Treasury Committee
Oral evidence: Reappointment of Graham Parker to the Budget Responsibility Committee, HC 580
Tuesday 22 July 2014
Ordered by the House of Commons to be published on 22 July 2014.
Members present: Mr Andrew Tyrie (Chair), Steve Baker, Mr Pat McFadden, John Mann, Mr George Mudie, John Thurso.
Witness: Graham Parker CBE, Member, Budget Responsibility Committee, Office for Budget Responsibility, gave evidence.
Q1 Chair: As you well know, we examined carefully your initial appointment on the grounds that your role is crucial to the underpinning of the independence and the perception of the independence of the OBR. Of course there was some concern at that time that you might, in spirit, not leave the Treasury but stay on there. In fact, Chris Giles, the Financial Times journalist, reported that you had been known formerly as “the wizard” in the Treasury for finding ways to flatter Gordon Brown’s forecasts and that you are now a poacher turned gamekeeper. Do you consider yourself a poacher turned gamekeeper, Mr Parker?
Graham Parker: Not quite. The role is quite different. I was always a little bit dubious about Mr Giles’ description of the wizard, because they were very much the Chancellor’s forecasts. My job was to make everything add up properly and be consistent with the assumptions that I had been given by Ministers.
Q2 Chair: However unlikely you thought those assumptions might be. Is that what I am gathering from your tone or not?
Graham Parker: I didn’t necessarily always agree with the assumptions I was given. I think that would be fair, but I had to make the numbers add up in a respectable and justifiable way. Now, of course, the assumptions are just ours, so it is much nicer. We take all the decisions ourselves so, in a sense, it is a very different role.
Q3 Chair: When you say “different,” you seem to be saying it is a much better role and you seem, therefore, to be supporting the political decision that was taken to create the OBR and remove this core assumption-making function from political control, which is right at the heart of whether we keep the OBR or go to some other system.
Graham Parker: Yes, I fully agree with that. It surely is much better to take the assumption making out of political judgment and let somebody like us do it transparently, objectively and professionally.
Chair: As you know, there were quite a number of people in all three main parties who took a different view, but it is important we have that view on the record.
Q4 Mr McFadden: Good morning, Mr Parker. It is assumptions I would like to ask you about. In his Autumn Statement in December last year, the Chancellor announced that he would abandon the current cap on student numbers and that this would be financed by selling the student loan book, and I am quoting from his statement now. He said: “The new loans will be financed by selling the old student loan book.” It was announced by the Business Secretary at the weekend that this sale has now been abandoned. How big a hole does this leave in the Government’s balance sheet?
Graham Parker: The first thing I would like to say about any sale of assets is that almost every asset has an associated income stream. When we put the sale of the student loan book into our forecasts, we also took out the associated income stream, because obviously when the loans were made by the Government, all the repayments will have come to the Government. What the effect of a sale would be is that you get upfront money, but you lose income further down the track. That is the first thing. What you have is higher public sector net debt. If you don’t sell it, you have higher public sector net debt in the short term, but you will gradually get it back.
Q5 Mr McFadden: I go back to my question: how big? The Government were expecting a receipt from this, which they are now not going to get.
Graham Parker: The assumption we made was that they would get £12 billion over five years, but there was an even bigger amount coming in later.
Q6 Mr McFadden: So you assumed. The Government said, and I quote from page 84, paragraph 216 of the Autumn Statement—this is the Government’s Green Book for the Autumn Statement not your forecast—“Over a five-year period the sale is expected to generate between £10 billion and £15 billion in sale revenues with a central estimate of £12 billion.” You thought that was a reasonable estimate.
Graham Parker: We did not have that much information to go on, because this was based on the actual book value in the accounts, but of course we thought it was a pretty reasonable estimate.
Q7 Mr McFadden: If you had not thought it was a reasonable estimate, presumably you would have flagged that up at the time?
Graham Parker: Yes.
Q8 Mr McFadden: The Government also said that the estimated cost of the policy that would be funded through the sale of this—the lifting of the cap on student numbers—would be £5.5 billion over several years. Did you or do you concur with that estimate?
Graham Parker: I am sorry; I can’t remember the actual details, but what we have done now in our recent Fiscal Sustainability Report was look at updated estimates of that. If I can find it in—
Q9 Mr McFadden: I don’t think there is that much dispute about the £5.5 billion. It is not something I am making up. I am taking this from the Autumn Statement.
Graham Parker: No, but we have just done an analysis of student loans for our Fiscal Sustainability Report, and there is a big difference between this year and last year for the actual additions to net debt because of the increased student numbers. We gave a full analysis in there of what we thought the cost is now.
Q10 Mr McFadden: Given that the sale is not going ahead and the estimated receipt, the central estimate of which was £12 billion, is not going to be in the Government’s books but the policy that this would fund is going ahead, which is the lifting of the cap on student numbers, where does this leave us?
Graham Parker: We will obviously have a proper look at this when we do the Autumn Statement forecast. In the short term, if nothing else changes, you would have higher public sector net debt, but over time this would gradually reverse because you would get the money. Now they are not selling it, the Government will get the repayments over time but over a long period of time.
Q11 Mr McFadden: Wendy Piatt, who is the Chief Executive of the Russell Group of Universities, has said that if the Government no longer intends to use the sale of student loans to fund the uncapping of numbers then, “We would urge them to abandon this idea.” Presumably, the OBR is not going to make a statement on the wisdom or not of policy, but you would make a statement on whether the numbers added up to pay for it.
Graham Parker: As I say, we will look at it properly when we do our next forecast.
Q12 Mr McFadden: That seems quite distant. We have had the announcement at the weekend from the Business Secretary. This means a forgone receipt of potentially £12 billion. Has the Government been back in touch with the OBR in recent days to ask you to look at the effect of this announcement?
Graham Parker: Not to my knowledge, but I have not talked to the chairman since the weekend.
Q13 Mr McFadden: It is quite a significant announcement that, as you said, in the short term at least will mean higher public sector debt, but they have not come back to you and said, “We asked you to cost or at least ratify our assumptions about the sale of this asset and the policy for which those sale proceeds would be used. We have now changed one leg of that equation by abandoning the sale.” They have not come back and spoken to you about that at all.
Graham Parker: No.
Q14 Mr McFadden: This leads to a broader policy question for me. You started in your response to the Chairman by saying, “The assumptions are ours now.” That is not strictly true here, is it? You were more ratifying the Treasury’s assumptions on both sides of this equation.
Graham Parker: On policy costings that is true.
Mr McFadden: So you are ratifying their assumptions.
Graham Parker: We are making them ours by putting them in our forecast.
Q15 Mr McFadden: Yes. You have said on both counts that you thought those assumptions were reasonable or, if they were not, you would have flagged that up at the time. Am I right?
Graham Parker: Yes.
Q16 Mr McFadden: Given that the OBR does this exercise twice a year for the Budget and the Autumn Statement, what does the OBR do when it costs a policy for a Budget or an Autumn Statement and then that policy is abandoned?
Graham Parker: We will take it out of our next forecast.
Q17 Mr McFadden: So you don’t do anything at the time? You wait until the next statement and then you do it there. Is that an arrangement you are entirely happy with?
Graham Parker: I think it probably has to be like that unless we are going to get into the business of doing more than two forecasts, and two is probably quite enough. What you do not necessarily know from the pure announcement is what the knock-on effects are on other things. You possibly can’t look at these things in isolation, so you would have to look at other big areas of the forecast and bide your time with doing that.
Q18 Mr McFadden: To complete this, if the sale of an asset for which it was estimated £12 billion would be raised is abandoned, that is going to mean a revision in your forecasts. Correct?
Graham Parker: Yes.
Q19 John Mann: Good morning. If someone else was doing the job rather than you, would it make much difference?
Graham Parker: It depends on who they were.
Chair: That could be answered in a range of ways. Are you irreplaceable? That is one aspect.
Graham Parker: No, I am certainly not irreplaceable. I think with the current setup there probably would not be that much change if just one person was changed at a time because the others could step into the breach in the short term. There would have to be changes probably because if it was not me the newcomer probably would not have so much detailed knowledge of the actual fiscal forecasting process.
Q20 John Mann: I look at your forecasts and the more years the OBR is there the more one can take issue, challenge, congratulate, or a combination, because there is a time series to analyse. Therefore, whether you are right or wrong has an importance, but in the political world when there is a time series that can be analysed that gives some political certainty. We are in fixed-term Parliaments now. How much of a consideration is that in terms of how the OBR operates?
Graham Parker: First of all, I can say we are probably never going to be right. It is just a question of how much we are wrong by and, perhaps more importantly, whether we are equally wrong in both directions. No forecast is ever going to be exactly right, but what we are trying to do is provide a central forecast and illustrations of the uncertainty around it. I think we have done reasonably well in that respect because we do this analysis every year. We will publish another forecast, the evaluation report, in October I think, where we do look at our forecasts. We present the time series ourselves and try to explain why things have gone wrong and then use that to improve the forecast. I think a very important part of our role is to publish our own evaluation of this and we do that every year.
Q21 John Mann: One of the beauties of certain institutions in the economic world is that, with so many variables and unpredictables, the longevity creates some certainty in itself. How much of a danger is there that the ethos of the OBR changes mid-term, by which I mean mid-Parliament, and then creates a political dynamic of itself?
Graham Parker: I am not sure there is that much of a risk of our ethos changing. The way we approach each forecast is pretty much the same. You try to look at it in a professional and objective way every time.
Q22 John Mann: The team is the same. Isn’t there a good argument for saying fixed-term Parliaments, fixed-term OBR; that the methodology ought to be there even for a period of, say, 10 years so that there is a consistency of approach for better or for worse and that any change of approach ought to be predesignated—as in, change of team, reassessment overall—rather than on the margins; otherwise a political dynamic comes into play? You are seen as independent and uncontroversial and that has been a success. That falls apart if there is perceived to be or could be argued to be a change of approach in the run-up to an election.
Graham Parker: I guess it might happen if you did have a change of personnel at BRC; certainly, if you had, say, two members change at once, there would be a risk of that. The initial contracts were staggered so that would not happen. We are trying to do that now so it won’t happen again, so that you would never get two people changing at once, hopefully, or you certainly would not plan to. On the methodology point, there is not just one methodology. There is the methodology, a macro-economic model and so on for the economic forecast, but each time we will look at different segments, so there’s a lot of judgments that go into that. On the fiscal side, there is a different model for every tax. You are never going to change all the methodologies at once; you just could not do it. I do not think there is a big risk of any wholesale change to the actual methodology of the office.
Q23 Steve Baker: Mr Parker, good morning. In your questionnaire, you note on policy costings that: “Most scrutiny and challenge is conducted via correspondence, however for more novel or complex measures formal star chambers are held where the estimates can be discussed in detail.” Can you tell us how often you have to resort to such star chambers, who attends and is it a debate with the Government or an interrogation?
Graham Parker: I am just trying to see if we have got anything here so I can answer the first one properly. From my memory, there were probably a dozen or so meetings on policy costings at the Autumn Statement. Who attends? It is us. There is usually at least one member of the BRC, which would be me and usually two or possibly three of us at these meetings. Quite a few of our staff attend. There will be somebody from the Treasury scorecard branch who co-own all the costings and then it is almost entirely the analytical officials from the Department who has produced the costing note. It is normally HMRC or DWP, and if it was HMRC, there would be an expert on the policy, such as a technical inspector of taxes.
Q24 Steve Baker: It sounds like a dozen large meetings over probably a fairly short period. Over what duration do they take place?
Graham Parker: We do try to spread it out as much as possible. We try to urge the Treasury to give us some of these measures fairly early on in the progress. For the Budget, hopefully, it will start pretty near the end of January or even February.
Q25 Steve Baker: How effective would you say that these meetings are?
Graham Parker: I think they are pretty effective. On the other part of your question, whether it is an interrogation or a discussion; it is like a challenge. They will go through the costing paper for us and we will look at each section and ask them lots of questions on each bit, so in that sense it is more an interrogation by us.
Q26 Steve Baker: How much time does the OBR spend looking into things that were not serious propositions or that were not later used?
Graham Parker: As little as possible. We now try to get the Treasury to give us an indication of how likely each one is to make it to the finishing point, but it is not perfect.
Q27 Steve Baker: I infer from what you have just said that it has been a problem in the past.
Graham Parker: Yes.
Q28 Steve Baker: Could you give us some indication of the extent to which it has been a problem?
Graham Parker: As I say, it has got better, but there have been a couple of times where I felt we have been wasting our time on too many measures. You have a meeting; they would do the note again; we would have more questions; and then, “Oh, we’ve dropped it.”
Q29 Steve Baker: Perhaps that is a subject to which we might return on another occasion.
Graham Parker: You mentioned the questionnaire. I was very conscious I could not answer your question properly in the questionnaire because Treasury still own the costing process. Where you asked for specific details, you have to ask the Treasury if you want to know and you will have to ask the Treasury about dropped measures as well, because that is something else they get a bit nervous about us talking about.
Q30 Steve Baker: Given that the Treasury still owns the process, to what extent has the creation of the OBR improved the overall quality of that process?
Graham Parker: I am not detecting any kind of political attempts to change the actual costings. The material we get is usually pretty good. There are gaps in some of it, which we tease out, and we do change quite a few of the costings. We have made a difference at the margins on that to the actual quality of the costings. They are probably more consistent with our forecasts. They take in more of the knock-on effects and other things than they would have done. I think because of the way we add this extra level of challenge that probably was not there before from people who are supposed to know about these things, it does improve it overall. It is still a very uncertain process.
Q31 Steve Baker: To what extent have you received policy costings that have been completely inappropriate?
Graham Parker: Sorry, I am not sure I quite understand.
Steve Baker: Just talking through this process and how the exchange has gone to and fro, to what extent have you received a policy costing that you have been able to look at and dismiss immediately as rubbish?
Graham Parker: I wouldn’t call it completely rubbish. That is why I am struggling. We have received some that are far too thin; they have almost presented the answer without giving us any evidence. That is the most extreme case of that. They almost say, “Here is the policy and here is how much it costs,” with not much information in the middle.
Q32 Steve Baker: Are the policies that are too thin frequently withdrawn by the Government?
Graham Parker: No. They just come back with the information we ask for, normally.
Q33 Steve Baker: Have you seen very many policies just simply withdrawn by Government?
Graham Parker: We do see some, yes.
Q34 Steve Baker: Could you put some kind of quantity on it proportionately?
Graham Parker: Quite small. A lot of measures get dropped, but that is not because we are—there have been one or two that have not been pursued in that fiscal event because of our interventions.
Steve Baker: Would you say it is single digits of per cent. as a proportion?
Graham Parker: Yes.
Q35 Steve Baker: In his letter to the Chairman of the Treasury Committee, Robert Chote noted that under the present arrangements a Chancellor might ask the OBR, “To certify the costing of a measure that the Chancellor intended to include in his party’s manifesto, so as to say that it had been endorsed by the OBR. Or he might ask the OBR to certify a measure put forward by an Opposition party, so as to say that the costing endorsed by the OBR was different from that put forward by the party in question.” Mr Chote says that would be an abuse of the arrangements and the OBR would make your concerns public. Have you seen any evidence that the Government has attempted such costings?
Graham Parker: Not yet, but I think if there was ever going to be anything like that, certainly for the Opposition party, it would have to be after the manifestos were published. If it was going to happen, I guess it would the fiscal event or two before the election.
Q36 Steve Baker: Just to be clear, you think it would only happen after the manifestos were published?
Graham Parker: If the Chancellor really wanted to put a different costing on something in an Opposition party manifesto it would have to wait, wouldn’t it?
Steve Baker: You don’t foresee circumstances where the Chancellor—
Graham Parker: There could be variations of that. If the Opposition party were saying, “We are going to do this now,” they could do it. We have not seen anything, but I thought the most likely time for it to be happening would be in the fiscal event.
Q37 Steve Baker: Mr Chote has said that the OBR would make concerns public. At what point do you think those concerns should be made public? To you, what would be a trigger event that would cause a public statement?
Graham Parker: If the Chancellor made an announcement—a public statement about us having said this—that would probably be the trigger for it.
Q38 Steve Baker: How would you make your concerns public?
Graham Parker: I guess probably what we would do, in parallel with what we did when the Prime Minister sort of misquoted us, is write to the Chancellor and copy it and publicise it.
Q39 Chair: That is a very interesting set of exchanges and, quite reasonably, you have said you do not have at your fingertips all the data required to answer some of the very specific questions. Perhaps it might be helpful if we set out on a piece of paper exactly what those questions are and you could come back to us and fill in the gaps and make any amendment to the oral record or supplement to the oral record you have just given us on the number of cases, the number of people attending meetings, these sorts of things.
Graham Parker: The number of meetings certainly. The people attending would be quite difficult because it does vary from meeting to meeting.
Chair: The cases where these proposals have not been pursued for forecast were also interesting. We will write to you rather than pursue the conversation now.
Graham Parker: Okay. Fine.
Q40 Chair: There was a second aspect to Steve Baker’s questioning, which he began with, which was about the quality of the certification process, and Pat McFadden touched on this. The alternative is to create a full, all-singing, all-dancing, Congressional Budget Office- style operation, isn’t it? The argument for that is that it removes this task of us relying on you—that is the three of you for the most part—to demonstrate that you really are independent. The disadvantage with it is that there will be a public expenditure cost because the Treasury will retain its own in-house function, so you will have two sets of people doing this work. You are clearly pregnant with quite a few thoughts. Your mouth was open while I have been trying to get to a question. Why don’t you say what you wanted to say?
Graham Parker: There would clearly be lots of public expenditure implications and a lot of duplication of effort, and you would also have problems about access to data. Clearly, at the moment, we are allowed access to all the Government data we think we need but not if it is covered by confidentiality. We can’t access individual taxpayer records, for instance, and I am not sure whether the CBO can.
Chair: I very much doubt it.
Graham Parker: That would be a big obstacle to doing some of the costings if that happened. At the moment, some of the corporation tax costings and some of the anti-avoidance costings rely on tax confidential information, which we certainly could not do by ourselves, and you would need a very much bigger organisation with access.
Q41 Chair: The bigger question I was trying to move towards asking is: is it worth spending this money to get a Congressional Budget Office-style operation or not? That is the question that every few years certainly Parliament needs to think about.
Graham Parker: My personal feeling is probably not.
Q42 Chair: Why not?
Graham Parker: I don’t think you would end up with anything much better than the current system with us being the gamekeepers.
Chair: The flipside of that point is we need to keep on relying on you.
Graham Parker: I am afraid so.
Q43 John Thurso: Can I ask you, first of all, regarding the Fiscal Sustainability Report, which you currently do annually and we recommended that that was really a bit too much and I think we ended up with biennial or twice a Parliament. What is your view of that?
Graham Parker: Certainly, you could make a good case for doing it only every other year, because our long-term projections are heavily dependent on the population projections that only come out every other year. There is a flipside. We do need some resource, some expertise on this in-house and it might be difficult to keep it if you were doing this only every two years. In a sense, it does not cost us that much to do it every year because we have a team—
Q44 John Thurso: You don’t have a team who only do that and they have one year off and—
Graham Parker: We have one or two people who are really expert in our cohort model, and they do other things as well.
John Thurso: Yes. So they could remain experts.
Graham Parker: Yes, but it is more difficult to remain an expert if you are only doing it once every two years.
Q45 John Thurso: I am not convinced of the argument that the best thing for doing it annually is that practice makes perfect.
Chair: Or that it is a way of occupying staff in between years when they might be producing something more useful, which is what you have just said.
Graham Parker: It does allow us to look at specific aspects. Every year, we do some new topics in it. This year, we have done an annex on student loans, which has all the analysis of why it is different from last year, which is partly because of the cap and partly because of revaluations and so on. You can use it as a vehicle for getting more stuff that should be of public interest out in the public domain.
Q46 John Thurso: What is the main benefit of the report?
Graham Parker: To look at the sustainability of the public finances over a longer term. That is what we are charged to do in the charter.
Q47 John Thurso: Let me move on to another question. In your questionnaire when we asked whether you would serve the full term, you said that you do not intend to serve your whole term. Why is that?
Graham Parker: If I do another three years, it will be something like seven and a half years in total, which is quite a long while. I think, by that time, I will be quite happy to go back to normal retirement.
Q48 John Thurso: Let me go on to the obvious question that comes from that—and this follows on from previous conversations—which is what work is the OBR undertaking to succession-plan both you and the team mix and the particular skills that have been brought together in this mix?
Graham Parker: It is quite difficult to succession-plan for individual members of the BRC because it is an open recruitment process. You just can’t promote people from within the organisation in the normal way. There is very much a separate process, so that will depend. What we are trying to do, as we said before, is we will try to stagger our departures as far as we can. Internally, we are doing quite a lot, because with a small organisation it is a problem. We are trying to make sure we have people who are sharing expertise and so on. We have been loaded with a few more tasks since we started, including monitoring the welfare cap, and that came with some money so we have some extra people. We are trying to increase the flexibility of what we do and what people do within the organisation, so that we will still have the necessary expertise, particularly on the fiscal side.
Q49 John Thurso: You made a very interesting point that I had not thought about before, which is that the three of you who come before us are public appointments and it is a public process and therefore you can’t succession-plan because you have to see what turns up in the process, but the rest of the team you can plan because that is a more normal organisational process. If I may say so, we have a particular set of interlocking skills that we see in the three that come before us that has made this first OBR quite a successful unit. To what extent is there a risk that we will not find a replacement for you or any other individual and that might significantly weaken the team? I don’t imagine there are people with your experience queuing up in the hundreds for this job.
Graham Parker: There is obviously a risk, but I think the organisation should be able to cope with it. We might have to rely more on the staff for certain things, for certain parts of the costing process for instance. We might have to rely more on the staff to do a lot more of the challenge themselves rather than Robert or me doing it. That is in part why this bigger organisation we have now, because of the welfare cap and things like that, does help us do this.
Q50 John Thurso: Would it be fair to say that in this first phase we have relied very heavily on the three of you at the top, but that with the passage of time, the skills that you brought in the formation should have been passed on into the culture of the team and it is the team that should be delivering it?
Graham Parker: You rely on us, but we rely on our staff already. We could not do what we do if it weren’t for the staff underneath us. There may be a case for you talking more directly to them in the future possibly; I don’t know.
Q51 John Thurso: Presumably, it is possible and wise within a public appointment process to ensure that you have identified some people who may wish to be considered and who have the relevant skills.
Graham Parker: Yes, I am sure that will happen.
John Thurso: It is not just an open-fishing competition.
Graham Parker: No. I am sure people have been encouraged to apply for things in the past.
Q52 John Thurso: How hard do you think it will be to replace the chairman?
Graham Parker: To get somebody as good as the current chairman, very hard. Hopefully, there will be suitable candidates out there.
Q53 John Thurso: Finally, you mentioned the welfare cap and your annual report notes that, from autumn 2014, you will be responsible for assessing the Government’s performance. How much progress have you made in preparing for those responsibilities?
Graham Parker: We do now have the additional staff in place. The last one joined about a month or so ago, so we have the resources in place and the work has started. As well as reporting on the performance, we will do a welfare trends report, so work has started on that.
Q54 John Thurso: I imagine this is something that very much plays to your skills and experience. In other words, have they put you in the lead on this?
Graham Parker: I think we will all have something to contribute on that, because certainly Robert’s IFS experience—
John Thurso: What would your particular role be?
Graham Parker: One of the most difficult areas we are going to have in this is in assessing what changes are the result of policy measures and what are the result of forecasting measures. That will be an area where I suspect I will have to get involved.
Q55 Chair: Again, a very interesting set of responses to some important questions. Can I just clarify one or two points? On John Thurso’s questions on the sustainability reports you said, “We need to keep the team together,” as one of the justifications for more frequent sustainability reports. Why do you need to do that? Why can’t you assemble a team of experts in this field every few years? After all, there is a heap of outside expertise in this field. Unlike the very specialist area you are in, there are loads of people who think about this in the academic and economic and wider forecasting community.
Graham Parker: I was thinking more of our actual running and maintaining our cohort model that we use for the long-term projections.
Q56 Chair: Do you need to maintain your cohort model for the long term projections?
Graham Parker: We certainly will need to use it every time we do a long-term projection, so we need to make sure it is suitably updated, yes. We have somebody who is quite good at doing that.
Q57 Chair: If I may say so, you are not presenting a convincing case for the staff retention point that you made on the need for frequent sustainability reports. It sounds as if the product, the need to produce sausages, is being driven by the number of sausage makers.
Graham Parker: No. Clearly, we have to produce an annual report at the moment. It is in the Charter for Budget Responsibility that we have to produce an annual report on the sustainability of the public finances. That is the overriding reason why we do it. I am just pointing out that there would some disadvantages, possibly minor disadvantages, to doing it every two years.
Q58 Chair: Even those population projections, to which you allude as a justification for doing it every two years, move around quite a lot too, don’t they?
Graham Parker: They can do.
Chair: Therefore, looking on a 50-year view of what changes every couple of years isn’t telling you very much, is it?
Graham Parker: They move around because there is more up-to-date information on life expectancy and on fertility rates, so if you are going to look 50 years ahead, you might as well use the most up-to-date information.
Q59 Chair: Might as well use it. Well, we are getting your degree of enthusiasm on that point as well loud and clear. There is also a question of whether you should be looking at a 50-year view or a shorter time horizon. Do you really think 50-year sustainability reports are worth much? For example, have you gone back and taken a look at sustainability forecasts written 50 years ago? They were written. Indeed, they were more common 40, 50 years ago than they are now.
Graham Parker: What we have done in this report is had a look at the past population projections.
Q60 Chair: That is only a small part of what I am suggesting to you.
Graham Parker: Yes. There was quite a lot of change in those, so we have looked at that. But this is a projection.
Q61 Chair: These are in the category of novels, aren’t they? They are in the category of fiction, really. Nobody has a clue what the world is going to look like in 50 years.
Graham Parker: I think that is a bit extreme.
Chair: Really? Have you had a look at—
Graham Parker: You can draw quite a few conclusions that are fairly certain about what is going to happen because of the age distribution and what is happening to life expectancy; the percentage of the population that will be of pension age will be higher.
Q62 Chair: I am not going to pursue this for very long, but isn’t a central failing of the long-term planning on pensions derived from catastrophic mistakes made by actions quarter of a century ago, never mind half a century ago, about exactly the questions you are just alluding to—life expectancy and morbidity rates as well?
Graham Parker: Yes.
Q63 Chair: On the forecast itself, these are now separated by only three and a bit months. To raise what you said earlier, I think your words were two forecasts a year is quite enough. What is the justification for two forecasts conceptually, as opposed to one forecast a year, and what do we get with two separated by three months that we would not get in practice with one a year?
Graham Parker: I think the main justification must be on the fiscal side. If the Chancellor wants to do two fiscal events a year, he probably needs a forecast of the public finances to make those fiscal policy judgments.
Q64 Chair: So it is a political driver? Was that a yes?
Graham Parker: Yes.
Q65 Chair: I am just trying to clarify that there is not an economic justification for this.
Graham Parker: Not that I am aware of.
Q66 Chair: That brings us on to the question of the quality of the forecast. Would you say that the forecast is the same quality, somewhat better or not quite as good as it used to be in the old days when it was in-house?
Graham Parker: We do look at this every year.
Q67 Chair: It is eventually susceptible to some statistical measurement, too.
Graham Parker: Yes. On the public finances, we are better than the historical average of official forecasts.
Q68 Chair: Is that because the element of aspiration that historically had been built into forecasts has been ruthlessly removed by a group of wholly independent OBR forecasters, or is it a spot of luck?
Graham Parker: I wouldn’t say it was either.
Chair: I think that is a fair reply.
Graham Parker: I wouldn’t like to estimate how much the aspiration you refer to contributed to past errors. We are still trying to do better, but our errors have been slightly lower on receipts and spending than the historical average. I think we are not doing badly but probably could do better.
Q69 Chair: Can I go back to this certification versus Congressional Budget Office model issue one more time? If we are going to stick with the certification approach, this Committee is going to need a good deal of reassurance every year that you have not gone native. Do you think it would be of assistance if you had a statutory duty to report any interference?
Graham Parker: We would report it anyway, so I am not sure it would make that much difference.
Q70 Chair: Do you have whistle-blowing arrangements in place at staff level? As you say, you rely on your staff. They are more likely to pick this up than you are initially.
Graham Parker: I have not looked at them personally, but I am pretty sure we do have those arrangements.
Chair: Perhaps you could send the Committee the whistle-blowing arrangements that you have in place. I think it would helpful to see them.
Graham Parker: Okay.
Q71 Chair: Finally, the greatest problem with forecasting is that they tend to be believed, isn’t it? Then, of course, because they are always wrong people say, “Either the Government or the OBR has got it all catastrophically wrong.” Isn’t that correct?
Graham Parker: The biggest problem is that they are believed as point estimates.
Q72 Chair: Therefore, isn’t a crucial part of your work to use your imagination, which is a tough challenge that I am proposing here, to find ways of communicating to a wider public that it is not that you are not doing your best; you are doing your best and they are very thorough pieces of work and they are an essential part of government because the Government has to take decisions on the basis of some assumptions, but nonetheless they are exceptionally uncertain and are highly likely to be wrong? Have you given further thought to how to try to convey that and constantly erode the inevitable political and other pressure to give a sense of spurious credibility to these forecasts?
Graham Parker: For every report we do, we emphasise the uncertainty. We always present the fan charts. I would be worried if the outturns were outside our fan charts. We present the fan charts, we do other ways of looking at—
Q73 Chair: The long tail risk in the fan chart is so wide it would be impossible.
Graham Parker: Quite. I would be quite worried if it was outside the plus 25% bits of it.
Chair: That is a narrower bit.
Graham Parker: In everything we do, we emphasise the uncertainty. Every presentation we do, we emphasise the uncertainty. All our working papers and so on include something about uncertainty. So we do try, but if you are a journalist writing an article about this, you are going to use the point estimates, aren’t you? I am not sure how much we could hope to remove the concentration on point estimates.
Q74 Chair: Thank you very much, Mr Parker. Unless colleagues have any other questions, thank you very much for coming in. It has been extremely interesting, as it always is when we see the OBR.
Oral evidence: Reappointment of Graham Parker to the Budget Responsibility Committee, HC 580 13