%; it is 56%. Of course, there will be fiscal events between now and three years down the road. At those, if it looks like the Government have been blown off track and plus 15 has turned into minus 15, that is probably going to generate changes in policy.
Professor Miles: There are, and we have mentioned some of them already. There is a spending review coming up. There are tough decisions to make about exactly which Departments are going to be squeezed somewhat to hit promises that the Government have made on defence, health and, to some extent, education.
Yes, I would agree with you that there are big decisions to be made, including on welfare. There is the Milburn report coming up, as well as the Timms review; they look like they are going to be sometime early next year. There are big decisions, yes.
Professor Haskel: Can I quickly add something that might help with the discussion? As well as talking about the £15 billion headroom, as David has just mentioned, and the 56% probability, we have also set out what headroom you would need to get a 60% probability or a 70% probability, to give one some sense of what you are asking, which is about what the uncertainty is around all of this.
If I read my notes correctly, to get a 60% probability, you would need £25 billion of headroom; to get a 70% probability, you would need £50 billion of headroom.
Professor Haskel: That tells you again that, in this noisy world that you are discussing, the headroom number is not that informative.
Professor Haskel: That goes back to what Tom was saying before about how this Committee has requested that number. Is that right?
Tom Josephs: Yes. We have a legal obligation to do our forecast based on announced Government policy, which is why the central estimate of headroom includes the assumption that the latest policy goes ahead. This Committee has written to us previously.
John Glen: I was not on it.
Chair: Dame Harriett was in the hot seat at that point.
John Glen: It just points out that, if the decision is as it has been for the last 15 years, it will be £5 billion lower.
Chair: It is worth just clarifying that the OBR has to act on the policy as the Government present it.
John Glen: It is as they present it to date.
Chair: To date, yes. Government policies can change, as we know, under different Governments and under the same Government, but that is the basis on which the forecast is written.
Professor Miles: Net migration, relative to where we were in March, is lower by about 70,000 a year over the next four years. It is not trivial; it will affect the labour force. Of course, it depends on what the people who do not come would have done had they come. We make the assumption, which seems to be in line with recent history, that they have around the same participation rate as the domestic working-age population. It is a slightly lower labour force than would have been the case, because over four or five years there are almost 300,000 fewer people, a significant chunk of whom would have worked.
GDP per person, though, which is a much better measure of the average standard of living of people in the country, is not any lower, because the people who would have come broadly have similar characteristics to the domestic population. On average they are a bit younger and more of them work, but on average they get paid slightly lower wages, and you could describe that as lower productivity. Those two things balance each other out.
The people who do not come would not have materially changed the average resources or incomes of people in the UK. Per-capita GDP literally does not move at all. Aggregate GDP would have been a little bit higher. That would have brought in a little bit more tax revenue, but then again 300,000 people will draw on public services.
Our view on that was that it did not do anything to GDP per capita, which is a much better measure of the standard of living of people than is aggregate GDP. The fiscal implications are relatively small.
Professor Haskel: I have said this to this Committee before, so forgive me for repeating myself. In practice, if you look at other institutions that have done a 10-year forecast, such as the Congressional Budget Office in the US, often if you look at their graphs of GDP, for example, they are quite detailed for the first five years, and then for the following five years they are just flat lines. There is a limited information set that one could get on that.
On the other hand, I am sympathetic to the notion that Tom was just talking about a little bit earlier on. Some of these policies take time to play out.
Tom Josephs: I can add a couple of things. First, the choice of the length of the forecasts for Budgets and fiscal events is for the Chancellor, not for us. The choice of the fiscal rule horizon is for the Chancellor. At fiscal events, this would be a decision for the Chancellor to take.
We do longer-term projections over 50 years, as you know, in our fiscal risks and sustainability report. One thing we could consider doing in those reports is looking a bit more closely at the next 10-year horizon, for the reasons you set out.
One piece of analysis that we did in this report that is relevant to this issue was that we looked at what was happening right at the end of our five-year fiscal forecast at the current time. Typically, towards the end of a five-year fiscal forecast, where you generally assume the economy is back at trend, and you assume that departmental spending tends to be rising in line with GDP, you would tend to see an improvement in borrowing, just from fiscal drag and the fact that earnings will be rising more quickly than tax thresholds.
In this latest forecast, before the impact of policy measures, borrowing was increasing a little bit. That is basically because we are starting to see some of these pressures that we talk about in our long-term report already starting to hit the public finances. Demographic pressure is pushing up pensions; it is pushing up borrowing at that time horizon already. We are starting to see a loss of tax receipts from things such as fuel duty due to the transition to net zero as well as the decline of production in the North sea.
We are starting to see a continual increase in debt interest costs due to the large stock of debt and the fact that the Government are having to roll over a lot of debt that they originally financed at lower interest rates; we project that they will have to refinance that at higher interest rates.
You are already starting to see some of those pressures emerging even at the five-year horizon, so if we were to do a 10-year forecast, you would probably see those pressures emerging even more clearly over that time horizon. It would likely show you an increasingly difficult fiscal environment.
In addition, two policy measures with direct impacts on the economic forecast were altered after it had been finalised. You say it makes a small difference, but we still do not know what those measures were. Could you share that with us and also explain to us what the level of difference was, even though it was small?
Tom Josephs: We mention in chapter 3 that small differences were made to the final parameters for the fuel duty policy. As it says, these were pretty small.
Tom Josephs: As I said, it was a very small difference. I will have to get back to you on what the two policies were.
Bobby Dean: You can write to us. Thank you.
Professor Haskel: I am not sure we do, because this is a complicated picture. David Miles should come in. The implications of the debt stance for yields and borrowing and so forth interact, in a way, with not only the gilt yields but growth prospects and things like that as well. There is a whole set of different circumstances that are pushing all these things in different directions.
We have a whole box at the end of the report, as you know—I think it is box 7—where we talk about the difference between the interest rate that Governments have to pay on borrowing and the amount of growth that affects the numerator of the debt-to-GDP ratio. It affects both the numerator and the denominator.
That is a very complicated picture. It is hard to get at all of that. At the heart of it is the question about whether one thinks that these yields, which are high in historic terms relative to the immediate past but not high relative to the longer term, are going to come back down again. That takes one to questions about what the future prospects are for productivity and things like that in the economy.
Professor Miles: I have two very quick points. First, some of the countries in the euro area that have higher debt than the UK—France as well as Italy stand out—are obviously issuing their Government bonds in euros. There is a sense, to use Mario Draghi’s words of however long ago it was—it was more than 10 years ago now—that, if a country in the euro area runs into problems, the ECB will do whatever it takes to stop that causing disruption in the euro area. Obviously, here we do not have that protection, as you could call it.
It has probably been the case that it is a consistent pattern, and not just in the most recent period when yields have gone up, that sterling gilt yields are a bit higher than euro-denominated sovereign debt. There may have been a time when Greece and Italy were actually higher than the UK, but in a way they were one of the things that prompted the Draghi statement about doing whatever it takes to stop this thing falling apart.
Some people look at where the yields are on gilts at the moment and say, “They have gone up so much in such a short period of time. Surely the next move is more likely to be down than up”. In some sense, the IMF has put the UK on the good step and the US and France on the naughty step. Will we come back down from here?
It comes back to something Jonathan just said about how you view 10-year gilt yields at the moment being up 5.3%. Do you view that as, “That is above where they should be, given that the UK is trying to do something about its fiscal situation”, or do you view it in a longer historical context and say that five and a bit per cent. on 10-year gilts is absolutely not unusual in history? There have been plenty of periods when inflation was lower and the fiscal situation was a lot better than it is today when yields were at that level or even higher. There is a lot to be said for looking at much earlier periods than just the last few years.
Professor Haskel: That is a forecast for unprotected Departments, as you have just said, which depends upon the assumptions you make about the protected Departments. I am not sure we have ever forecasted unprotected Departments, have we? I am not quite sure. Forgive me; I am turning to you, Tom.
Tom Josephs: As you know, the current spending review period runs until 2028-29 and 2029-30. After that, the Government just set an overall assumption for total spending. That estimate of real-term cuts for unprotected Departments is based on what that assumption leaves to those unprotected Departments after you have taken into account some spending pressures and commitments, mainly in big Departments such as health, defence and education.
It is the case that, in the past, Governments have reduced spending as a share of GDP, so it can be done. It is essentially a choice for Governments to take as to spending allocations, but what we are doing here is just highlighting the kinds of risks and pressures to the current spending assumptions.
Tom Josephs: Defence is factored into that calculation. Underpinning that calculation is an assumption that defence rises to 3% by 2030-31. The increase in defence spending is part of the protected Department calculation.
Tom Josephs: On the efficiency targets that this Government have set, we also flag that as another source of potential pressure to those spending numbers, in the sense that, if the efficiency targets that the Government have set are not met, then that obviously increases the pressure within existing departmental spending settlements.
If you look back at previous efficiency targets that previous Governments have set, it is often the case that they have not been achieved.
Tom Josephs: We do not take a view on whether those efficiency assumptions will be met because, as I say, beyond the spending review period, we basically take the Government’s assumption for overall spending. That is what we put into our forecast because, beyond the current spending review period, that is government policy. They set an overall spending assumption as policy.
In all of this analysis, we are highlighting the kinds of risks and pressures around that assumption, one of which is the achievement of the efficiency targets; another is all the policy commitments that you are talking about. Ultimately, it is a government decision as to what spending allocations they will make when they come to the next spending review.
Tom Josephs: That is right.
Chair: We can do a back-of-the envelope calculation. We know it is going to be challenging.
I just wanted to check whether that underspend being lower is down to project progression or higher levels of inflation, or a bit of both? Do you have any insight?
Tom Josephs: We have not done that kind of detailed assessment because we do not look, project by project, at what has led to the higher capital spending than we previously estimated. It is basically just a top-down assessment based on the progression of capital spending to date through the year.
It could be a combination of both of those things. There probably has been some increase in costs due to higher inflation, but also it is likely just that Departments have been able to progress projects more quickly than we expected.
We made the original assumption that Departments would underspend their capital projects by quite a significant amount, based on looking back at the past, at what had happened in previous periods where capital spending had been increased very quickly. We found that, in the past, Departments had underspent very significantly, but it seems that this time that has not happened to that extent.
Tom Josephs: You would need to confirm that with the Treasury.
Chair: We have the Chancellor next week.
Tom Josephs: I do not recall that being the case, but you should check with the Treasury.
Chair: That is a question for Government.
In terms of the upside and downside risks in your assumption, you think it could be potentially, if it carries on at the same trajectory, as much as £25 billion higher than you forecast. It is paragraph 5.22, if you are looking for it. Then, if it returned to pre-pandemic levels, it could be £13 billion less than forecast. These are just huge numbers.
In terms of those assumptions, we hear from the Government that they are planning to look at welfare spending “in a Labour way”, in the Prime Minister’s words. That involves employment support programmes. Have you evaluated any employment support programmes in those numbers?
Professor Haskel: That is a very good question. No, we have not. Sorry to repeat this, but were the Government to come forward with some particular plans and all that, we would evaluate them in the normal way.
Professor Haskel: I am not sure.
Tom Josephs: In the past, we have evaluated the potential impact of some employment programmes. We have a bit of an evaluation of how two of those programmes have performed against our original evaluation in the report. One of the policies was just not implemented.
Tom Josephs: Yes, and the other one was delivered but with significantly less support than originally envisaged. That just highlights the fact that it is quite difficult for us to make these assessments of the impact of departmental spending programmes such as these employment support programmes, because so much of it depends on subsequent delivery and funding. In these cases, we found that big changes to delivery funding subsequent to the initial announcement meant that the programmes looked very different from the ones we originally scored. That is going to make us think very carefully.
Tom Josephs: Yes, exactly. We have to have quite a high bar in the future for including impacts on these sorts of programmes in the forecast.
Professor Haskel: Paragraph 3.35 is the paragraph that Tom was talking about where we looked at the extension to Talking Therapies, which we estimated would increase employment by 10,000, but that extension was not delivered. The Connect to Work programme turned out to be about a third lower than had originally been projected. That may add a bit of colour to what Tom was just saying.
Chair: There are some very important lessons for Ministers there about what they promise and what is delivered. I am sure they are listening.
I wondered if you had made any assumptions in your economic forecasting around whether effectively taking capital from the private sector and giving it to the state-owned banks is likely to be a more or less productive use of capital.
Professor Haskel: No, we have not thought of it that way. We have just tried to do the fiscal arithmetic around the extra revenue that it would generate and the extra spending. That is all worked out.
We did a bit of work on the extent to which we thought that the cost of bank lending might be affected, but it is a tiny number, if I recall correctly.
Tom Josephs: That is a slightly different thing.
Tom Josephs: The £190 million that you mention is an estimate of the degree to which revenue from that measure would be lost because banks will shift profits overseas. Then the impact on bank lending is a separate issue.
Dame Harriett Baldwin: You did not think it was material.
Professor Haskel: Correct, yes.
You have said that you think it is not possible at the moment to assess the extent to which that is going to add to growth. Do you have any idea when you might be able to do that at all?
Tom Josephs: The plans the Government set out at this event were just initial plans at a fairly high level. That is what they are saying themselves. Further firm details will come in subsequent announcements.
This is just consistent with the usual approach we take to assessing policy. We only assess policy when there is enough final and firm detail to allow us to make a reasonable assessment of the potential impacts.
On devolution as well, it is fair to say that the impact, if you are thinking about the economic impact, is likely to be one that is pretty difficult to assess, because it will be very dependent not only on the final policy that the Government announce, but also on how it is then implemented at the local level. It is very dependent on decisions that local leaders take, local capacity and all those kinds of things.
If you look at the economic literature on this, that is basically what it says. The economic impact is very dependent on delivery, quality of local leadership and all that kind of thing.
Tom Josephs: No, and it will probably be quite challenging to do that even when the final decisions have been taken.
Tom Josephs: Again, it very much depends on the final details of this. If you look again at the literature on fiscal risks associated with devolution, there are definitely potential benefits from local leaders and local areas being able to better target spending in their particular areas, but there are also risks around the potential for tax competition, a race to the bottom, competition between different regions and more complexity in the tax system.
There are definitely risks around revenue volatility at the local level. A narrower tax base potentially means more volatility, and that can be a challenge for local administrations to deal with. Again, that all very much depends on the final design of the system.
Tom Josephs: I would not say so at this stage. As I say, we will look at the final details, the overall package and how it all fits together to make that sort of assessment.
Tom Josephs: It varies from policy area to policy area, but typically we would need to see firm details of the rates that will be set, the actual amount of funding that will be devolved and the final details of any revenue equalisation type arrangement that the Government are considering. Those kinds of things have currently only been spelled out at the high-level framework type detail.
We heard very strong evidence yesterday from witnesses about the lack of accountability at the moment. We do not know what will come out in the paper. Is there any message you have for Government about how that will help you assess whether every tax pound is being spent as well as possible?
Tom Josephs: Our job is not really around the accountability in the system; that is more an NAO area. Our assessment will be on any fiscal and economic implications.
I would definitely emphasise the point that you are making. This will be very much dependent on final decisions that are made at the local level. A good example may be devolution of tax powers to Wales and Scotland. It took around nine years from the original announcement of devolving tax powers to Scotland before Scotland actually took some decisions to change tax rates.
Chair: So it is quite a long way off, potentially.
Tom Josephs: In Wales, it was about seven years before they got the powers, and they have not used those powers yet. If we had tried to make an assessment back when those original announcements were made of what the impact would be, you can see that it would be very difficult to do.
Tom Josephs: This is in terms of that fairly narrow behaviour that I was just describing around potentially shifting profits overseas.
Tom Josephs: That effect would be lower for a temporary increase than it would be for a permanent one.
Tom Josephs: No, we just assessed the policy as the Government put it.
Yesterday, we heard some very trenchant evidence from Helen Miller at the Institute for Fiscal Studies about founders’ relief. She described it as a dreadful policy. I am someone who represents a lot of founders who have then sold out. Some of them have stayed around, but some have sold out to big American companies and made some good money out of it, sometimes founding another business, I should say. We can see that there are challenges with how you do this.
What is your assessment? Have you modelled the responses to this new policy? Do you have any assessment about whether you think it is going to grow businesses, or do you agree with Helen Miller’s analysis?
Tom Josephs: As you know, it is not for us to comment on the merits or otherwise of the policy.
Chair: Indeed, no. It is just on the modelling and what you think might happen.
Tom Josephs: For us, this new founders’ relief is a fairly small measure. Across the forecast, we estimate that it only costs around £200 million or so. It is not a huge change and therefore is not one that we assessed in detail in terms of the economic impact that it would have.
Tom Josephs: It is obviously an important policy for those people who are directly affected, but in terms of the overall fiscal forecast, it is not a huge policy and therefore not one that we looked at in detail in terms of the economic impact.
The Government also announced a reduction in the allowance for the BADR relief, which has a roughly offsetting cost. There was more than one CGT measure announced.
Tom Josephs: As I say, at this Budget the measures were relatively small-scale in terms of the overall fiscal impact. Therefore, we did not make a detailed assessment of the economic impact that they would have. That is not to say that they will not have important impacts for those people or businesses that they are particularly targeting, but from a macroeconomic perspective—it is our job to look at the macroeconomic and fiscal impact of these measures—we did not make that kind of assessment for these measures.
I have asked some parliamentary questions that have highlighted that a lot of the people who are being asked for the first time are over 80 or 90; I had some people over 100 recently. They are suddenly being sent these self-assessment forms for the first time in their lives. How realistic is it that £3 billion of extra money is going to be raised by HMRC resources being boosted?
Tom Josephs: It is definitely an area that we highlight as very uncertain. There is quite a lot of risk around the estimates that we make of the additional revenue from compliance measures and measures that aim to reduce tax debt. We have quite a lot of analysis in the book about the way in which we assess individual measures of that type to try to ensure that the estimates are central.
We do a lot of work with HMRC to think about, “Okay, this measure will target this non-compliant section of the taxpayer population, but that population might move on to find different ways to avoid tax”. We adjust the costings for that sort of thing.
We also look top down at what each measure means for the overall tax gap—the overall difference between the tax that HMRC collects and the tax that it could theoretically collect.
Dame Harriett Baldwin: It is an enormous number.
Tom Josephs: The difference is the tax gap. That has been relatively stable in outturn over the past few years. Our forecast assumes that it will fall because of all the additional resources that HMRC is putting into tax compliance. It will fall to a level that is not unprecedented. In the recent past, we have seen it at the levels that we are projecting that it will get to in the forecast. We are reasonably confident that that is a central forecast, but, as I say, there is definitely a lot of risk around that. Those measures are particularly difficult ones to cost because they rely on assumptions around how people will respond to the measures that are very uncertain.
Tom Josephs: The measure on self-assessment payments being made in a more timely way in the year, rather than with a lag, as is the current approach, is the one that raises the most in this Budget from 2029-30 onwards. Most of that additional revenue is based on analysis that suggests that, because people are paying in a more timely way, they are less likely to build up tax debt than is the case when they are paying with a big lag.
Tom Josephs: Yes. As you say, there is also a bit of an effect from tax being dragged from the future into the forecast period, but most of it is around those assumptions about having lower debt and more recovery of debt.
Chair: Can I thank our witnesses very much indeed for their time? That is the members of the Budget Responsibility Committee. The chair is Professor Jonathan Haskel and the other members are Tom Josephs and Professor David Miles.
The transcript of this session will be available on the website uncorrected in the next couple of days. Thank you to our colleagues at Hansard. Thank you to our colleagues at Bow Tie for the broadcasting.
For those of you who are followers of the Treasury Select Committee, our next session is next Wednesday with the Chancellor of the Exchequer, the right hon. John Healey MP. Thanks to the witnesses for providing both the forecast and today’s session, as we prepare to question him about his Budget.