Public Accounts Committee
Oral evidence: Managing government borrowing, HC 74
Thursday 7 December 2023
Ordered by the House of Commons to be published on 7 December 2023.
Members present: Dame Meg Hillier (Chair); Sir Geoffrey Clifton-Brown; Mr Jonathan Djanogly; Peter Grant; Sarah Olney.
Gareth Davies, Comptroller and Auditor General, National Audit Office, Simon Reason, Director, National Audit Office, and Marius Gallaher, Alternate Treasury Officer of Accounts, HM Treasury, were in attendance.
Questions 1 - 96
Witnesses
I: Richard Hughes, Chair, Office for Budget Responsibility.
II: James Bowler CB, Permanent Secretary, HM Treasury; Sir Robert Stheeman, Chief Executive, Debt Management Office; Dax Harkins, Chief Executive, NS&I.
Report by the Comptroller and Auditor General
Managing government borrowing (HC 1658)
Examination of witness
Witness: Richard Hughes.
Chair: Welcome to the Public Accounts Committee on Thursday 7 December 2023. Today we are examining how the Government manage their borrowing. We are not the Treasury Committee: we are not discussing the policy and whether borrowing levels are right or wrong; we are looking at the cost-effectiveness of how the Government’s borrowing systems work. That is done through two main bodies, but before we go to that panel, we are pleased to have in front of us Richard Hughes, chair of the Office for Budget Responsibility, so that we can get a perspective on this from outside the Departments and bodies concerned.
We are going to have half an hour or so with Mr Hughes, to talk about how we handle this important issue. Of course, borrowing has gone up, and the cost of borrowing has gone up, so this is a very interesting time to have this session, and we want to get into the important issues.
Q1 Peter Grant: Good morning, Mr Hughes. First, could you briefly outline, in terms that we would understand, what role the Office for Budget Responsibility plays in Government borrowing?
Richard Hughes: Sure. We are the Government’s official economic and fiscal forecaster. In the run-up to fiscal events—both Budgets and autumn statements like the one we had a few weeks ago—we first put together a macroeconomic forecast for the UK economy, looking five years ahead. Then, on the basis of that, we put together a fiscal forecast—we forecast Government tax revenues and Government spending.
That, at least in recent years, has left a deficit that needs to be financed—an amount of borrowing the Government need to do to finance the gap between receipts and spending. We turn what is Government borrowing into something called the Government cash requirement, which is one version of borrowing. There are lots of different versions of borrowing we can get into if you like, but the one that matters for the Debt Management Office and colleagues at NS&I is the public sector net cash requirement, which is how much cash the Government need to raise over the next five years in order to finance the gap between revenue and spending. We work closely with the DMO to provide them with our best estimate of what that cash requirement is likely to be. They turn that cash requirement into a remit of finance, which they raise from financial markets. For the DMO, that is in the form of gilts and Treasury bills; for NS&I, it is in the form of NS&I products.
Q2 Peter Grant: What challenges arise from the fact that borrowing and overall debt are higher than they have been for quite some time and probably higher than would have been expected not so long ago? Does that make your job more difficult, or is it the same job but with bigger numbers?
Richard Hughes: I think it makes everyone’s job more difficult. It is always easier to manage finances when you have less debt, because it gives you more options and choices. In the very short term, in a country like the UK— with a few exceptions—the Government does not struggle to raise the finance it needs to finance its operations from year to year. In the short term, having high levels of borrowing or debt does not pose a particular challenge from a fiscal management or forecasting point of view.
But in the medium term, having high levels of debt certainly means that more of your fiscal space is taken up by having to service that cost of debt, because of the interest you pay on the stock of debt you have already accumulated, particularly in an environment where interest rates are highly volatile, which they have been—they have risen a lot in the last 18 months—and where inflation is also volatile—we have seen double digits in recent years, which we have not seen in over 40 years.
Much of our debt is inflation-linked debt. That makes forecasting the inflation component of the public finances more challenging, and that constrains the Government’s choices when it puts together Budgets, because the more your fiscal space is taken up by interest costs, the less you have to spend on health, education and other areas, or the less you can give back in tax cuts, if that is what you want to do.
In the much longer term, in addition to doing our five-year forecasts, we try to forecast the long-term trajectory of debt to see where the public finances are headed over a 50-year period. In that context, you have to think about a much wider set of questions, like what is happening to the age of the population, what pressure that puts on the health service, and what the working-age population is earning in terms of the taxes to fund these things.
Q3 Peter Grant: How do you respond to the suggestion, which has been made in the Chamber of the House of Commons on more than one occasion, that there is not much point in having all these economic forecasts, because the only thing they have in common is that they always turn out to be wrong?
Richard Hughes: All forecasts turn out to be wrong: that is true of weather forecasting, political forecasting and economic forecasting. The question you have to ask is whether your forecasts are more accurate than those of others producing economic forecasts. You can only really benchmark yourself against other people who are forecasting in the same volatile environment.
The average errors in our forecasts are a fraction of the actual volatility and out-turn data economically. In that sense, the biggest challenge all forecasters face is volatility. Within that volatile environment, you do your best to provide a central and accurate forecast. Every year, we produce what is called a forecast evaluation report, which looks back at our forecasting record for the two previous years. More recently, we produced a much broader study that looked at our entire forecasting record since we were established in 2010. It showed that our forecasting record compares favourably with that of the Treasury, when it did the forecasts in-house. It also compares pretty well with those of other independent forecasters in the UK and our counterparts elsewhere in Europe.
So our forecasting record is pretty good by domestic and international standards, but no forecast is ever perfectly accurate, just as no weather forecast tells you exactly what the temperature will be at the end of the week or what the rainfall will be on a given day.
Peter Grant: Thank you. I am aware that we don’t have a great deal of time, so I will finish there.
Q4 Mr Djanogly: Good morning. Could I go back and look at the operational side of things a bit more? I will start with paragraph 2.3 of the NAO Report—you mentioned most of this—which states: “The government’s borrowing process begins with OBR’s forecasts of the government’s cash needs…for the following financial year.” It then states: “The Debt and Reserves Management team within HMT works with the DMO and NS&I in preparing their remits, including with DMO to decide how best to structure and profile borrowing.” Do you feel that that process works well from the OBR’s point of view and, generally, looking over the shoulder of the Treasury and NS&I?
Richard Hughes: I think it does work well. We have very good, collaborative relations with our colleagues in the DMO, and we rely on the infrastructure that they have: in essence, they have a big model of every single gilt that we have ever issued and what the impact of different interest rate assumptions might mean for the cost of those gilts over the long run. In that sense, we rely a lot on the infrastructure that they have in putting together our five-year forecasts.
We also work very effectively together in looking at the risks around those forecasts. In addition to producing a central forecast on which the Government base their fiscal policy, we also do a lot of work—an increasing amount of work—to look at the risks around that central forecast: alternative assumptions for interest rates, inflation and other kinds of shock that the public finances might face. We rely a lot on the expertise, analysis and modelling capacity of the DMO to tell us what happens if there is an interest rate shock. For example, if interest rates turn out to be 100 basis points higher, what would that do across the entire maturity of the debt portfolio over a period of 50 years? The DMO has some very sophisticated, state-of-the-art models that we use to produce that kind of scenario analysis. I think the collaboration with the DMO is excellent.
Q5 Mr Djanogly: In operational terms, are there processes that you think could be improved?
Richard Hughes: There is certainly one area—and I think DMO colleagues would share my desire for more information on this—around who owns our debt. We have a very good grip on the debt that we have issued: its composition, its maturity and the risks around it. We know, down to the last penny, how much of it is owned by the Bank of England. We know, generally speaking, how much of it is sitting in the hands of pension funds and insurance companies in the UK.
But where those gilts are held abroad, we know relatively little, because the debt is difficult to find. We know it is held abroad and, in some cases, we know who the nominated owner is. But given that, nowadays, about a quarter of our debt is owned abroad, you would feel better if you had a better sense of who those people were, because that debt is likely to be the most volatile element of your debt holdings.
Q6 Mr Djanogly: On a day-to-day basis, do you feel that you have a good working relationship with the Treasury, the DMO and NS&I? If there are things that need to be sorted out, are they?
Richard Hughes: Yes, very much so.
Q7 Sarah Olney: Could you talk a bit about quantitative easing, the impact that it has had on the ability to manage debt, and the risks that it represents?
Richard Hughes: In some ways, quantitative easing is incredibly complicated, but in other ways it is incredibly simple. From a public finance point of view, the simple way to think about it is that it was the purchase by the Bank of England of fixed-rate gilts that the Treasury and DMO had already issued, in exchange for variable-rate reserves that the Bank of England issues.
Its effect on the public finances has been, in effect, to swap a long-term liability for a very short-term liability. In making its purchases, the Bank of England replaced a liability that had an average maturity of 15 to 20 years, as our gilt stock did, with a liability that has, in effect, an overnight interest rate, because as soon as the bank rate changes, the interest rate on bank reserves also changes. It was swapping one asset for another, and its effect on the public finances was to shorten the average maturity of our overall liabilities considerably, because more of them were at a floating bank rate and less of them were at a fixed gilt rate.
Q8 Sarah Olney: What impact has that had on the cost of borrowing?
Richard Hughes: The fiscal impact of quantitative easing is a function of two things. One is the interest rate differential between what the Bank of England pays on its reserves and what the Treasury would have paid had the gilt still been issued. That interest rate differential between bank rate and gilt rate is one way in which you make interest rate gains or losses on the instruments. The other way in which you make losses are valuations on the stock when it is either sold or redeemed. In the initial phases of QE, and up until very recently, QE was actually making a profit—quite considerable profit.
Q9 Sarah Olney: Profit for the Treasury?
Richard Hughes: Profit for the Treasury. Between 2009 and 2022, it made cumulative gains of £124 billion, and that was on a stock of just under £900 billion-worth of QE. Up until very recently, basically before the bank rate went above the average gilt rate holdings, it was making a significant net profit. It has now begun to make losses, because that interest rate differential has gone the other way and the bank rate is now higher than the average rate that the Bank of England receives on its stock of gilts. The APF has begun to make interest rate losses, as well as making losses on the sales or redemptions of gilts as they go back into the market, because the gilts are now worth less when they are sold than their value when they were held.
In our most recent EFO, we did a long-term forecast all the way to the end of the APF, when it has finally reduced all its holdings; in the long run, we expect it to run a net loss of £126 billion over its lifetime. It is still making a profit overall on the net interest margin—it is making around £63 billion—but it is making £56 billion of losses on redemptions and £132 billion of losses on active sales. So, over the long run, the APF is making a loss over its lifetime. Up until now, it has been making gains; over the remaining period, you expect it to start making losses on both the interest rate differential and disposals of the assets.
Q10 Sarah Olney: Just to clarify, you said that the overall loss over that full length of time is £126 billion.
Richard Hughes: That’s right.
Q11 Sarah Olney: When does that loss impact? If we start making losses on the APF in 2024, will that impact the money available for departmental spending, or is it not that simple?
Richard Hughes: The APF is fully indemnified by the Treasury. One real advantage of the UK system is how transparent we are about these things. In other countries, these losses accumulate in the balance sheet of the central bank, which at some point may need to be recapitalised by the taxpayer. Central banks are always indemnified by taxpayers, whether you write it down or not. If central banks make big losses, like the ECB or the US Treasury, taxpayers at some point lose out. We are much more transparent about that in this country, because we have an explicit indemnity and we report in very great detail on when losses arise and are made.
It is the case that, as the interest rate differential declines, it eats up margins that Governments have against any borrowing that they have. Any difference in bank rate versus gilt rate affects borrowing every year. The losses on the APF also affect whatever objective the Government has to get debt under control, because these are all transactions that, in the end, affect debt.
Q12 Sarah Olney: Can you talk a bit about quantitative tightening as well? Obviously, the Bank of England is now starting to sell on the debt that it took off the Treasury, and that will be competing in the market with new debt that is being sold. What impact will that have? Is that additional to the £126 billion impact that we discussed, or is that contained within it?
Richard Hughes: The £126 billion is over the life of QE and QT, so that is the entire life of the APF. That is it stocking up and stocking down. On the assumptions that we took for the unwind of the APF, we basically took the public pronouncements of the Bank of England on the rate of disposals over the coming five years, which we then project forward into the future. In that sense, that encapsulates what are, in effect, the losses on QT over the life of the APF.
In terms of what pressure it puts on the gilt markets, it is the case that the gilt markets are now being asked to absorb gilts from two different sources: primary issuance from the Debt Management Office, as well as the Bank of England selling gilts back into the market that it has been holding for monetary policy purposes. What that does to the volume of gilts going on to the market is that what is called the Government’s gross financing requirement peaks at almost 8% of GDP next year—for the UK, that number is usually below 5%. It has got close to that during other periods, like the pandemic and the financial crisis, when the Government was issuing large amounts of gilts to finance a big gap between receipts and spending. What we have not seen in the UK is that number sustained for a long period of time where you have sustained high levels of primary issuance by the DMO. That is partly because the Government are borrowing a lot of money and the DMO is rolling over gilts that are coming to maturity in the middle of this decade. So a gross financing requirement of almost 8% is historically high for the UK. If you look at other countries in the G7, though, countries like the US and Japan have gross financing requirements much in excess of that, either because they have, as in the US, much higher primary issuance—the Americans are running big deficits at the moment—or because, as in Japan, they are just rolling over a very large stock of debt.
Q13 Sarah Olney: Are we unique, or comparatively unique, in having this situation of quantitative tightening happening at the same time as primary issuance?
Richard Hughes: No, it is happening in most countries now because most countries are tightening monetary policy at a time when fiscal policy is still recovering from the pandemic. Government borrowing is still relatively high—historically, relatively high—in most G7 countries, but their central banks are also now putting assets back into the market. So, in some ways, all G7 countries face this challenge. The financing pressure we are putting on the markets is historically high for the UK but, relative to some other G7 countries, which are rolling over a lot of debt in the markets, like the US and Japan, it is not quite as high.
Q14 Sarah Olney: Do you expect the 25% foreign ownership that we currently have to increase as a proportion, because of the market stress?
Richard Hughes: I think it has turned out to be a safety valve for the UK gilt market. It has grown steadily over time. It used to be about half that amount, so we used to have about 12% or 13% foreign ownership of debt—debt that was in private hands. It has grown over time as, basically, our debt stock has grown over time.
There is another thing that has happened in the UK over the last two decades. A major end holder of gilts is defined-benefit pension schemes; those schemes have got close to being fully funded to match all their liabilities, and their liabilities are moving into payment, so you would expect there to be less strong demand from the domestic DB pension fund sector. That means that, for the volumes we have to put out, as well as the likely volume of demand, we are more likely to have to keep looking to the foreign markets to take down the stock of issuance.
Q15 Chair: This is not our area of expertise—the Treasury Committee obviously looks at the policy on this—but the Bank and the DMO sending gilts into the market at the same time seems an immediate area of potential risk. From what you have said to Ms Olney, you are quite relaxed that, even though this is happening, it is manageable within the world market.
Richard Hughes: What we have seen so far, apart from what happened last autumn, has been not a lot of disruption in terms of the volume and stability of the gilt markets. Even in the context of recent months, the Bank of England are saying not only that they are going to be selling £80 billion-worth of gilts into the market from the APF, but, “Actually, we’re going to push it up to £100 billion.” You have not seen a big gilt market reaction to those volumes of both primary issuance from the DMO and gilts coming back out of the APF.
Q16 Chair: You seem very calm here, Mr Hughes, about all of this.
Richard Hughes: I am not usually calm about things but, so far, a combination of fiscal policy recovering from a pandemic and monetary policy trying to unload gilts as a tool of monetary policy has not put undue pressure on the gilt markets. That is not to say the gilt markets have not been under stress; they certainly were during the pandemic, and they certainly were this time last year, but that was for different reasons from what we have discussed so far.
Q17 Chair: We will probably be asking our next witnesses about foreign ownership as well. Do you see any risks there? You have said that you are not concerned about what is pretty much a doubling of foreign ownership. There must be risks, because we don’t know where the money is, and different global geopolitical risks could have an impact. Is that something the OBR has considered?
Richard Hughes: It is definitely a risk because, by their very nature, foreign investors are less committed to holding sterling assets. They are interested in having diverse portfolios. They are interested in rates of return. What this means is that what matters is the quality of gilts as a substitute for US treasuries, German Government bonds and Japanese Government bonds. In that sense, our profile as a nation, as an investable proposition, matters a lot more than if you are borrowing from domestic pension funds, which are just looking for sterling assets to match sterling liabilities. So this kind of capital is, almost by definition, more fickle and more flighty than domestic investors.
Q18 Chair: You have talked about the transparency of the copper-bottomed guarantee or gold-plated guarantee by the Treasury. Does that mean that UK Government gilts are a more attractive investment for some of these overseas investors?
Richard Hughes: I would not necessarily draw that conclusion. It would be nice from a risk management perspective to know who the owners were, to get a better sense of how fickle and how flighty they are. If these are long-term pension fund investors in other countries, they are probably looking to hold for the long term. If they are just looking for a quick return on an interest rate that looks a bit tasty compared with US Treasury and German Government bonds, you would worry a bit more about where they might go if those interest rates were to change.
Q19 Chair: And, of course, there is no visibility on any of that.
Richard Hughes: Yes.
Q20 Sir Geoffrey Clifton-Brown: Your forecasts are extremely important, because at each fiscal event you are asked to adjudicate on what is proposed. How do you look to constantly improve the accuracy of those forecasts?
Richard Hughes: Every year we produce this document, which is called the forecast evaluation report, and it looks back in great detail. Everybody has forecast errors; we have forecast errors. We look at where those errors were generated, why they happened, and how we can improve our forecasting models and methods to take better account of any risk factors that we should have been cognisant of. For those who want to read more about our forecasting performance, I can definitely recommend this working paper, which looked back at our forecast performance over the past 13 years—since we were set up—and where we benchmark ourselves against other countries and other forecasting institutions, to see how we fare and how we could improve. We are always identifying areas where we can improve the way we forecast things, and we are always improving the way we model both the economy and the public finances.
Challenges arise because the economy is very volatile, and we have to try to have a central forecast through some very choppy weather. Challenges also arise because we are forced to rely on stated Government policy when we put our forecast together. As we all know, there are patterns of behaviour in Government whereby they say they are going to do things and then everybody knows they are not going to do them, but we have to believe them when they say it. Fuel duty is the classic example: the Government says every year that it is going to index it to RPI, but it gets frozen every year. The Treasury Committee eventually got fed up with this and instructed us to do the forecast on two different bases—one is frozen, the other indexed.
We have a bigger problem on the path of public spending. If it is not a spending review year, the Government tends to have a rather tight path for spending on public services. When spending reviews come around—at least recently, they have chucked loads of money into the pot when it becomes time to allocate it out to health, education, transport and other areas. That is to the tune of £20 billion or £30 billion, so it could drive up to a £20 billion or £30 billion error in our medium-term forecast for public spending, because before a spending review happens, the Government says, “We’re going to run the public finances really tight after 2025,” and when 2025 comes around, suddenly there is more money on health, more money on education and more money on transport.
Q21 Sir Geoffrey Clifton-Brown: I do not want this question to be taken as a criticism of you or your organisation, because you were not alone in this. The Bank, the Treasury and everybody else got the inflation forecast very wrong in 2020. I accept the effect of the pandemic, but what lessons are to be learned in terms of major, catastrophic events and, when you see these emerging, adjusting your forecast sufficiently to be able to deal with that?
Richard Hughes: This was one of the big topics in our last forecast evaluation report, where we had an analytical box breaking it down. The global forecasting community got a few things wrong. One was about the pace of the recovery in demand, relative to supply, coming out of the pandemic. Demand recovered much faster than supply. That is especially true in the UK, where demand came roaring back. The supply of goods was constrained because Asia was still locked down. In the UK, the supply of labour was constrained because it turned out that more people left the labour market and did not come back. That generated both international and domestic inflationary pressures that I think we all underestimated.
Thinking back to the time of the pandemic, what we were all worried about was that demand was not going to recover and that people were so traumatised by the pandemic that they were not going to go back out to the shops—that they were going to hang on to lots of savings, because they were so worried about how it was all going to unfold.
As vaccines were rolled out, people were able to get back into the shops. That demand really came roaring back, not just here, but in the US and lots of other places. That was the first surprise. The second surprise was the Russian invasion of Ukraine and the fact that energy prices shot up. That was also something that, unless you were an incredibly good geopolitical forecaster, you probably would not have predicted happening, at least looking a few years out. That came as a shock to the global energy price.
The third area where we have all learned lessons has been in the spill-over of energy prices on to the rest of the economy. We knew it was going to be important. We had looked back at the ’70s and looked at the pattern of higher oil prices speaking through to the rest of the economy. What we also knew, however, was that we are a much less energy-intensive economy than we were in the ’70s, so we thought that the effect was going to be significant but less. At the time, we thought there was going to be about a 25% spill-over of energy price inflation on to whole-economy inflation. In practice, it has turned out to be more: I’d say a 50% spill-over.
Q22 Sir Geoffrey Clifton-Brown: This is absolutely fascinating. I will be in trouble if I ask too many questions, but how much international benchmarking of your forecasts do you do? How do other similar organisations in other major economies do compared to you?
Richard Hughes: We do a great deal. We compare ourselves to all other European official forecasters because we want to benchmark both our economic forecasting record and our fiscal forecasting record, and so we have to find other fiscal forecasters. There are lots of charts you can have a look at if you like. We are, in the near term, one of the best official economic forecasters in Europe; over the medium term, some people beat us out, but we do benchmark ourselves against our European partners, and we compare pretty favourably on our economic forecast. There are more challenges on the fiscal side, partly for the reasons I highlighted, which are that it depends on whether the Government stick to what they say they will do with public spending and other areas.
Q23 Sir Geoffrey Clifton-Brown: A final question from me: the Bank reacted pretty quickly in terms of monetary policy when inflation went up. Now that inflation is coming down, will the Bank act quickly enough to adapt to that condition?
Richard Hughes: This is definitely not a question for me, I am afraid. If you had the Bank in front of you, I am sure it would give you its views, but I cannot comment on monetary policy.
Sir Geoffrey Clifton-Brown: Fair enough. It was worth a try.
Q24 Chair: You have an interesting perspective overseeing and forecasting, as you say, fiscally and economically. The Debt Management Office has had a pretty critical role in issuing gilts and securing borrowing on behalf of Government up to its remit. It has a challenge because it will have a change of leadership after 20 years of stability. Does that worry you?
Richard Hughes: The DMO is a world-class organisation. Robert has been there for a very long time and has made a very big impact on the organisation. Succession planning is always a challenge, but he has very good deputies, who we work with very closely, and there are also people in the markets who admire and value the work the DMO does who could be potential candidates. Robert has been good enough to give a fair amount of notice of his departure, which means you can find a good range of candidates. We have our own succession challenges in the OBR. There are not that many macroeconomists in the UK willing to take the Crown shilling—
Chair: In other words, the pay cut.
Richard Hughes: In some cases. But we have managed to find some very excellent people to fill the BRC, as well as senior posts in the OBR. Because the DMO is so well regarded, people want to work there—maybe even at a salary sacrifice.
Q25 Chair: Do you have anything to say about National Savings & Investments?
Richard Hughes: It is a smaller but still important part of the Government’s financing. To be very frank, because we deal in hundreds of billions and sometimes trillions, it makes a relatively modest contribution to the kind of work that we do around forecasting the financing requirements, as well as the risks around that. Having a diverse range of possibilities for financing yourself is a good risk management practice.
Other countries, like Italy, rely a lot more on retail savers as holders of Government debt, whereas we tend to rely more on the DMO as a primary issuer into the gilt-edged market makers’ market, and then it finds its way into people’s pension funds. In other places, like Italy, there is a lot more direct issuance straight to households. The NS&I is the route through which we tap the retail market, but it is always good to have several routes into the market.
Q26 Chair: We are looking at how the Government is managing its borrowing, and you have given us some indication of your views. Is there anything that keeps you awake at night? Do you have any areas of worry about Government borrowing, given that we now have two lots of sellers in the market, with the Bank of England now selling some of the gilts it owns? What are the areas that worry you?
Richard Hughes: As I said, with the exception of last autumn, when you saw signs of significant stress in the gilt market, I don’t think the Government struggles to finance itself in a given month or year. In the medium term, you can certainly see in the public finances that having a higher level of debt and higher interest rates is constraining the Government’s fiscal choices very considerably. You saw that in the last autumn statement. Just in the space of the six months between our March and November forecasts, interest rates went up by more than 100 basis points, and that added £12 billion to the cost of servicing the Government’s debt stock. That is £12 billion it didn’t have to spend. If it wanted to do more tax cuts, it could have done more tax cuts; if it wanted not to cut the real value of public services spending, it could have done a lot less of that. Instead, all that money went on debt interest. In the medium term, the fiscal choices that the Government faces are now much more constrained by its debt stock.
In the longer term, we have a large debt stock now and growing pressure on public finances going forward, because you have an ageing society, rising costs of healthcare, pensions and other things, so all the pressures on debt are in the upwards direction. In the past, we had a demographic dividend from more and more people coming into the work, a geopolitical dividend from defence spending falling and making space for the welfare state to grow and the tax burden to come down, and relatively high growth rates. All those tailwinds that we used to have supporting the public finances and fiscal policymaking are now headwinds that we are confronting with a debt-to-GDP ratio that we haven’t had since the 1960s. Those sorts of things keep me up at night. We write a lot about them and try to raise awareness of them.
Chair: Thank you for what you do, because it does cast light for the average citizen, as well as for Members of this House and Committees.
Q27 Sir Geoffrey Clifton-Brown: This may be another question that is out of order, but let’s try it. The Bank is selling low coupon gilts that it accumulated under quantitative easing, and the debt management market is going into the market and buying higher coupon gilts. What is the sense in that? Why not simply transfer those low coupon gilts from the Bank to the Debt Management Office?
Richard Hughes: When we think of the public finances, we just consolidate everything, so the Bank of England is just a wholly owned subsidiary of the Government—UK plc. If you sell the gilt now, you will make a loss, but if you hang on to it until redemption, you will make a loss on the interest rate differential, because it is paying an interest rate back to the Treasury that is lower than what the Bank of England is paying out to the creditor. So the longer you hold on to it, the more interest losses you make. If you sell it now, you will make a loss on disposal, but at the point of maturity, it is going to have to be replaced by a higher-yielding gilt anyway, so in some ways the timing of when you sell a particular gilt matters less than you think, in terms of the loss you might make on it. If you hang on to it in the public sector, you are just making losses on the interest rate differential, and if you sell it, you crystallise that loss now but you are not making interest rate losses in the future. Ultimately, the Government is a net borrower and it needs to roll over this stuff, and if you hold on to it to maturity, it is going to have to be replaced by something else.
Q28 Sir Geoffrey Clifton-Brown: Ultimately, with all these institutions, what really matters is the cost of servicing borrowing to the taxpayer.
Richard Hughes: Yes, exactly. Our net interest costs on the whole stock of public sector liabilities—Bank of England liabilities and gilts outstanding—is pretty high these days, both at Bank rate and at the market gilt rate.
Chair: Thank you very much indeed for your time, Mr Hughes. You are very welcome to stay for the next session if you wish. Your transcript will be available on the website uncorrected in the next couple of days, should you wish to read what you have just told us. We will be producing a Report on Government borrowing in the new year. It was great to have you as a witness today.
Witnesses: James Bowler, Sir Robert Stheeman and Dax Harkins.
Q29 Chair: Welcome to our second panel on managing Government borrowing. Government spending has succeeded income in every year since 2000-01, and in all but five years since 1970-71. In any year, of course, where the Government borrow more than they repay, their stock of debt increases. We are here today to talk to the Treasury and the two bodies responsible for arranging the borrowing on behalf of Government and how that is managed. We are talking not about the policy—that is a matter for the Treasury Committee and other parts of this House—but about the mechanics of how the system is working and what the Treasury is doing to keep on top of this and ensure that it is working well.
I am pleased to welcome James Bowler, permanent secretary to the Treasury, and Sir Robert Stheeman, chief executive of the Debt Management Office. As I indicated in our first session, he has been in that post for 20 years now.
Sir Robert Stheeman: Twenty-one!
Chair: Twenty-one—forgive me, I cut you short. I also welcome Dax Harkins, who is the chief executive of National Savings & Investment. NS&I is the retail offer that individuals, like all of us in this room and out there, can buy savings products from.
Sir Robert, you are standing down very soon. You have been in post for 21 years, which is one of the things that really impressed us while we were preparing for this. What has kept you from going into the private market but instead working for the public sector in this way? Presumably, from your background, you could be earning a lot more money somewhere else. What has made you stick with the DMO?
Sir Robert Stheeman: It is a good question. I was about to say flippantly that no one has actually offered me a job, but it is probably too late for that.
As Richard hinted at earlier—I say this hoping that my eventual successor hears it, whoever he or she might be—this job is, for anyone who works in markets, arguably just about the most interesting job around, full stop. It is this unusual interaction where you are involved in government, in policy and in delivery on the one hand, but on the other hand you are also interacting directly with the financial markets, which was my personal background; I did that for over 20 years. You have a vantage point that is, in my opinion, unique.
It has also helped that the job has changed fundamentally. When I started, we were borrowing nearly £50 billion in the first full year that I was there, and the entire value of the gilt portfolio was £300 billion. This year, we are borrowing nearly £240 billion, and the entire value of the gilt portfolio is approximately £2.5 trillion. That gives you an idea of how the job has changed, but that itself generates its own interest.
Q30 Chair: When you put it like that, jaws were dropping around the table at the scale of what you have taken on. You have been there for 20 years. There are very few roles like that: the C&AG has a 10-year public appointment, which I think is the longest fixed public appointment. Yours was not a fixed position. What if a successor were appointed and were to stay for two or three years, or four or five? Is there an optimal minimal time you think someone should be in the post for, to create confidence in the market and so they can understand the role?
Sir Robert Stheeman: It is a good question. I would say that it depends very much on the background of the individual and whether, like me, they have come from the private sector. One of the reasons I am going is that I feel my excuse, “Please don’t ask me, I’m new”, is wearing a bit thin, but it does take an awfully long time. I thought I knew a lot about bond markets when I came to the DMO. I was challenged intellectually in a way that I had not anticipated. I happily admit that it took time to understand how the office itself actually functions and how the gilt market functions; I was familiar with sovereign bonds, but not in the UK quite so much as other international bond markets. Understanding how government and policy making work, developing relationships—those all take time. Arguably, if it were someone internal or from elsewhere in government, the learning curve would also be steep, but it would be a little different.
I do think that it is important that both someone in my position and staff in the DMO should ideally be in post for long periods, although maybe not quite as long as I have done. We are a specialist agency and we need to develop specialisms, skills and experience; we rely hugely on that experience to make certain judgments, and that comes with time. I do not want to put a number on it, but a certain period of time is very helpful.
Q31 Chair: I understand that an advert has been put out for your role, so I hope people are thinking about that.
May I ask James Bowler, the permanent secretary, what the timeframe is for appointment? Sorry, it sounds like we are shooing you away, Sir Robert, but it is important; we will get to the reason why we are worrying about it later.
James Bowler: It is probably worth saying a couple of things. The timescale is that Robert moves on in the middle of next year. We hope that that gives us a sensible time to find the right candidate and get them appointed well ahead of time, for a sensible transition. I should pay tribute, too, to the incredible work that Robert has achieved.
The Treasury, the DMO advisory board and I have had succession as one of our main things to focus on in the enormous job that the DMO does for some time. I am grateful that we are doing this in a really orderly fashion.
It is worth saying—this is also difficult to say in front of Sir Robert—that we have upgraded the role to a director general post and increased the salary accordingly. [Laughter.]
Chair: Count it as a success, Sir Robert.
James Bowler: That is in terms of the base pay, with a view to the change in the last 20 years that Robert has just underlined. We are looking to make sure that we are as attractive as we can be to both the public sector and the private sector.
Chair: Thank you very much. We will go into more detail about the management of the organisation later.
Q32 Mr Djanogly: Good morning. Building on Sir Robert’s staggering figures—well, they’re not really his figures—I notice that Government debt interest payments are an estimated £110.6 billion in 2022-23 compared with £60.8 billion in 2021-22. Mr Bowler, how do you define success for Government borrowing?
James Bowler: If I can just paint the picture, debt has risen considerably, in particular since the middle of the 2000s, due to a set of global shocks. Our debt as a proportion of GDP was some 40% to 45% in the mid-2000s. We had the financial crisis; the proportion jumped to 70% to 75%. We had the covid pandemic; it jumped to 90% to 95% of GDP, which is similar to similar countries around the world. The level of debt and the level of borrowing is considerably heightened. It is worth saying that that is not at the foot of the DMO. The DMO is a price taker: it takes the cash remit that we give it, which the OBR helps us forecast, and it is a price taker in the market.
When you ask whether the DMO has done a good job, it is difficult to say that there is one figure that tells you the answer to that. The objectives for the DMO—I can come on to NS&I—are to raise the net cash we need to minimise costs and risk. Cost and risk are the two things here. As it does that, it is balancing the demand out there in the market—I said that it is a price taker. It is looking at the strategy of how to sell our debt. We look at yields and at risk premia across the maturities to do that. It looks at risk, and the key thing there, which we will probably come back to quite a bit, is that we have a deep and liquid market. It is balancing those things out.
The NAO Report, which I applaud, sets out that when you look at this, there is not just one metric you can point to that measures success. Each year, we look at the start of the year and at how the auctions go through the year. There are all sorts of measures there. Then, at the end of the year, we evaluate and see what lessons we can learn to set the remit the next year. We are looking at a range of metrics across those spectrums.
Q33 Mr Djanogly: You have just paraphrased your debt management report from March of this year, which says that the objective is “to minimise, over the long term, the costs of meeting the government’s financing needs, taking into account risk, while ensuring that debt management policy is consistent with the aims of monetary policy.” How comfortable are you with the current costs of meeting the Government’s very significant financing needs?
James Bowler: I will answer similarly to Richard: I think we have deep and liquid markets, and I think the markets are functioning well. The DMO is doing a deeply sensible job. I am comfortable that we are able to finance the Government’s needs. The DMO always meets its financing requirements, so I am pretty comfortable there.
Would I prefer debt to be at a lower level of GDP, and would heads of Treasuries around all the developed world prefer lower levels of debt? Yes, we would. Why? I have talked us through the shocks we have had to our economy. Looking forward—Richard Hughes gave evidence on this—do we have space for the next shock? Do we have space for the levels of public services and the level of tax we want to provide? Do we have space for the transition of the economy to net zero? We would all like more space, but these shocks are not significantly different for the UK and for other countries. Not least due to the excellent work of the DMO and the NS&I, we have really strong functioning markets, and I am comfortable that the Government can get the financing it needs.
Q34 Mr Djanogly: Working together—I will perhaps take the issues in turn on this one—what would you do to minimise the costs of meeting these financing needs?
Dax Harkins: We have a number of metrics we agree with the Treasury and Ministers each year: service delivery measures. A couple of them are particularly pertinent to measuring our effectiveness. The first is the achievement of the net financing target itself, which this year is £7.5 billion, plus or minus £3 billion. We also have an efficiency ratio, our cost of managing those funds, which is currently under 7p per £100 of investment.
We also have a value indicator. It is not a service delivery measure, because there are challenges around that. It is directly impacted by gilt yield, which is outside our control, and we would not always be sensible to react to those. But we do have a measure of that, and we involve it in all our pricing measures. That is currently in a very healthy position.
Q35 Chair: You don’t have a value indicator currently?
Dax Harkins: We do have a value indicator. We removed it from a service delivery measure, for the reasons I have outlined. However, we still measure, monitor and use it in all our decisions. We share it with the Treasury and the Minister when we agree our rates.
Mr Djanogly: Sir Robert?
Sir Robert Stheeman: I want to answer by pointing to something that James said about the market and efficient markets. Meeting the debt management objective is about not just cost minimisation, but trying to minimise risk. Both of those are over the long term. I stress that because it means that sometimes we have to make certain trade-offs and judgments when it comes to issuance strategies. The most important thing for us to meet that objective is to support—as far as we can, within the limitations that we have—a deep, liquid, well-functioning bond market. That allows the DMO to access funding for Government whenever we need it, ideally for as short or as long as we want, in all conditions. That is of primary importance.
As important as the debt management objective itself is, even more important is providing funds for Government, full stop. That is part of what we refer to as our cash management function. A lot of people assume that what we do is purely about gilt issuance. We also have, below the radar, a very significant cash management function, which is about smoothing Government cash flows as much as possible. The transaction volume there is staggering.
All of that means that sometimes we have to make decisions. I will give you a good example and be as jargon-free as possible. In a normal world, one might expect yields on short-dated gilts to be less than yields on long-dated gilts. In other words, we would have an upward-sloping yield curve. If that were the case, if we were purely about cost minimisation, we would borrow everything short. However, if we were to do that, we would maximise or significantly increase risk, so we have to make judgments about how to spread that out. A good example is that back in 2000-01, the DMO issued everything in longs—in other words, 15 years or more. That was everything. We didn’t issue any shorts or mediums. We only borrowed £10 billion that year.
This is something that Richard referred to. If you have a lot to borrow, your choices are sometimes constrained. All of this is to emphasise the long-term point: we need to ensure a deep, liquid market, because we never quite know when and why we might need it. The financial crisis is a good example. The pandemic is a very good example. Even what happened last year, although the markets were indeed stressed, is another good example. We need to have an efficient, flexible market.
Q36 Mr Djanogly: Thank you. Mr Bowler, paragraph 2.15 of the NAO Report says, “Government’s debt management objective is high-level and not directly measurable, and HMT and the DMO do not have measurable success criteria and indicators allowing a quantitative assessment of whether government is meeting the debt management objective.”
James Bowler: To illustrate why—and I hope I get this right—so far this year, we have sold £155 billion of debt, and the yield on that is about 4.4% on average. Last year, the yield was 3.18%. It would be harsh for me to say that the DMO has become less cost-effective or hasn’t done its job as well, because it is a price taker and the market has determined that price correctly. If I tried to set that kind of thing as an indicator of when the DMO is doing its job, that would be unfair.
Instead, we have a whole set of metrics that we look at. What you are looking at is yields and risk premia across all maturities, and the extent to which we are allocating our debt in the right areas. I will give you an example of when we changed that: in 2018, we reduced the number of index-linked gilts that we issued because of the inflation exposure we had.
The point that the NAO Report rightly raises about how we measure performance is a reasonable one, and we are not unaware of it. Spurred on by the NAO Report, we will look at how everyone else does this and at whether the metrics we are looking at have any gaps in them. We will try to ensure we bring all that together so that we can have a better overview of performance in accepting the NAO’s first recommendation.
Q37 Mr Djanogly: Thank you: that is the answer I was looking for. Moving on, how do you evaluate past performance? In particular, how does that feed into future decision making?
Dax Harkins: I talk about the service delivery measures we have—the clear measure that we go against there, and the value indicator that we look at and review. Those all get built into our annual process when we agree the net financing target with the Treasury and the Minister, taking into account market conditions, customer behaviours and the competitor landscape to work through.
We also look for ongoing improvements to how we manage that process. We have had an independent review of our forecasting capability to help change our processes, models and ways of evaluating. We have looked at implementation pieces this year in a very dynamic marketplace, with lots of base rate changes. We look to tighten the approval process of our rate changes so we can be more responsive and more reactive in the way we increase our interest rates.
Sir Robert Stheeman: I would just like to make the point that, even if there is no single metric by which to measure our performance, I would not want the Committee to think that we do not care about performance. We care deeply about it. We have some 12 pages of reporting and analysis on our performance in our annual report, which we make sure we publish. That is something really important to note.
As James mentioned, if we are looking at past performance, what we do—this perhaps answers the question directed to me—is analyse all our individual operations. As we are raising financing, for me an important sense check is: are we doing so consistently, on terms that feel right and are in tune with general prevailing market conditions? Does that give me the sense that the financing programme itself is robust and sustainable? Are there any signs within the market of stress as a result of what we are doing? For instance, if we feel we are issuing one particular maturity a little bit too much, and our primary dealers—the so-called gilt-edged market makers—tell us “We are having trouble absorbing this,” we must reflect on that, and try as necessary to feed that into the following year’s advice.
I think a good example from the past is that we have gradually shifted our issuance patterns a little bit away from just focusing purely on the long and very long end, to those areas of the market where demand is demonstrably strong or even stronger—that might be shorter down the curve. It is important that we are always issuing into strong demand, so we get the better price.
Q38 Mr Djanogly: Thank you. Mr Bowler, if we take the financial crisis or the pandemic response, what lessons can we take?
James Bowler: Well, it is worth saying that the DMO raised £485 billion in the pandemic; that was triple what it originally thought it was going to do in 2020-21. We talk about costs and that risk element—making sure that you have deep and liquid markets so that you are ready for shocks—really plays out.
Q39 Mr Djanogly: Have you been spending a lot of time evaluating that?
James Bowler: We have looked really closely at that. Having issued all of those things, there is then a refinancing that comes down the line. We are looking at all of that. It is worth putting this in context: this is because Governments around the world took the decisions they did to support the economy and ensure even worse issues did not come into play. But yes, we have looked at the lessons there. We may come to it: in autumn last year the DMO’s remit was increased and, even in stressed conditions, it was selling successfully into the market.
Sir Robert Stheeman: There is a constant dialogue with the primary dealers—these gilt-edged market makers.
Q41 Chair: Just to be clear for people who might not follow, these are the ones who will then sell on to the—
Sir Robert Stheeman: Exactly. We appoint them, based on a number of criteria, specifically to distribute the supply that we need to distribute to the entire investor base across the world and, importantly, to provide prices to that entire investor base—be those bid or offer, buying or selling, prices—at all times to encourage liquidity. That is vital. For us to do that and to know what they are doing, we speak to them very regularly to get a sense from them of what the market trends are and how investors are viewing gilts. That informs our views.
When it comes to the actual auction process, that is a slightly different thing. GEMMs have exclusive bidding rights at auction—we only sell directly to the GEMMs at auction—and in return we demand from them that they contribute to the market. But they submit their bids blindly; they are in direct competition with each other. We publish statistics after each auction, but not the names—we are the only ones who will know who the successful bidders are, apart from the individual successful bidder. We will publish the statistics around the auction. What was the highest price? What was the average price? What was the lowest price? There are lots of bids in each auction that are simply unsuccessful: they were not at the right price. Sometimes they were deliberately low because a bank might think it could pick up cheap gilts on the day, but the market did not allow it to, and the others bid a higher price. But it is a blind process—it should be a blind process.
Q42 Chair: Okay. So you do not feel that there is any conflict in terms of the value you are getting by having conversations with the market?
Sir Robert Stheeman: No, absolutely not, because it does not affect bidding behaviour. What we want, and what we need, is for the bidding process to be as competitive as possible. Ultimately, that is another way in which we try to secure value for money.
Q43 Sarah Olney: I had an interesting conversation with Mr Hughes about quantitative easing, and he told me that the cumulative net lifetime loss of the quantitative easing process is going to be £126 billion. I have written that down next to the extract from the OBR report accompanying the recent autumn statement, and I notice that the figure has been revised from £63.4 billion in July, when the OBR last did their assumptions, so it has doubled in that time. I obviously missed the opportunity to ask Mr Hughes about that, but perhaps I can ask you about it, Mr Bowler. Obviously, this is about different assumptions, and I accept that this is not a sudden massive increase in spend, but what are the factors that have such a big impact? What are the underlying assumptions, and how can they change so significantly?
James Bowler: Inflation and interest rate expectations are the answer. I agree with everything Richard said on QE and QT and with the numbers he used. The only other number to tell you is that the transfers to date, as we moved from quantitative easing to quantitative tightening in February 2022, is £38.2 billion.
To give this its full context, it is important to say that quantitative easing and quantitative tightening are part of monetary policy, so they are decided entirely and rightly by the Bank of England. The reason it did quantitative easing was that interest rates were very low, and it was trying to use another a method to reduce interest rates and stimulate the economy through the financial crisis, and then through covid.
As you go through quantitative tightening, it has the effect of increasing interest rates: there is more supply, so the price falls and yields go up. As the Monetary Policy Committee sets interest rates each month, it is looking at the interest rate—it sets the bank rate—and at what to do, periodically, in quantitative tightening. The committee looks at the two together. That is entirely independent from me and the Treasury, and it is right that we leave the committee to decide on that independently.
The difference between the March forecasts, the Budget and the autumn statement is that inflation and interest rates have gone up, so our forecast for debt interest has gone up quite a bit.
Q44 Sarah Olney: I am just going to interrupt, because it says here that the estimate was in July. We have had high inflation and high interest rates for some time, so they do not really explain the big increase between July and November.
James Bowler: I think they peaked in August.
Sarah Olney: Okay.
James Bowler: So there was quite a change between March and there. I do not have the page in front of me. The point I was going to make was that our debt interest—which is very much driven by market interest rates, which the Bank, to a certain extent, set—went up by £22 billion between the Budget and the autumn statement for ’23-24. You will see a similar impact on the overall lifetime cost of quantitative tightening.
Q45 Sarah Olney: I obviously accept that you are not responsible either for the Bank of England or the OBR, but if we assume that August was the peak, are we saying that we would expect that £126 billion to be lower in the future?
James Bowler: No. There are three numbers to have in mind. First, £124 billion is what was transferred to the Treasury in excess cash between 2012 and 2022, when, if you like, quantitative easing was making a profit. That changed from May 2022, when the Bank determined that it was no longer going to issue further gilts and would start tightening. Some £38 billion has moved from the Treasury to cover the costs of that to date. What the OBR is forecasting is the lifetime net cost of all of it; it is saying that it is £126 billion, and that is based on the OBR’s expectations of inflation and interest rates going forward. So that £126 billion is a forecast, and as Mr Hughes said—
Sarah Olney: All forecasts are wrong.
James Bowler: I was interested in his weather analogy. So that figure will change, but I cannot predict that it will be smaller or larger. That is the OBR’s central forecast.
Q46 Sarah Olney: Even though it differs so enormously from earlier in the year.
James Bowler: Yes, but that is because its central forecast for inflation has changed from earlier in the year.
Q47 Sarah Olney: Fine. Sir Robert, as we were discussing with Mr Hughes earlier, there will be a time when the quantitative tightening will put gilts on the market at the same time that you are trying to raise new money with an issuance of new gilts. What will be your—or, dare I say, your successor’s—approach to the risks of flooding the market, if I can put it that way, and the risks that that creates for Government borrowing? How will you manage that?
Sir Robert Stheeman: It is a very good question. I am glad that Richard Hughes phrased it in the way he did. I need to acknowledge that a situation where we and, in this case, the Bank of England—in other words, the authorities—are both sellers is a specific dynamic in the market, which, until effectively the beginning of this year or the end of last year, the market had to get used to.
It is important to understand that what applies to the whole gilt market and financial markets here applies to other Government bond markets around the world where you have QE and the potential for QT. What they focus on in particular is what we refer to as the net supply. For instance, in the various periods when QE was running, you had the Bank somewhere as a buyer in the market, and we were out there selling. There was a certain equilibrium. The market looks at net supply, and it saw that there was relatively little net supply.
This year, net supply is, by a large margin, at a historical high. That is a challenge for the market, on two levels: it is about absorbing all of that and about establishing the right price at which to take it down. That is not just about the supply; it is also—this plays rather nicely into James’s point—about both QE and QT being monetary policy. It is important to understand that the biggest single determinant of gilt yields is almost always—not always—monetary policy or, to be precise, expectations in the market around monetary policy. One of the reasons why yields have risen this year is not just because of the increased net supply, but because this is all seen as part of the tightening of monetary conditions in general, which goes hand in hand with what the Bank is doing elsewhere in monetary policy.
The challenge specifically for us is to make sure that our operations continue to work as much as possible and as effectively as possible, so we talk to the Bank very explicitly about operational co-ordination in terms of our sales. We have been doing that since the very inception. Actually, we spoke to the Bank years ago about this to make sure that, when the time came, there would be as seamless an operational flow between us as possible.
Just to be clear, we do not talk to the Bank about monetary policy. The people we talk to very often do not even know what the decisions of the MPC are going to be. We do talk about how they see market conditions and how we see market conditions, and we talk about scheduling particular operations. For instance, most of our gilt auctions—not all, but most—happen on Tuesdays and Wednesdays. As far as possible, the Bank will avoid those days in a week or, if it cannot—because it also has to be active—it will choose operations that are very different from ours. If we are selling, for instance, a short-dated gilt on a Tuesday, the Bank will not be selling a short-dated gilt on a Tuesday; it will choose a different maturity.
Q48 Sarah Olney: So there is no direct competition on any given date?
Sir Robert Stheeman: We try to avoid it. It could be difficult always to rule it out in the future, depending on how much the Bank might choose to sell. In all of this, though, it is important to understand that we, as the Government issuer, are first mover. Once we have decided what our auction calendar looks like, we discuss that with the Bank so that it can factor it into its conditions subsequently—it can react to that.
James Bowler: I was just going to build on the point about evaluation. The Bank very much looks at its impact. The September minutes of the MPC concluded that the Bank’s selling into the market was not disrupting the market to date, and I think that informed its future decisions. They and we are not just doing this blind.
Q49 Sarah Olney: We can see from figure 11 of the Report that the redemption profile of the DMO shows that redemptions are peaking in the next financial year and the one after it. What challenges does that represent, Sir Robert? I assume that a lot of this is related to the covid borrowing, so obviously that is the point where that is now falling.
Sir Robert Stheeman: It is, and it does present challenges. I referred earlier to the cash management function within the DMO. It presents challenges for the cash management function, because what we need to do is to make sure that, on any redemption date, sufficient funds are there to repay a bond. The money markets, while very large in volume, do not have limitless liquidity. Indeed, I would argue, and I think so would many others, that the UK sterling money markets are less liquid post financial crisis than they were beforehand. So what we have to do in terms of meeting those redemptions is to think how, given the constraints under which we operate, we can best manage cash flows in such a way that we know that we can try to meet those redemptions effectively and with a very high degree of certainty. That is probably the single biggest thing.
We can also do other things. For instance, close to a redemption, we can schedule an auction. So, yes, we are paying back on the one hand, but we are also raising new funds on the other. We do quite a lot of that, and it is important to note that it is the nature of the market that, when a gilt redeems, it is not as if investors leave the market at that point; they want to buy something else. This is, again, where we have to factor into our decision making what is the most suitable issuance programme, given what is happening in terms of net flows into the market.
Actually, we have a very, very big focus on redemptions next year—I happily admit that. The market knows that, and we have to be very careful that they do not game that knowledge in some way. We are already conducing certain transactions, as you would expect, to ensure that those cash flows and those redemptions are all smoothly managed, but the scale of them has increased hugely, so I would say that it has become more challenging than it used to be.
Q50 Chair: This must be the same internationally, because there will be other Governments that will have done similar short-term gilts as a result of covid, so is the market going to be flooded?
Sir Robert Stheeman: I would say that, internationally, it is a much bigger problem than it is here, not least because of the very long average maturity that we have in our debt portfolio. As you are probably aware, it is over 14 years.
Q51 Chair: So that balances out the short-term problems over the next couple of years.
Sir Robert Stheeman: It does in general. What it means is that, yes, we have a lot of redemptions to manage but, compared with our international peers, it much less. We will obviously have to see what next year’s financing requirement is, but redemptions in some major European countries, depending on what the fiscal situation might be, can often be 40%, 50% or 60% of what the overall annual borrowing programme actually is.
Q52 Chair: And ours is?
Sir Robert Stheeman: It depends. It is creeping up. I would need to check the exact numbers, which I do not have in my head.
Chair: You can send that, but what is the ballpark figure?
Sir Robert Stheeman: Less than that.
Chair: Let’s speak to the permanent secretary.
Sir Robert Stheeman: But it is larger than it was. That is an inevitable consequence of how much we have had to borrow. Again, it goes to the point about the short term and the long term that I made earlier. We try to ensure that we are making the right trade-offs in terms of what we issue. If we were to borrow everything short, which is what other countries do to some extent, we would have a much greater challenge. With the limits on sterling money markets that I described, that is not a challenge I would particularly like to have.
Chair: Okay, thank you. That is very helpful.
Q53 Peter Grant: Sir Robert, we have spoken about the redemption peak that is coming up over the next couple of years. How much of that is down to short-term borrowing to cover the cost of covid, and how much of it was going to happen anyway?
Sir Robert Stheeman: I do not know exactly what the figure is, because we get an overall funding need for the year from the Treasury, but when you talk about redemption specifically and how much we have to refinance, I would have to look at the exact profile of the bonds we were issuing. Given that the high borrowing during the covid period was in 2020-21, that suggests that we are already starting to have to redeem some of the very short-dated gilts we were issuing—the three-year ones—now and especially next year. We are also having to redeem those gilts that we issued much earlier—for instance, in the financial crisis. Actually, gilt sizes—the size of an individual gilt—have increased markedly over the last 15 years or so. Previously, £20 billion or £25 billion was the biggest size; now, an individual line of gilts can be as much as nearly £40 billion. These are very, very large flows that we are having to manage, but it is all part of the redemption management.
James Bowler: Perhaps the best way to answer your question, if I may—
Chair: It is just worth pointing to page 7, paragraph 9, which gives some summary of these figures.
James Bowler: What were we going to borrow anyway, and what is the impact of covid? The financing requirement for the DMO tripled to £485 billion due to covid, so you can see the difference from covid.
Sir Robert Stheeman: But some of that we would have borrowed not just short term; we would have borrowed that in five years, in 10 years, in 30 years and even in as much as 50 years.
Q54 Peter Grant: Looking at the chart that the NAO has included in the Report—I realise that these things change every day, so the numbers will be a bit out of date—it looks as if, compared with 2023-24, there is about an extra £21 billion or £22 billion in 2024-25 that would have to be redeemed. In the following year, there is about £15 billion on top of that. I do not know if this is a question for you first, Mr Bowler, but is it possible to estimate what the bottom-line impact of that is on the Government’s revenue budget? When the Chancellor sets out to create the Budget for 2024-25, how much of a hole do they have to fill simply because of the increase in debt redemption and refinancing?
James Bowler: Probably the best way of answering that—I am not entirely sure exactly which chart we are looking at—is to look at debt interest payments that the Government makes. Mr Hughes talked about this. Debt interest payments are the amount of debt that the Government as a whole must pay on all of its debt, however it finances it. Debt interest rates rise, particularly when inflation and interest rates rise. We talked about QE earlier. On debt interest and the Budget for this year, 2023-24, the autumn statement said that debt interest this year will be £116 billion; in March this year, that was £22 billion lower. So that is a measure of how much it costs. Debt interest is not a Department, but people often make the example that, were it to be a Department, it would be the second biggest Department out there. I do not disagree with Mr Hughes that the more debt interest you pay, the more constrained the Government is on its fiscal policy going forward.
Q55 Peter Grant: The other event we saw recently that caused quite a bit of jitters in the financial markets was the plan for growth, or mini-Budget, in September 2022. What impact did that ministerial statement have on the cost of meeting the Government’s financing needs?
James Bowler: In September 2022, the plan for growth/mini-Budget had the impact that the net financing requirement for the DMO rose by £72 billion. I think that is the answer to your question. Obviously what happened next was that the autumn statement later reversed a number of the decisions—the majority of the decisions—and that then reduced the net financing requirement, so it was lower again.
Q56 Peter Grant: Has that caused any longer-term impact on the market’s view of the United Kingdom?
James Bowler: The markets were under a period of stress. You saw that by the prices and yields relative to other countries changing. Yields were higher in the UK than in other countries. We talk about deeper liquid markets and being able to be ready for risks and shocks, and the Debt Management Office was continuing to issue auctions at various yields. Then LDI stress emerged in the market too, which came out of that. I can happily talk about that if you want me to.
Q57 Peter Grant: Clearly, it is up to the Chancellor of the day to decide when they are going to make a Budget announcement and what is going to be in it. Is there more we can do to make sure that if there are ministerial announcements that might cause concern in the markets, that concern does not have the impact on public finances that it did last time?
James Bowler: Yes, certainly. It is widely deemed to be the case that you would not want to go ahead with very large fiscal events without them being accompanied by an OBR forecast to put them into context unless you are in extreme circumstances. Equally, we have updated the memorandum of understanding between the Treasury and the OBR. If we are changing the remit to the DMO in particular, we would want the OBR, even outside of a complete fiscal event, to assure us on the central Government net cash requirement we have; we have updated our memorandum of understanding for what happens outside fiscal events, if you like.
Q58 Peter Grant: Realistically, can Parliament actually prevent a Minister from making a Budget statement without an OBR assessment?
James Bowler: Can they prevent—
Chair: Can they prevent any Minister saying what they want?
James Bowler: That’s for you guys.
Q59 Chair: I wanted to come to you, Mr Harkins, because you are a smaller player in the borrowing market, but it is very interesting because you are facing the retail side. You are in an open competition with all the normal lenders. Your profile of savers is a higher age range. We noticed this and wondered what you are doing to try to encourage younger people to save with NS&I?
Dax Harkins: It is a good point. Like all financial institutions, we need a pipeline of new customers. In our new strategy in 2018, we identified this as a key theme. Since then we have taken a number of actions to try to address it. We launched a new junior ISA. We changed and made more flexibility in the rules on who can buy gifts for children with premium bonds.
Q60 Chair: Can you explain that last one—are you trying to make younger people buy these rather than grandparents?
Dax Harkins: Previously it could only be grandparents and parents who buy a gift for a child. Now we can have godparents and aunties and uncles and so on, and that has been very popular. To coincide with that, we reduced the minimum investment in premium bonds from £50 to £25, and we have run many gifting campaigns and other campaigns to engage with younger people. Really positively, over that period we reduced the average age from 57 to 49. We will look to carry on that in the next strategy.
Q61 Chair: What is the profile of savers in each age range? That is the average age, but how much have you increased the younger decile?
Dax Harkins: Generally, there will be older savers within there. This will have brought savers in at the much younger end, so the children have to go through there.
Q62 Chair: So you haven’t got figures for how many younger people you have encouraged?
Dax Harkins: I don’t have the age breakdown on me, but those campaigns are all geared towards children and much younger investors, and that would have had the impact on the balance there.
Q63 Chair: Do you think you have enough flexibility to increase and decrease your savings rates?
Dax Harkins: The short answer is yes. I would point to this year—a really dynamic interest rate environment, lots of base rate changes. We have been really responsive in reacting to them to increase our rates over that period. We have gotten lots of good media coverage and from our customers. In a really challenging market, we are now on track to achieve our net financing target of £7.5 billion plus or minus £3 billion.
Q64 Chair: You had this particular bond you issued that had a 6.2% interest rate. Just to be clear, because people following this might not realise that you are not capped in the same way that private banks are on the guarantee, you have an uncapped guarantee for any saver, which is pretty attractive and different from the rest of the market. That 6.2% bond was over-subscribed; you had to close it. Do you know the profile of people who took that out? Because it was quite a good deal.
Dax Harkins: I would look at it in a slightly different way. Generally, a fixed-term product will go on the market and be taken off sale after a period. Generally, over this past year, it has been for a shorter duration than the five weeks we had it on the market. We had it on for a long period, and we had 225,000 customers who subscribed to invest, so I think it was a good length of time for customers to invest. The key thing is we needed that because, despite all those interest rate rises over the previous year, we had seen unplanned defunding for several months over the summer because of the various pressures in the marketplace.
Chair: Presumably, people cashing in their premium bonds in order to get it.
Dax Harkins: Yes, repayments particularly for people using it for living expenses. We also saw a shift to fixed in the market. The good news is that that did what we needed it to, and we are now on track to hit the net financing target.
Chair: Okay, so you really use it to rebalance the portfolio.
Dax Harkins: Yes, in a very dynamic marketplace.
Q65 Chair: When you decide on that interest rate, how do you work out what is going on in the rest of the market, because you could really destabilise the rest of the market if you are not careful? What mechanisms do you put in place?
Dax Harkins: That is a key point. We work within an operating framework that balances those perspectives—the perspective of our savers and customers, the perspective of taxpayers on interest cuts, and the stability of the wider financial marketplace—so it is a key consideration within that. We took that into account throughout this. A key mechanism in balancing those is real transparency on our net financing target. We talked about the £7.5 billion plus or minus £3 billion through the year. We report quarterly on where that is going, so the market can really plan around it. As I say, even after the one-year bond, we are still firmly within the remit of that target.
Chair: You had that big remit increase during covid, which was—
James Bowler: £6 billion to £35 billion.
Q66 Chair: Yes, it was really huge, and you have made £24 billion, which is pretty heroic in a year, but also potentially very destabilising for the markets. What did you go through then to do that? That was at rapid pace with a big increase in remit, even though you did not hit the full remit. Did you get any kickback from the private sector?
Dax Harkins: I think we would all accept that they were very atypical circumstances with very unusual requirements at that point in time. We stepped in to help the Treasury with some of its funding needs—
Q67 Chair: No one is doubting that. We can see exactly why the Treasury—You are part of the borrowing requirement; you needed to do it. There was the furlough and all those things to pay for, but it is really about what the rest of the market was thinking about this. We have seen in other areas—for instance, some of the loans during covid—that, because there is cheap borrowing effectively, there is a risk of destabilising the lending market. How do you ensure you are not doing that in the retail sector?
Dax Harkins: In normal circumstances, we sit and work through all those logistics. We look at market share and competitors’ rates.
Chair: But you had to do this at pace.
Dax Harkins: These were not normal circumstances; that would be my observation. The way it played out was that, over that period, all our competitors were reducing their interest rates. We did propose reducing it if we wanted to remain within our standard remit. We made a decision with the Treasury that it would be beneficial if we held them. In hindsight, the inflows that we generated and the pressure we came under was really a challenge for us. Once we got to the end of that period, and we had a clearer view of how the wholesale markets were working and how we might manage that, we reduced our rate, which caused an equal outflow at that stage in time. It was a really challenging period for us.
Q68 Chair: What would you do differently if you had to give yourself or a successor advice on what to do in a pandemic and that sort of situation?
Dax Harkins: We have all learned many valuable lessons from all that. The first is that we need to work through the demand implications and impacts, and we can make some simple choices. If you choose to fund large sums like that through fixed term, you have far more stability and you don’t get the outflows when you have to change your rate, so you can manage it that way. We would have a much more nuanced discussion if we were to repeat that, but—touch wood—I hope we won’t.
Q69 Chair: One hopes we won’t be here when the next pandemic happens, so how are you going to make sure those lessons are retained inside the Treasury, Mr Harkins and Mr Bowler?
Dax Harkins: I will go first. One of the key lessons that we have built into our Rainbow programme is that we still have a dependency on legacy. When you get large inflows and transactions, it can switch, so it can depend on people, paper and physical locations. The transformation and the new operating model will give us far more scalability and resilience through true digital processes so that people don’t fall out. There will be slicker ways of assisting people through them. That will absolutely give us more scalability.
Q70 Chair: So you will be able to deal with the influx.
Dax Harkins: Yes.
Chair: Okay. We will come to the Rainbow project later. Mr Bowler?
James Bowler: I would make two observations. In terms of the quantum of savings, of course there was a huge net saving during covid. A lot of people were unable to spend as much money as they normally would on going out and doing things, so they saved a lot more. A lot of that flowed in, and that helped the financing.
Then there is the operational side. It was very difficult for NS&I to deal with that customer demand, not least with having to work through covid with its own staff as well. The changes we are making to NS&I should make that easier and, it its worth saying, more attractive to younger people because—to go back to your earlier point—they want an app and something digital.
Q71 Peter Grant: Given that we have been discussing NS&I, I should say that my investment with NS&I as a proportion of my loaned assets is probably significant enough to be placed on the record. Sadly, it is not very significant in the context of the numbers we were talking about earlier.
Sir Robert, we previously had a briefing from the National Audit Office about the actual process that is gone through every time you hold an auction for gilts. How do you make sure buyers aren’t co-operating behind the scenes in order to manipulate that process and prevent you from getting the best deal?
Sir Robert Stheeman: It is an important question, and I would note a couple of things. As I indicated earlier, we insist on the blind nature of bids that are submitted in an auction. It is essential that that happens in order to avoid any collusion. We have to rely on the fact that all our primary dealers are regulated organisations, and that manipulation and collusion are illegal and against the law. We make it absolutely clear that we expect our primary dealers to adhere 100% to all regulation and the applicable laws.
As I say, we are not the regulator ourselves, so we cannot actually police that, but the way we design our issuance programme and auction processes deliberately minimises the chance of manipulation or collusion. That doesn’t mean that I can tell you categorically that it has never happened in the past or that it might not happen in the future. I can’t say that because, for any crime, who knows what people will ultimately do?
We take clear steps. When we hold our auctions, we always make sure we hold them at the time of the day when the market is, we hope, at its most active—busiest, deepest and liquid, with a lot of trading going on. All these things arguably minimise the chance that collusion or manipulation might take place. The blind bidding process and all the other things are critical to that. But can we rule it out? It is always an issue.
The one thing that I should perhaps also say is that this is a very, very deep and liquid market; most Government bond markets are. The deeper a market, the more trading goes on, the more liquidity there is in the market, the more difficult it is for a single player to start manipulating it. It doesn’t make it impossible; it makes it that much more difficult.
Q72 Peter Grant: Earlier this year, the Competition and Markets Authority issued a provisional finding against five very major banks—Citibank, Deutsche Bank, HSBC, Morgan Stanley and Royal Bank of Canada—all of which had been exchanging sensitive information around gilt auctions and other activities. First of all, how long did that take to come to light? How long had it been going on before it was detected?
Sir Robert Stheeman: I believe that the period that the CMA is investigating dates back 15 years at its farthest point. I think they are looking at a period in 2008—
James Bowler: 2009 to 2013.
Sir Robert Stheeman: Literally nearly 15 years ago. That is something where they have asked us to supply them with certain information to help with their investigation, already long before the investigation was made public. But I don’t think I should comment on any kind of ongoing investigation. You should really probably put that question to them.
I would say one thing. Again, I cannot rule out the possibility that bad behaviour exists and might even persist, but we are talking about a period in the past. Anyone who is familiar with the financial market system—for all its flaws, I would suggest that the level of compliance and the focus on compliance issues that exist in wholesale financial markets today is a lot greater than it used to be.
Q73 Peter Grant: Thank you. I am not going to ask you when you first knew about this investigation, because obviously you would have known about it before it was made public, but since you became aware of the concerns that were being raised, without going into details, has that indicated to you that you needed to make any changes to your own processes, to prevent it from happening again?
Sir Robert Stheeman: Good question. We have considered that very carefully: “Is there something that we should be doing differently in the way we hold our auctions and organise them?” So far, we have not changed anything.
I do not want to preclude the outcome of whatever this investigation might ultimately reveal, but it is not obvious to us that, if you like, there is a fundamental flaw in the system, so nothing at this point has actually been changed.
I should say that we remind very, very frequently all our gilt-edged market makers of their obligations to observe relevant legislation and the regulations at all times. They know how important that is. There are not just huge financial risks for them were that not to be the case but huge reputational risks as well.
Q74 Peter Grant: Mr Bowler, we heard earlier on that about 25% of Government debt is held overseas. How much of that do you know where it actually is?
James Bowler: That is the case. I think we are second in terms of the most overseas thing in the G7 to France. Mr Hughes set out the desire to know more about where it is located. There are limits—I guess legal and commercial limits—at the moment as to being able to tell exactly where those are. It’s a traded commodity and you tend to know—as I understand it; Sir Robert might add more—which institutions do it. But obviously these things can get sold on and sold on and sold on, so it’s difficult. I can’t say that we have a very accurate picture of that.
Sir Robert Stheeman: Richard Hughes made a couple of interesting points about foreign ownership. He noted that it has increased, which obviously it has, but I would also point out that that increase took place primarily from the period when, at its low point, it was at about 15% or 16%—I think that was the figure he cited. That was the case when I started. By 2009-10, there was already 25% foreign ownership, according to statistics from the Office for National Statistics. It has stayed approximately at that level since then—a bit higher, a bit lower. Interestingly, it has not increased specifically in the aftermath of the financial crisis or, especially, during the pandemic period.
On foreign ownership, I hope he will forgive me, but where I would potentially take issue with Richard Hughes is that he referred to foreign ownership as likely to be the most volatile element of the investor base. This is my personal opinion: I am not convinced that that is the case. Foreign ownership needs to be seen in the context that sterling has become, over the last 20 years, much more of a reserve currency compared with what it once was, albeit on a limited scale. The IMF publishes the COFER statistics, which now say that that sterling, as a currency, accounts for about 5% to 6% of global reserves. That is up from about 3% when I started.
I would suggest that an overseas central bank, a sovereign wealth fund or a life insurance company located in Asia is not necessarily—to use his word—flighty. It is not necessarily more volatile than any other part of the investor base. We always assumed, right up until the aftermath of the so-called mini-Budget last year, that it was the domestic pension industry that was the ultimate buy-and-hold part of the investor base, and that had held true. It was the case but, as we also all know, that proved to be arguably almost the more volatile portion of the investor base at that point.
Richard also said, quite correctly, that it is part of a diverse investor base. This is where we need this balance. We need those foreign investors to be interested in our market and to participate because it acts as a safety valve, to use his words.
Q75 Peter Grant: What can you say to reassure this Committee and other MPs that that debt is not somehow finding its way into the hands of people we really would rather not be in debt to?
Sir Robert Stheeman: Ultimately we cannot know, unless we are told explicitly and it is revealed to us, who the ultimate beneficial owner of any gilt is, but then no country can.
Think about a gilt that we might auction on a given day. We might receive a bid for £100 million from a single bank. If that bank sells it on—the same day, the next day, the following week or whenever—to another market participant and the gilt is sold on and on, we will not necessarily know, if these gilts are held in nominee accounts, who the ultimate beneficial holder is. However, I think it is also important to emphasise that they are the ones giving us money; we are not giving them money. They are investing in us. That does not mean that they are necessarily an investor that you might deem acceptable, but it is not as if we are somehow beholden to that particular buyer.
Peter Grant: Mr Bowler, do you have anything to add?
James Bowler: These are traded commodities, so there is a free market.
Q76 Peter Grant: Is that kind of activity covered by international sanctions? If someone acting on behalf of Vladimir Putin tried to buy UK Government debt, would the sanctions regime say to the person selling it to them, “You are not allowed to sell it to them”?
Sir Robert Stheeman: Yes. I should say that the sanctions regime was applied directly following the invasion of Ukraine last year. Ultimately, as I say, we do not sell directly, but I know for a fact that the registrar and particularly the clearing houses looked very closely at the whole question of adherence to sanctions. They are very well aware of what that means and of the risks involved if they, for instance, found that payments were going in some form or another to a foreign regime. So that is being applied, but it is further down the food chain, so to speak. It is away from us.
Q77 Peter Grant: Just as a matter of accuracy, the invasion of Ukraine happened in 2014.
Sir Robert Stheeman: But the sanctions regime was applied and tightened in particular last year.
Q78 Mr Djanogly: I want to go back to the production of the analysis of Government borrowing needs by the OBR, the DMO and NS&I. Mr Bowler, how does HMT challenge all the reports that you get?
James Bowler: How do we challenge where the DMO might plan to make its issuances?
Mr Djanogly: If you have three reports coming in, how do you deal with them?
James Bowler: We have an excellent debt team in the Treasury. It sits within our economic and fiscal team, with sits under our second permanent secretary. Its role is to be the gatekeeper for both the agencies here, and it works with them both to stress-test and challenge appropriately in three ways. This particularly applies to the DMO, but also to the service delivery measures of the NS&I. It discusses and agrees a remit at the start of the year; it looks at how that has gone during the year, and how much the net financing is, or the auctions that are being undertaken; and it looks at the debt management report and the end of the year, from which Sir Robert was reading out the performance criteria. It is about stress-testing what has happened and what we might want to change. Is that a helpful answer?
Q79 Mr Djanogly: Yes. Do you review that process?
James Bowler: Yes.
Q80 Mr Djanogly: Regularly?
James Bowler: The three moments in the year are under constant review, and then we have more generic framework reviews tailored to both organisations less regularly—we had one in 2021 and one in 2020—about whether our relationships and the governance of each agency are correct.
Q81 Mr Djanogly: I put this question to the OBR, so I will put it to all of you: do you feel that the process is working well, or are there ways in which it could be improved?
James Bowler: I think it is working really well. I am always keen for us to learn and look at peer organisations around the world. At the moment, because I think it is working well, my focus is more on operations. It is on sensible succession in the DMO and the operational change that we are delivering in NS&I.
Sir Robert Stheeman: I underline that the DMO was set up for a reason, namely as a specialist organisation; I have used that term a lot. Our staff are specialist and they have very specific skills, which are probably not normally found in the Treasury, but that is why we sit where we do and the Treasury sits where it does. That expertise is—I hope I can say this—very valuable to the Treasury. It is also a reflection on the quality of the staff that we have. Genuinely—I would say this, wouldn’t I?—I think they are world class. I like to think that Treasury colleagues know that. We talk to our colleagues in the Treasury about some of these matters literally daily. We are in constant dialogue, not just in the run-up to fiscal events, but about all sorts of things. We are a repository of specialist skills that is there for the Treasury to use.
Mr Djanogly: Thank you. Mr Harkins, have you anything to add to that?
Dax Harkins: James has gone through all the pieces. The only other element I would add is all our pricing changes. We work with the debt management team, going through all the indicators that we have discussed, and then they are approved by the Minister as well. It is an ongoing, iterative process, with full visibility.
Q82 Chair: I want to turn to you, Mr Harkins, and ask about the Rainbow programme. This is to move from having Atos, one supplier, to multiple suppliers, so that the issue when you have a sudden surge of savers is better handled. But you have already had to extend the Atos contract, which was due to expire next year, for at least another year. Can you tell us a bit more about why that is, and how much longer you could extend the Atos contract for if there continues to be a problem with delivering the Rainbow programme?
Dax Harkins: The first part of that question is about the main driver of the need to extend. One of the procurements for one of the elements of that multi-source was on the digital experience layer. We ran that procurement. We had a good competition; we had a good number of bidders of the right type of quality coming through. Really surprisingly, they all submitted non-compliant bids. That is highly unusual and, I think, reflects the different risk appetite in the market that we saw over the period we ran it. We did a full “lessons learned” on that. We made a number of changes. We changed the procurement method. We used a framework. We changed some of the requirements around security; we kept it secure, but we give a bit more flexibility—
Q83 Chair: The bidders submitted bids that they obviously didn’t know were non-compliant, so something was wrong. What happened?
Dax Harkins: Well, I would argue the case that they submitted bids that did not meet all our requirements. That is highly unusual—really unusual.
Q84 Chair: You didn’t see it coming?
Dax Harkins: We did not see it coming.
Q85 Chair: So what have the lessons learned showed you about that?
Dax Harkins: There were a number of key changes we made with the lessons there. We changed the procurement mechanism. We used a framework that was a bit more definitive in how they would reply, so it gives a bit of a narrower band. We got a bit more flexibility in some of the requirements that caused the challenge. Some are security pieces. Some of the big vendors had had some challenges with those, so we give more flexibility. The positive piece is that we have now completed that procurement. We were able to announce yesterday that IBM was the successful supplier.
Chair: Yes, we had the press release. Thank you.
Dax Harkins: It’s good that we learned the lessons there. We have now completed all the procurements; we have announced all our suppliers, and we start to move to delivery.
Q86 Chair: What’s the timeframe now? You have already extended the Atos contract. When will that have to stay until?
Dax Harkins: We have extended for 12 months. It goes to March ’25. Each of the bidders has developed their plan in isolation, so at the moment we have a loose plan. We are going through the process of creating an integrated plan. There will be some movement as dependencies go through. At the moment, it is intended for us to be able to deliver all that next year, ’24-25. However, we do have the contingency of an additional 12 months should we need it. That can be scaled down to just being parts of the service that we may need as a contingency if we run into any issues.
Q87 Chair: There is no cliff edge for Atos?
Dax Harkins: There is no cliff edge.
Q88 Chair: So you’re going to ease this in. What will be the difference for consumers? If I were trying to buy premium bonds or a savings product from NS&I under the new system, what would I see that was different?
Dax Harkins: There are many changes that we will look to do here. We have talked about the resilience and scalability. That means that when we get big inflows, you will still get the same quality of service to go through, rather than some of the pinch points that we have had. There will be new mobile apps, with more functionality and flexibility. We will have much more control as to how we can change and flex that and give customers different journeys. We will be rebuilding all the digital journeys from the ground up. This will help customers to stay in, which is a challenge now—they often fall out into paper.
Q89 Chair: So it’s a bit like other banks, basically. You are coming into the modern world, and might attract younger savers.
Dax Harkins: Yes, that will be a key benefit also.
Q90 Chair: You sound very confident. It’s easy to sound confident in front of the Public Accounts Committee, but bear it in mind that we will hold you to those dates. You’re not a regular witness, so perhaps you don't appreciate that, but—
Dax Harkins: I am really pleased that we have completed all the procurements. This is a huge milestone. It has been several years in the making to get through there. I don’t doubt this is a big, complex programme. There will be moving parts; there will be challenges, but I am confident we have contingencies to manage any of those challenges as they arise.
James Bowler: You have obviously seen similar IT-based issues around Whitehall. Another part of the solution here is that we are upskilling and upping the size of the NS&I workforce quite considerably, not least because there has been a reliance on consultants, with the cost and short-termism that that creates. We have had Atos as the incumbent; going to a multi-source requires NS&I to be a much better IT customer.
Q91 Chair: Have you been working with the procurement profession in Whitehall?
Dax Harkins: Yes. We are engaged with various complex transactions, because there are big, complex transactions there and we have a number of people in the Cabinet Office supporting us right now.
Q92 Chair: I have one general question and one specifically about your departure, Sir Robert. You are going, but you have a team that is not, dare I say it, in the first flush of youth. What is the risk of the rest of the team also retiring in the next few years? How long will your successor have to work with those who are experienced in order to embed that experience?
Sir Robert Stheeman: I cannot honestly speak for individuals, and I should say that one of them is in the room, so I have to be very polite. It is true, however. It is the flip side of experience—an institutional memory—which is incredibly important.
There is inevitably a risk. There are limits to what we can do, but we are doing what we can in recruitment and talent building, not just in the level immediately below me but throughout the organisation. To some extent, my decision to leave in the immediate future was also taken with that in mind, in the sense that the four of us in our small executive committee—the top team—are all slowly getting on, but I am getting on more than anyone else. It is probably the best thing for the organisation that I go, so that, ideally, they will be around for a little bit longer to help pass on that institutional memory and foster talent and growth further down the organisation.
Q93 Chair: So you are not going idly; you are going with a very clear plan, from what you are telling us.
Sir Robert Stheeman: Yes, and by announcing my departure well in advance, I hope we are doing what we can to make the entire process as smooth as possible, with an eye to the future and effective handover. I will mention something else with regard to my position and effective handover. When I joined, I was offered the ability to talk with my predecessor. I am now offering the same to any candidates who are there, because it makes the whole thing better informed.
Q94 Chair: Absolutely. Having done some of those appointments, I completely appreciate that.
I have a final, big question. We are looking at the cost to the taxpayer of managing the whole process of borrowing. Borrowing has to happen, for the reasons we know—I suppose there might be a world where it doesn’t, but I can’t envisage that at the moment. What is the biggest driver of the future taxpayer cost of Government borrowing, Mr Bowler?
James Bowler: Because we have looked quite a lot at the next five years, and we have talked about debt interest and all the rest, let me link up with what Richard Hughes said. The biggest long-term driver will probably be demography and the impact it has on healthcare and social care, which is our biggest spending item.
I commend the OBR’s fiscal risk report to everyone—I think he had it in his hand—which looks right out to 50 years from now. It shows that, if you carried on with the exact same policies we have, the borrowing and debt situation becomes pretty unsustainable. I think a big challenge for future Governments and future officials is going to be plotting a path that looks at not only changes to demography, but things such as the move to net zero. We have a tax system that gets loads of taxes from fossil fuels at the moment, and that will rightly go.
Q95 Chair: So you are saying that a lot of it is very policy-driven, which is fair enough.
James Bowler: Absolutely. There are global trends in things such as demography.
Q96 Chair: Sir Robert, from your perspective, having run the bigger chunk of Government borrowing for the last 20 years, what do you think is the biggest driver of future taxpayer cost? How would you minimise it?
Sir Robert Stheeman: I try to split these things up into things that you can influence and things that you cannot influence. With the latter, clearly the biggest driver of Government borrowing—I indicated this earlier—is around expectations of interest rates. With debt sustainability, the big question is clearly growth, but that is not for me.
With what we do and how we do that, it is all about ensuring that we have effective, deep, liquid markets and that our market stays attractive to investors, wherever they might be—across the globe as well, if necessary. It is about ensuring that the market is one that they want to participate in, that functions well and that is healthy. Some of that is within our gift; some of that is not within our gift. But for me, that will be the biggest single way in which we can ensure that costs are at least minimised over the long term.
Chair: Thank you very much indeed, and thank you all for your time. It has been very interesting for us as a Committee. We do not delve into this area that often. We work with our sister Committees, but often this would be seen as more for the Treasury Committee. We are obviously not discussing the policy, though we touched on the challenges there at the end.
The uncorrected transcript of this session will be available on the website in the next couple of days, and we will be producing a report on this in the new year. We wish you all the best with project Rainbow—particularly you, Sir Robert, as you move towards retirement next year after 20 years of service to the taxpayer.