Treasury Committee
Oral evidence: Reappointment hearing for Chair of the Office for Budget Responsibility, HC 459
Tuesday 15 September 2015
Ordered by the House of Commons to be published on 15 September 2015
Members present: Andrew Tyrie (Chair); Mr Steve Baker, Bill Esterson, Mark Garnier, Helen Goodman, Stephen Hammond, George Kerevan, John Mann, Chris Philp, Mr Jacob Rees-Mogg, Wes Streeting
Questions 1-79
Witness: Robert Chote, Chair, Office for Budget Responsibility, gave evidence.
Q1 Chair: Good morning, Mr Chote. We are here to consider your reappointment. As you know, we have a statutory duty to make up our minds whether you should be reappointed, just as we have a statutory duty to think through whether, were the Chancellor to decide so, you should be dismissed. We are able to protect you from dismissal; we are also able to decide whether or not to appoint you. With that in mind, the independence of the OBR is crucial. Indeed, that is why we are involved in your appointment to the degree that we are. It is to that that I would like to turn this morning in the light of The Times’ release of information under FOI that was put in the paper on Monday. Under the arrangements that were put in place for contact between the Treasury and the OBR, a log of substantive contact was drawn up. That was also in response to one of the Treasury Select Committee’s recommendations in 2010. That log excludes the contacts. Why is that? Should we widen the log?
Robert Chote: The log refers explicitly to contacts between us and ministers, ministerial private offices and special advisers. It does not apply to the exchange of information at a working level.
Q2 Chair: Should it?
Robert Chote: No, I do not think it should.
Q3 Chair: Why not?
Robert Chote: There is a much larger volume of interaction at a working level and we rely on a candid flow of information in both directions. We get useful intelligence from that—often deliberately, sometimes inadvertently—which I would be loath to lose.
Q4 Chair: On the specific material that has been released, phrases like “rather convoluted” and “top-sliced” were removed. I am not going to take you through these all in exhaustive detail. These contacts are supposed to be limited to corrections of errors of fact and not to stray into opinion. Do you think these are merely requests for factual changes?
Robert Chote: Drawing a line between fact and interpretation is difficult to do. Personally I am relaxed about the idea of particular officials—in this case we are dealing with a relatively junior official—wanting to offer us unsolicited drafting advice. We would be in a pretty weak state if we were not able to be robust to that. Sometimes the things you get are sensible and useful; sometimes they are not. At the end of the day, we write what we want, and the record shows that we do not fight shy of using phraseology that perhaps the Chancellor would rather we did not.
Q5 Chair: Why do you not just give the Committee some examples of that to illustrate what you consider to be your own independence over the last five years?
Robert Chote: If you take the couple that you mentioned, on the point of “rather convoluted” I felt that rather than use that sort of phrase the better thing to do was to demonstrate it by releasing a paper showing how this ridiculous formulation had evolved over the whole period of the Parliament, which is what we did. I felt that made the point clearly enough. In terms of whether we were underplaying the impact of the spending cuts—in this particular episode we are talking about the December 2015 EFO—that was front and centre in the executive summary. We referred to spending falling to its lowest share of GDP since the 1930s, and I suspect that is not phraseology that the Chancellor would have chosen for himself. Similarly, I do not think they were very pleased with “rollercoaster” as a description of the path of departmental expenditure limits in March. Some people suggested to me that they felt that was inappropriate; I said I thought that the language was necessary to create an impression to a wide audience of stakeholders.
Q6 Chair: You have given us a few examples there. Why do you think this official in the Treasury, by his own admission, said that he was not limiting his comments to matters of a factual nature—I will not read it out, but that is what he said in his own email amounts to—when he should have known very well that that is exactly what he should be doing?
Robert Chote: They know very well that we write what we like and if they want to offer views on that we are robust to it and we will say what we want. I feel entirely relaxed about it.
Q7 Chair: So, Treasury officials have made a mistake but you have fended them off. Is that what you are telling us today?
Robert Chote: No. The memorandum of understanding and the distinction that makes is an important backstop for dealing with what I would regard as being a serious interference. Serious interference I would regard as being suggestions to change things that are repeated, suggestions that are accompanied by menaces or threats, or that sort of thing. It is not to preclude somebody offering a bit of unsolicited drafting advice.
Q8 Chair: What kind of threats do you have in mind?
Robert Chote: In 20 years as a journalist and at the IFS writing about Treasury policy, I have been treated to the “we know where you live” and “you will never work in this town again” routine more times than I care to remember. It did not affect me then; it would not affect me now. If I was being put under serious pressure by the Chancellor, by private offices or by senior officials, I would tell them to buzz off and I would tell you what was going on, but I am not going to sit here and beat up on a hardworking Grade 6 for having the temerity to offer us unsolicited drafting advice.
Q9 Chair: So you do not think that the Treasury official made a mistake by saying that he was straying beyond making factual comment.
Robert Chote: Not a mistake that I feel offended or moved to alter my actions by.
Q10 Chair: You have described your battle-hardened nature there. Have you not also made the point that Mr Page, the Canadian, made in his report about your organisation: that it is very dependent on a robust top man? You have been through, as you have put it, 20 years of warfare from time to time and had these terrible experiences where they say things like, “We know where you live”. I must admit I have never had that myself, although I do occasionally get these calls myself, but I have had other efforts made from time to time. Do you think that your successor—a subsequent person—might find it more difficult to fend it off if they have not had that sort of experience?
Robert Chote: I am sure there are plenty of people who could do it. I certainly feel that that experience was quite useful as having toughened the skin in preparation for this job.
Q11 Chair: I think we had better check out this independence point, just to make absolutely sure. I would like your agreement to supply us with all exchanges that took place between receipt of the pre-release text of the Economic and Fiscal Outlook and its publication just under two days later.
Robert Chote: Yes, that would be alright.
Chair: I think we had better see all correspondence and a log of all contacts of any type. I realise that this might involve a bit of bureaucracy, but I think we had better see it and reflect on it. I will ask one of the Committee staff to go through this in some detail with you; then I will have a private discussion with the Committee and decide whether we need to take further hearings on it.
Robert Chote: Fine.
Chair: You are happy with that.
Robert Chote: Yes. From our point of view, that is fine. Whether the Treasury would have any issues—I would doubt it.
Q12 Bill Esterson: Good morning, Robert. I was just curious about what kind of pressure you had been put under by ministers or senior officials.
Robert Chote: During this period—the last five years?
Bill Esterson: Yes.
Robert Chote: Very little. In a sense, the surprise is the fact that we have got five years in and I have not read—that may just be because I do not read widely enough—a single disobliging newspaper article referring to “sources close to the Chancellor” saying we are clowns or frauds or doing things inappropriately. If you had asked me five years ago what the nature of this relationship could be, I would have said I would be astonished not to have that. Partly—and other fiscal councils have found this—when you are in the first flush and you have the people who set you up running the relationship to begin with, there is a tendency to deal with that relatively carefully, but whatever you think of the policies, the Chancellor has always dealt with us very straightforwardly and very courteously. As I say, I have not had the threats that five years ago if you had asked me—you probably did ask me—I would have said I probably would have had.
Q13 Bill Esterson: Do you think the public perhaps would have greater faith in you—I am not saying they do not have any faith in you—if there had been those sorts of criticisms? Do you think there is a danger that the lack of that sort of vibrant exchange might suggest that the relationship is too close?
Robert Chote: That vibrant exchange will be taking place behind the scenes, as it were, anyway. The public can take reassurance from the fact that we say things that are inconvenient to the Government of the day when we feel that is appropriate, not just for the sake of it. There have been numerous occasions on which we have said that the Government was not on course to achieve the particular targets that it had set itself in the previous Parliament. Rightly, it is for elected ministers to decide whether to accept that and change policy as a result of that or to say, “Well, okay, we accept the OBR’s view but we think it is better to miss the target”, so that is useful. As I say, what I have tried to do and what I think we have done in the EFO is to recognise the fact that Governments have particular objectives in the fiscal path that they are setting out, which include the formal fiscal targets but also include informal things. The Chancellor may be very concerned about whether you still have a surplus in a particular year or whether the deficit is moving in a straight line over a particular period. To get all of that to add up, assumptions need to be made about paths of spending and particular measures. What we need to do is to point out what is having to be assumed and what is having to be done to ensure they can tick all those boxes.
In a sense, achieving all those various objectives is like stamping down on the lumps in a carpet: you stamp, stamp, stamp but there is one lump in the corner of the carpet that just gets bigger and bigger. Our job is to say: “Yes, they have achieved all of these things, but look at that lump over there in the carpet”. That is, for example, why we emphasised the fact that there were a variety of objectives that the Government clearly wanted to get at in the March 2015 Budget but that the only way to do that was to end up with what looked like a rather strange profile for departmental spending and announcements of asset sales. It is not just saying either, “You have achieved this” or “You have not” but explaining the mechanics of how you got there. That is what builds up a robustness. It is not just saying, “We have ticked this and said, ‘Yes, they are fine’ or ‘No, they are not’”; you have to explain, if they are doing it, how they have done it.
Q14 Bill Esterson: It is very difficult for most people to understand economics and forecasting. How well do you think you have explained to the public, and how well do you think the public have understood, what you do?
Robert Chote: You are dealing with, in a sense, two groups of people. There are the cognoscenti, who are regular consumers of the forecasts that we are producing. For them, the key thing is to be able to demonstrate to their satisfaction why a forecast has moved from the one you produced at time T to the one you produced at time T+1—so, explaining why the income tax forecast has moved and what the impact of policy measures has been. That is the key to demonstrating that you can explain—people might not agree with your analysis—why it is that things have moved and that this is not politically-motivated wishful thinking reflected in the new forecast. So, there is a group of people for whom the difference between the last forecast you did and the next forecast you did is key to that level of understanding.
There is—the Chairman has referred to this in the past—a much wider proportion of the population who are not regular consumers and readers of Annex B of the EFO. There is not the amount of empirical evidence on this. The Times did a Gallup thing some while ago showing what I thought was surprisingly high awareness of the work that we did. I cannot remember the precise figures. There are always people who say, “Well, you forecast it would be X.X last time. It has not turned out to be correct. How do we know you are going to be right next time?” and you have to explain to them that you do not know you are going to be right. We have done our bit, but, as the Chairman said, there is always an impetus to try to do more there.
Bill Esterson: You have to balance the needs of these two groups. That is the challenge.
Robert Chote: It is one issue that comes down—this came up with the Ramsden review—to the structure of the document. I am conscious that it is a weighty read, but we have a selection of audiences that are fascinated by four pages in it and are very upset if those pages are not there, but do not care about the remaining 190. You cannot emphasise what everybody is interested in all the time; you just have to make sure it is comprehensive and you can pull the key highlights out for key audiences.
Q15 Bill Esterson: Are there things you would like to do over the next five years to improve understanding among the general public?
Robert Chote: Yes. One thing that we started off doing in the last Parliament but extra fiscal events and now the Ramsden review got in the way a bit is producing more non‑technical publications. We have done a brief guide to the public finances and road-tested that with people who are teaching economics in schools, etc. They have found that to be quite useful. We could do more of those sorts of things. The other possibility, again with more resources, is that you could use the website more imaginatively. We have somebody who has just arrived who is hopefully going to be able to help us use data visualisation stuff more effectively so people can get at some of the key numbers more easily. We could marshal together information—not just the stuff that we have provided but stuff that we know our consumers are interested in—so you do not have to chase around to find HMRC’s documents on the tax ready reckoners and those sorts of things. With the time and resources, there are things we could do.
Q16 Bill Esterson: During 2013-14, your forecasts overestimated productivity and wage growth and underestimated employment growth. Is there a danger that in less good times forecasters are over-optimistic and, conversely, you become more pessimistic in better times?
Robert Chote: There are always those dangers. It is a median forecast; there is a 50% chance, in our own view, it is wrong in either direction. If you are looking for the single greatest forecasting challenge that we and everybody else have confronted over the last five years certainly, it has been the productivity puzzle. There have been issues about the pace of economic growth—we were too pessimistic to begin with, then too optimistic, then too pessimistic, and now the economy is chugging along at the 2% to 3% that makes you look like a good forecaster but it is just different circumstances—but a consistent path through that has been the productivity puzzle. Employment has performed far better than you would have anticipated even if the good fairy had come and told you exactly what was going to happen to GDP growth. That is then linked to what has happened to wage growth, and that then is linked to what happens to things like the effect of tax rates on labour income. The difficulty with that is: if you have a puzzle like that, do you assume that that is the new normal and things are never going to return to normal, do you assume we have lost a chunk of activity and we will return to the sorts of long-term growth rates we had in the past but never make up the difference, or do you assume that we are going to have some extended period of really very strong growth that will get us back up to the line we were on to start with? I do not know what the eventual answer to that question is, and there is clearly a very high chance that we could pick the wrong alternative. That is why we have to set out what difference it would make to the fiscal position if we are wrong in those different directions.
Q17 Bill Esterson: Stephen Nickell used the term “pretty dire” to describe the consequences of a failure of recovery in productivity growth. Are you intrinsically reluctant to make judgments that lead to use of those sorts of terms when you are forecasting?
Robert Chote: I would not say so. The judgments that we have made about the loss of potential GDP relative to the pre-crisis trend I think many people, certainly before the event, would describe as “pretty dire” under any circumstances. It all depends on your level of direness. No. We give our best judgment on the basis of the information as we have it, but—and this is one of the first things I said when I was at this hearing five years ago—what we have done is to say, “Look, it is not enough just to say this is your best view; you need also to say what the alternative states of the world are that could emerge and what difference it would make”. There are some alternative states that do not have a great deal of impact on the fiscal position; there are others that do. We need to spell that out.
Q18 Mark Garnier: Could I just turn to the quality of the data going into your estimates? There has been some criticism of the ONS that they are not particularly good at their job. What is your feeling about it?
Robert Chote: There are areas where we have had doubts over particular sets of data at times. I look forward with great interest to seeing what Charlie Bean’s review concludes. As I have said here before, for economic forecasters to complain about the outturn data is like sailors complaining about the sea; that is what we have to sail on. Measuring a large, complex economy is not a straightforward thing to do. There are areas where quality assurance could be stepped up. The ONS has taken a lot of steps in that direction and I am sure there is more it could do.
Q19 Mark Garnier: To use your simile there, if a sailor is worried about the sea, they tend to go and sit in the clubhouse drinking pink gins while the weather gets better. What are you doing at the moment? Are you dealing with the choppy waters or are you sitting around drinking pink gins?
Chair: Are you describing your sailing habits there, Mark?
Mark Garnier: Possibly.
Robert Chote: The Chancellor and Parliament instruct us to go out on the seas at times not necessarily of our own choosing, and we then do it to the best of our ability.
Q20 Mark Garnier: Let us hope you have not drunk too many pink gins before you go out there. There is a serious point: at the end of the day, if you are getting duff data coming in, then clearly you are going to have trouble coming up with decent data. Are you getting involved with the ONS and also trying to help them find where they are getting it wrong?
Robert Chote: Yes. We have a series of interactions with the ONS on a regular basis. There are user groups for people who are using particular bits of data. Occasionally there are things where we as users are looking at the numbers and they do not seem to make sense, and we can bring them to those groups and see whether other people are having the same sort of experiences and the ONS can work on that. Our experience has been that the ONS are very receptive to helping when those sorts of things can be surfaced. But data are revised and, however good they are, they can be revised quite significantly. One of the most interesting things about one of our documents that I would describe as a niche publication in terms of the demographic it pitches to, our Forecast Evaluation Report, is looking back at how the story that you had for explaining why things had turned out unexpectedly one year after the event can then turn out to be completely different two years after the event. We thought some while ago that the reason the budget deficit had not performed worse as the economy failed to recover as quickly as we anticipated is because the nominal picture was more robust than the real picture, and that was what the outturn data suggested. Go on a couple of Blue Books and you suddenly discover it is not the relationship between real and nominal but the composition of nominal. The reason that the deficit did not deteriorate more in the 2011-12 period was that the bits of nominal GDP that were weakest relative to expectations were the bits you do not get much money from.
Q21 Mark Garnier: Fair enough. But the important point is you are getting involved with other commentators in terms of trying to drive better outcomes from the ONS.
Robert Chote: Yes.
Q22 Chris Philp: I would like to ask about the timing of your reappointment. I am aware that the deadline was 7 September, about a month before the term expires, and the announcement was made on 3 September. Although that timing met the statutory requirements, can you comment to the Committee on whether, from a practical and operational point of view, you feel that one-month notice period is adequate, speaking both personally and on behalf of the organisation?
Robert Chote: Speaking personally and I suspect as well on behalf of the organisation, a longer period of knowing what the outcome was going to be would be helpful. If the Chancellor decided not to reappoint me or you decide to boot me out, then time is short to get a replacement in place to take over. My two fellow members of the Budget Responsibility Committee were reappointed much earlier before their terms came up, and I had assumed that I would be either reappointed or booted out on the same sort of timetable, but the Treasury was concerned that with the election intervening they did not want to make an appointment prior to the election that only took effect after the election. My entirely personal view is that it would probably be better in future for my successors, if that situation arises again, for there to be some informal or formal consultation between the parties prior to the election so that there is somebody that everybody is happy with and you can do that in good time, but I am not a constitutional lawyer and there may be objections to that.
Q23 Chris Philp: Were you given any informal indication prior to 3 September about your reappointment? If so, when?
Robert Chote: Nobody came over and dropped any black bin liners on the desk. There was nothing pointing particularly in one direction or another. I was happy to wait. I had hoped that I had done a reasonably good job and that that might be viewed positively, but there are no guarantees.
Q24 Chris Philp: You were given no indication prior to 3 September.
Robert Chote: No, I had no conversation with the Chancellor about whether he wanted to reappoint me at any period prior to being reappointed.
Q25 Chris Philp: Or with anyone else.
Robert Chote: No. I was asked whether I would want to carry on. I said yes, and the person who asked me that did not then say, “You will be lucky”, or anything like that. I might have taken some reassurance from that.
Q26 Chris Philp: Yes. When was that question posed?
Robert Chote: I cannot remember.
Chris Philp: Roughly. Which month?
Robert Chote: No idea. Certainly at some point prior to the election and after the election, but there is no specific—
Q27 Chris Philp: You mentioned the sensitivity with a new appointment or a confirmation crossing an election period, which I think the Committee would understand, but the election was on 7 May, so there was a period of four months—the months of May, June, July and August—during which an appointment could have been confirmed after the election but prior to 7 September. In the future, might that be an appropriate timeframe?
Robert Chote: It might be, but I am also conscious that the Government had a substantive Budget package to get through in that period, so I suspect people’s minds were occupied on other things. As it then moved later, once we were getting to around this time, I suspect that, perfectly reasonably, they thought it made sense to announce the reappointment at the same time as the publication of the Ramsden review, as there was a logical assessment to do at the same time.
Q28 Chris Philp: Aside from those sorts of considerations—elections and third-party reviews—what do you think would be an appropriate period of notice, both from the point of view of your prospective successors and also from the point of view of the organisation?
Robert Chote: You want to do it in such a time that if a new person is coming in they have a chance to bed themselves in and know the ropes before the next fiscal event is coming up. If the other two members of the Budget Responsibility Committee have been there for a while, that would provide you with greater reassurance on that score, but it would be very difficult for somebody suddenly to be parachuted into this job when you are already in the middle of a Budget or an Autumn Statement forecasting round. For example, the Government will have to find a replacement for Steve Nickell, who said he would serve three years and those three years end next autumn. Identifying somebody in good time for them to be confident of being able to start and be ready for bedding in for the Autumn Statement—
Q29 Chair: What is that timeframe? That is the question. Is that three months? Is it six months? What is the minimum that you think is reasonable?
Robert Chote: If you were doing an appointment then, I would have thought it would be a good idea to have you holding your confirmation hearing before Parliament rises for the summer to appoint somebody who is taking over in the autumn, ideally, rather than to be doing it now—if you did not have an election in the way.
Q30 Chair: If you do not mind, Chris, I just want to go back to the point you made a moment ago. You have not really given a full explanation for why the Government could not have thought through what it wanted to do on the basis that it did get elected and then appoint you immediately after the General Election.
Robert Chote: You would have to ask them what their thinking was on that.
Chair: You have given them a sort of apology for it by saying there might have been some symmetry with publishing it at the same time as the Ramsden review, but that is pretty thin, is it not? The truth is that your announcement could and should have been made much earlier.
Robert Chote: That would have been my preference. Whether you would have objections to them announcing it when Parliament is not sitting, I do not know.
Q31 Chris Philp: Just to conclude the Chairman’s question in terms of the number of months in advance it would be helpful to have any member of the Budget Responsibility Committee’s appointment announced, could you just give us the minimum number of months that you think would be practically, operationally and personally reasonable? You have hinted at it, but it would be helpful to have a number.
Robert Chote: As I say, if somebody is leaving in the autumn—if it is being renewed at this time—which is at the moment the timescale on which the three of us are operating, having a situation where you have had a chance to consider them before Parliament rises in the previous summer would be logical.
Q32 Chair: It is nice of you to consider us, but that is not the question. The question is: how long do they need to bed themselves down into the job?
Robert Chote: That would be reasonable enough, because you then have the summer to read yourself in and the early period of the autumn.
Q33 Chair: So is it three months or six months?
Robert Chote: Well, you rise in July, so—
Chair: Do not worry about our timetable. I did not ask you about our timetable. I am asking you about your working methods. How long does this person need to bed themselves into the job?
Robert Chote: I would say an appointment three or four months before commencing the next forecasting round.
Q34 Chair: How long is the forecasting round?
Robert Chote: Six to eight weeks prior to the Autumn Statement.
Chair: So you are talking about three months plus six weeks minimum.
Robert Chote: Yes.
Chair: Four and a half months. Thanks. We have the answer.
Q35 Chris Philp: Just on a slightly different topic before finishing, there has been some debate over the past few years about which top rate of tax optimises the tax yield. The tax rate has come down from 50% to 45%; there has been some debate about whether it might yield more at 40%. Have you done any work yourselves on that question?
Robert Chote: As you will appreciate, we are required by you only to look at current Government policy, so we would not explore alternative policy options on that basis. The last time we went through this sort of analysis, which was when you had the move from the 50% rate to the 45% rate, the work that HMRC did, which we scrutinised, suggested that the revenue maximising rate on incomes of above £150,000 was, as I recall, about 48%. That was based on the sort of taxable income elasticities that the IFS had used in the Mirrlees review, so that seems pretty respectable. Having said that, if you look back at the work that they did, the old phrase “uncertainty” I am afraid comes back and the number could be very different from that. That is why the move from 50% to 45% had at the time a very small revenue impact, because you were moving from slightly above the revenue maximising level to slightly below the revenue maximising level and, as I said, you were wandering round the summit of the Laffer curve and it did not make a great deal of difference between the two. That may have moved on, and if there was another policy put before us to change the upper rate again, then you would need to go through that whole analysis again and there may be things that have happened, in policy terms, in the distribution of income or in the international literature on labour elasticities that would lead you to a different answer, but that is roughly what it was last time we looked at it.
Q36 John Mann: Morning. Brian Clough and Peter Taylor—impossible to replace. Is Graham Parker irreplaceable?
Robert Chote: He is going to be very hard to replace. He has a breadth and depth of experience across the revenue and welfare forecasting terrain that few other people I can think of have. When the interim OBR was recommended, one of the few key recommendations I made to Alan Budd was, “If you can get Graham to do this job, please do”. I was delighted when he carried on, and he will be a big gap to fill when he goes.
Q37 John Mann: Will it have to be a Treasury insider who replaces him?
Robert Chote: Not necessarily. Graham was not an entirely Treasury insider; he had been a Revenue and Customs—Inland Revenue in those days—person as well. There is a value to having somebody who knows where the bodies are buried in some of these policies. There is a value in having people who have seen the last time arguments were put up suggesting, “This is going to be fine” or “That is going to have some sort of impact” who can say, “This was mooted 20 years ago and that is not quite how it turned out”. The number of people outside Treasury, HMRC and DWP who would have that breadth of experience across those areas I would imagine is pretty small, but you could, for example, have an IFS-style person who would be able to engage on this on a more academic level rather than on years of experience. Graham has done a wonderful job and he has made my life enormously easier doing it as he has done.
Q38 John Mann: I recall in the early 1990s the University of Warwick had what seemed to be a pretty much open source piece of software that allowed anybody to put in their own Budget and see how different economic models would respond and what people would be seeing in terms of unemployment, inflation, etc. In terms of how the public sees your work, do you think that kind of a model ought to be widely available now so people can test out their views on the economy and see what the general economic consensus—or, indeed, a specific model—says will happen?
Robert Chote: I remember the days. The plug was pulled on the funding of the Warwick macro bureau. At the time, one of the values of that exercise was that you had five or six reasonably well funded, constantly maintained macro modelling operations underway—not just NIESR but London Business School and other people like that: Patrick Minford at Cardiff and Simon Wren-Lewis was doing one as well. Funding that sort of macroeconomic modelling by people like the ESRC is less fashionable than it once was, and that exercise disappeared partly because there were fewer component organisations to compare. Jeremy Bray used to be very engaged in this sort of thing as well, as you will know. Restarting that exercise, you find yourself with fewer people to compare. It was an exercise that was primarily of benefit to practising macroeconomic modellers to compare the outcomes they got from running things through. You could write a newspaper story that “there is a difference between how all these models regard the likely consequences of a rise in VAT” or something, but it was primarily an academic exercise.
In terms of the public ways of doing this, you can try to have web-based games where you can be your own Chancellor. When I was at the IFS, we had a thing called Be Your Own Chancellor. You could stick in a variety of policies and it would spew out some distributional consequences and some economic consequences as well. It is not an entirely happy experience running those. They take a lot of maintenance. Keeping them up to date is not straightforward. There is a tendency, as there sometimes is with, for example, the ITEM Club, to assume that once you have the model, if you shove in a set of your numbers and I shove in the same set of numbers, you are going to come up with the same answer. That is not the way modelling works. Models are basically a constraint for ensuring that the exercise of judgment is internally consistent; it is not something that generally gives you the answers to the most interesting questions that you want to ask.
Q39 John Mann: No, but it is very useful in looking at options and debates of what some people will say are the consequences. There may be more radical economic solutions floating around and there is going to be more pressure, perhaps, for people to look at what the majority of economists would conclude are the outcomes from those economic models. Would it not be useful and perhaps take pressure off someone like yourself if there was such a tool available, even to members of this Committee?
Robert Chote: Because we are constrained from looking at policy alternatives, that pressure is not on us in the first place. You have places like the National Institute that will look at these, and I am conscious—as an Islington North resident, I am seized of the importance of considering radical options at all times—that some of those things are easier to model than others. How straightforward it would be to plug those sorts of things into a conventional model and get a sensible answer out I would not like to say. It is why I think it is important for the ESRC in particular to be funding adequately the sorts of independent bodies that can do this sort of work: the IFS on the fiscal side; the Centre for Economic Performance at the LSE; the National Institute, etc. Having that infrastructure supported in independent institutions I agree would be a very good thing.
Chair: A lot of my colleagues want to chip in, which you should take as a compliment, Robert.
Q40 Mr Jacob Rees-Mogg: I think this is more a congratulatory viva than a reappointment hearing. Can I ask you about the proposal that the OBR takes on more responsibilities—a fiscal risks report and, in that awful phrase, greater stakeholder involvement, which I think means more PR? I have two thoughts. First, does this not distract you from your main duty that you are doing very well? Is there not an advantage in the OBR sticking to what it does and not wandering off into other things? Is the value of your independence really not in what you are currently doing on budgetary responsibility?
Robert Chote: Clearly, that was one of the arguments that the Ramsden review put forward specifically for us not doing distributional analysis. You can strongly make the case that that is just taking us into a different bit of territory from that which we occupy at the moment and it is not clear that we could do that to anybody’s greater satisfaction than the IFS can do it already. That argument applies quite strongly there. In terms of doing costing of Opposition policies, that is, as it were, using our existing set of skills, so it would not be diverting us into new intellectually advantageous or disadvantageous byways; we would be applying the skills and the techniques that we use now to a wider range of policies. There, the concerns that were expressed, as the Committee has discussed before, are genuine ones. One is it pulls us into much more politically contested territory; secondly, it has consequences for the way in which political parties and the Civil Service behave at least as much as it does for us.
On the ones you mentioned, we do a lot of work on fiscal risk at the moment. It comes back to the discussion of uncertainty we were having a moment ago. I view the fiscal risks report as partly a way of marshalling the analysis. We do that already—for example, kicking the tyres of the forecasts in the Economic and Fiscal Outlook and the examination of contingent liabilities and balance sheet features for the whole-of-Government accounts in the Fiscal Sustainability Report—and I am sure there are opportunities to add new things and do different things. A variety of other countries do these and we can learn from what they have done and pull together a good product that is based on what we are doing at the moment but can expand on that. It is still core to the mandate.
On stakeholder engagement, the Chairman has made the point on numerous occasions of the importance of us being able to explain the difficulties and the practicalities of the sort of job we do to a wider public, not just to the cognoscenti who pay regular attention. I think there is a value to that. I have made it very clear that I am only going to do that additional activity if we are resourced to do it. I am not going to take people away from the core resources of getting the job done properly in order to devote them to that, but if the Government wishes to provide us with some extra resources to do that, I am very happy to do it.
Q41 Mr Jacob Rees-Mogg: That leads on to the next question. In terms of extra resources, you have 19 people at the moment. How many more do you need? I take it you can do the fiscal risks report with your current team; that does not need an increase in the size of bureaucracy. It is the engagement issues that would increase it.
Robert Chote: The fiscal risks report would do. Although the recommendation is that we step down from doing long-term projections every year, the review concedes that the scope for freeing up staff from that is relatively limited. What I think we will do is the sorts of things we included alongside the long-term fiscal projection—for example, a detailed examination of North Sea oil receipts. You can do special projects in the off years rather than that. I do not have an aggregate number—we have to have this discussion with the Treasury during the course of the Spending Review—but if you go through and look at the various things that they have said, “We would like you to do more of this; we would like you to do more of that,” and the recognition that we need to bolster the resources we have to do the existing activity, then it is a substantive increase in resources and people. We have to have a negotiation with them about how much money there is available and what exactly we want to do and they want to do. That is a delight to unfold over the next few months. I made it very clear when the Ramsden review was underway that I hoped they were not going to recommend things that they were not willing to fund. Now we will find out.
Q42 Mr Jacob Rees-Mogg: Do you have an idea of how much more money and how many more people you want?
Robert Chote: No. As I say, not in aggregate. It depends in part, if they would like us to do more stakeholder engagement, on how much more stakeholder engagement. It is a relatively flexible view, but it is significant; this is not something you can do by bolting on two or three more people.
Q43 Mr Jacob Rees-Mogg: Moving on, you run the healthiest operation, as far as I can tell, in the whole of Whitehall, with only 0.2 days’ sickness—illness, really—per annum, against the sickly Treasury at 3.6 days and other Government Departments, who must all have the plague, at 7.2 days. Is this because your work is interesting and fulfilling and therefore people do not pretend to be sick when they are perfectly well, or is it because you eat your Weetabix?
Robert Chote: A rigorous programme of morning calisthenics for all members of staff. I hope it is because people find it an interesting and stimulating place in which to work. It is particularly striking that you get those sorts of results at the same time that we have a situation where our Range Es are working far more overtime relative to their main roles than the equivalent officials in the Treasury are. That is one reason why we need some more resources—for that to continue to be the case and to maintain it. One of the nice things about being a relatively small organisation is that you can build a good working environment. You are not part of a huge bureaucracy in which you are the one person who is doing the house price equation; you are part of a team. Also, there is the fact that we are independent. One of the biggest surprises, comparing what I expected at the beginning coming into this job and what has been the outcome, is that I thought, “I have come from the IFS. How am I going to build a culture of independence from the top down? It is like battery hens being released from the cage; these are people who have come out of the Treasury. Are they are going to spread their wings or are they just going to cluck around and stay in the same area?” but that was a complete misjudgment of human character. There is nothing they like better than poking their former colleagues with a stick, and it is my job to take the heat if things get difficult, as we were discussing earlier on. I hope it is that we are doing a worthwhile job and people like doing a worthwhile job in an environment in which they are paid attention to.
Mr Jacob Rees-Mogg: It is so striking that it is almost worthy of a report in itself that your levels of illness are so low compared to other people. I accept they are a small number.
Robert Chote: Yes; it is a small sample size. I would not want to claim credit where it is probably not due.
Q44 Mr Jacob Rees-Mogg: My final point is: what do you do if somebody in the small team is ill for an extended period? What sort of cover do you have?
Robert Chote: This is clearly an issue. If a lynchpin person in the forecasting round becomes ill four weeks before the Budget, then that is a serious issue. We do have arrangements with the Treasury; there is a network of people, for example, who can operate the Treasury model and who have been through this round before. They may have gone off into other jobs; it is our alumni association, essentially speaking, as much as anything else. There is an understanding with them that if it came to it we could draft somebody back in. Normally you do not want to do secondments, but if needs must, we can haul back somebody who used to be the lead macro forecaster and is now doing Scotland or something like that. That is accepted. Touch wood, we have not had that as an issue yet, but it is a risk that we are alive to and have plans in place for.
Q45 Chair: But you have operational independence over your budget, so you can set aside cash and go and recruit without having to go and ask anybody.
Robert Chote: You could do, but if you lose somebody four weeks before the Budget you cannot run a recruitment exercise and find the right skills.
Q46 Chair: Yes. It is a bit disappointing that, like all quangos, you have only taken five years to start bidding for more money, Mr Chote. All quangos tend to expand and you seem to be demonstrating that tendency. You say, “We need more resources”, because it will take work away from your core function in order to perform this other work, but is it not true that the forecasting cycle is all back-end-loaded around the winter and that some months are relatively quiet and in fact they are a period in which you can do some gentle recuperation and longer-term thinking? Is that not one of the reasons why the switch from one to two years for the long-term work might create quite a bit of spare resource from time to time?
Robert Chote: It will certainly be helpful. As you rightly say, we have bursts of activity in what one might think of as harvest time but there is a lot of work to be done on the farm in the other period.
Chair: Both forecasts come within four months.
Robert Chote: Yes. For example, the summer—this one has been curtailed because we had the extra fiscal event—is the time when you can go back to the macro model and say, “Our household debt projections have not been performing as well as we would like. Let us go back and look in detail at those sorts of equations,” before you are in the thick of the battle trying to run it in the forecast. In the forecast round, things pop up and you say, “Oh yes, we really must have a look at this when it is quieter”, and we do that. It is about having a bit more capacity to do that and a bit more rigour in saying, “Are we kicking the tyres of the bits of the toolkit and the bits of the machinery that we need to?” There is more scope for that.
Chair: I can hear the subtle spending bid coming through.
Q47 Mr Steve Baker: Good morning, Mr Chote. Another fascinating conversation, as ever. I am struck that you are exceptionally good at a task that is possibly impossible and absurd, but nevertheless I do think it is something of a congratulatory viva.
Robert Chote: Likewise.
Mr Steve Baker: I understand you have an iterative dependency on a number of Government models. Is that correct?
Robert Chote: Yes.
Mr Steve Baker: To what extent do you co-own those models with other parts of Government?
Robert Chote: At the end of the day, the forecasts that we generate are our forecasts, so if we want them to change an equation or to make a particular interpretation of outturn data, they have to do that. If they want to maintain a different model with different sets of assumptions for their own purposes, that is fine, but at the end of the day we have the right to get the models and the outcomes we want.
Q48 Mr Steve Baker: I am just trying to get to the bottom of that. It is possible that other Government Departments, like the Treasury, the DWP or others, might end up maintaining multiple models, one in order to give you what you want and one for their own separate purposes.
Robert Chote: It is possible. Certainly in terms of the main macro model, maintaining that is quite a labour-intensive exercise. In principle, we could run away with it and, if we had the resources, maintain it and develop it in a different way. I do not think we would say we have a monopoly of wisdom over how this thing should operate; we are sitting down with other people who are using this as a toolkit and reaching a shared view on where you need to scrape the barnacles off, which bits of the machinery need to be replaced and what you want to have a look at. That is beneficial. You also have ITEM using it externally, which is another source of intelligence coming back.
Q49 Mr Steve Baker: I think we expected that there was one model and that you collaborated in its development. What I think I am hearing from you is there are multiple models and quite a flexible exchange of views about how they should be maintained.
Robert Chote: On the macroeconomic side, there is the “main” model. We have other bits of toolkit. For example, when we are doing the scenario analysis, it is too time‑consuming to run the whole model four times to generate yourself different outcomes, so we have a small model, which we did develop in-house, which allows you, in a relatively quick and dirty way, to come up with alternatives and to plug in the impact that changes in monetary policy will feed back, etc, without running the whole thing over again. There is then a whole series of other models that are specific to particular flows of revenue or flows of spending. There is a model that helps you generate a forecast for debt interest payments; there is a model that generates an outcome for housing benefit expenditure; there is a model that is generating VAT; there is a model that is generating an estimate of stamp duty. Again—this is one of the things that the Ramsden review helpfully highlights—when using those, from time to time you do want to devote a bit of effort to saying, “Let us look at this. Perhaps we could publish the details of it. Perhaps we could run some more simulations on it that we would not normally have time to do that would make that more robust and more understood by stakeholders.” That is helpful.
Q50 Mr Steve Baker: A wide range of questions present themselves. I imagine that there is a sea of spreadsheets involved in all of this.
Robert Chote: We are not light on spreadsheets, yes.
Mr Steve Baker: Is there any software that is not a spreadsheet that gets used routinely? Is there any formal software other than Excel spreadsheets operated by non-software-engineers?
Robert Chote: You are taking me beyond my areas of technical expertise. Yes, I am sure there are other statistical packages that are used in addition.
Mr Steve Baker: But mostly what you and your team see is Excel spreadsheets.
Robert Chote: Yes.
Q51 Mr Steve Baker: Can you give me some indication of how many?
Robert Chote: My times as a practiced Stata operator are a long way in the past.
Mr Steve Baker: Of course. But could you give us some indication of how many spreadsheets are involved in this process?
Robert Chote: I could not. I could find that out for you. We did have an internal audit process that was going through and basically checking that we were using all our spreadsheets properly and that that technology was well set up. If you are interested, I can get some details of that.
Mr Steve Baker: I am interested in an order of magnitude. Is it tens, hundreds or thousands? I remember looking at one large international conglomerate that seemed to have hundreds of thousands of spreadsheets involved. I would be very interested in an order of magnitude. Let us just park that for a moment.
Robert Chote: Yes. If you are keen on pursuing that, I am sure I can provide you with more information than you want.
Q52 Mr Steve Baker: Whether it is tens or hundreds of thousands would be interesting. It does not sound like there is a lot of tension or head-banging over difficulties in maintaining these models or agreeing directions of travel. I am glad you shake your head, for the record. Are there any particular areas where you have priorities for the development of the main UK macroeconomic model?
Robert Chote: One thing we are looking at at the moment, as I just mentioned, is household debt. We have tended to produce forecasts showing that household debt was likely to rise more than has turned out to be the case. This is something that the Treasury has been engaged in and interested in as well. We can have useful discussions with the Bank, who do their own work in this area, and we can compare them if we are shoving in roughly the same sorts of assumptions as are coming out. That is probably the obvious live example of something that we are looking at at the moment. On the fiscal side, we are looking at whether we have the VAT model working satisfactorily.
Q53 Mr Steve Baker: There are two areas of particular interest to me. One is monetary policy. I have recently written to the Governor and pointed out that monetary policy has not changed during his tenure, but what has changed are efforts to manipulate economic expectations. I wonder how you might model those efforts to manipulate economic expectations. That is one area.
Robert Chote: To the extent that they are reflected in the yield curve, which is one obvious area where that would come through, what we have done is to run the model taking the assumption that monetary policy follows the path that is set out in market expectations. If you are looking at the debt interest forecast, it is reflecting what is moving at different points in the yield curve, which is in part a reflection of what you might regard as attempts to manage expectations—or expectations managing themselves—of where monetary policy will go.
Q54 Mr Steve Baker: What I am driving at is that what the Bank of England says—indeed, what all central banks say—now seems to be a predominant factor in what happens in the economy that goes far beyond where the yield curve is. I am inclined to believe that it is impossible for you to model those things.
Robert Chote: Do you mean in terms of impact on animal spirits, investment intentions and those sorts of issues?
Mr Steve Baker: Yes. If the Bank were to give some indication about where interest rates were likely to go, we would soon see stock markets dancing.
Robert Chote: Yes. On equities, we basically assume that equity prices move in line with growth in money GDP. I would not claim that is a way to go out and make money if you are trying to judge it for yourself, but, as a relatively neutral assumption from our point of view, it makes sense. As you say, if the Bank is saying things—that it will move the yield curve around or move the exchange rate—those things can be reflected in the models. In terms of near-term economic developments, which, frankly, are not of enormous importance for the medium-term fiscal analysis that we focus on but people are obviously interested, survey data is important as a cross-check and as a further guide to how you think data is likely to move in the near term.
Q55 Mr Steve Baker: The other area of interest is the output gap. We have sat here on other occasions talking about the output gap. If I were to caricature perhaps a little my understanding of the conversation, I think we agreed that the output gap was a crucial figure but that it was almost certainly wrong. That is broadly correct. Is there work that could be done to improve the modelling of the output gap? If so, given that it is of crucial importance to your forecast, where would you like the model to develop to try to improve output gap forecasting?
Robert Chote: What we have done is very much in line with the recommendations that the previous iteration of the Committee made. I am very wary of the idea that there is one correct way to do the output gap calculation and that we need to be searching for that Holy Grail and investing all our time and effort in that. The sensible thing to do is to look at a range of ways of estimating this and to balance those off and to take a judgment about what looks sensible over all of them. I am afraid experience suggests that once you alight on a particular measure and a particular methodology, you look at it and it starts to give you answers that you are not happy with. I am wary of an approach that says, “We have said we are using that methodology so we must not move away from it”. I prefer what some people will regard as being an undesirably ad hoc approach of saying, “Let us look at what all these different indicators are telling us and draw our best conclusion from all of them”.
Q56 Mr Steve Baker: Where I propose to leave this is to ask this question. You said earlier that models are a constraint on judgment. Do you think that politicians, the public and commentators have an adequate understanding of the extent to which what you do is uncertain and the extent to which it does rely on the judgment of the individuals and the team as to what is going on and what is likely to happen?
Robert Chote: I think that is pretty well understood, but then I go and do a TV interview and somebody says, “You said this last time. How do I know you are right next time?” which shows they have not understood that. That is always an exercise in pushing water uphill. We have set out a much more explicit discussion of uncertainty. It is easier for us than it is, frankly, for politicians and for ministers. Conviction forecasting is not something, fortunately, we have to engage in. I can say, “We were looking at this. Frankly, we have changed our mind. We have reached a different judgment.” That is easier than in a political environment in which you said, “I promise you that the outcome is going to be this”. It is harder to row away from that. That is one of the advantages of moving to an independent institution, but it does not make the job of doing it any easier.
Q57 Stephen Hammond: Good morning, Mr Chote. Most of the questions I was going to ask have been asked by Mr Garnier rather more eloquently, but, continuing his analogy, you have passed the pink gin test and you are out on the water. You will have seen the Governor said on 6 August that “no statistics are perfect”. Just so the Committee is clear, any sailor would like to know which set of statistics or which direction is most likely to be right, so is there any group of statistics provided by the ONS that you as a body are more concerned about and you have asked them to do more work about—or that you are doing work about?
Robert Chote: There are two issues there. There are the ones that matter to us more than others, which we are keen in, and there are ones that are difficult. If you look at the nature of the revisions that you have seen over the period of the last five years, one area that has clearly proved extremely difficult to measure has been business investment. We had long discussions with Mr Kerevan’s estimable predecessor, Stewart Hosie, about, “Why has business investment not picked up? You were saying this was going to move ahead” and then we had a set of Blue Book revisions that showed it had been roaring away for three years previously and we had not been aware of that.
Another area where we have to worry more than most consumers is that we are fundamentally interested in the nominal measures of income and expenditure, whereas most people are more interested in the measures—real GDP; the headline things. If you are interested in the amount of tax revenue you are going to have coming in, you are more interested in things like nominal consumer spending and nominal profits. We have to do estimates of Government spending and turn those into real measures, so the deflators—the prices of these things—matter a lot as well. Some of the reason why you get a revision in the path of real investment is not because you have changed your mind about the total amount of nominal investment but because you have reached a different view on the way the prices of investment goods are moving, and that can cause changes.
One area of forecasting error that we took perhaps longer than we should have done to get on top of was the fact that cuts, or restraint, in nominal Government spending did not show up as a drag on real GDP to the degree that we had anticipated via real Government spending because of the way in which the price of Government output is measured. Quite a lot of Government output is measured directly. Education output is the number of pupils going through school. If you cut nominal education spending, that does not show up as a cut in education output. If you stopped educating every child whose name ended after M in the alphabet as your way of saving money, it would do, but if it does not, it may show up as a quality change in education provision that does not show up as a real change in the national accounts. We took longer, as I say, than we would have liked to get on top of the degree to which that was likely to be the case, so, even though we were over-optimistic about GDP growth for a while, we were overly pessimistic about the degree to which the consolidation programme was going to squeeze GDP via a direct impact on Government output.
Those are just a couple of examples, but now there is quite a lot of interest in the current account picture. Much of the recent deterioration has been on the flow of income rather than the trade deficit. What is going on there? Is that being accurately measured? How worried, therefore, should one be about a current account deficit of 6% of GDP? There are different things you worry about at different times. The way you identify what look like problems when you are forecasting and working with numbers is to notice that they just look odd and to look at them harder.
Q58 Stephen Hammond: I was interested—this goes to the heart, perhaps of this issue—in the Governor’s other comment, which was, in terms of the GDP number: “We forecast what is [expected] to come out [and] we forecast what we expect it ultimately to be measured as”. Do you think he was just saying that the forecasting art is so difficult that it is sophistry, or is it he thinks there is some systemic bias in the numbers?
Robert Chote: Because they back-cast? Yes.
Stephen Hammond: Yes, because they back-cast as well. Is that just getting two bites of the cherry? Is it sophistry? Is it the fact that they have identified some systemic failing in the data set that they do that?
Robert Chote: I think it is less that they have identified a systemic failure in the data set than that they think that they have identified a regular pattern in the relationship between early estimates and later estimates that may be partially explainable by things like looking at what business and other survey data is telling you at the same time, and that therefore they think about, on the basis of those relationships remaining true in the future, how the past is likely to be revised to reflect that. That is an exercise that is easier to engage in if you have the number of staff that the Bank of England do than with ours, and we have decided not to go down that route. We take the ONS’s outturn data as being the best available estimates while being entirely conscious of the fact that these things can be revised. As I say, the Forecast Evaluation Report is a very useful tool for going back and discovering that you were wrong, but wrong not for the reason that you thought you were wrong a year and a half ago.
Q59 Wes Streeting: I want to ask you about the issue that became quite partisan in the run-up to the General Election, which is whether the OBR should have a role in costing the policies of Opposition parties ahead of a General Election. In your letter of 15 January 2014 to the Chairman of our Committee, who has a longstanding and very wise position on this, you noted that there are issues that could be overcome, but they include careful planning and adequate resourcing, not just for the OBR but also for the Civil Service and for the parties. At this point in a Parliament, where we are some way off the next General Election, we do not know what the view of the Shadow Chancellor is as yet and the Chancellor may well move on to bigger and better things or to the departure lounge of his political career, depending on how things pan out for him, this is a good opportunity to just look at the issue purely on its merits and in terms of some of the practical issues that exist around it without risk of breaking lots of eggshells. When the Treasury undertook their review of the OBR, they rejected the idea it should cost the policies. Were you surprised by that conclusion?
Robert Chote: No, I was not. The Chancellor was not a supporter of the idea prior to the election. He said that he would come back and look at it again. I would have been more surprised if he had changed his view than I am that he stuck to the previous one.
Q60 Wes Streeting: Do you think it was the case that the Chancellor’s view very much coloured the outcome, or the judgment, of the review?
Robert Chote: It is a Treasury review; it is not pretending to be an independent review in that sense. If I look at the review, as I say, in principle you can see good arguments for doing this, but there are considerable practical difficulties. I would probably order my reasons to be anxious about this somewhat differently from the way in which the Treasury did in the review. They say it would not have a great deal of impact on fiscal credibility. I think it could have an impact on fiscal credibility. There is the argument that the IFS does it perfectly happily. I think you could bring more to it if it was on a level playing field, which is the way initial analysis is done. On the other hand, they very rightly place high weight on the logistical and practical difficulties that this would involve. The other fiscal councils that do this work are much larger and devote a much larger proportion of their resources to it. The particular regime that we have, in which we are a relatively small body and we have civil servants who are basically doing work for us commissioned, in a sense, by us in areas like HMRC and DWP, would make this much more complicated, because you are in effect having to make a decision about what access you provide Opposition parties to the resources of those departments, albeit mitigated through us. That is not the same as having 115 people at the Dutch office being able to do all this work in-house themselves. The other thing, which we discussed in the last Parliament, is the timetabling issue. If you were going to do this properly, parties would have to make up their minds what they were going to stick in their manifestos a lot earlier than they currently do and a lot earlier than they currently show signs of wanting to do.
It is an interesting thought experiment to go back to previous elections and say, “How might the world have looked if we were operating this?” That raises the issue of what sort of questions you would want to ask. Would you want, as Ed Balls suggested, for us simply to produce an analysis of particular individual policy measures of the sort that appear as lines in the scorecard in the Treasury Red Book or the Autumn Statement, or would you want us to make broader statements about an overall fiscal path? What would you want the OBR to say about the Chancellor’s statement in the Budget of March, when he said, “If we are elected alone, we are going to do £30 billion of consolidation, made up of the following proportions of tax avoidance and welfare, etc,” without setting out what those measures were? What would somebody wanting us to fulfil this role want us to say about that? Would you want us just to say, “He said that is what he would do, but he has not explained how”? Would you want us to say, “Can you really raise £5 billion from anti-avoidance measures after two years?” Quite how long that particular piece of string is would, I am sure, result in a lengthy move from a decision in principle to do this to designing the practicalities.
Q61 Wes Streeting: Given some of those variables and practicalities, do you think the right range of stakeholders was consulted during the course of the Treasury’s review?
Robert Chote: They talked to people who had done it in other places. They talked to journalists and think tank people. I do not know whether they talked to political parties about it. I suspect not.
Wes Streeting: No, they did not.
Robert Chote: My sense of the discussions was that quite a few people shared the concerns of the Page review, partly because of the challenges, I suspect, of how you get from, “This is a good idea in principle”, to precisely how this would work in practice, which would be contentious enough before you even got to applying it in a particular election environment. I think they found widespread recognition of quite how difficult that would likely be.
Q62 Wes Streeting: Did Sir David Ramsden speak to you during the course of the review about the OBR costing Opposition policies?
Robert Chote: Yes. We had a general discussion. As they had meetings with sets of stakeholders, they had a meeting with us, in which we commented on the draft thoughts that they had had and the other issues that we had raised and thought were important.
Q63 Wes Streeting: What was the view you expressed at the time?
Robert Chote: On the manifesto costing stuff explicitly?
Wes Streeting: Yes.
Robert Chote: My view has not changed, really, from the letter that I sent to the Chairman in the last Parliament. I can see the attractions of doing this in principle but the logistical consequences—not just the cash sum but the change this would require—are considerable and, as I say, at least as considerable for the Civil Service and for the political parties as it would be for us.
Q64 Wes Streeting: Taking into account all of those practicalities and the need to make sure it is adequately resourced and, crucially, properly planned, on balance, do you think those changes outweigh the cost in all sorts of terms?
Robert Chote: At the end of the day, it is for Parliament to decide whether it wants us to do this. It is potentially very valuable, and I am delighted it is my successors who will have the responsibility—
Chair: I have tried that question before, Wes, and I did not get anywhere.
Wes Streeting: Nicely sidestepped again.
Robert Chote: I shall be delighted to see my successors try to tackle it at the appropriate moment.
Q65 Chair: Before bringing in George Kerevan, just one clarification of one point you made. You said you would be able to bring more to it than the IFS. You used to work for the IFS, so this is hardly me having a swipe at them. By that, you mean you could do a more thorough job, and a job also that would carry more public credibility, do you not?
Robert Chote: I do not know about carrying public credibility. The range of measures on which we, via HMRC, for example, on the tax side, are equipped to reach a judgment is greater than the range for IFS.
Q66 Chair: So, overall, you are going to do a better job.
Robert Chote: The IFS will be as capable as we are of judging the impact of a change in the VAT rate or in an income tax rate or in tax credit rates. On some detailed reform of insurance taxation, we would be interrogating detailed stuff from HMRC and that is not an area where, when I was at IFS, we would have felt we had the expertise to reach a judgment.
Q67 Chair: But then there is the credibility of the overall conclusions.
Robert Chote: Yes. That is vested partly in the credibility of the institution, which is in the eyes of you and other beholders. Some people have objected to this on the grounds that, “But surely the answers to this are going to be uncertain. You do not know whether this is going to be correct. That is a reason not to do it.” That would be a reason not to do most of what we do anyway, so I do not find that very compelling. One attraction of us doing it is that if a party gets elected, it is going to be us who are looking at the policies when they are in Government, so it may be wrong but at least it is going to be consistent with the approach that we took when they were in Opposition.
Q68 Wes Streeting: Very briefly on that, from the public point of view—I would be interested in your view—I think they would welcome a General Election debate that is about a battle and contest of ideas rather than the plausibility and the credibility of the policies being put forward. Do you think from a public point of view it would be far better to have an independent verification that the policies could be deliverable so that, when the public are casting their votes, instead of having to decide for themselves who they trust more not to fiddle the books—I am not sure the answer to that is anyone—they can think instead, “Of these policies, which are independently assessed and verified as deliverable, and for which the costings are there, which do I prefer?” Do you not think that would be beneficial to the public?
Robert Chote: Yes, I think that is helpful. The other thing is you have to look at the sorts of things that parties promise in the run-up to elections. They do not typically set out 70 policies of the sort that you see in a Budget Red Book. Some of it falls into areas of public services spending that we do not look at in a micro area anyway, and some of the promises are of the sort that the Chancellor made in the Budget speech of saying, “Rather than the current plan laid out, I am going to do X more of welfare and Y more of” something else. You cannot shed a great deal more light on that, because it is not a concrete set of measures. Whether the public wants you to be setting out some colossally detailed set of measures or wants to be inspired with a few key messages and something written on the back of a card—you are the experts in that, not me.
George Kerevan: The media like numbers.
Robert Chote: Yes.
Wes Streeting: We cannot foresee our next manifesto at this time.
Q69 George Kerevan: Morning, Mr Chote. You do not do a macroeconomic forecast for Scotland. Indeed, you say you cannot because of the paucity of data. As a result, your forecasts of tax yields in Scotland are done by extrapolating the share of UK taxes. A number of questions come to mind as a result of that. As we devolve more and more taxes to Scotland—ditto the other regions and nations in the UK—surely that starts to undermine your methodology of simply taking historical tax share to do your fiscal forecast. Beyond that, if you cannot properly forecast the fiscal shares in the regions and nations, surely that starts to undermine your whole-UK fiscal forecast. It is not just a devolved issue; it is about the weight you put on your overall forecast for the UK. If that is the case, does that then also lead us to ask whether or not you should be doing more macroeconomic modelling of Scotland and then the other elements of the UK?
Robert Chote: You are right; this is going to be one of the key challenges of the coming five years, as to how not just we but collectively we and the Scottish institutions as well address the task of forecasting and scrutinising the evolution of public finances as devolution extends. I suspect that the need for a Scottish macroeconomic forecast is less key than some of the substantive issues that arise with individual forecast measures. One area where, for example, the shares process will become less important is the income tax forecast, because at the moment HMRC does not identify a particular taxpayer as being Scottish or not, whereas it will do, which means that we will no longer need to try to work out what the share was from looking at the Survey of Personal Incomes, which is a good thing. You do then have to make a judgment about whether you think that the movement in labour income—in earnings—is going to be substantively different in one direction or another in Scotland from the rest of the UK. You could say, “Do you need a fully articulated macroeconomic forecast for that or is it better to look at a UK-wide thing and to get supplementary data?” We are going to have to make those sorts of decisions a long time before the data machinery is available.
There are other areas, like, for example, the land and buildings transaction tax, where the Scottish Government has designed and is operating a new tax system of its own. Ironically, subsequently, the rest-of-the-UK measure got rather closer to that after the event, but that then does present you with a whole new tax stream to look at. For that, it is very important that we should have a constructive relationship with the Scottish Government and the Scottish Fiscal Commission, respecting each other’s independence but I hope recognising that we all have a shared interest. I have a responsibility to the citizens of Scotland as much as to anywhere else to keep the UK’s public finances under scrutiny, and sitting down around the table and thinking, “Do we have a shared view on what is the best way to try to model this?” and having a disagreement because we had a different view of what source data you should look at for the distribution of Scottish house prices three years ago is probably not a world you want to be in. It is very important to look at those sorts of issues.
Q70 George Kerevan: So, you are discounting the need at some stage for a separate macroeconomic model for Scotland.
Robert Chote: I think it will be a long time before that is the most cost-effective way of addressing these sorts of issues, but I certainly would not rule it out.
George Kerevan: So it is a cost issue.
Robert Chote: For example, if you are looking at understanding the movements in stamp duty or property transaction tax, the distinction that matters is often not Scotland versus the rest of the UK; it is London versus Scotland and the rest of the UK other than London. Just saying that Scotland behaves differently economically from the rest of the UK and that is the distinction you need to worry about most—in discussions of shares, for example, around the property transaction tax, it is as much about whether we should be worrying about London looking odd relative to the rest of the UK as it is about whether the rest of the UK looks odd relative to Scotland.
Q71 George Kerevan: Accepting that is the line of march you would initially go on, if you then multiply by the other devolved jurisdictions—we have devolved corporation tax in Northern Ireland, etc—and accepting the devolution of fiscal responsibility is the line of march, no matter how fast we go, then you are pyramiding up the amount of work your Department is doing in working out all the individual taxes, so you still have a resource issue.
Robert Chote: We do. The Chairman says we are behaving like every quango in pointing out future resource needs. That one has been pointed out. I have cited in my questionnaire exactly your point: you can see the direction of travel, but knowing exactly how much of an issue that is going to be and how quickly is quite hard to do. The other issue that we have, which is yet to be determined, is: for those taxes that are operated, designed and run in Scotland, is it going to be in the fullness of time the Scottish Fiscal Commission that produces the official forecast for the Scottish Government? Is it going to be the Scottish Government? What is the relationship going to be between Revenue Scotland and the Scottish Government and the Scottish Fiscal Commission? I give evidence to Kenneth Gibson’s committee in Holyrood. We have had very interesting discussions with them on these sorts of issues, because at the moment the relationship is the Scottish Fiscal Commission saying whether a particular set of Scottish Government forecasts are reasonable or not, but Kenneth’s committee has suggested you might want to think of a model that is more like ours, in which the commission takes more responsibility for the forecasting. Working out exactly—coming back to the dreaded phrase “stakeholder”—what the range of stakeholders is and how we can best work together while respecting everybody’s independence is going to be a considerable challenge.
Q72 George Kerevan: I was going to come on to that, but to finish off the forecasting issue, I suspect, from discussions I have had over the years, that ultimately the Scottish Government is going to want to have a macroeconomic model. There is a constant debate about this. Would you be open to exploring the creation of such a macroeconomic model for Scotland?
Robert Chote: Absolutely. We have very good relations with the Scottish Government and with the Scottish Fiscal Commission. When we are doing our forecasts for the Scottish tax receipts, we invite the Scottish Fiscal Commission and the Scottish Government officials to be present, as Treasury and HMRC people would be. I am very keen that that relationship works as well as it can.
Q73 George Kerevan: Going on, I noticed in your reappointment questionnaire—a fascinating document—you were asked: “What risks do you identify to the OBR achieving its aims over your next term?” You say, “Operationally, I think there [are] also potential risks from the expansion of devolution” and you cite “stakeholder engagement”. Do you want to clarify that?
Robert Chote: That is coming back to the point I have just made. It is hard to know how quickly this is going to go and, as I say, what the eventual relationship is going to be between the Scottish Government, Revenue Scotland and the Scottish Fiscal Commission in terms of who is doing the monitoring and the forecasting for the purposes of the Scottish Government budget. In terms of the nature of interaction we would like to have, the approach I have taken is that I am very keen for the Scottish Government and the Scottish Fiscal Commission to be involved in the meetings and to have discussions with us around our forecast. I stand ready for us to provide any advice that the Scottish Government, or whoever is doing the forecast, would find helpful when they are doing it. At the end of the day, it is in our shared interest that even if we do not have the same forecast—and there is no reason why we should necessarily—we should be able to explain to your and to your Holyrood companions’ satisfaction why it is they are different. If we do not understand that, that is a bad position to be in. If this all moves to welfare policy as well, then you have a whole new set of potential—
Q74 George Kerevan: What is the timescale and structure for these discussions?
Robert Chote: In the near term, I know the Finance Committee at Holyrood is continuing to look at the future relationship. There is legislation—I am not sure if it is drafted—coming up on the Scottish Fiscal Commission, so there is a live discussion, which I have offered my thoughts on, in terms of that. Once that is decided and we know what the relative responsibilities are going to be, that is an appropriate point for us to sit down and to come up with a memorandum of understanding on what sharing of information or expertise we all think is appropriate. It is a bit hard to go into that straight now; we need to wait for Scotland to decide exactly how they want to arrange this. It is going to be fluid. As I say, if you end up with more devolution on the welfare side, then there will be a question about where that is going to be managed from and who we need to interact with on that score. At the moment, we have a memorandum of understanding that covers our relationship with the Department for Work and Pensions as well as Treasury and Revenue and Customs; potentially we would need to be replicating that as well. We are remaining in close contact, giving evidence regularly and contributing as we can to the committee activity up there, and, as I say, involved in informal engagement. The land and buildings transaction tax is a good example. We are in a transition period in which you have to forecast it one way because it is a new policy. What I am keen is that when we know what the steady-state evolution of information is going to be on which we are judging outturns and looking at the future, we have something that works to everybody’s satisfaction, if that is possible. As I say, it is not for me to say; if they do not want to have that relationship, I cannot force anybody to.
Q75 Helen Goodman: Just flipping back to the questions that Mr Philp asked about high rates of tax, my recollection is that when the top rate of tax was cut from 50% to 45% in 2010 the 50% had only been in operation for a year or a few months. That is right, is it not?
Robert Chote: I cannot remember precisely. The other complexity is that the changes were pre-announced in both directions. One of the big complexities about working out this is that people took action to anticipate pre-announced movements. If, in addition to working out precisely where your Laffer curve peaks, you have forestalling and reverse forestalling activity, it makes a very difficult job even more so.
Q76 Helen Goodman: Thank you very much. I would like to ask you about the Chancellor’s new proposals on running a permanent surplus as long as growth is more than 1%. The previous fiscal target included some cyclical adjustment but the new one does not. Does that mean it will be easier for you to judge whether the new target is being hit?
Robert Chote: Mechanically it will be easier, in the sense that there is a relatively straightforward definition of normal and non-normal times—basically, a sort of four-quarter movement in GDP. That will fall straight out of the outturn data. In that sense it is straightforward, in that we do not have to calculate an output gap ourselves and make a judgment on that sort of basis. That would certainly be true. What it is not going to mean is that we can drop the need to worry about what the output gap is. At the end of the day, we have to make a judgment on how much nominal GDP and the components of nominal income and nominal expenditure are going to grow over the next five years, consistent with the Bank aiming to achieve its inflation target. It is very hard to do that without implicitly or explicitly making a judgment on what the trend growth rate of the economy is and how much spare capacity there is at the moment. If we never had to worry about the output gap again, that would be delightful; this policy, whatever its other attractions or lack of attractions, is not going to allow us to do that.
Q77 Helen Goodman: The Chancellor’s political intentions behind making this change are clear. Have you understood whether he also had some economic arguments for doing this?
Robert Chote: I think it was partly a desire not to be as dependent on movements in estimates of the output gap. I think it was partly on a view that you tend to have periods in which GDP growth chugs along at 2% to 3% and then periods when you are in a recession or going into a recession; you do not have very extended periods of it growing by about 1%, which would clearly make this rather awkward to look at. I think they felt that this was a simpler way of saying, “Look, there are two states of the world—there is chugging along reasonably happily and there is being in trouble—and a relatively simple broad-brush rule that draws that distinction would be helpful”. It is obviously for them to say exactly why they have chosen to do it this way. When you have slowdowns in growth, you also have the difficulty that there are some recessions where the growth numbers just fall away very quickly and there are some where it slows in a path. In an environment in which you are squeezing out inflation through a gradual tightening in monetary policy, it might be a smoother transition; if you have a commodity-related or a financial-crisis-related change, you can be in the situation you had in spring 2008, when everybody thought the economy was still going to be growing for the next two calendar years and, click, it did not.
Q78 Helen Goodman: That is interesting. I must say, not surprisingly, you have thought about this more than I have, and I had not thought of that. In a world where growth might be fluctuating around the 1%, how can a judgment be made about what to do with the public finances?
Robert Chote: The way this rule is specified is fairly mechanistic, in the sense of doing a sort of four-quarter comparison, so if either in outturn or in the forecast that is below 1%, then the Government is able to run a deficit at that stage. The rule is set up in a sense of not being overly prescriptive about exactly what you do at the point you get into that worse state of nature, perhaps in the recognition that every time somebody has designed a rule in the past you have ended up having to bin it, but it is basically saying, “Look, this is what we want to do in normal times, and if we hit some non-normal times it is reasonable to run a deficit and we will explain then in greater detail what the transition to a new normal would look like”. Clearly, one challenge is if, God forbid, we are in a world in which the trend growth rate is a lot lower than it has been in the past, then this distinction between most of the time you are growing at 2% to 3% and sometimes you are falling off a cliff will be less clear and this will be more awkward. But that is not what we are forecasting and it is not what most people are expecting.
Q79 Helen Goodman: I had not thought of this before either, but you said four quarters, either in the forecast or in the outturn. We were having some discussion previously about the fact that figures get revised. What happens in a world where you thought that you were above the line for the whole of the four quarters and then you discover that you were not—or, indeed, vice versa?
Robert Chote: The rule is clearly specified that if you are revised back out of trouble, that is not the point. You take the view that if, in the outturn data at a particular time or in our forecast data, you trigger into abnormal times then you move into the abnormal times world. If a month later the ONS decides that we were never in abnormal times, that is spilt milk; you are in the new state of the world, and the Government would have an easier job, presumably, explaining that you are back on normal times again.
Chair: There is quite a bit that we need to think through on this one.
Helen Goodman: There is quite a bit to think about with this, yes.
Robert Chote: The issue about how much of the time you are going to be there or thereabouts is quite important. I think the analysis that they have done—and if you look at the charts that we have produced—as I say, there is this general view that most of the time you are in the sorts of growth rates that we have now, or more rapid if you are recovering from a downturn, or you click into it. As I say, the greater danger is if you have a significant slowdown in trend GDP growth, in which case you spend more of your time at that rate than you currently anticipate.
Chair: That is extremely interesting. Thank you very much for your evidence this afternoon. It has been thorough, informed and, as usual, detailed—and also frank in places where it needs to be frank. We are very grateful.
Oral evidence: Reappointment hearing for Chair of the Office for Budget Responsibility, HC 459 21