Welcome to the Treasury Committee on Wednesday 20 May 2026. Today we have one of our regular sessions with the Bank of England, particularly relating to the Monetary Policy Committee’s last decision, on 26 April, which was to hold interest rates at 3.75%. I want to welcome our witnesses. We have, of course, Andrew Bailey, the Governor of the Bank of England. He is joined by Sarah Breeden, the deputy governor, whose particular focus is on financial stability at the Bank of England; Dr Swati Dhingra, who is one of the external members of the Monetary Policy Committee; and Dr Catherine L Mann, also an external member of the Monetary Policy Committee. A very warm welcome to you all. It is pretty clear about the decision to hold, and thank you for the thoughts that you have provided the nation with on that, but obviously, the big issue in town is what is happening in Iran and the middle east. I want to get a quick sense from each of you of whether that was the main reason that you decided to hold and what your concerns are about the long-lasting impact of the situation in the middle east on inflation. Perhaps I will start with you, Governor.

I was looking back, because I had forgotten that the last time we were here was four days before it began, and none of us at that hearing had any idea what was going to happen. It really has been the dominating change in the landscape since then. You may have seen that we had an inflation number out this morning. I remembered back to the hearing we had then when I was saying that I thought it was today that we would be back around target. I think everything we have seen since, including today’s number, justifies that judgment. Unfortunately, we have had events since then that have changed the picture. We are also in a world which is very unpredictable and uncertain in that sense. The way I look at it is that there are three starting things that condition our judgment on this. The first is, how long is it going to go on for? The second is, whenever it does stop, what lasting damage has been done to the energy supply infrastructure, and how long is it going to take to put it back in place? By the way, neither of those two are things we can influence, but they have an impact on the third one, which is the one that we can: what will be what we tend to call the second-round effects of all of this in terms of the inflation picture? That depends on a number of things. It depends on those two things—on the impact—but it also depends upon, as I am sure we will come on to, the condition of the economy and the context in which we are operating. Certainly for myself, that is the framework of judgment that we are operating in. What I would say is this. In terms of the judgment we reached at the end of April, where you get to depends on where you start, as it were. As I said at the last hearing, and I may have said it at a previous one as well, it was a reasonable expectation prior to this all happening that we would probably cut once or twice this year, and the market was pricing that. The market is not pricing that now. Effectively, that was taken off the table, and it was really taken off the table with the March decision—quite reasonably. If you start there, effectively we have tightened policy, because we have removed the expectation of a cut before you start talking about what might happen next. You can see that in the way the curve has moved. You can see it in the increase in mortgage rates, for instance. That tightening also gives us, I think, some time to assess. The whole decision for me is also reflective of that point—very high uncertainty and very high unpredictability. But we start in a position where, as we will no doubt come on to, we have a somewhat softening picture for growth and a somewhat softening picture for the labour market, none of which is really inconsistent with what we were talking about last time we were here. That is the context, so I think a hold decision was appropriate at this point.

Dr Mann, you had some interesting things to say in the minutes about yours being an “active hold” position. What do you mean by that? Can you expand a bit on why you voted to hold and the impact of the Iran war?

My strategy for monetary policy decision making all along has been activist. “Activist” can mean moving in increments of 50 basis points or more, which I have done both up and down at various times over the last five years, but it also means you can hold for longer. That is also something I have done, even as there have been other members of the committee who have moved in a 25 basis point increment. What am I looking for now, and what are my concerns? On the issues regarding labour market slack and pricing power, prior to the middle east conflict developing, I had thought, “Well, I will be close to a position where labour market slack might in fact be sufficiently in evidence as to reduce demand for goods and services, so that would limit the pricing power of firms.” That was where I was last time with a hold—or two times ago. Last time, at the last, most recent meeting, where I sat was slightly different. The first point is that, on firms’ pricing strategies, in the face of a common shock, which this energy price shock has been, firms tend to pass that through to a greater degree on the upside, and not so much on the downside, so there is both a pass-through as well as an asymmetry with regard to ups and downs. I thought that it was important to consider the possible moves going forward. The second issue, though, importantly, is on what the dynamics for inflation look like in the second half of this year. Why is that particularly important? In the past—or, at least, in the last two years—I have not been as concerned about the labour market. Yes, it was softening; yes, wages were coming down and were becoming closer to what we would consider target-consistent. However, it is also the case that inflation expectations play a very important role in wage setting. There is a seasonality in wage setting—we kind of had some in April, and that was relatively soft because, of course, inflation had been coming down at that time. But when I look into 2027, and the possibility that wage bargaining in the first quarter of 2027 is going to be on the back of relatively high inflation prospects for the second half of this year, that is something that makes me worry about an embedding of both the cost structure for firms and their price setting, as well as inflation expectations and inflation outturns in the second half of the year becoming embedded in wage developments for next year. That puts me in the position of looking at those two elements going forward as being particularly of concern. However, I have to acknowledge that financial conditions have tightened dramatically, and that plays an important role in the transmission of a hold, or whatever the committee decides to do; it is against this landscape of the broad set of financial conditions, and we have to—or I have to—take that into account in making my decisions going forward.

Thank you. That is very helpful.

I can explain my decision, particularly with respect to how I was voting in February. Had we been in the position that the middle east crisis had not happened, I think I would have felt fairly secure regarding the way the nominal dynamics were going, in terms of the weakness in the real economy, which meant we were not going to see another flare-up in any near term. Also, compared to some of the peer European countries, which look very similar in terms of wage inflation or services inflation despite much lower rates there, we were not seeing any kind of derailing of the disinflation that we were seeing here. Of course, all of that now has to be completely revised, because I would say that some of my worst-case scenarios have probably unfolded, or are looking like they might unfold. Unfortunately, I do not know how to put probabilities on these events—whether we will be in scenario A according to the MPR, or in scenario C. One thing is pretty clear, which is that there is a lot of tightening already in the system. At 3.75%, even by the most generous estimates—which I do not necessarily find particularly credible—we are really restrictive at this point, and the economy looks weak enough that it does not look like we are going to see some kind of consumption-led boom at the moment. Given all of that setting, it looks like there is enough in the system out there that, even if we see somewhat of a scenario B unfolding, we should be able to control inflation. I am more worried about the fact that we are now faced with a much sharper policy trade-off: if inflation happens now and we need to tighten in a more severe scenario, we are going to have to pay a much higher cost for that, for the obvious reason that the economy is in a weaker place and monetary policy is already tighter than when the 2022 shock happened. In that sense, there is more uncertainty about which way things will unfold. Broadly speaking, even in a fairly severe scenario, we have enough tightness in the system, and the economy is in a normalised position compared with where it was in the pandemic, so we should not see the disinflation process derailed and things will normalise in due course. But I cannot rule out the possibility that these twin energy shocks, happening on the back of each other, have repercussions for where inflation goes. That is why we have to do our job more strictly than before.

Finally, Sarah Breeden.

Like Swati, I have found it helpful, when thinking about what to do after this shock, to have in mind the world that I expected to be in prior to it. At the February monetary policy report, I expected inflation to come down to 2% and, with wage growth already target consistent and with little sign of pricing power in a weak demand environment, I expected inflation to stay there. With that context, and the Bank rate already restricted, I would have expected to vote for one or two more cuts this year. In this environment, which is, as Swati and others have said, completely different, we have already seen a tightening in financial conditions. The yield curve was 60 basis points higher in April than in February, and it is 80 basis points tighter now. As I say, that was from an already restrictive policy position. In a world where we are trying to offset second-round effects, I think that, like Swati just described, we have had sufficient tightening in the conditions that matter for businesses and households to be able to offset the current risk of second-round effects. We do have to be alive to the risk of a bigger shock and it having a bigger impact on second-round effects, but we are well placed at the moment to judge how that plays out, rather than needing to move quickly to do so. If it does look like we are moving to a prolonged conflict, with more meaningful second-round effects, we will obviously have to be ready to move swiftly and, if necessary, forcefully.

Before I hand over to Ms Yang, Governor, you said that the second point was about the lasting damage to energy supply infrastructure. Where do you get intelligence about that? The last meeting was probably too soon to get it, but given that that is a very real concern, do you get intelligence on it, and does that feed into your thinking?

Yes, it does. We get it from a number of places. We talk to people in the industry, who are obviously much more expert than we are. I talk a lot to my counterparts in the Gulf countries. I see a lot of them, and saw them yesterday, actually—Saudi Arabia, UAE and Qatar. They are obviously on the ground and know exactly what has happened. The assessment is that there has been more physical damage to the gas supply infrastructure than the oil supply infrastructure. Oil would take some time because, as we always hear on the news, ships are in the wrong place and the whole supply lines have been disturbed. The Saudis have been quite effective. They cannot do everything, but they have been quite effective. They have one pipeline that goes across Saudi Arabia to the Red sea, and they are using that to its full capacity at the moment, I think. It does not replace all the lost supply, by any means. The attack on the gas field in Qatar in particular has damaged some quite key pieces of infrastructure that are apparently quite technologically complicated and will take some time to replace.

Sarah Breeden, when you come to your next meeting, you will be considering all of this.

I was just going to say that that intelligence really fed into the design of our scenarios. Scenario A is the one that we usually use, based on what the financial markets are telling us about the future, but given all the intelligence we got from the range of contacts that the Governor described, we thought it was prudent to put in a different scenario.

I will just add two points. First, and this is reflected in scenarios, I think the futures curves are pretty benign when you sit them alongside what I was just saying about the damage to the infrastructure in the gas market in particular. We had one scenario that follows it, but then we have scenarios that aim off that. That is important. On the other hand, importantly, when judging it compared to four years ago, this is much more of an oil shock than a gas shock globally. Although there is damage to the gas field, it is a smaller part of the overall global picture. That is relevant when you look at the relative movements in the Ofgem cap. They are much smaller than four years ago, and the reason for that is that far more of it is in oil than gas and gas is a bigger part of the UK’s story.

I have a question about your responses to potential new data and new measures. In our last session, we talked about how the Government’s intervention on energy bills, in terms of fiscal policy, had brought down headline CPI. Given that all three of the scenarios that you have projected have quite a significant output gap, do you think there is more policy space for such measures that bring down CPI to have a positive impact in terms of your mandate on curbing inflation? Or do you think that they would have bad effects in terms of stimulating too much demand?

Obviously it is for the Government to take those decisions, not us. We have our tools to do monetary policy and tackle inflation, and we do not consider things that could be done outside our framework. That is important. I do not want to decry the Government for looking at them or not, but there were some very big measures taken in the past. Admittedly, as I have just said, it was a different situation in terms of the scale of the issue, but with very big fiscal measures, the cost comes back to you later. The Government are right to think about these measures, but they are also right to think about them in that context because eventually the cost does come back.

There may be fiscal and other measures—we will talk about food later in this session—but what do you see as the purely macroeconomic impact of that reduction on CPI in terms of how it affects the output gap and inflation?

I think Catherine referred to this earlier: inflation expectations are an important part of this story. If you go down the rank order, you have household expectations, business expectations and financial market expectations. They have pretty much moved in that order. The biggest move has been household, followed by business and the lowest has been financial markets. They are influenced by these things. As Swati was saying earlier, this is the second energy shock in four years and that will weigh on households’ thinking, no question about that. I think something was announced on the delay in fuel duty at Prime Minister’s questions a couple of hours ago. Those measures are relevant and we will factor them in now that they have been announced, and they may be relevant to that inflation expectations formation process.

From a macroeconomic viewpoint, we have quite a bit of information in the household cost index about the different income deciles in the economy and how the prices of the different components of their baskets change. When the last shock hit, we were able to evaluate the extent to which the different programmes that were put into place at that time would affect different income deciles. We could also use that to evaluate the extent of the change in the overall CPI, which was an important feature of the energy price guarantee—instead of 14% inflation, it was 11%. We could track how those different programmes would percolate through the income deciles. We can do the same type of exercise this time with a little bit more granularity once it becomes clearer what the programmes might look like. We would be able to take that into account when making our judgments.

Has some of that modelling been published?

Back in the day? Yes, it was in the MPR.

If I may, let me just illustrate this point about energy. The current Ofgem cap is £1,641. The latest market estimate for the next quarter is £1,850. Had there not been Government intervention I think the estimate of what it would have been at the beginning of 2023 is £4,279, so that gives you a sense of the comparison.

If I may add a couple of quick points to broaden the question out a bit, we already saw what happened with the interventions that the Government did with the monthly inflation number that has come in today, which is that many of these administered prices have taken some of the pressure off the headline inflation. Also, if you look at the past energy shock, one of the key reasons we saw this sharp drop after the really big energy price spike that happened very quickly is because a lot more supply came on the market, which in this case was largely driven by the US coming into the LNG market. More broadly, supply side policies are particularly useful in being able to alleviate inflation pressures without putting excess pressure on output.

It is an important distinction—you are right to point to it. Our interpretation is that short-run inflation expectations are heavily influenced by what inflation is today. If you ask people what they think inflation will be in a year’s time, it is quite heavily influenced by what they believe inflation to be today. As you go further out, that is not the case; hence the distinction. We look at further out very carefully, because you would expect it to be anchored more to the inflation target—not precisely necessarily, but anchored more to it. You do see that, but we watch this very carefully.

I don’t think they are de-anchored at the moment, but it is something—I think Catherine described it very well earlier—that we watch very carefully. There is a tension here. As I said earlier, we are seeing softening conditions in the real economy, but we have, because of this energy shock, seen an upturn, particularly, as you said, in short-run inflation expectations. That is a point of tension. On the point about second-round effects that I made, which of those two things is going to have more weight?

Wages, of course, are the other part of this. We have seen, as with the numbers that came out on Tuesday—yesterday—a continued gradual reduction in increases, particularly in private sector wage settlements. I would also say that we have another way into that, which is, again, survey evidence on wage increases. Our own agents always do one at the beginning of the year, and they did one at the beginning of the year, but that was before any of this started. We get the decision-maker panel survey, which is more benign so far, but we will have to watch that one very carefully to see what it shows. Catherine mentioned that there is a seasonality to wage increases. A lot of the private sector wage settlements have been done this year. We will get information, but it will be a while before we get a lot of information on that front.

On your previous question about pricing and its relationship to expectations, an important ingredient in the relationship between wages and prices, particularly in expectations, is what our research has identified as threshold effects. Consumers pay a lot more attention to food and energy—we know that. When inflation as an aggregate gets above, say, 3.5%, there is quite a bit more attention in general to the inflation process. It starts to look like it gets embedded in price setting by firms, because firms increasingly set their prices based on the state of the economy, and inflation is one measure of the state of the economy. Of course wages tend to get settled as well on the state of the economy, including the expectations. Again, I am concerned, on the private sector side, about wage setting in the first quarter of next year, which is the season for most of the private sector wage setting. Some of the research that shows up in the boxes in the MPR noted the extent to which certain characteristics of different groups key off either inflation itself or bargain against expectations, as about 80% do. It is very important for me to look at the second half of the year. You say, “How do you look at the second half of the year? It hasn’t happened yet.” Again, our decision-maker panel has one-year-ahead price expectations, one-year-ahead wage prospects. We can track the extent to which firms are pretty good at estimating their future price and wage outturns. They have been pretty good at that. That to me is forward-looking information, of which I can make some assessments. With regard to wage settlements and different groups in the labour market, we have to look very carefully below aggregates. Most of the softening in the labour market that we have seen so far is in a couple of categories, particularly hospitality/retail and information technology. They account for most of the numbers of unemployed. To me, that means that the labour market softening, while true, is not broad-based, in my view. That is not broad-based. When we think about different groups, hospitality and retail have less wage flexibility on the downside, because of the approaches to minimum wage. That means, to the extent that there is adjustment, it is going to be in some other margin than wages. It could be profit margins, it could be numbers of people. Of course, there could be capital labour substitution and productivity enhancements as well. But it does mean that, when we think about the relationship between the labour market and the wages, we really have to get into a much more disaggregated analysis.

Right. That is consistent with my assessment of the labour market, yes.

I look at it regularly. You are right that, if you were basing your decisions on money and the money supply, you would not see inflationary pressures. I think it is always sensible, though, to look at it in the round with other things. It is one of the things that influences. It is also important to look at the stories that lie behind that. At the moment, no, it is not a flagging pressure in that sense.

Turning to some of the household finance issues that are likely to be a real concern to our constituents, we are already hearing about some of those worries. To Mr Bailey first, how do you think food and other prices, which are obviously very important to household finances, are likely to be impacted, compared, for instance, with the 2022 fuel prices hike?

I think it will happen. Today’s number had surprisingly benign food price inflation. It came down, which was a little contrary to what we were expecting. The way I look at it, there is a sequence of effects in terms of the feed-through of an energy price shock. The first, which we have obviously already seen, is the one that comes through particularly at the pump, as it were. Pump prices adjust very quickly, for instance. The second one in the UK is where the UK is a bit different. By the way, today’s number looks quite benign, but we have to take this into account because it pushes in the opposite direction. The way in which the Ofgem mechanism works, because it is a quarterly reset, means there is a delayed reaction, if you like. It is going to go up. The inflation number we had this morning reflected the fact the Ofgem cap came down in April. Now we know it is going to go up in July. In fact, we pretty much now know how much it is going to go up because we are through the observation period. So that is second. The third one—this is where we get to food—is what we tend to call the indirect effect. In other words, that is where energy is a component and a contributor to the costs of other products. Honestly, one of the things I learned four years ago was just how large an energy share there is in things like processed food production and distribution. For some of it, we are getting up to numbers like 30% of the cost is energy. Obviously, therefore, energy prices do come through. How they come through and when they come through depends upon the business energy supply costs that producers of food, for instance, are experiencing. We go around the country a lot. We talk to a lot of companies. A lot of companies obviously take out energy contracts. They therefore hedge to a degree. It depends on what the nature of the contract is and it also depends when they reset the contract. I always ask companies and I can tell they are either going to slightly smile at me, because they have managed to reset their contract just before the increase happened, or they are going to look extremely pained, because they are about to reset their price. You do not know until you talk to businesses which situation they are in, but it will come through. It is really too early to start. I see a lot of speculation about what food inflation is going to be later this year. I think it is very hard at this stage to really put any good figures on that.

Dr Dhingra, there was some suggestion from the Bank’s agents as mentioned in the Monetary Policy Report that we could be looking at food price inflation later in the year of 6% to 7%. Would that be a figure you would be expecting might come through?

The important thing to keep in mind here is that it is much less—it is almost an order of magnitude smaller than the food inflation we saw in the past. That is really important. What precisely the number would look like is really hard to tell. Even if you were to trace what is happening to, say, global food prices from where we are going to import things, those are moving in all sorts of directions. I would not want to hazard a guess on that one. What the Governor outlined about the sequencing of how these effects work is something that we should expect to happen in the future. We are very alert to the point that there will be food price inflation and we are going to have to think about that.

Mr Bailey, has there been any contact from the Treasury to discuss with you, either for information or on any other basis, any steps that Treasury might be looking to take to intervene around food price inflation at all?

No. I have seen the stories on this, particularly earlier today. We have had no contact with the Treasury on that. I am afraid that is question you would have to ask the Treasury.

Would any of the panel wish to comment on whether there are any interventions that the Treasury might undertake, particularly around food pricing, which might help to reduce the future level of inflation going forward?

I am looking at Dr Mann, because she is not an employee of the Bank.

I am not going to comment on any of the interventions, but I will comment on this: if we compare going into your grocery store today versus four years ago, you go into the grocery store today and there is the little label “Matching Aldi” or “Matching Lidl”. Four years ago, you did not see that type of competition undertaken in the grocery stores. You did see empty shelves, but you did not see that competition, where, if you want to do comparison shopping, it is right there and you can do it.

The big difference to four years ago, going back to Swati’s point, is of course that Ukraine is a major global food supplier. That is a big difference. We are talking about an energy shock here. Four years ago, we had the Ukraine war as well.

That is helpful. I have a question about our exposure to the jet fuel market. All the indicators show that the UK is a particularly large importer of jet fuel. Do you think this is likely to have a significant effect on inflation in the UK as compared to other territories?

It obviously affects transport prices. It affects airline prices. That is one of the more volatile components of the services index. I do not in any sense pretend to be an expert on this market, but again, going back to the Chair’s question earlier, we do talk to a lot of people about the energy markets. I think you are right that in the sort of rank order of things where we are experiencing scarcity, jet fuel was probably at the top. What I have picked up more recently is that, first, there has been a switching to alternative locations of supply of jet fuel, because jet fuel comes from other parts of the world. Secondly, there has been some switch in refining activity to produce more jet fuel. As I say, I am not the expert, so I do not want to give false comfort here, but my sense is that there are measures being taken globally that are easing that position.

I have one final question, Dame Meg, if I may. On the length of time for which we are likely to be experiencing the shock that comes from the closure of the Strait of Hormuz, if the closure ends reasonably soon, how long do you, as the MPC, think the effects from the closure are likely to last?

That goes back to my point about there being three parts to the story. That is the second part. What we are told is this. In oil, it would be a matter of a number of months, and a lot of that is to do with getting the supply chain back in order, getting the boats in the right place and so on. With gas, I think it is quite a bit longer than that. It could be a year or two, or more, because of the need to rebuild the infrastructure. Swati, you were saying earlier that more gas—LNG—has come on to the market since the shock four years ago. By the way, I should say that before any of this started there was, if anything, a surplus of supply in the oil market. It is a question of how long it takes to get back to that point.

I just want to press more on my colleague’s line of questioning about the Treasury’s measures. Would it be normal, when the Treasury is looking at interventions that change prices—lowering the cost of living—to have that conversation with you to understand your reaction function to different measures, or is that not part of the institutional set-up?

Well, it is normal in the sense that if it is part of a Budget, there is a process for having those conversations, and they would keep us informed. I did not know that the Prime Minister was going to say what he said in Prime Minister’s questions today, but that is all right—I do not need to know. We now know, and we will take it into consideration.

Understood.

Does anyone else want to comment?

In the long term, you could think about what that co-ordination should look like in a world of big supply shocks. I think that is something for Parliament to think about.

We have taken the hint.

Can I ask just one question? It may be a daft question. When I read people like Professor Nick Butler, formerly of BP, and Professor Birol of the International Energy Agency commentating in the FT, they suggest that reserves are actually becoming very tight and that we are sailing into one of the biggest energy crises than we have ever seen—even bigger than what Chancellor Healey and your predecessor dealt with in the mid-’70s—and yet the general discussion here is about 6% inflation. Is there is a risk that we are missing the risk of a very significant shock that goes way beyond what you are describing, Governor?

Let me be honest. The longer this goes on, the more difficult it gets—there is no question about that. I was at the G7 on Monday and Tuesday with the Chancellor. The head of the International Energy Agency was there, and he talked us through his latest thinking. He has made the point a number of times that there are obviously drawdowns on the world’s reserve supply going on, so the system is, in a sense, in deficit at the moment. That deficit is being made up by drawing down the world’s reserves. The point that he has made is that it would be easy to think, if you are not a technician in the market, that you can fully draw down the world’s oil reserves, but I don’t think you can do that. I don’t think they are all available, because there are various technical reasons why you cannot do it. He is a real expert on this, and he sensibly introduces a note of caution on this front when he talks about it.

Governor, you are a noted economic historian—

Not noted. Other economic historians would dispute that, I am sure. Anyway, I will take the compliment.

Well, I have noted it. Could you point to any historic examples where the state setting the price of food has been an effective economic intervention?

Well, I am sure there are many in some countries. I think the question you to have think through in this sort of thing is, “Are you doing it for some well-grounded, very temporary reason?” If you start doing it as a matter of course, you are artificially moving prices relative to costs, and that is not a sustainable thing in the long run. There may be reasons for doing it in the short run, but it needs to be thought through.

May I put the same point to Dr Dhingra? You often have talked about the global chain in your evidence to us. How practical is it in the modern world for one Government effectively to encourage price caps or whatever mechanism they use, when the food supply chain is not in the control of one sovereign nation?

I often get asked this question, and I do not have the solutions. I only have a sense of what would be economically logical. Thinking about where to find sources of supply when you have a crisis is key. That is one thing that we can think about doing. A lot of our food is imported, but many other countries do not face that problem. On the specific point about price controls, lots of countries have tried it out, though not necessarily successfully. The biggest problem with these sorts of controls is that you are dampening the price signal—precisely what you want people to react to. As a market economy, you can do that for a little bit, but how low can you let it go on? I grew up with price controls on food; to some degree that was very successful in being able to cure famine and poverty and various things like that, but at the same time it ended up creating a highly distorted agricultural sector in India. In that sense, do it with a lot of caution and thinking behind what it is trying to target.

There is, of course, another area where it is being used and where I think it is more justified, which is energy. You have other sources of energy where the cost has not changed, but energy prices have. Governments see that as somewhere where they take measures, but that is a different situation because the costs have not changed.

I thought Dr Mann made an interesting point just now about the competitive nature of the supermarket sector in the UK, but I will change the subject completely to another area where the Government have intervened and changed the price of something, which is the price of employing a young person. That has gone up by about £4,000 a year. We have seen a huge increase in young people who are out of work to levels that we have not seen since 2009. Dr Mann, could you share with us how the Monetary Policy Committee is looking at the labour market statistics at the moment? It has been a problem area. There has been difficulty looking at those statistics. You have not been getting very accurate information from the normal sources. Could you share with us how you are thinking about impacts like that? There has clearly been a loosening of the labour market. There are fewer people on payrolls than there were a couple of years ago. Are you factoring that into your decision making, and if so, how?

As you can see from the boxes in the MPR, there is a quite a bit of research being done by the staff using the granular data across different sectors with the wages associated with those sectors—the Beveridge curve and the clustered wage analysis. I will not ask you whether or not you have read all the boxes in detail—

All your forecasts do show unemployment going up, including the central forecast, don’t they?

I think I have said before that, when I evaluate the state of the economy, I tend to focus on the granular information that I can get from firms with regard to their pricing strategy, their labour employment strategy, their production prospects and what they believe about demand. I use that information to judge the state of the economy far more than I use the aggregate data, and it is the aggregate data that go into the forecast. Although that is the forecast that has come out, I tend to look at other information that I think is a better indicator of prospects for the economy. For example, I think that a very important distinction for understanding the behaviour of the economy is the difference between GDP and market sector output, because it has a direct impact on employment prospects. There has been a widening gap between those two different measures of economic activity. It also shows up in terms of employment in the public sector versus employment in the private sector. Last year, when there was quite a bit of a deterioration in employment in the private sector, employment in the public sector basically countered that almost entirely. Things have been different towards the end of last year and going into the quarters here, but that type of analysis is, to me, indicative of what the prospects for the economy are. For example, private sector wages are at 3.2% or 3.5% from DMP, through the agents and through the various measures of private sector wages, and public sector wages are at 5.2%. The new national minimum wage for those aged 21 and above is at 4.1%. I find bringing these types of assessments to bear to my judgment is the way I evaluate the state of the economy.

Sarah Breeden, you were nodding along when I was mentioning some of the problems with the labour market statistics. Are you comfortable now that you are getting the right kind of information when you are making these decisions about the state of the labour market, and how do you see the softness of the labour market affecting the pass-through of inflation into second-order effects?

I very much agree with Dr Mann’s comments about needing to get below the aggregate data and understand what is happening in different industries and in different bits of the age distribution and skills distribution. I personally find it incredibly helpful to go out and talk to businesses about what they are actually doing with their hiring intentions. What has really come through when talking to them is that, as people leave, they are not replacing them. That means their productivity levels are going up; that is a good news story, but it is a problem for entry-level jobs or for people coming into the labour market, which I think is the counterpoint to your point about youth unemployment. We take that into account when we are thinking about what is happening in the economy as a whole. It is not obvious that monetary policy is the only response or indeed the right response to that, but it is part of our decisions.

Indeed. Governor, we heard from the chief economist last time about how he empathised with young people because it was something that he was seeing in his own family. I just wondered what assessment you have made of the impact that artificial intelligence might be having on those entry-level positions, but also how you then take your assessment of the situation for young people in entry-level positions and how that feeds through in terms of what you need to do through a monetary policy response because of the impact on inflation, if any.

Artificial intelligence is another essentially supply-side change in the economy. I think you can divide it into four parts, or at least I tend to lean on this. First, it can destroy jobs, or in other words, remove jobs. An example of that is that if you create a chatbot to speak to customers, then you will reduce customer service staff. It can create new jobs. An example of that is data scientists. If you vastly increase the capacity to analyse data, you will create the demand to do that by humans—I think a lot of this is humans working with technology. The third one is where it increases productivity, and that productivity is essentially shared between machines and humans, as it were. For instance, if it can process medical analysis a lot more quickly and process a lot more of it, you can increase the demand for doctors because then you have to do something—obviously, you want to act on it. There is a much weaker form of that, which is job-negative, where actually it is doing it all and the only employment effect is the person who is keeping the data centre going. That is the fourth one. There are two rather more positive ones and two frankly negative ones. Really, it then comes down to two things. The first is, what is the balance of those forces? You mentioned economic history. This is what in economic history we tend to call a general-purpose technology innovation. We have had steam engines, electricity, internet, ICT, and I do think the next one is going to be AI—probably AI and robotics, in some combination. These are the things that I think do move productivity growth, and this is really important for the economy. Each of these has had a different effect in terms of employment. You cannot generalise across these things; they are different, because they are structurally different. It is down to A, the balance of those four things, and then B, the sequence of those four things, because they can come through in different sequences. General purpose technology innovation historically has taken a long time to come through. It may come through quicker this time, because it is obviously moving very quickly, as we are seeing in other parts of our world at the moment. The classic one is that it took eight decades between James Watt inventing a steam engine and it coming through in the UK productivity numbers. They are slow, because they usually require other innovations to make them actually work on site, as it were.

Is the Monetary Policy Committee now getting good enough data on the labour market? That was a problem that you pointed to for a long time.

I think the ONS is still very much working on it. I think it has achieved quite a lot already. As my colleagues were saying, we are great consumers of data, so we search for all the data we can get. I would very much echo what Sarah was saying, because I am in the same position, and I think we are all in the same position: we get very good information by going out on the road, as it were, with the agents. As Sarah was saying, we are in a much lower turnover labour market at the moment. That is one of the very clear messages we get: people are staying in jobs—they are not leaving jobs. This goes back to your point about youth unemployment. As Sarah was saying, youth unemployment is particularly affected, because it is people who do not have jobs to start with who are coming into the job market, and they are going to find it more difficult.

Might I add one other thing that I have found interesting watching? Net desired hours—the amount that people want to work more and are unable to—has risen recently.

They want to do more hours?

They want to do more hours. That is an indication of slack in the labour market and therefore the reduced likelihood of there being pressure on wages. That has certainly been something I have followed.

Dr Dhingra, on the labour market specifically, did you want to add anything to those points?

Yes, a couple of points. One is that I think the situation has improved, both from the fact that the LFS is now available in some form, and also from the way the labour market has turned out to be—a lot of different indicators from different data sources are pointing in the same direction. That gives us a lot more confidence in thinking that what we are seeing is not an artefact of the data for that month or that quarter; it is much broader than that. That is also taking into account what the deputy governor just said. I will add one more point, which is that I think the last point you raised about youth unemployment and the effect of AI is really important. From what the staff have shown us, they did some really good analysis showing that youth unemployment seems to be rising globally. That could be a cyclical thing, which of course is very important for the conduct of monetary policy—to what extent is it a purely cyclical phenomenon that the people who come in last will be the ones who probably get fired the first when the economy is in a weaker position? The IFS has done some very good recent analysis on this, which is suggesting that there is a cyclical component, but that does not seem to be the only story. There seem to be some more structural factors, such as health issues more broadly. It does not seem to jump out that it is to do with the minimum wage or with NICs. You can segregate the workers you are looking at by the age profile that gets really hit by the minimum wage or by NICs—for example, the 21-year-old threshold. You can also look across different regions. There are some regions—Southend being the famous one—where a lot more people are on the minimum wage than in other regions of the country. You do not necessarily see a strong correlation—in fact, almost none—between unemployment changes that have happened recently and the minimum wage or, to that degree, even NICs. But it is much more important to have that broader conversation about the extent to which AI or some of these other global phenomena are driving what is going on.

That is consistent with the view that AI is taking out more entry-level jobs. If you take lawyers, for instance, it is taking out the entry-level jobs because they were far more information processing jobs, if you like.

I am going to push back a little bit—do you see the differences of views in the committee? You can have a little taste of what happens around the table. If you look at our agent reports, disaggregated by sector, on the effect of the different components of the minimum wage and so forth, you will see that they are saying there is an impact—

In hospitality.

Yes, in hospitality in particular. The reports say that more research needs to be done, but to say that it is a global phenomenon and the UK is just the same as everybody else is something that needs a lot more consideration. On AI, it is interesting that across the US, the UK, Germany and Australia, and a variety of research on each of those elements, 90% of the sources, when you ask firms, say there is no impact. It may be that the manifestation is, as the deputy governor said: “We’re not going to hire somebody if someone retires.” That is not exactly no impact, but it is not like AI is taking a job; it is just replacing it. The last point I would make about AI is that we are using it internally on the agent reports. They extend across many years and various regions, and we have notes of the different questions we have asked and the responses and so forth. All that is now in a database and it is queried on questions of particular interest to the committee.

I am going to move on to one last question of particular interest. On 1 July, the Government are bringing in a 50% steel tariff. Has the Bank done any analysis of what the impact might be on the UK economy and on inflation of the 50% steel tariff from 1 July and the reduction in quotas? If not, perhaps you could follow up with a letter.

I think we will follow up with a letter. We have obviously looked at the whole tariff question generally for more than a year now. I have not looked at steel tariffs—

This is the UK bringing in a 50% steel tariff on imports of certain grades of steel.

Tariffs are suddenly jumping around rather a lot at the moment.

I certainly recognise that there has been a greater focus from the businesses we talk to on driving productivity improvements. I use the example that as people leave, businesses are not replacing them. Automation is happening, and there is the use of AI, as we have talked about. I can definitely see that that is a theme through our conversations with businesses. I do not know whether that equates to 1.6%, and how sustainable that is, but, as the Governor was saying, AI has the potential to be the next big thing in terms of enabling businesses to do more with similar inputs.

I want to turn to the growth outlook. Before I ask my question, I think one of the things that people will have noticed is the great amount of volatility in the predictions and the numbers. For March, I think the GDP figure from the ONS was an expansion of 0.3% but, as has been reported, Reuters polled economists and a 0.2% contraction was expected. We see often-restated figures all the time. What do you think about how we account for that and minimise it? The latest monetary policy report says that underlying growth remains pretty flat. Why do you think growth look so meagre?

As I have already mentioned, there are multiple ways to calculate growth. We can look at output growth, at income growth, at growth at the top line of the economy or at decomposition, to look at the market sector, where we think pricing strategies might be particularly important for generating inflation, for example. I did a speech not too long ago looking at consumption growth, which has been remarkably flat, basically since covid; notwithstanding increases in real income, consumption has been basically flat. Consumption is of course an important driver of overall decisions by firms to add workers or produce more; there is no point doing that if the consumption looks relatively low. I also want to continue with my theme of not only needing to look at growth in a disaggregated way to evaluate the state of the economy and appropriate monetary policy decision making, but having a number of statistical methods where we put in high-frequency data, as well as the normal statistical things that come from official statistical agencies, to evaluate the prospects for the economy. One of the things that has been most notable about this so-called quantile MIDAS statistical methodology is that rarely over the last six to nine months has there been a centre of the distribution of the data. If all the data was telling you the same thing, it would look like a nice hump, and we would have some tails, but the middle would be like, “Okay, I am pretty sure this is going to be my growth rate.” That has not been the case for, basically, six to nine months. Instead, our statistical methodology of putting in lots of data gets a double hump, meaning there is not a lot of mass of the data telling us the mean. It is saying, “Well, it could be higher, it could be lower”, which requires us to look into those two humps to get an indication of what the data are—these disaggregated data, some high frequency, and some the standard statistical stuff—to evaluate our prospects for growth.

What do we do about it then? The country sees the growth figures, for which we as politicians generate great big headlines, one way or the other, but they seem pretty meaningless, because they are often restated. If they had been true to what they actually were on the restatement, they would not have generated anything like the headlines. The same worked when I was in government, so I am not making a particularly political point.

I will offer two thoughts: one is general and the other specific. I would therefore be a bit cautious about commenting on any particular single growth number. As you say, they do get revised. To go back to the first part of your question, which was interesting, it does look like there has been some change in the seasonality pattern of growth. James Benford at the ONS has published a really good blog in the past couple of weeks, taking this on. It looks like there has been some change between the fourth quarter and the first quarter. That may be down to the fact that there has been a shift to one fiscal event a year. The ONS think they have done the seasonal adjustment, or quite a bit of it, but you obviously need a run of data to do that. But they are very much on to it, which is a good thing. As he said in the blog, if you take the fourth quarter and the first quarter, and put them together to get a growth figure, actually it washes out.

That just smooths it, so this notion that somehow we are 0.1% higher than other countries in the G7 is—

It looks like growth is stronger at the beginning of the year than it is throughout the year—

That has been a pattern for a few years—

And he said, “Look, there may be a reason for that, and we are on it,” as it were, in terms of diagnosing it.

Dr Dhingra, is there anything you would like to say on this subject?

You asked, “What should we do about it?”, so I can plug our report, which we worked on for two years—the Economy 2030 inquiry report. There are 10 recommendations in there.

When we look at this situation—that we have the lowest fiscal headroom among the G7, at a time when gilts are at their highest since 1998—it feels like we are in quite a vulnerable position if you look at the two key drivers. What can you say about the best-case scenario for underlying GDP growth up to, say, 2030, given the context that we find ourselves in?

I will say two things. As you know, I am obviously very cautious about what I say about fiscal policy, so I will only say this: there is an issue for many countries in terms of sustainable fiscal policy, and that is in good part because we have some pretty big headwinds—as you know from your time in government—running the other way. The ageing of the population is a big one, and the situation with defence is now another big one. These are running in the opposite direction. Many Governments have used fiscal policy more actively, and you can say that that is because we have had much bigger shocks going on—covid, Ukraine and now another one—but that raises questions about long-run sustainability, which the OBR has said a lot on. That is important. The gilt market actually reacted very positively to the Budget at the end of last year, because it saw it as putting fiscal policy on a sustainable track. If it was followed through to 2030-31, which was the outer point, it looked like a sustainable track. Some of the reaction we have seen subsequently, post the outbreak of the conflict, has obviously been the gilt market saying, “Well, that does not look so sustainable now.” That question has come up. On the second part of your question, I would go back to the earlier question on AI. I do not want to constantly talk about economic history, but a lot of the longer-run growth is caused by the supply side; it is the supply side that does it. Technology is an important issue, and it is quite possible—I think this is a strong argument—that globally we have been in a period between two of these longer-run technology innovations. We had the internet and ICT; the contribution to growth ran out, so we have had lower potential growth now for 15 years or so. It looks like AI will probably be the next one. But they do take time to come through.

On that point—this is my last question—we are often told that we are in a much weaker position in terms of our energy costs, but in terms of AI, as a strength and momentum for the economy, how would you characterise where we stand relative to other economies in the G7?

At the moment, the US stands out relative to everybody else. The US has a head start on this, certainly in terms of the amount of investment in the US economy going into the tech industry. Obviously, many of the very big global tech companies are US-based. It is not the UK standing out from the rest; if anything, it is the US standing out from the rest. The question, of course, is how quickly we catch up.

It is important to think about adoption separately from building the AI technology—about making sure we have the necessary foundations in place to enable UK businesses to adopt AI, regardless of who the model provider is. We have been doing a lot in financial services, as you know, to try to enable the responsible adoption of that. Ensuring that that happens economy-wide is the best foundation for the growth that you are talking about.

Can you be optimistic about our relative position, notwithstanding the US’s leadership?

I can tell you that the feedback that we get from financial institutions is that our approach to supporting them in adopting AI is enabling them to do it well.

It is fair to say, though, that if we look at the last round of IT hardware and software, sluggish productivity growth in the UK has a long history. It is true that there is a broad-based productivity slowdown around the world, but the UK does stand out. That has to do with adoption not being broad-based. It would be unfortunate if this next round of innovative capability is also somehow not fully adopted by business.

You mean that outside of financial services is where the sluggishness exists.

Yes.

Catherine is right. An interesting area is robotics. On the face of it, the UK does not have a particularly good story to tell on that front.

I will start with my second-favourite topic, the gilt market. Dr Mann, you gave a speech a few days ago, talking about the increase over the past few years in hedge fund trading in the UK gilt markets. What impact does that have on volatility and on monetary policy, from your perspective?

The increase in hedge fund trading in sovereign obligations is not just for the UK; this is a broad-based evolution towards the importance of that group of financial institutions for sovereign debt markets. Now, the UK is particularly, perhaps, dramatic in terms of the change from the preferred habitat long-term holders of pension funds towards the hedge funds, because of the LDI situation. It has been particularly dramatic for the UK. In terms of volatility, it is important to distinguish between price volatility and quantity volatility. The research that was identified in the speech last week focuses on the extent to which hedge funds tend to be more elastic—in other words, in their responsiveness to buying new issues or QT options. Their elasticity is higher than a preferred habitat type of investor; in other words, you do not need as much of a change in the rate of return on the instrument to make it attractive for them to hold it. For the new issuance, it is beneficial to have hedge funds be part of the financial institutions at your auctions. That has been found to be true in Canada, in research at the Bank of Canada as well. That is the good news type of story. The more challenging news is that there are many sources of volatility that change gilt valuations. It is not just issuance; there is a lot of stock out there and there are a lot of factors that are going to affect the gilt price. Under those scenarios, where you have more volatility coming over the transom from other places, you end up with more elastic responses by hedge funds, which will alter the quantities that they are willing to have. That will create additional volatility in gilt prices. Again, this sort of result is consistent with what the Bank of Canada has found as well.

Does that have any impact on monetary policy?

In my view, it does. In my view, volatility of the yield curve is something that I have to incorporate into my assessment of financial conditions, against which I choose a decision about monetary policy. Volatility, to me, is a very important component of the immediate decision. It is also relevant for thinking about r*. You do have to think about it a little bit—not the big r*, but the little r*. Volatility is an essential part of the monetary policy transmission mechanism, for mortgage rates that are going up and down with the RAS curve. Yes, that is very important for the decision making of households. Do I mortgage now? Do I remortgage? Do I have to, or can I wait a little longer? Volatility is a key ingredient of the overall state of financial conditions.

Do you see volatility as introducing more uncertainty, or is it directional, if you see what I mean, in terms of rates?

I think there is some asymmetry, in that hedge funds are more likely to buy in on the way up and sell quicker on the way down. But there are asymmetries everywhere. One of the things we have learned in the past four years is that many responses of economic actors are not the same, depending on what the state is. Those are factors that we have learned a lot about and that, at least in my case, I incorporate into my decision making.

I want to build on that slightly and take it into our financial stability work, if you don’t mind. Hedge funds are price sensitive, leveraged and international, which means that their behaviour in markets will be different as a result. There is potentially more volatility, and higher risk premia on the back of that. In a financial stability context, we have been talking about trying to improve the resilience of the gilt repo market as the foundation for the resilience of the gilt market. We issued a discussion paper about that last year, and we are working with the Treasury, the Debt Management Office and the FCA as we do that. That would be a natural counterpart to these risk premia: how can we introduce more stability and resilience in the market?

That takes me to my next question, on everyone’s favourite topic: quantitative tightening. The MPC is due to make a decision later this year about the pace and scale of that. In explaining your decision from the previous year, each of you had slightly different ideas about how quickly the pace of QT should go. I would love to hear from each panel member very quickly whether the recent changes in inflation and gilt market movements have affected your views on that.

That was probably one of the hardest decisions I have had to make. In August, we were on a loosening cycle—or, rather, an un-tightening cycle—and at the same time we were doing QT. You might say that that is not a consistent strategy, but of course we have a legacy asset on our balance sheet, which came with the expectation that that was for a particular purpose and that we had to take it down. That is what motivated me to vote in that direction, keeping in mind that we use the interest rate as an active tool because we understand the impacts of that much more carefully than we do of what quantitative tightening does. I do not fully agree with everything that we write in the MPR. You can disagree with some of the precise numbers about how much tightening has been caused by the quantitative tightening that we did. Broadly, we could say that we were in a place where interest rates could have offset a substantial portion of the tightening that was coming from QT. That is a little bit of what my thinking was at the time. It was important to get the asset purchase facility down, but at the same time we needed to keep in mind the tightening effects of the policy.

Looking forward to this year, with the changes in inflation expectations that you have now—

I am still weighing what the possibilities are at the moment.

I think Swati has set out the framework very well. We have not started this year’s thinking yet, so we will get to that.

I am sure you are all thinking all the time.

Well, yes, but specifically on the decision. Obviously, there are, in a sense, two decisions: a decision on the total and a decision on the mix. Last year, we skewed the mix away from longs—long-dated gilts—and I think that was the right thing to do. We do not have any more long-dated gilts to sell this year, by the way; they are all done. We will come back to that question, because it is important. As colleagues were saying, we are not seeing disorderly conditions in the gilt market at the moment. That is an important point to make. The other thing that I would say, which goes to the other question—the cash-flow question—is that we have done a lot of work to build and enhance the reports that we publish. What we have tried to do is capture something very important in there. The UK has a history—a good history, in this respect—of issuing longer-dated debt. I think it is right that the DMO has shortened it, but even so, our debt has a longer maturity than other G7 countries. That is a good thing. Successive Governments have got a lot of benefit from this. During the QE period, you were locking in lower yields—particularly when we had a flat yield curve—for longer. The Government benefits from that. QT effectively takes part of that benefit back, because we are reversing QE, and in doing so we are taking interest rate risk off our balance sheet. As we have said before, if you don’t do that, you pay the carry cost of having the gilts on your balance sheet for longer. We have sought to incorporate the effect of that benefit into the calculation. Interestingly, it basically leaves the cash flow pretty neutral at the estimated end of the life of the operation.

For me, there are two things that I have in mind when I approach our QT decisions, and as the Governor says, we have not got there yet. The first and most important thing is that, for monetary policy purposes, I want to take the portfolio down to zero. That is because I want to make sure that we have the headroom to be able to use it again should it ever be necessary, rather than it forever ratcheting upwards. As we do so, we have to have an eye to the impact on the market functioning. As the Governor said, the gilt market has functioned through this period but, as we just talked about with hedge funds and long-dated investors, the nature of the market is changing. The changes that we made last year in skewing our sales as we were taking our portfolio down to zero were our contribution to trying to best manage that.

In terms of my assessment of the appropriate strategy for QT, the most important thing to consider is that my instrument is bank rate, and QT affects other points on the curve. I evaluate how those two things, which are happening at the same time, affect the financial conditions to which businesses and households are responding. The basis of the research that is being done to evaluate those relationships is the key element. Again, we of course have to acknowledge that those two things, which is what we do, are in the context of a global financial market, and that the spillovers from the conduct of monetary policy, fiscal policy and overall macroeconomic developments, not to mention what is happening in terms of middle east warfare and so forth, are extremely large.

Our predecessor Committee in the last Parliament wrote a report on quantitative tightening, and we described it as a “leap in the dark”. Obviously, we have now been in this leap for a few years, so we have a little bit more information about what the real-life impacts of it are. ING bank published a report this week saying that they thought that the quantitative tightening that the Bank is undertaking has increased gilt yields by eight tenths of a per cent. That sounds substantial given that the evidence that you gave us at the beginning of the “leap in the dark” was, I think, that you did not think it would have a material impact—that it would be less than a quarter of a per cent in terms of monetary policy equivalent. What do you think the real-world impact has been?

I see the Governor wincing. What phrase would you use there?

How much more are people having to pay on their mortgages?

I am happy to write to you. I looked at that ING report, and I think that that 80 basis points is actually the total change in swap spreads. That incorporates a lot more than QT, and I think they do make that point. I would be happy to write to you—maybe that is best, because it is all quite technical.

Will you update us on what impact the Bank thinks quantitative tightening has had in the real world?

The annual review always does that, and we will do that again. You will get that. I will happily write to you on the ING point because I think that is a bit misleading.

They are wrong? Okay.

No, I don’t think they are, actually. I think they described it in the report rather differently and it got taken into a rather different context.

We will have a conversation about an article when we have all that in front of us. It is difficult for us to have a discussion about an article that not all of us have read. Dr Mann, you and Yuan Yang had a very brief exchange about mortgages, and you talked about the volatility. One of the interesting things about the mortgage market now is that a lot of people on fixed rates will move off. I know that the banks were looking at their five-year fixes coming off on to what would now be higher rates. How do you factor that into your planning when you are looking at what is happening and the impact on mortgages? That has a real run-through effect into the wider economy in terms of less disposable income, and so on. Dr Mann, as you were the first to talk about it, would you like to expand on that?

There is a box in the MPR that looks explicitly at this question about how much consumers might be affected by the energy price shock as it transitions into the mortgage market and at the cash-flow channel, which is the refinancing and how much you have to pay out on your mortgage. The five-years are in a worse place; the two-years are in a better place. We have all the information on every single one of them, so we can get those numbers at a granular level. There is also the relationship between housing prices and using your house as collateral against which you might do some borrowing; that also has implications. That effect is also relevant and is pointed out in the box. We have a pretty good handle on this; I argue a little bit about this collateral channel and the extent to which it is actually being deployed and plays a role in the consumption strategy of households, but we cannot argue with the cash-flow channel, because it is there in the numbers.

Perhaps on that, Governor, there is also the fact that in parts of London, for instance in my own constituency, one-bedroom flats are typically going at 2015 prices because they are less popular for various reasons. That is also factoring into the mortgage market: people have mortgages on those flats and then cannot sell them for the rate they bought them at. It is not always negative equity, but is not the inflationary property scenario that my generation have grown to see as the norm for many years. Do you have any comments on that?

Yes, we do. We have seen some softening in house price rises, certainly. The system now runs on much bigger buffers, so it is not negative equity in the sense that we were seeing around the time of the financial crisis, but we do factor that in, because it will be relevant to people’s thinking. I was just going to add that I saw there was an article in the FT either yesterday or the day before that suggested that mortgage rates in many countries have gone up by more than the increase in market rates. That set my antennae going, but we checked and I think it is not the case in this country that the increase in two-year and five-year mortgage rates is actually larger than the rise in the market curve.

So that is not a big worry to you.

I very much agree with Catherine. She set it out well in terms of financial conditions; that is the way we look at it. But I wanted to address that particular point, because I know that it was picked up in the news: I do not think it is the case in this country that we have seen mortgage rates going up out of line with market rates.

I would add one point to that. Dr Mann made the point that we have good data on this. It is probably the bit of the transmission mechanism that we have the clearest read on. As she said, there is a debate about whether the collateral channel is operational or whether people would borrow more if house prices were higher, but the mortgage cash-flow channel is a clear part of our understanding of what might happen to consumption. As we put in the MPR this time, the increase in rates that we have seen since the conflict began will reduce consumption by 0.6 percentage points. That is a chunky weight on the economy.

I will add one thing. We had a long discussion about this, since we need to be prepared for every eventuality: to what extent is the mortgage channel going to be different this time compared with, say, 2022? There is granular, detailed data that can also be linked up to consumption patterns to some degree through, for example, credit card transactions. We can see clearly that the consumption channel was working in terms of the monetary transmission mechanism through the mortgage cash-flow channel and that it looks pretty similar now, even though there has been quite a shake-up in terms of who is on fixed, variable, two-year or five-year. It looks roughly like we have settled at the same kind of place.

Just for clarity, I think that the MPR report in February points out that for new mortgages, the increase in rates has been 100 base points on average. Is that in line with your expectations? There was a lot of speculation priced in about what your decisions would be. Does that increase reflect what you would have expected?

Sorry, are you talking about the increase since the conflict broke out?

Yes. The new mortgage rates—around 100 base points.

It is a bit less, but around that.

Let me have a go at answering that, and you can tell me whether I have or not. What we have seen since the conflict is that, for example, two-year mortgage rates have gone up by about 90 basis points and five-year by about 70 basis points. That is consistent with the underlying moves in market interest rates rather than the rates charged by banks having widened. That is worth 60 basis points on consumption, but it comes on top of a monetary policy stance that was already weighing on consumption to the tune of 80 basis points. On where we started this discussion about how much we need to further tighten monetary policy, as opposed to being able to hold where we are, I think the two numbers of 80 basis points of impact on consumption in February and a further 60 basis points since give you a sense of how much monetary policy is already weighing on economic activity and so offsetting the second round.

I just want to add that there was a point after the March decision when I felt that the market had gone too far in terms of its reaction. There might be a number of reasons for that. We had to take cuts off the table, which was quite an abrupt thing to do, but this was an abrupt event. Going back to Catherine’s points, there was a lot of hedge fund positioning for us to cut—not unreasonably; that seemed a reasonable assumption to make. That hedge fund positioning came off, though, quite abruptly. Again, that caused the market to move. There was a point when the market was pricing us to raise rates four or five times this year. I have a high bar for going in and saying this, but I certainly used the words, “You’re getting ahead of yourselves here.”

Deputy governor, I believe you mentioned that you are a professional catastrophiser—you should try being a Labour MP with two teenage children. You also said the thing that keeps you awake at night is the likelihood of a number of risks crystallising at the same time: major macroeconomic shock, confidence in private credit, AI and other risky valuations on stock markets and so on. On a scale of one to 10—where one is very relaxed and 10 is extremely worried—where would you put yourself on risks to financial stability at the minute?

I think we as a committee have described them as material and elevated, so we are at the top end of the distribution. On the things that you picked up on in your comment just then, the fact that risky asset prices appear not to reflect the risks in the economy that we have spent the last session talking about and the fact that private credit is a new form of credit that is important to the real economy that has leverage, is opaque and has complex interconnections. We do not have good visibility of that at the moment. Those things do keep me awake at night. What gives me some reassurance is that the core of the banking system is resilient. We have been here with our Financial Policy Committee hats on and talked to you about how we have stress-tested the banks. The core of the system is in good shape, so maybe I would give it a seven or eight out of 10 on that basis.

I will put my other hat on: I chair the global Financial Stability Board, for my pains, because I do not have enough to do. There are five things that we are focused on. One is asset valuations, particularly the AI implications of that, and sustainability. The second one is private credit, which Sarah just referred to—by the way, particularly in the US, which finances a lot of the investment. The third one, which we have talked about a lot, is hedge fund leverage, particularly in the core Government bond markets. The fourth one is the introduction of new forms of money, particularly things like stablecoin. The fifth is the other big thing happening at the moment, which is this whole thing around Mythos and the question of the increase in potential cyber-risk as a result of the power of these models.

I think you mentioned, deputy governor, that one of the reasons to sell down the Bank’s holdings of gilts is to give more headroom to deal with financial instability in the future. Another tool in the locker is interest rates, but we have a less benign inflation environment. At the same time, Government finances are stretched. Debt-to-GDP is about 100%, and we are not the only G7 economy in that state—Japan has over 200% and the US over 100%. We are in a very different environment from the one we were in in 2007-08. How confident are you that we have the headroom to deal with a financial shock, given we are at seven or eight on the scale?

I look first at the core financial institutions that matter for the provision of credit to businesses and households—the banks. I think we are in a good position on those. The other important thing is the resilience of the gilt market. As I mentioned in our discussions earlier, the participants in that market have changed. There is a need to build further resilience in the gilt repo market to underpin that, and we are pushing ahead with that. We have a plan of work that will enable us to be in a better position, and we have use of our balance sheet if necessary. We have introduced, for example, our contingent non-bank term repo facility, which is there to step in in the event of gilt dysfunction if we need to do so. We continue, with our financial stability hats on, to explore all the other risks that the Governor just highlighted, with a particular focus on private credit at the moment. We are doing an exploratory exercise on that, and we will be able to tell you more about that at the half year.

There is a lot of work to look at this. One of the themes from previous discussions is the need to understand much more about private credit, which probably requires international solutions. Are we ahead of the curve on that, or do you feel that we have some catching up to do and actually we all need to roll our sleeves up, nationally with the Treasury and internationally as well, to look afresh at financial stability?

There is a danger that I am marking our own homework here, so please take this comment with that in mind. The Financial Policy Committee is doing a good job of picking up a rock and looking underneath it at areas where the new risks are occurring. We did an exploratory exercise that looked at how the behaviour of the gilt market is different now that we have different participants in it, and we have a plan of work on the back of that to improve resilience. We are doing work on private credit to try to better understand that. We have work in train on Mythos and stablecoin. I think we have good actions in place in order to fix the areas of vulnerability that we are concerned about. The other important thing is that we talk about these things publicly. Our first line of defence is the financial institutions in the market, and that is a big part of why we put these issues out there in the public domain.

Are you content with the reactions of the financial institutions? Do you feel that they are derisking sufficiently and taking steps that you would wish them to take?

I think the fact that the banks pass our stress test gives us some grounds for reassurance, but there is always a need for banks to continue to double-click on these areas of interest.

There are two big questions in that world. First—this is what the systemic exercise is addressing—does everybody properly understand the interconnections between private credit and other parts of the system? They are interconnected, even though the interconnections have changed in a way. Private credit is an important part of the whole system, because it is funding a lot of investment in tech. The interconnections are important. Secondly, do the people who own it understand the characteristic of the thing they own? If they do, the reactions will be easier and smoother. If they do not, they will be more abrupt.

A final question from me. We have mentioned Claude Mythos and developments in AI. Let’s say I was out in the Calton in my constituency on Saturday, and someone asked, “What’s all this AI and Claude Mythos? What’s the risk to me?”. Are you seeing a risk to things like payment systems—a shock that could affect people withdrawing cash to go to Morrisons in Calton on a Saturday morning?

Claude Mythos and the other models like it—it is not the only one—are a big deal in terms of increasing the capability to identify and exploit vulnerabilities that exist in technology already. Our understanding of it at the moment is that, if you are a well-defended institution, you should be in a good place, because you will be able to see the model coming and ensure that it does not get in and exploit those vulnerabilities. The biggest banks and the biggest payment systems—the core of the system—should be well placed to defend themselves against such an attack, but be assured that we are working with them to make sure they are. Also, we have to be ready for the wave of patching that will come their way, as all these vulnerabilities are identified and require software to be updated to ensure it is more resilient in the future.

Of course, this is not limited to our world; let’s be very clear that this goes across the whole landscape. Our world is obviously an important part of the world, but this is about IT systems in all parts of the landscape.

That’s right. It matters for energy, health and air traffic control.

“Should be” means they probably are, and the fact that you are working with them suggests to me that the Bank of England is not entirely content yet, but people are. There is a difference between “should be” and “are”, isn’t there? Sorry to nit-pick.

No, no—it is a reasonable point to make. We have been working for many years with banks to get the basics of cyber hygiene right. What you saw with JLR, Marks & Spencer and the Co-op is that this is a difficult thing to do across the entirety of your waterfront. It is much harder to defend against everything than to find the single weak point to get in. We are working with firms to make sure they are ready for that.

Thank you very much, deputy governor.

May I thank our witnesses very much indeed for their time? We had a very interesting discussion of the latest interest rate decision back in April and, particularly, the impact of the war in the middle east on UK energy and food prices. We also covered the MPC’s outlook for unemployment and wage growth, concerns around labour market data—a perennial issue here and for the Bank—and the impact of AI. We also covered the MPC’s analysis of recent economic growth, the pace of quantitative tightening and the mortgage market. I thank our witnesses: Dr Catherine Mann, Dr Swati Dhingra—both independent members of the MPC—the Governor, Andrew Bailey, and the deputy governor, Sarah Breeden.