Examination of witnesses
Elizabeth Truss MP, Sir Robert Stheeman, and Steve Farrington.
Q32 The Chairman: Chief Secretary, Sir Robert, Mr Farrington, welcome to the Economic Affairs Committee. As you know, we have just started what I hope will be a short‑ish inquiry. Perhaps I could begin by asking the first question, which is this: does the Treasury agree with the chair of the UK Statistics Authority that RPI is an inadequate measure of inflation?
Elizabeth Truss: We certainly agree that it is not the preferred measure of inflation. CPI is a much better measure of inflation, but RPI continues to have its uses and is widely used in a number of private and public spheres. While we agree that it is not the preferred method, and we are seeking to move away from RPI, we nevertheless think it continues to have uses.
The Chairman: Why would you want to use a measure of inflation that the statistics authority, which is under a statutory obligation to provide proper statistics, believes is inadequate?
Elizabeth Truss: RPI is baked into quite a lot of contracts and things such as pension arrangements and index‑linked gilts. People have invested in index‑linked gilts or their pensions on the understanding that they would be uprated by RPI. They have entered into those arrangements with knowledge of the difference between RPI and CPI.
While ideally we want to move away from RPI towards CPI, as we have done, for example, in business rates very recently, we do not want the perfect to be the enemy of the good. Sometimes, when Governments conduct tidying‑up exercises, they can find there are unintended consequences of doing that for people who have RPI‑linked investments. There could be problematic consequences. Of course, there could be difficulties for the public finances, where we have various items of income that are RPI‑linked. It is not a cost‑free choice to move away from it.
The Chairman: Surely this is not about cost; this is about principle. When people enter into contracts that protect them against inflation, they are entitled to expect a measure of inflation that is correct.
Elizabeth Truss: There are different ways of measuring inflation. There are debateable issues. For example, RPI contains a measure of housing cost whereas CPI does not. We would agree with the statistics authority that the gold standard is CPI, but I would certainly argue that people have invested in index‑linked gilts, for example, with the understanding that those would be uprated by RPI. That was priced into their decisions when they made those decisions. RPI is a measure that is still produced by the statistics authority. I agree it has lost its gold-standard status, but it is nevertheless a measure that people entering into those contracts or paying those prices understand.
Q33 Lord Sharkey: Several of our witnesses have said that there is at least a case for adjusting RPI to correct for the 2010 changes to the clothing price collection. But Sir David Norgrove told us that the Chancellor would refuse a request to make improvements to the way RPI is calculated. Is that the case? If it is, why?
Elizabeth Truss: It is a decision for the ONS to make in the first instance, and it ultimately has to put that decision to the Bank of England, to see if it has significant impact. Then it would be a matter for the Chancellor. I cannot prejudge what the Chancellor would decide in any given instance. The issues I have just been talking about would come into consideration—the potential fiscal cost or money that would have to be found elsewhere. There is also the impact on private individuals’ pensions, for example.
Lord Sharkey: Is there a case for adjusting RPI to correct for what is known to be a failing in terms of the 2010 change?
Elizabeth Truss: The statistics authority should be constantly reviewing how it calculates RPI, making sure it does that in the best possible way and, as the process is every year, checking that with the Bank of England. If it is such a significant change that it is essentially making it CPI, that would clearly be problematic because RPI itself is laid out in primary legislation. But, fundamentally, the statistics authority has the power to do that; it has the power to look at it.
Lord Sharkey: It has the power to make a recommendation on it. One of our witnesses told us that he did not believe the correction to clothing price collection would trigger any of the concerns of the Bank of England or the Treasury. Is that right?
Elizabeth Truss: It is difficult for me to prejudge it before those methodology changes have been put forward. It would need to go through the process of the Bank of England in the first instance. I would suggest that a change completely to the level of CPI or to indexation like CPI would be quite significant.
Lord Sharkey: Do you accept that the RPI is flawed?
Elizabeth Truss: Yes. I accept that it uprates too rapidly. There is a fundamental statistical issue with the way it is calculated. But I am saying that everybody who invests in, say, index‑linked gits linked to RPI understands that. It is priced into the market, essentially. It is not a perfect measure of inflation, but it is an understandable measure and people understand how it is calculated.
The Chairman: Do you think the commuters who see their rail fares increase by RPI understand and accept that as well?
Elizabeth Truss: The specific issue we face with rail fares is that the costs of railway provision are, by and large, uprated by RPI. If we were to be in a position of uprating fares by CPI but seeing the costs raised by RPI, further money would have to be found from the public purse to pay for that. In an ideal world, I would like to see the costs lowered and rail delivered in a more efficient fashion. I have been in discussions with the Transport Secretary about precisely that issue. But we cannot be in a situation where the costs are rising by RPI linkages but the revenue is not rising. In that case, we will end up further subsidising the railways and putting taxpayers’ money at risk.
Lord Sharkey: I want to go back to Sir David Norgrove. He is convinced that the Chancellor would say no to any request that he made. This seems to me to be a fundamental problem. As you say, it is the duty of David Norgrove and his people to make a recommendation, and he is saying precisely that it is a waste of time and he is not going to do it. That seems an unsatisfactory way of dealing with a known flaw in the index.
Elizabeth Truss: As I said, RPI is laid out in legislation. That is a fundamental issue. It needs to be in line with the way it has been legislated for. It is difficult for the Chancellor to say yes or no to a request that has not been submitted.
Lord Sharkey: I agree.
Elizabeth Truss: I am suggesting that it is the role of the ONS to put forward that proposal. Then due consideration can be made of that proposal, and the benefits and costs weighed up. You will be aware of the ruling by the Bank of England: is it sufficiently deleterious to make a difference? That process has not been gone through.
Lord Sharkey: Does it not concern you that the person charged with making that recommendation, if he thinks it is appropriate, has already decided that the Chancellor will say no, so he is not going to do it?
Elizabeth Truss: If there are reasonable changes to be made that remain within the spirit of the legislation, I would suggest that that is something for the ONS to consider. It is a matter for it. The whole point of an independent body is that you cannot have government Ministers telling it what to do.
Lord Turnbull: I want to come back to the question of rail costs being uprated by RPI. What are rail costs? There are wages, interest on capital and fuel. That approach is completely wrong. Which of the railway industry’s inputs are uprated by RPI?
Elizabeth Truss: Take wages, for example.
Lord Turnbull: They are not; that is the whole point. They are not uprated by RPI. Sometimes they go up more quickly; sometimes they go up less.
Elizabeth Truss: I am telling you about the specific case of the costs of Network Rail, which is one of the determining factors of the costs of the railways. Those costs are going up in line with RPI.
Lord Turnbull: They are not going up in line with RPI, but they are going up. They may go up more, or they may go up less. They are not linked to RPI. That is the statement you made, and it is wrong.
Elizabeth Truss: I am very happy to provide you with further information, Lord Turnbull, but there are input costs, including wages, that are rising by that.
Lord Turnbull: They are rising, but they are not necessarily rising by RPI; therefore, there is no necessary link between RPI for fares and the costs of the industry. They may go up more; they may go up less. It is just a very slack way of describing it.
Elizabeth Truss: There are costs in the rail industry that are linked to RPI.
Lord Turnbull: I would like to know what they are.
The Chairman: I think we have covered the point.
Q34 Lord Lamont of Lerwick: RPI is sometimes described as a “legacy measure”. One of our witnesses, Jonathan Athow, talking about RPI, said that there was no need for it. But he said, “We describe it as a legacy need. We very much hope that RPI quickly becomes a legacy measure, used for fewer contracts”. That is not really sustainable, is it, given that there are index gilts that do not mature until 2068? This is more than a legacy measure, is it not?
Elizabeth Truss: It is certainly true that we continue to offer gilts on the basis of being linked to RPI rather than CPI. That is because of work the Debt Management Office has conducted, which suggests that that is better value for the public purse. As I said, in an ideal world, we would have everything on CPI measures. That is not where we are at the moment. We are moving towards that, but it is important that, in trying to get the tidiness of having a single inflation measure, we do not end up incurring costs to the public purse and potentially incurring legal action from private citizens who have bought into RPI‑related contracts.
Lord Lamont of Lerwick: You could issue CPI‑linked gilts, could you not?
Sir Robert Stheeman: In theory, we could. No decision like that has been taken. I would emphasise the Chief Secretary’s point and say that the fact we use RPI still for inflation‑linked gilts is not an expression of the view that we may have on how good or not RPI actually is. It is because that is where the market currently exists.
Lord Lamont of Lerwick: I wonder whether you think the statistics authorities have any conflict, possibly, in the use of RPI with their legal duty under Section 7(1) of the Statistics and Registration Service Act 2007, talking about safeguarding the production and publication of official statistics that serve the public good.
Elizabeth Truss: I do not believe they have a conflict, no. This is a clear statistic. We may agree that the methodology is not as accurate in terms of reflecting the inflation of goods as CPI. However, it is a well understood and properly calculated index, which is used in a number of different scenarios, underwriting private pensions, gilts or rail fares and rail costs. All those things are properly provided by the ONS. They have been clear that it is not a gold‑standard statistic, but nevertheless it has a widely understood methodology.
Q35 Lord Tugendhat: We are in a very odd situation, are we not? On the one hand you accept, as indeed does almost everybody else, that this is an extremely imperfect measure, that things would be different in an ideal world and so forth. Yet we not only continue to use it; we continue to issue debt that is linked to it. Can you envisage circumstances in which RPI could ever be abolished?
Elizabeth Truss: Government is always dealing with legacies. We never start from a greenfield site. We are always in a position of deciding whether to move to something new and whether the costs of doing so are outweighed by the benefits. I can certainly envisage at some point in the future that RPI would not be used any more, but it is probably in the relatively distant future. Lord Lamont has mentioned the fact that gilts have already been issued to 2063. A gradual path of moving towards a single measure is the right way of doing things.
We have not yet reached the tipping point where it becomes efficient to move to a single measure. That is what Sir Robert was saying about the Debt Management Office. If at the moment we were to offer RPI index‑linked gilts alongside CPI index‑linked gilts, it would cost more to the public purse, because you would have a fragmented offer. We always have to consider what the overall impact would be on the public finances and the private sector economy before we make that full move.
Lord Tugendhat: I would interpret your reference to 2063 as meaning that, for as long as there are index‑linked gilts in existence, it will not be possible to abolish RPI.
Elizabeth Truss: No.
Lord Tugendhat: As index‑linked gilts are continuously being issued, the date goes on and on into the distant future.
Elizabeth Truss: I will defer to Sir Robert on the details of index‑linked gilts, but there could be some point in the future where we switch to CPI for index‑linked gilts. Then you have the legacy of the remaining RPI‑linked gilts. I am suggesting that we have not yet reached that point, which is something the DMO has looked at.
Q36 Lord Burns: Could you take us through the argument as to why the fragmentation of the offer means that it would cost the Government more if one had a situation where some of the gilts were indexed to RPI and others to CPI?
Sir Robert Stheeman: Fragmentation of any market that one hopes is deep and liquid is generally not a good thing. We try to do everything we can to encourage a deep and liquid market in all gilts, including index‑linked gilts. This is something that came out in response to the consultation we had in 2011 on potentially issuing gilts linked to CPI. Were we to do so, one of the issues clearly raised by the market, which is not necessarily a killer argument, was a concern that, if we had two types of index‑linked gilts out there, we would have a fragmented market, and with that comes cost. I am not saying that cost is insurmountable. There may well be a time in the future when we have to live with that cost, because we want to transition to something for which there is superior demand.
Lord Burns: Have you worked through this? Do you have a paper? Do you have a worked-out paper that explains why it is going to cost more and how much more it would cost, or is this simply a finger in the air?
Sir Robert Stheeman: Purely in terms of fragmentation, it is a sense—and I happily admit that it is no more than a sense—that, especially for something as arguably complicated as inflation‑linked debt, you need as far as possible to have a market with one clear point of reference. As I say, that does not mean you cannot have two. But, if you had two, if one of those were to be CPI and if you had the situation you currently have where demand for RPI heavily outweighs demand for CPI, it is very likely that we would achieve a poorer price for CPI‑linked issuance than we would for RPI‑linked issuance.
Lord Burns: There is a definition that says these are legacy issues. Normally with legacy issues, in my experience, you say, “We will continue to use them going forward for as long as we have to use them, but we will take every opportunity that we have now to move to a newer set of arrangements”. What is puzzling the Committee is that, even in those situations where there seems to be an opportunity to find these legacy issues and say, “From now on we will do something different”, that opportunity is not being taken.
Sir Robert Stheeman: Let me try to answer that very clearly. The reason that it has not been taken yet—and we have been very clear that we keep this under review, and we are open to taking this forward at one point in the future—is, to repeat, that demand for RPI protection is much greater, so there is a better price point.
Lord Burns: Can you tell us what would create the better circumstances that would mean that occasion was the time to make the change? You are further in. If you are talking about fragmentation, the longer you leave it, the greater imbalance there will be between this huge stock of outstanding debt linked to RPI and the amount of new debt linked to CPI.
Sir Robert Stheeman: The answer is to have a greater amount of pension liabilities linked to CPI or whatever measure of inflation is used in the future than there are currently issued to RPI. You need to have a core demand to build a market up and to make it sustainable over the long term. From our perspective in terms of debt management principles, that is absolutely key because we benefit from having this deep liquid market in something.
Lord Burns: This is a legacy issue that we are trapped in and cannot get out of.
Elizabeth Truss: It is just a long‑term issue, but we are seeing an increased take‑up of CPI‑linked private pensions. There are signs, and the Government are moving to CPI in more areas of our operations as well. There are now more assets linked to CPI. We have to think about what the immediate cost of a “big bang” switchover now would be and whether that is worth it. In our judgment, we have not yet got to that point.
Lord Burns: We are not talking about a “big bang” switchover here; we are talking about switching over the new flow of debt to CPI. This discussion we have been having, as I understood it, was not about switching everything that is currently outstanding to CPI; we are talking about changing the new flow, which is not a big bang. That is just a decision that says, “From now on, we will use CPI”.
Sir Robert Stheeman: That may, in my opinion, very likely occur at one point. The point is that it needs to happen at the point that it is cost-effective to taxpayers.
Lord Lamont of Lerwick: Could I go back to this point about fragmentation increasing the cost? You talk about a deep and liquid market, but is it just one market? Surely there are different markets with different maturities. You are continually looking further out and further out. With very low inflation, there have been big step changes in fixed‑interest markets all over the world. When you get an extension into a new maturity, another 10 years added on, surely you could at that point introduce a CPI index‑linked gilt that would not be in competition with those of earlier maturities.
Sir Robert Stheeman: We might very well want to do that. We would also want to see, though, that there is demand for that particular type of protection. To me, that is key. The Government would also need to be quite clear as to their preferred method of measuring inflation in the future, because it is not just a question of CPI. CPIH has also been mentioned. The market needs clarity on something like that.
Q37 Baroness Harding of Winscombe: Let me first declare my interest as a non‑executive director on the Court of the Bank of England. I wanted to follow up on your point, Sir Robert, about the cost to taxpayers. There has to be a tipping point when a statistic has so failed in its core purpose, and has diverged so far from an accurate assessment, that we risk society losing faith not just in that metric specifically but in the whole stack of cards in total. If I hear you correctly, if you only think about the immediate cost to taxpayers, the cost to society of citizens losing faith in indices is discounted. At what point do we reach the tipping point where we should seriously consider changing this, even if there is a cost to taxpayers, because there is a much bigger societal cost?
Sir Robert Stheeman: In fairness, it is not the role of the Debt Management Office to make that trade‑off, because one trade‑off is a cost-benefit trade‑off; the other one, indeed, is almost a question of political economy. At what point is that necessary? It is for others, not so much the Debt Management Office, to make that choice. However, in my personal opinion, it is when and if there is clarity over a new index and the preferred measure of inflation going forward, and when we see a significant number of liabilities linked to that measure in the pension industry.
Do not forget that we issue these things not because we in ourselves have a desire to issue any one particular index. There is a real need for hedging out there, coming primarily from the private sector. The private sector is willing to pay us a premium for that. At one point that may change but, if you view gilt issuance as the issue, I would suggest that something else has to come first, which is the whole question around what index is being used to measure liabilities in particular and contractually pay out liabilities in the private sector. To me, that is an absolutely key question.
Baroness Harding of Winscombe: Might I ask the Chief Secretary, as the politician in the room, to answer an essentially political question? At what point does this start to damage the reputation of all experts, as a key index that has lost its way?
Elizabeth Truss: We have not reached that point. There are genuine fiscal costs at a time when the Government do not have a surplus. We are still running a deficit, so the fiscal cost is a significant question. There is a wide understanding, among those who are purchasing the index‑linked gilts, about what RPI and CPI are. I do not think there is a question about whether people out there are buying things they do not understand. It is widely understood. It is a factor of matching the assets people have with the liabilities they have, which is why people want to invest in index‑linked gilts that are linked to RPI.
If there was evidence that this was causing a public trust problem or even an economic problem, I would be very interested to hear it. But I think RPI is priced in when investors make those decisions. Nobody is pulling the wool over anybody’s eyes. We have been honest about the difference between RPI and CPI and the fact that CPI is a better measure of inflation than RPI is, but we have this legacy of using RPI, and the costs of moving away from it are significant for both the taxpayer and the individual investor. It is better to be open about that.
As I was saying earlier, Ministers and officials often have a desire to tidy things up, and there can be unforeseen consequences of seeking to do that. I question how big the problem is and whether the solution that is being proposed of a rapid move to CPI might be worse than the problem people envisage in the first place.
Q38 The Chairman: I have one point on cost. Chris Giles told the Committee that we were dealing with a £400 billion market, and a 1% difference between CPI and RPI would save £4 billion a year on debt interest payments for the Government. Do you agree with that?
Elizabeth Truss: No.
The Chairman: Why not?
Elizabeth Truss: When people invest in RPI‑related gilts, they are factoring in the fact that RPI is likely to rise more quickly than CPI. It is baked into the market.
Sir Robert Stheeman: I completely agree.
The Chairman: You completely agree with what?
Sir Robert Stheeman: I agree with what the Minister has just said.
The Chairman: That is good.
Elizabeth Truss: It is controversial.
Sir Robert Stheeman: If you are talking about what is known as the wedge between the two, the market is pretty sophisticated. The market will figure out immediately that it is going to receive less in terms of CPI than it would were it to receive RPI, so they will pay us less, if you are talking about future issuance. That is not the end of the world, but the market would factor it in. It means that cost savings arising as a result of that are likely to be relatively small, if any. If he is talking about a figure of £4 billion, I do not know exactly where that comes from. What would be very significant, which is quite clearly not on the cards at all, would be to change the indexation of existing gilts. That would result in a very significant difference, but no one is realistically suggesting something like that.
The Chairman: We shall see.
Baroness Bowles of Berkhamsted: I am not suggesting that. Has there been any analysis done, in terms of the bond purchase, as to how much of it is for hedging of pensions in particular, given that a lot of those will be linked to RPI? I fully understand the hedging. Has any effort gone into nudging more pensions, in terms of new contracts, because you probably cannot change all the existing ones, on to CPI? I fully understand that, if you are offering the market RPI or RPI minus 1%, the prices and what you get will be different, corresponding to what you pay out being different. But what we are getting at here is an element of fairness and reality. I am not sure the average person, when they are thinking about RPI, necessarily knows the difference between RPI and CPI, even if their pension funds do. We want to get to a fair, understandable and genuine measure that does what it says on the tin. Can we do something about the new contracts?
Sir Robert Stheeman: I am not sure that we, from where we sit, can do anything about new contracts. It is worth noting that there has been a growth in CPI pensions over the last years. This goes back to the point some of your colleagues have made about a transition, at some point, to us issuing using a new index. The more demand there is for that sort of thing, the more interested, frankly, the Debt Management Office gets, because we have a cost‑minimisation objective. What we can do in terms of encouraging private sector contracts and pension agreements to use CPI is probably not too much.
To give you a sense of the two different markets, the PPF at the end of October last year suggested that total defined benefit pension liabilities stood approximately at £1.7 trillion. In November last year, a private sector bank, NatWest Markets, made an estimate in something it put out that there is £100 billion to £150 billion of existing CPI liabilities in pension schemes. That gives you a bit of a sense of the difference in the pension fund industry between the very large RPI market out there and this much smaller but growing CPI market. If CPI were made into the Government’s main preferred statistic, you would see even more.
The Chairman: Could you explain what you mean by “the Government’s main preferred statistic”?
Sir Robert Stheeman: I mean the Government’s preferred method for measuring inflation.
The Chairman: Does that mean abandoning RPI?
Sir Robert Stheeman: Not necessarily, no. The word “abandoning” worries me. Do you mean abolishing it? The Government do not use RPI, I believe, except for linkers, in terms of paying too much out elsewhere. There has to be clarity about this. Is it CPI or CPIH?
The Chairman: I really wanted clarity on what you meant by the phrase you used.
Sir Robert Stheeman: I meant whatever index turns out in the future to be the one around which things coalesce.
Q39 Lord Turnbull: Chief Secretary, you downplayed the view that there was a public trust problem. There are two areas of concern that diminish public trust. One is the inherent feature of parts of the RPI where, if a price goes up in one period and goes back to the same level it was at, or down and then back to the same level it was at, it is not calculated as zero change. That is a serious statistical flaw.
Secondly, the Government are widely perceived as playing arbitrage: “Heads, I win; tails, you lose”. When they are paying out money, they use CPI; when they are charging people, such as rail fares and student loans, they charge RPI. That is a serious question. There is a strong case for standardising all new government dealings on CPI. Then they will be seen as playing fair.
Elizabeth Truss: I agree with you about the statistical imperfections of RPI. I agree it is not as good a measure as CPI for precisely those reasons. I would point out to you that on business rates, for example, we have moved from RPI to CPI, despite those being a revenue raiser. We are moving in that direction when it is affordable. The issue here is the public finances and what the public finances allow us to do. There are other areas. For example, fuel duty is RPI‑linked, but it has been frozen since 2011. There are a number of areas that are linked to RPI, but we have decided to freeze them or not take them forward.
In any of these decisions about public policy, we have to look in the round at all the different policies that affect people through the income distribution, through different stages in their lives. They are not divisible, in my view. For example, planning policy affects the price of houses. Tax policy also affects people’s incomes at various levels. These policies in respect of whether we use RPI or CPI are not divorced from the rest of government policy-making and the overall distributional impact.
I have said that RPI rises more quickly. It is not our preferred measure; it is statistically imperfect. But if we were to change something that is currently linked to RPI to CPI, the fiscal reality is that we have to find that money from somewhere else in the government budget. As Chief Secretary, I constantly face the issue of finding money from elsewhere in the government budget. That is just something we have to be aware of. In the case of rail, which we were talking about earlier, should rail fares rise by CPI rather than RPI? That is money that would have to be found elsewhere.
Q40 Lord Kerr of Kinlochard: I have a question, if I may, on the reference to the fuel duty. What about students? Most student loans are not paid back.
Elizabeth Truss: Yes, quite.
Lord Kerr of Kinlochard: They are RPI‑related. Is that because the total sum, not the sum expected to be paid back, comes into the government accounts as income?
Elizabeth Truss: Philip Augar is currently conducting a review of higher education funding. This whole area is something we are looking at very closely. The way student funding is currently set up is that it is an income‑contingent loan. It is only when you are earning £25,000 per year that you have to pay it back. RPI is part of the way that that overall funding was calculated. The reality is that student loans come at a significant cost to the Exchequer, as I am sure you are aware, because of the number of loans that end up getting written off.
Lord Kerr of Kinlochard: I understand the argument that it is all one ball of wax, but when one sees an anomaly one has to start somewhere. It is a bit strange that the interest you charge on student loans is related to RPI. You uprate people’s pensions related to CPI.
Elizabeth Truss: Yes, for public sector pensions.
Lord Kerr of Kinlochard: Exactly, so it depends on whether they are paying you, in the case of the students, notionally in most cases, or whether you are paying them. Is that not a bit odd?
Elizabeth Truss: On business rates, where the businesses are paying in—
The Chairman: We have got the point on business rates.
Elizabeth Truss: That is good. I am just saying that it is an example of where we are moving in that direction. We are moving gradually in that direction.
Lord Kerr of Kinlochard: Are you going to move on fuel duty?
Elizabeth Truss: That is a matter for the Budget, as I am sure you would expect me to say. On the matter of student finance, this area is entirely under review. Any comments the Committee makes, I am sure, will be closely followed by Philip Augar with a view to his review.
The Chairman: We spent a year on this subject and just published a report last month. You might find the section on student loans and student finance of interest.
Q41 Lord Layard: I want to come back to the Chief Secretary’s remark that so many contracts are written around RPI, private as well as public, that, in some sense, you cannot really abolish it. But you can change how it is calculated. It seems that there is quite a good argument for saying, “People want to be protected against inflation. We should be measuring inflation in the correct way to protect them”. That seems a rather straightforward argument.
If we are using an index that does not provide the correct protection, it is rather a difficult position to defend. You could just, as you have mentioned already, calculate RPI as CPI. You could call it RPI but calculate it as CPI or CPIH. There would be then these capital losses. A previous witness pointed out to us that the capital losses would be more or less the same as the capital gains that occurred when the ONS decided, after all, to keep RPI. This is part of the chances of life. It does not really have a clear ethical implication one way or the other.
Suppose the ONS plucked up the courage to recommend the change and the Bank of England referred it to the Treasury. How would the Treasury go about analysing it? What would be the procedures? What would be the role of the inflation tripartite group? How public would any of this analysis be? We note that the minutes of that group are not currently published.
Elizabeth Truss: Should a proposal come forward, it would have to be analysed with respect to the implications on the public finances, but also the legal implications of people entering contracts with the expectation of a particular type of inflation index and then finding that inflation index had been changed significantly, were the ONS to suggest a measure that was essentially similar to CPI. One of the factors here is that the RPI measure is baked into primary legislation, so it would most likely need some kind of legislative change. As you will be aware, the legislative timetable is currently busy, so that would be another factor to be considered.
It is back to the fundamental point you raised in your question: when people have entered into arrangements based on RPI, are they doing it on the basis that they understand the way that the RPI metric is calculated, even if it is not perfect, see it as something that rises more quickly than CPI, which it does, and bake that into their calculations, or are they simply looking for a way of protecting against inflation? That is a critical question.
Lord Layard: Could you mention something about the inflation tripartite group?
Steven Farrington: Let me tell you a bit about that. The inflation tripartite group, which I occasionally attend, is simply a working-level group between the ONS, the Treasury and the Bank of England. It is not responsible for any of the more conceptual or higher‑level discussions about relative developments in RPI and CPI. Those sorts of discussions take place separately at the two advisory committees the ONS has set up. There is a technical working group and a stakeholder group, which I attend. The Bank of England also attends both those committees. Those committees then make recommendations to the National Statistician on technical grounds, and they take in the views of stakeholders from the stakeholder group, which represents a much wider group of people than just the Treasury and the Bank of England. Those recommendations go to the National Statistician, who then makes those decisions. The inflation tripartite group plays an overall role in those decisions, and it is more of the working‑level relationship between the Bank, the Treasury and the ONS. The minutes of the groups that advise the National Statistician are all published.
Q42 Baroness Harding of Winscombe: I would like to explore a little more how we get out of the trap, as Lord Burns described it, where we all agree that RPI is an imperfect measure but we cannot work out how, even beyond 2068, we could change. I would like to understand to what extent you think the Government are able to change the inflation index on existing index‑linked gilts from RPI to CPI, specifically with regard to some evidence we had from Chris Giles, who told us that, apart from three issues of index‑linked gilts that mature in 2020, 2024 and 2030, all the more recently issued gilts have no constraints. The prospectuses say they would be linked to an index that the Chancellor considers “continues the function of being an officially recognised index measuring changes in the level of UK retail prices”. On his read, it would be possible.
Sir Robert Stheeman: Can I try to answer that? This can get complicated and technical, and I will try my very best not to be. It is quite important to be clear about some things. If you are talking about changing the way RPI is calculated, that would have significant ramifications for the market, but it is arguably doable, and the market would price it accordingly. That was also behind Lord Layard’s question. If you are talking about abolishing RPI and choosing a new index, again, that in theory could happen. But, as long as RPI continues to be published—according to legislation, one should also point out—we have a contractual requirement to link our payments to RPI. We cannot unilaterally switch index‑linked gilt payments, for instance, to a new index if RPI is still in existence. That would be a breach of contract; it is a breach of the terms of the prospectus, which currently allow us to issue only RPI, as long as RPI is there, obviously.
The clauses you have just referred to in the prospectuses refer to what happens if RPI ceases, for whatever reason, to exist. Then there is, indeed, this process, but, as you have correctly noted, there are currently only three out of 29 inflation‑linked gilts that have this clause whereby the Bank of England would determine whether it is materially detrimental. Then it is for the Chancellor to decide whether he or she wishes to allow that change to go though. That is only for three gilts, the last one of which matures in 2030. Recent gilt prospectuses have different language.
The Chairman: On that point, I am now quite confused because, Chief Secretary, you used the phrase “baked into legislation”. My understanding is that the legislation says this: “The board must under Section 20 compile and maintain the retail prices index and publish it every month. Before making any change to the coverage or the basic calculation of the retail prices index, the board must consult the Bank of England as to whether the change constitutes a fundamental change in the index which would be materially detrimental to the interests of the holders of relevant index‑linked gilt‑edged securities. If the Bank of England considers that the change constitutes a fundamental change in the index which would be materially detrimental to the interests of the holders of relevant index-linked gilt-edged securities, the board may not make the change without the consent of the Chancellor of the Exchequer”. You say that the fact you have to publish the RPI is baked into the legislation, but the legislation does not prevent you from altering the RPI to make it a more effective measure. Can you explain why you are shaking your head?
Sir Robert Stheeman: I completely agree with you.
The Chairman: You shake your head when you agree.
Sir Robert Stheeman: No, it does not say that. You are absolutely right.
The Chairman: If the index is inadequate, which is what the statistics authority is telling us, the fact that the RPI is baked into the legislation is surely irrelevant. It can be changed.
Sir Robert Stheeman: It can be changed. They are the ones who have chosen not to change it.
The Chairman: Who is “they”?
Sir Robert Stheeman: It is the statistics authority. They are the ones who have chosen not to change the methodology of calculation.
The Chairman: The reason we are doing this inquiry is because we got the broadest of broadest of hints from the Governor of the Bank of England that we should do it when he gave evidence to the Committee. When we had witnesses from the statistics authority here, they told us that they think this is a rubbish index that needs attention. If one was being uncharitable, one would say the blockage was the Treasury.
Elizabeth Truss: No proposals have been put to the Treasury.
The Chairman: You would welcome proposals, then.
Elizabeth Truss: I did not say that.
The Chairman: I know you did not.
Elizabeth Truss: It is a matter for the statistics authority. What I have pointed out are some of the potential flaws of changing an index that is widely understood. Should the Bank of England say, “That fails the test and it is detrimental”, you are potentially damaging the interests of index‑linked gilt holders.
The Chairman: Right, so you would not welcome a proposal.
Elizabeth Truss: It depends what the proposal is. There is no proposal at present.
Lord Sharkey: Minister, could you come back to the original question?
Elizabeth Truss: I am pointing out the potential flaws there could be, should a proposal emerge.
Lord Sharkey: But a proposal will not emerge under the current chair. He has said it would be a waste of time because the Treasury would turn it down, so we are stuck.
The Chairman: Is he under a misapprehension?
Elizabeth Truss: You are asking me a hypothetical question before we have seen the analysis or the proposal.
Lord Sharkey: No, I was not asking—
Elizabeth Truss: I gave you an indication earlier in this hearing of the types of issue we would look at and the balances that would need to be weighed up in terms of the fiscal cost and the impact on those who had bought RPI‑linked products. Those are the types of things we would look at.
Q43 Lord Turnbull: This is probably for Sir Robert. You used this phrase “fragmented offer”. If you have a gilt offer that is very different from all the rest, I can see the problem. If, someday not far in the future, you started to issue gilts linked to the CPI, two things would happen. It would be large because we borrow large amounts of money. It is not a small offering that is so small it is not liquid. But all the stocks you issue are auctioned, so the immediate effect is that the price paid would be different between an RPI gilt from the existing stock and a CPI gilt in the new flow. I reckon the two would pretty immediately converge to something like the same real price. You made the argument that there was a cost to the Exchequer; I do not see it, in an auction world.
Sir Robert Stheeman: I am not quite sure I understand what you mean about the two prices converging. Which two prices?
Lord Turnbull: The real rate of return on those two gilts would converge.
Sir Robert Stheeman: The point is that the market, indeed, can price in this differential, and it would do so immediately. But there would then be one market out there for those bonds still linked to existing RPI, and there would be another market out there for those bonds linked to CPI. That is completely possible.
Lord Turnbull: Both of them would be large. In the fixed gilt market, you have different coupons for different maturities, and people find a way of producing some kind of—
Sir Robert Stheeman: They arbitrage them.
Lord Turnbull: The same thing would happen, as long as the two kinds of index gilts were running alongside. The discount or premium would adjust so that the real rate of return on those two instruments would be pretty much the same. There is no tax wedge to distort this. I cannot see why the Exchequer would find itself increasing its debt servicing costs by introducing this.
Sir Robert Stheeman: There is a cost arising from fragmentation, because one of the ways we issue any gilt, including nominal or conventional gilts, is to try to build these up in large size so the whole market achieves what we would describe as a liquidity premium. In other words, we are extracting value from the market through having a large, liquid market. I am not saying that we could not extract any premium. I suppose I am saying that we would perhaps not benefit to the same extent from having one single market as we would otherwise. It would potentially be more costly to have a fragmented market. It is not an insurmountable cost.
Lord Turnbull: I am questioning whether that difference is significant, given the fact that the prices of the new gilts would be auctioned and would be on sale alongside the remaining stock. Before long, this second block of gilts would also become significant in size.
Sir Robert Stheeman: One of the things we would have to take into account is, for instance, how much Government choose to issue in inflation‑linked debt in general. We would then need to think about how much, if any, we would still want to allocate to RPI gilts. We have been quite clear that we retain an open mind on the possibility of doing this. For us, the question is fundamentally one of timing. From where we sit, in order to minimise cost, we would prefer to see more of a critical mass of demand for CPI protection than is currently out there. That is why I mentioned these two different numbers, in terms of the potential size of one market versus the other. Will that eventually come? It may well come. We have said that we will keep this under review. Just because we concluded back in 2011 not to do it, it does not mean we would not want to do it in the future.
Elizabeth Truss: Sir Robert has talked about the potential costs of doing this. At the moment, I struggle to see what the benefits are. Where are the economic benefits in moving?
Lord Kerr of Kinlochard: You save on the coupon.
Elizabeth Truss: We are saying the market would adjust to reflect that. To me, there is the cost of moving. There is the fact that, as Robert outlined, at present many of the private pension funds are in RPI‑linked funds rather than CPI‑linked funds, so there is potentially less demand for it. It is not clear to me that there is an economic benefit. To make a move like that, I would want to see a demonstrable economic benefit.
Q44 Lord Kerr of Kinlochard: I will try to put it very, very simply. You say it is government policy to move away from RPI to CPI as much as possible.
Elizabeth Truss: Yes.
Lord Kerr of Kinlochard: This is described as a legacy issue. But so long as Dracula is undead, it is not a legacy issue. In 2063 or 2064, out they go. At some stage, if it is to become a legacy issue, you will have to be issuing something on another index, which at present looks a bit like CPI. How will you know when is the moment to stick the stake through Dracula’s heart?
Elizabeth Truss: We were asked a question earlier by Baroness Harding about how we are proposing to do this, and my answer is “slowly”. At the moment, we are not at a point when there are obvious economic benefits to doing it.
Lord Kerr of Kinlochard: How will you know when you are? How will you know when that point is?
Sir Robert Stheeman: One of the roles we have in the Debt Management Office is to talk regularly to the market. We consulted in 2011. I see no reason why we might not want to consult again on this issue when the market is giving us a clear signal that demand is now really sufficient to get this market seriously going.
Lord Kerr of Kinlochard: It is government policy to try to move away. Therefore, you ought to be trying to encourage the market, surely, not waiting for a signal from the market.
Elizabeth Truss: To be clear, government policy is to move from RPI to CPI but not at an economic cost.
Q45 Lord Layard: One of the obvious changes—we should call it a “change” rather than a “benefit”—is for private pension funds, which in many cases are struggling with low interest rates. That is a very big economic issue. You might not know which way to say it was balanced. I do not personally think the recipients of pension protection have any very clear idea as to which is the better form. They want to be properly protected. Why do we not help the pension funds to be a bit more solvent?
Elizabeth Truss: I would argue that it is a matter for the pension funds to make the decision on what terms to offer the people investing in their pensions.
Sir Robert Stheeman: Those pension funds are the ones that tend to have their liabilities linked to RPI. Us issuing, for instance, CPI would not necessarily help their solvency.
Lord Layard: I was thinking of recalculating the RPI as CPI.
Q46 The Chairman: Sir Robert, could you answer a daft bloody question from me? My understanding is that RPI went off the rails in 2010 on this clothing index issue. All the people who had index‑linked gilts that were associated with RPI, which, depending on whom you talk to, is now thought to be between 0.7% and 1% higher as a measure of inflation than CPI, have gained a huge windfall gain, have they not?
Sir Robert Stheeman: You could argue that.
The Chairman: Why would the Government want to proceed with a policy that is giving those institutions and others a windfall gain because it is based on the index being inadequate? That seems to me to be the key question. Am I missing something?
Elizabeth Truss: I believe the problem with the clothing has been fixed.
Steve Farrington: The wedge will persist over time. You are right: at the point at which the wedge widened, it is the case that existing holders of gilts will have benefited from the increase in the wedge. That is correct.
The Chairman: What would be the amount by which they have benefited?
Steve Farrington: I could not tell you that number off the top of my head.
The Chairman: Perhaps you could write to the Committee and let us know.
Steve Farrington: Yes, sure.
Lord Turnbull: Why do you think this clothing index issue has been fixed?
Elizabeth Truss: The statistical issue remains—I acknowledge that, as I said earlier—between the Carli and the Jevons methodologies of calculation. I believe the specific way the basket was changed in 2010 was changed after that. That is my understanding, reading the evidence you have had previously.
The Chairman: If you have read the evidence, you will see that Chris Giles estimated this at £1 billion a year. Do you disagree with that? You have read the evidence. Do you disagree with it?
Steve Farrington: That £1 billion per year is a different number. I am not quite sure if the £1 billion a year is the correct number. I would need to check.
The Chairman: Perhaps you can write to the Committee.
Steve Farrington: Yes.
Q47 Lord Burns: When the Governor was here, he gave us the distinct impression that he envisaged there would be a transition period over which he would really like to see a change to this. The impression I have gained today is that you do not envisage changing the treatment of the existing gilts that are in the market at all at any point. This is not a question of giving a warning of 10 years or 20 years ahead and saying that we will gradually phase out the RPI.
The position you are taking is this: “They are baked into the contracts; these are the contracts. This is the basis upon which we sold them and we are going to see this out. We will maintain an RPI until those contracts are closed”. We happen to be adding to them, but that is an issue we dealt with earlier. Am I correct in assuming that you do not envisage any change in this? You are not thinking in terms of transitions. You are taking the position that this is the contract and we are sticking with it.
Elizabeth Truss: I believe the Governor’s point was about new contracts being issued. Is that correct?
Sir Robert Stheeman: I believe it was.
Elizabeth Truss: It was not about existing contracts. The Governor’s point was really around what we have just been discussing: at which point—I do not believe we have reached it yet—should we switch or, at the very least, offer CPI‑linked gilts? That was the point we were discussing and the point the Governor made. I believe his point was that we should set a 10‑year horizon.
Sir Robert Stheeman: I am not totally sure, and I do not really think I should speak for the Governor. He referred to a transition, which was more in general about the use of CPI and transitioning to that, saying that the current situation is not a satisfactory one. That is pretty obvious. If I were to make one comment—I hope I am not exceeding my brief—it seems to me that, if the situation is genuinely deemed not hugely tenable over the very long term, there must be a point in the future where you would want to transition everything.
From where I sit, from the narrow but very important perspective of the index‑linked gilt market, I would have a personal preference, if I am allowed to say that, to see the composition of RPI, over a very long time, and with sufficient notice and warning to the market—that is absolutely key—so that everybody felt this was the right thing to do, changed and improved. Otherwise, indeed, you will have that legacy issue until 2068.
Lord Burns: This is exactly what has been concerning a number of us on the Committee: the idea that we are going to have to keep this index of RPI for so many years because of existing contracts, but we have no appetite for improving it or, when errors or issues come to light where the methodology could be improved, we are not prepared to take them because, somehow or other, this would be interfering in some of these processes. It is not that we just had a legacy index; we had an index that was regarded as beyond its sell-by date, and yet we were not proposing to improve it when circumstances required it, even though we were going to face another 50 years, or whatever it is, of having to use it.
It is this that really got us into trouble in some of our early discussions when we were taking evidence on this. We seem to be saying, on the one hand, that there is a problem with it and we will have to continue to use it, but on the other hand we do not want to improve it. This looks to be rather curious public policy.
Elizabeth Truss: In terms of the improvements to the index, that is a matter for the ONS to propose, which we will consider, bearing in mind all the potential detriments that are part of that process. It is important to separate out the discussion we have been having about gilts and the gilt market from issues essentially of public policy, which are the fiscal considerations the Treasury has to take into account when indexing various duties and costs that the Government bear. For me, those are two different issues, and we are gradually moving towards CPI on the latter issue. That is where we want to get to. We are going through a period of fiscal consolidation at the moment, which makes that a tricky thing to achieve, but we are moving in that direction.
The matter of the gilts is very much linked to the development of that market and how we make sure that people who have set contracts in RPI do not suffer a significant disadvantage. That is not to say that errors cannot be addressed. We talked earlier about the clothing error, which has been addressed. If you are talking about fundamentally changing RPI so it is CPI, that is something different entirely.
Q48 The Chairman: I have what the Governor said to us in front of me. I do not know if you had a chance to read this, but, coming in at the middle of the story, he was referring to the problem we have just been discussing. “This seems to me to be one of those issues that we have to decide at what point in the future that change becomes acceptable”. “There is a short-term argument that of course it is better to continue issuing them on RPI because of the liquidity of the benchmark. I understand all those points, and the DMO does a fantastic job”—you liked that bit—“but we would not want to be in the same position 10 years from now”. “With these very difficult decisions—I have seen this in the past; I saw it in Canada—in the end you have to pick a date, and it tends to be seven, eight or 10 years down the road, at which point you will have transitioned off and then work back”. Do you disagree with him?
Sir Robert Stheeman: No, I do not see anything in that that actually contradicts what I have said.
The Chairman: I was not suggesting it contradicted or supported what you said. I am just asking whether you agree with the Governor’s view that you have to pick a date, and it has to be seven, eight or 10 years in advance.
Sir Robert Stheeman: I agree that we could issue gilts with a new index. If no change is made to the current way RPI is calculated and the CPI—or whatever it is—market and demand for that product continue to grow to make it worth while, we should be issuing that new type of gilt at one point. But we will have to service the RPI‑linked gilts until 2068 regardless.
The Chairman: That is understood.
Q49 Baroness Bowles of Berkhamsted: I want to go back a little to pensions, because the Government have changed state pensions and public sector pensions to be uprated by CPI. As we have we discussed, some private pension companies are now creating CPI‑linked products. In terms of solvency, I suppose it has to be the case that, unless you change to, say, CPI on the outstanding liabilities, there is no gain, because you get the market adjustment on the new side for the different pricing and so on. On that basis, I am struggling to see how you ever get what the Minister said would be economic gain, unless you followed the Carney advice to say, “We are going to do it in 10 years and work back”. If you say, “We will wait for the economic gain; we will wait for the market”, the market is really not going to be created while you maintain everything trundling along as the status quo.
Elizabeth Truss: We are seeing a growing market for CPI‑indexed pensions. That is happening anyway.
Baroness Bowles of Berkhamsted: Maybe they will do that because they feel they can get certain improvements in their solvency if their assets are index‑linked and their liabilities are in CPI, but that is going to take perhaps quite a long time to show benefit. Do you see that as what the insurance companies are trying to do?
Elizabeth Truss: As I have said, the CPI is a more desirable index for future costs that pensioners will face, so that makes absolute sense. That is why we have moved the public pension to CPI, although I would point out that we have a triple lock, so it is whichever is the highest of CPI, 2.5% or wage increases. We have taken a decision in the public sector to do that, but it is ultimately up to the people purchasing the pensions and the pension providers to make those decisions. There is clearly a trade‑off. Given that RPI rises more quickly, you would assume that would be a more expensive product.
Sir Robert Stheeman: You are making a very valid point when you suggest that the question from where we sit is not so much what we choose to issue, but there is almost a prior question: what is the nature of the liabilities of the insurance sector and the pension fund sector, and how do they want to hedge that? If the nature of those liabilities is still RPI‑linked, you will have this dichotomy for a very long time. It is not insurmountable, but it sounds messy to me.
Baroness Bowles of Berkhamsted: But there is going to be a transition period. Maybe 10 years is an adequate transition period for them to make sure all their new contracts are on CPI.
Sir Robert Stheeman: It is a fair point. Most new pension fund contracts that reference inflation, as far as I am aware, are indeed using CPI. So far, that is still, relatively speaking, a smaller market.
Baroness Bowles of Berkhamsted: That is for 20 year‑olds.
Sir Robert Stheeman: If they are lucky enough to have an index‑linked pension, yes.
The Chairman: Can I thank you, Chief Secretary, Sir Robert and Mr Farrington? That has been a really interesting session, and we are very grateful to you for helping to explain such a complex subject. We hope to be able to produce our report before the autumn. Thank you very much.
