Examination of witnesses
Jonathan Athow and Sir David Norgrove.
Q14 The Chairman: Sir David and Mr Athow, you heard the previous session. Thank you very much for coming and listening to that and for giving evidence today. Lord Burns will ask the first question.
Lord Burns: As you listened to the previous session, perhaps the easiest way to start is to ask how far you agree with what the witnesses said about the extent to which the RPI is a good or a poor measure of inflation, the extent to which there are obvious methodological problems with it, the extent to which there may have been errors in the past and how they should be treated, and whether you agree with them about the options for moving forward.
Sir David Norgrove: That is a very long series of questions. Why not start with the problems with the RPI? I agree wholeheartedly with what Paul Johnson said: that there is a series of problems here. I have been very clear that it is an inadequate measure of inflation.
Lord Burns: Given that it is a poor measure of inflation, what is the case for simply leaving things as is?
Jonathan Athow: In 2012, the then National Statistician ran a consultation on the future of RPI, of which you have heard. There were a number of respondents to that. We have talked about gilts and the way RPI is baked into that. A number of other users came forward who said that RPI was central to many of their private sector contracts.
At the time, we were told that tens of billions of pounds of corporate bonds were linked to RPI and had similar conditions, and if the RPI was changed they could be redeemed. So it was hard-coded into the system, and if we made those changes it could have much wider-reaching implications.
The key thing was that there were people who had spent money on index-linked instruments who had paid for them with the expectation of RPI as it was at the time, so there was an existing group of users who valued RPI and who had invested in it on that basis. So the decision was taken at the time to leave the RPI as was and to keep it on life support at a minimum level of updating. That was the argument at the time.
Since then, a policy has developed of saying that there are better alternatives—as the previous witnesses said, we currently put much more weight on CPIH and CPI as stronger measures—and encouraging people to use those, whether in the private sector or in government.
Lord Burns: Do you accept that on a methodological basis the way in which the CPI is constructed, with the weights et cetera, is a better method than the one used for the RPI?
Jonathan Athow: Yes. One thing I would amplify is that the focus has often been on the formula effect. Clearly we do not see the Carli formula as being the strongest of them. There will be people who put different weights on different factors, and different bits of evidence will point in different ways. We think that the weight of evidence points against the Carli, but we do not think that the way in which housing costs, for example, are treated in RPI is very strong. It blends different approaches, and issues such as the weighting and the fact that the spending of certain households is not included in RPI mean that it is not particularly satisfying either. We think that there is quite a fundamental set of issues with RPI.
Lord Burns: What about the 2010 change to the clothing-price collection of which we have heard? People now describe it as an error and that the rate of change has produced some bizarre numbers when you look at the levels. Therefore it is really quite puzzling to imagine that one could recognise this effect and simply not do anything about it.
Jonathan Athow: At the time, the impact on RPI was completely unexpected. The problem with the previous way in which the clothing price was collected was that the clothing market has changed. You increasingly hear phrases such as fast fashion—items come into stock for two or three months and go out again—so it is very difficult to collect a price over time.
That was creating lots of problems with the clothing price collection, so we asked the people who go out to shops to collect prices to slightly change the way they collected them. We expected that to correct for some of the issues that we had. As Chris said earlier, the way the change to how the data was collected interacted with the Carli formula to give this result was completely unexpected. The change was made, as I said, to try to make clothing price collection more representative of what people were spending their money on, but it has this unintended effect.
Lord Burns: Has it changed the level once and for all by increasing it, or is this an ongoing issue that will continue in circumstances to produce bizarre results?
Jonathan Athow: Indeed. It is an ongoing consequence. Every year, the measure of inflation using RPI is higher, so the effect is permanent and, as I said, one that we have continued to investigate. It was completely unexpected at the time, but it has this permanent ongoing effect.
Lord Burns: Given that we think that the weighting system is not as good as it should be, and given what we know about this issue and about clothing et cetera, and no doubt there are other problems, is it really tenable just to leave this to wither and believe that it will continue to be an active index for another 50 or 60 years? I cannot think of any other aspect of statistical life where one would accept that as the only way forward.
Jonathan Athow: We have taken comfort from the fact that there has been some movement away from the RPI. Recently, business rates moved from being uprated by RPI.
Lord Burns: But this is snail’s-pace correction.
Jonathan Athow: The Chancellor recently appeared before the Treasury Committee saying that the direction of travel was away from RPI but that he was having to think about the fiscal consequences of that. I think the direction of travel is clear, but it is probably for others to say whether we are moving at a snail’s pace.
Sir David Norgrove: I agree. I am not sure that it is sustainable until 2068. Methods of collecting and analysing prices will change significantly. If the RPI cannot change in the same way, it will diverge ever more. I am not sure from that point of view that we can manage it.
Lord Burns: So do I take it that you both feel deeply uncomfortable with the position that we now find ourselves in?
Sir David Norgrove: Yes, definitely. The big one is the bond market and how we move away from the RPI there given the quantity, not just of gilts but private sector bonds.
Lord Burns: I was involved in these issues at one stage of my life—I think Lord Lamont referred to it earlier—when I tried to persuade the Retail Prices Index Advisory Committee. I do not know whether it still exists. I was told that we would have to buy back all the bonds. At that stage, the market value went the other way. It would have ended up costing the Government somewhat. Will there ever be a better time to do this, when the chances of the market wanting to sell us these bonds back is extremely unlikely?
Jonathan Athow: I think we are bound to think of the statistical issues and leave others to think of the public finance consequences.
Sir David Norgrove: When the Governor appeared before this Committee, he talked of a length of notice—so making the change in 10 years’ time.
Lord Burns: But announcing the change at an earlier point.
Q15 Lord Darling of Roulanish: Sir David, you were admirably succinct in reply to Lord Burns’ first question on whether you agreed with what Paul Johnson said. You said that you agreed with everything. I am puzzled: you are the chair of the United Kingdom Statistics Authority. Your job, briefly put, is to maintain the integrity of the statistics that this country produces.
If that is the case, and given everything that we have heard this afternoon, not just from the previous two witnesses but from the two of you so far, what is the statistics authority doing about it? Should you not say to the Government, “Look, this has to stop. You’ve got to do something about it if it’s a statutory remedy that you need”?
I am surprised that you are confining yourself to agreeing with the witness this afternoon; you should be saying something.
Sir David Norgrove: We have said it. I have said so very publicly, as did my predecessor. We have not been reticent about it.
Lord Darling of Roulanish: But have you gone to the Chancellor and said, “You need to change this”?
Sir David Norgrove: I could go through the stately dance of going to the Bank of England and saying that we want to change it. It would then say, “It’s a material detriment. Go to the Chancellor”. The Chancellor would then refuse to change it. We may need to do that, but I would much rather work with the people concerned to see whether we can come up with a solution.
Lord Darling of Roulanish: What work are you doing then to come up a solution?
Sir David Norgrove: We are having discussions, have had discussions in the past and continue to have them with both the Bank and the Treasury.
Lord Darling of Roulanish: It sounds a bit “Yes Minister”-ish. Having been in that position in the past, I understand why you might be saying it. You will understand just from the last hour or so that it is becoming abundantly clear that what we have is just not satisfactory. Surely you should be doing something more proactive than just having discussions.
Sir David Norgrove: We could write letters. I do not know that that would do anything more than have discussions. People are very clear about the problems, and the Chancellor has acknowledged that.
Lord Darling of Roulanish: But you are effectively saying that nothing is going to happen.
Sir David Norgrove: No, I am not saying that. The probing from this Committee and others is very helpful in this. I feel that things are now beginning to move more quickly than they have since 2013.
The Chairman: Moving more quickly to what?
Sir David Norgrove: I hope to finding a solution, particularly in relation to the future of the bond market.
Q16 Lord Lamont of Lerwick: How important is it to have housing included in an index?
Jonathan Athow: It really depends on your use of the index. Many people judge an index to say whether standards of living are rising or falling. That would be one use of it. In that case, you want a measure that captures the whole of people’s spending and consumption, and in that sense you would want a measure that included housing in it.
One of the challenges about encouraging people to move away from RPI—it was a legitimate challenge early on—was that there was no alternative that included housing costs. We had CPI, which in many ways is a good measure but does not have housing costs. When people were looking for something to replicate the coverage of RPI, it was not available.
Now there is, with CPIH. CPIH has been given its national statistic status. In that sense, we think it is important to focus on that wider basket if you are interested in living standards. If you are interested in what is happening with interest rates, the Bank of England’s remit for monetary policy is set in terms of CPI, so you would be interested in that, but CPIH is a more comprehensive and stronger measure.
Lord Lamont of Lerwick: Right, but you began your reply by referring to coverage, as though there was some criterion other than coverage.
Jonathan Athow: No, that is the main one. If you were looking at living standards, you would want something that included all of it.
Lord Lamont of Lerwick: Yes, it would seem to follow that whatever the methodological deficiencies between RPI and CPI, the CPI is also deficient in a completely different way.
Jonathan Athow: Indeed. As I said, the CPI has some strengths, but the fact that it does not have owner-occupied housing costs is its major weakness. That was why we developed the CPIH: because we think it is more comprehensive and therefore stronger.
Sir David Norgrove: The Governor said in his evidence to this Committee that he would hope in due course to look towards moving to the CPIH for inflation targeting.
Q17 Lord Layard: Can you clarify the legal situation to us? If you think that the CPIH is the best measure—coming back to the integrity point—what is to stop you keeping the RPI and using the methodology of the CPIH to calculate it from now on? You have changed in the past the way you calculate the RPI. Why can you not change it again to the way in which you calculate the CPIH?
Jonathan Athow: The 2007 legislation sets down two steps if there is to be any change to RPI. The first stage is to determine whether it is a fundamental change to the methods or the data collection. That is determined by the Bank of England. If it is fundamental, it also has to determine whether it is detrimental to the owners of index-linked gilts: that is, does it reduce their payments? In which case, it is then remitted to the Chancellor for him to decide whether the change can be made.
Lord Burns: Is there sterling advice that can be received at any stage in the process?
Jonathan Athow: Yes. We have advisory committees at the moment. They advise the National Statistician; it is the National Statistician who in law makes the decisions here. Obviously, that then needs to be approved. Any change like that, even if it is a very simple swapping of CPIH for RPI, would require that process to be gone through and a potential sign-off by the Chancellor.
Lord Layard: Supposing that our group in this Committee recommends that that is what you do, can you say a little more about the politics? One obvious advantage is that pension funds would presumably become a lot more solvent. That is a huge advantage.
Jonathan Athow: Yes.
Lord Layard: They would suffer on gilts, but overall they would become much more solvent.
Jonathan Athow: Yes, but many millions of pensioners who currently receive RPI-linked payments would get less of an uplift.
Lord Layard: As a recipient, I have no idea on what basis I will be uplifted. If I am told that I will be uplifted on the correct basis, I do not see how I will have any complaint.
The Chairman: There is a man who has never stood for election.
Jonathan Athow: In preparation for this Committee, I looked back on the consultation responses. Many people believed that the way RPI was constructed was more representative of their cost of living. As I said, a number of private sector companies have lots of bonds linked to RPI. If we made changes to the RPI, they were potentially in the position of having to redeem those bonds. The consequences here are quite wide-ranging. RPI is baked into a lot of private sector contracts, so a change to RPI could have far-reaching ramifications.
The Chairman: You have obviously thought about this very carefully. Would you be able to give the Committee a note on these consequences? am not sure that we can go through all of them today.
Jonathan Athow: Yes.
Lord Layard: That was my last point.
Q18 Lord Turnbull: Mr Athow, in answer to Lord Burns, you said that people bought indexed instruments on the basis of what they were at the time. The Giles article says that they did not buy them as they were but on the basis of the prospectus as it was at the time. There was a change in 2002, but his claim is that since that time, nothing in the index links says that issuers cannot change the price to which they are linked.
Is that correct if these two assurances have to be provided by the Bank of England? Is it fundamental, immaterial and detrimental? Are those two pieces of information that they have to take into account, or are they in a sense vetoes on a change?
Jonathan Athow: Clearly, as I said, the Chancellor would make the decisions. In that sense, he has the right of veto over the change if those conditions were met.
On the prospectus, especially earlier ones, I cannot remember when the last ones were issued. I think it was about 2002.
Lord Turnbull: That is what he said: that 6% of them are grandfathered, protected. For the rest, the Government have the power to make the change.
Jonathan Athow: That is true, but private sector bonds also have similar conditions. Some of the respondents to the 2012 consultation identified those. They are on the public record. Companies such as Network Rail said that they had considerable bonds with similar conditions to the 2002 index-linked gilts, although they did not provide the prospectus.
Q19 Lord Turnbull: Do you think it is impossible for the pension funds and the people holding the pensions to reach the kind of agreements that are going on all the time? Employers are revisiting the terms of their pension funds. They just bring this third factor into account.
Jonathan Athow: That was the nature of our statement in 2013. We encouraged people to say, “We should be moving away from RPI. It’s not a good measure. There are now better measures that you can move to, and we will provide that direction of travel”.
We also recognise that there were long-standing contracts where a change to RPI could be very disruptive. That was our statement, encouraging everybody—the Government and the private sector—to say, “Start moving away from RPI. It’s not a good measure of inflation. There are a number of problems with it. Use a better alternative”. That has basically been our position since January 2013.
Lord Layard: I think I have asked my question.
Sir David Norgrove: I should declare an interest.
Lord Layard: So should I. We all do.
Lord Burns: I have an index-linked pension.
The Chairman: If you did not have to go through the Chancellor and you had freedom to do what you thought was right, Sir David, what would you do?
Sir David Norgrove: Abolish RPI.
Jonathan Athow: If there was no user need in the way that has been specified, we would not want to continue with RPI. As I said, in 2012 we heard that, certainly for the foreseeable future, there was no need for RPI. Without that user need, we would not want to use it, because we do not think that it is a good measure. As we set out at the time, we consulted on this and looked at those user needs. We describe it as a legacy need. We very much hope that RPI quickly becomes a legacy measure, used for fewer contracts.
The Chairman: Does that mean that you would do nothing?
Jonathan Athow: As I said, if there were no user need—
The Chairman: We are where we are. You are putting a conditionality on it.
Sir David Norgrove: My answer was probably too short, from that point of view. I think I would give notice, in the way that the governor suggested, of probably about 10 years in the expectation that, over that time, prices would adjust and people would be less hard done by.
Jonathan Athow: I think it is a reasonable course of action to think about those arrangements. It gives people enough time to negotiate new deals.
The Chairman: Suddenly, that has stimulated some questions.
Lord Darling of Roulanish: Chairman, you asked the witnesses to provide evidence of the pros and cons. I think we would find it very useful to get some quantification on the numbers that would be affected, because this is politically highly sensitive.
I am also interested in the melioration that we were just talking about. If you were to say, “This will happen in 10 years’ time”—that is probably the longest horizon you can use—it might be possible to quantify what the effect might be.
The Chairman: It is difficult to see why we can say that no one will be allowed to sell cars with internal combustion engines when this is not a possible remedy in future.
Sir David Norgrove: There are certain areas where we would struggle to find numbers, such as where people have bought RPI-linked annuities. I am not sure that there are data available on that kind of thing.
Q20 Lord Burns: Chris Giles mentioned earlier the statement that was made, which then brought about the sharp improvement in the gilts market. Can you remind us of the strength of commitment in that statement? Was any length of time associated with it, or did it give the impression that the decision was final?
Jonathan Athow: It was not caveated, so it was very much presented as a final statement that the RPI would continue and be maintained through routine changes. I think that was the phrase. We can provide you with the details.
Lord Layard: Sir David, why did you say that you would abolish the RPI? That would leave a great vacuum where people have contracts related to the RPI. Why would you not do what I think Chris Giles suggested, which is to turn the RPI into the CPIH?
Sir David Norgrove: We could rename CPIH as RPI. That would be much more technical and we would want to consult on what made sense from the point of view of various contracts. There is an honesty about keeping CPI and CPIH and doing away with RPI, rather than just renaming it.
Lord Turnbull: I am puzzled by your choice of second best. You seemed to say, “I would abolish RPI, but if I can’t I’ll stick with the present scheme”, whereas the two previous witnesses said that you could correct the strappy tops problem and its application of that method to those kinds of prices, which would leave you with an RPI that was a good deal better than the present one. I would have thought that was within the bounds of reasonable methodological change.
Jonathan Athow: If we were to try to do that, it would certainly trigger the same clauses that would lead to the Bank of England and the Treasury having to look at it. There are two reasons why that is not particularly desirable. One is the question whether we could actually achieve it. Could we go back to the 2010 clothing price correction? The clothing market has moved on. Could we try to wind back to the 2010 method? I do not know whether that is possible. I am also reluctant about half-fixing it. It is like having a completely broken down car and putting new tyres on it.
Lord Burns: It is still better to drive the whole car with new tyres than no car with bad tyres.
Jonathan Athow: I am also thinking of my successor’s successor’s successor and wanting to put them in a place where they have the best methods over the long term. Therefore, trying to fix one part of it when there are so many shortcomings is challenging.
Lord Burns: This is the danger of the best driving out the good: saying that we must wait for the complete solution, otherwise it is not worth doing anything about it. In observing politics, that often strikes me as a dangerous task, because the “no change” can go on for a good deal longer than you ever imagined. Therefore, it is better to do a certain amount of maintenance when you get the opportunities than just to let it be.
Jonathan Athow: It would not be the best being the enemy of the good. I would not describe an RPI rewound to 2010 as anywhere near good. We would be baking in—
Lord Turnbull: It would not be rewound; it would be taking whatever the RPI is now and then adding its progression thereafter on a better basis.
Jonathan Athow: Indeed, but we could end up with new problems from price collection. We talked about clothing. Many markets are going through a similar change. We could end up with similar problems in future—you would not necessarily be future-proofing it. As I said, we are starting to see movement. The Chancellor has said that he wants to move away from RPI.
Lord Turnbull: I do not buy this argument that other changes will come. You as the guardians of the methodology should keep adapting this index in the light of the way markets and products develop over time. You cannot simply say, “We’re not going to change because there will be future developments”. There are bound to be. You are bound to have to come up with modifications. This is another in that series.
Jonathan Athow: Indeed. If this was a change to the clothing price collection, for example, we might have to make a change to other price collections. That could exacerbate the problem. All I am saying is that you might try to correct for one problem—
Lord Turnbull: You would not exacerbate the problem, you would have corrected one problem. Another problem arises and then you bring in a correction for that. So year by year—it may not be every year—you are gradually keeping the methodology up to date with the way in which the market is operating.
Jonathan Athow: I fear that the RPI would then collapse under the weight of the sticking plasters that we had applied to it, because essentially that is what you would be trying to do.
Sir David Norgrove: If I may, I think this conversation is showing us the difficulty of maintaining this index over the next 50 years and the need for us to look for a more fundamental set of changes.
Q21 Lord Tugendhat: Given the deficiencies of the two indices that have emerged this afternoon and given all the changes that are taking place in the economy, epitomised by Airbnb, Deliveroo and all the rest of it, is it your impression, taking the two together, that the rate of inflation is underestimated or overestimated? Are prices going up so far as ordinary members of the public are concerned? Do you feel that we are underestimating or overestimating?
Jonathan Athow: You are certainly right that the changes in the economy are making prices harder to measure. The quality of goods—a smartphone changes all the time, with new cameras, new screens—is making it harder. We try to adjust for the fact that there are new features on phones and such things. We adjust for quality. In that sense, we are trying to pick up those elements. We probably do not do it perfectly.
The question whether we are under or overestimating is challenging. Are we trying to leave people no worse off adjusting for all this quality? Those would be the sort of questions asked. Certainly, we see the RPI as being likely to be too high compared to a measure such as CPIH. We think that RPI, other things being equal, overstates inflation, but there will be people who argue about whether CPIH properly adjusts for the quality of new goods and new services. Some people argue that inflation might be a little lower if we were fully to adjust for the new quality of the all these digital services.
Lord Tugendhat: My gut feeling, which I have no confidence in whatever, is that actually prices have not been rising as fast as the indices indicate; that they have been rising more slowly, taking into account all the factors that we have been talking about. I do not know whether other people’s gut feelings would be the same, but I am trying to understand where your gut feeling would be.
Jonathan Athow: Given my job, I try to avoid gut feelings and to focus on statistics and facts. Some people have made that case, but we also get a number of people who argue exactly the opposite: that the way we treat some of these elements of quality change understates inflation. Some people worry about those issues. I do not think the view would be universally shared.
The Chairman: In the previous session, at which you were present, I asked whether people agreed with the Governor of the Bank of England that there was a 0.7% difference. Do you agree with that?
Jonathan Athow: Again, it is very difficult to come up with a definitive view.
The Chairman: Is he right or wrong?
Jonathan Athow: That is a pretty good estimate.
Q22 Lord Lamont of Lerwick: Going back to something you said earlier, the idea that you might keep the index and the level of the index and then build on top of it by a different methodology strikes me as bizarre. Have there ever been changes in the history of the RPI that would be comparable to changing the Carli methodology? Obviously, you make changes in the RPI to do with changes in spending habits—new products come on to the market that were not there before. In that sense, you change the RPI. Have there been other changes of a methodological kind that would be comparable to tampering with the Carli method?
Jonathan Athow: The only change that I am aware of was a change made in the early 1990s to the way in which housing was done in the RPI. I do not think there has been such a change in the past 25 years. I may be unaware of some changes; we can let you know if there was anything comparable. There have been changes when new data sources and new data collection methods have come in, but there has been nothing comparable to swapping the Carli method.
Lord Lamont of Lerwick: That makes the case against the point that Lord Layard was arguing: that you should not abolish but change. You would have to abolish.
The Chairman: On that note, I thank you very much for coming to the Committee and helping us with this issue, which is considerably more complicated than perhaps some of us imagined when we started this inquiry. Thank you very much.
