Item 2 is an evidence-taking session for our inquiry into proposals for an independent fiscal body. I welcome Dr Jim Cuthbert, Professor Campbell Leith and Professor Jeremy Peat. Good morning to you all. I understand that no one has a statement to make, because you have all made submissions, on which we will ask you questions. Questions will probably be specific to one individual on the panel, but other panellists should feel free to contribute their views. I will start the questioning and then I will open the session to other members.
I thought that, like all the submissions, Jim Cuthbert’s submission was excellent. “Missing the Risk” is an in-depth paper, which I very much enjoyed reading—it provides great food for thought. In the introduction, you say:
“the second main conclusion of this report is that the OBR should be radically changing its emphasis. It should be putting much less effort into forecasting, and much more into assessing the potential vulnerabilities to which the economy is subject”.
Am I right in saying that you feel that a Scottish fiscal body should have a different focus from that of the Office for Budget Responsibility and should not focus just on predicting Scottish economic prospects more accurately?
When such a body is set up, the remit is critical. What should not be overemphasised is the forecasting role, but what should be emphasised is the role of assessing risk. I set out in my paper why the forecasting role is relatively meaningless.
I did not mention in the paper something that points in the same direction, which is another difficulty of forecasting—the difficulty of calling turning points. It is quite easy to see that something like the banking system in the 2000s cannot continue on its current trend for ever and that something will change, but realising that is different from being able to say when the change will occur. When a forecast is made, the turning point must be called, which is almost impossible to do.
The remit is very important. I do not know whether the committee has seen the response that I got from Robert Chote to my Jimmy Reid Foundation paper. The clerk has seen it, but I did not ask him to circulate it, although it is available to be circulated if members want it.
The response was positive, although Robert Chote did not fully agree with me. He made a comment that is significant in the current context. He said:
“Some observations on your broader points, in no particular order:
1. We have to emphasise a central forecast in part because, like it or not, the primary task we have been given is to judge whether the government has a better than 50% chance of hitting its targets, which implies comparing those targets to a median forecast.”
In a sense, he is saying that, given the OBR’s remit, it must go down the road of forecasting. I would say that a lot follows from that.
The remit is vital. It should not debar a body from forecasting, but it should not make forecasting the primary role, and it should make assessment of risk a major role of the body.
Yes—that is quite interesting. We have not seen that paper yet, but the clerk will circulate it to committee members.
What do other witnesses think? Is the OBR remit, for example, too narrow? Could or should the Scottish remit be wider?
The OBR remit is among the tightest of any fiscal council in the world in that it must give a purely positive analysis of the fiscal situation. It cannot look at alternative policy scenarios—it cannot make recommendations that policy A is preferable to policy B—but most other fiscal councils can do that to some extent. Were I designing a remit for a Scottish OBR, I would broaden its remit substantially relative to that of the OBR.
I agree. First, Jim Cuthbert is right that the uncertainties around any forecast are inevitable and considerable. I think that it was Eddie George who said:
“When it comes to forecasting, there are only two kinds of economists, those who don’t know and those who don’t know that they don’t know.”
It is the latter type who is always the most dangerous. You must accept that there are uncertainties particularly but not solely around turning points, so no one should take any forecast as gospel and inevitably correct. I like the approach of scenario-based analysis, in which the risks to the forecast are looked at on the basis of a range of scenarios, with that work informing decision making instead of in any way thinking that that dictates decision making. The uncertainties are too great to permit that.
I also agree that expanding the remit makes sense. I am speaking on behalf of the Royal Society of Edinburgh. In our submission, we suggest that, in addition to the forecasting role, an additional task should be considered, namely looking at what we call
“strategically important forward-looking issues and commitments.”
Analysis of those topics should be carried out. We gave the implications of demographic change as one example.
That role should be kept wholly distinct from the forecasting work, but having an independent body to undertake such work, and using the same sound base of skills to form a body that can undertake the dual tasks, would provide a much more valuable input into the Parliament and decision making more generally.
A Scottish OBR should have a wider role. It must have some form of forecasting role, but it should be made to fully accept the scenario and risk issues. It should also have an alternative role in which it is more thoughtful and considers matters in more depth, with the results fed in to inform decision making.
Paragraph 8 of your submission says that the RSE recognises that
“concerns have been expressed as to whether the forecasts provided by the UK OBR fully capture the distinct circumstances and drivers in Scotland.”
What does the RSE think about that matter?
We have not gone into depth in looking at the full detail of the OBR forecasts. However, from talking to Robert Chote when he is up here and from listening to the evidence that he has given to you in the past, I think that it is evident that the OBR does not necessarily have the full understanding of all the niceties of the issues that are around and that, inevitably, if its main work is at the UK level, its work at the Scottish level may be part of a narrower element of its remit and therefore may not necessarily fully capture all that should be captured.
If one adds a task for a Scottish independent fiscal institution alongside the work of the OBR, the interaction between the two bodies would have to be looked at very carefully. One does not want to have competing views that are used politically to beat different Administrations over the head. One could perhaps allow an IFI in Scotland to drill down in more depth into the issues that matter, whether in the context of the Scotland Act 2012 or under any further devolution or constitutional change that takes place.
How you set up the relationships would need a lot of care, but having a body based in Scotland that is doing work on the Scottish economy as part of its day-to-day activities would allow a better, more informed view on issues on which the Scottish aspect is critical.
One issue to do with newly devolved taxes that has been of concern to the committee is the fact that the OBR has, in effect, just been extrapolating UK figures. Particularly with the land and buildings transaction tax and landfill tax, there does not seem to be any correlation between what the Scottish Government believes is likely to be raised and what is forecast by the OBR, simply because the OBR does not feel that it is able to make more nuanced forecasts. Is it your view that there would have to be an agreement between the two bodies in a devolved situation so that we had just one joint forecast?
It is difficult to see how the two bodies would interrelate while we were in a period of devolution—and increased devolution, if that was to take place on the fiscal side. Disagreement would be difficult, because there has to be agreement on what the tax take is likely to be, which feeds into calculations of one type or another on deficits and the like.
I suspect that, if a role was taken over by a Scottish IFI in the area that you mentioned, for example, it would probably determine that sufficient data, or sufficient analysis of data, was not available to the OBR. It might well undertake or commission more detailed work, which the Scottish IFI would then utilise in consultation with the OBR to come out with more refined work on Scotland that started from the base where the OBR left off but which used the more detailed, Scotland-specific data and analysis. One would hope that a combined view of the two bodies would then be produced that took a central OBR position but added that much clearer and more carefully articulated Scottish dimension.
Professor Leith, you specifically touch on the matter in paragraph 8 of your submission, where you state that a key question about a Scottish IFI and the OBR is
“whether they operate cooperatively ... or provide separate, potentially divergent, forecasts of common items.”
You add:
“One advantage of the former approach is that the two bodies share information and expertise in producing their consensus forecast”.
Just to pin that down, are you suggesting that the OBR would continue with its broader UK forecast but, on Scotland, it would work together with the Scottish IFI to provide a joint report, and that it would be agreed beforehand that they would do so? If so, what would happen if there was a divergence of viewpoints between the organisations? They would not necessarily agree on the forecasts.
My view is that the two organisations should work together. As soon as there was a disagreement, the media would focus on that, whereas the job of an independent fiscal council is to hold policy makers to account and not to get into a dog fight with another forecasting institution. Disagreements should be dealt with between the two organisations in the process of producing a joint forecast and, ideally, they should be resolved before the forecast is completed.
It might help the situation if the status of the forecast was made absolutely clear. One of the suggestions that I made in my OBR paper is that, when the bodies present their forecasts, they should present them more as projections, saying, “This is our best view of what’s going to happen, conditional on certain assumptions.” The two bodies would produce things that were closer to projections than forecasts. A projection is a view about what is going to happen that is conditional on certain assumptions. If both bodies made it absolutely clear what the conditioning factors were—they could say, “These are our conditioning assumptions, and this is our best guess”—it would be clear where any disagreement arose, and one could see that the OBR was taking a different view about factor X or factor Y.
I would hope that, in terms of methodology, the two bodies would be able to agree on a model so that it did not come down to a difference in the model that was being used, and that, if there were any differences, they would just be due to specific assumptions that could be identified and on which people could take a view. That would be important information for policy makers. If the OBR was saying one thing and the Scottish body was saying another about what was going to happen on a particular factor that can be influenced by policy, policy makers would need to pay attention to that. If the assumptions were clear, the whole process could be valuable.
09:45
Can I add one small additional point? I agree entirely with the suggestion that Jim Cuthbert has made, but it is also important that the model that is used is wholly transparent, so that others can see it, can know what data and assumptions are built into it, and can carry out their own calculations and some what-if work on the basis of it. It must be transparent so that everything is set out and available to be tested.
Dr Cuthbert, in the third paragraph of the introduction to your paper, you say:
“the OBR’s treatment of risk is grossly inadequate”.
Your paper explains in detail why you believe that, but can you expand on that now for the record?
As I have already said, part of the problem goes back to the OBR’s remit. Any body will have relatively limited resources, and the OBR clearly commands some good, high-quality resources, but it is a relatively small body. It has to produce a detailed forecast because of its remit, so it must put a lot of resources into that, and I do not think that it has the spare capacity to do a thorough assessment of risk. The way that the OBR looks at risk tends to be by looking at past forecasting errors and at a small number of different scenarios. That is all valuable stuff, and it should be doing that, but there are big things that it is not taking account of. A classic example identified in the paper is the whole question of quantitative easing. We are in uncharted territory with quantitative easing as regards when and whether it is unwound, and the unwinding is likely to be a difficult process, as we saw recently when it was suggested that America might start tapering off quantitative easing. The suggestion shook world markets quite profoundly, so it had to back off.
Similar risks attach to the whole process of quantitative easing here, yet the OBR just makes the bland assumption that quantitative easing will, in due course, start to be unwound in the middle of the forecast period, and that it will be unwound without disturbing the futures market and what is predicted for interest rates. That is a big assumption for the OBR to make, and it illustrates how what it is doing is dependent on a number of very large assumptions. Other risks have to do with external shocks, and I do not need to enumerate the potential external shocks that are waiting in the world just now or describe the risks surrounding the occurrence of external shocks and those surrounding the robustness of the UK economy when coping with such shocks. I will not go into the same level of detail that is set out in the paper but, given the increase in public sector debt since 2008, the robustness of the UK economy to respond to a major external shock must be called into question.
There is a whole number of such factors that are not dealt with and which would be difficult to deal with in a forecasting context, but it would be useful to have some appreciation from the OBR of the likelihood of those risks, how vulnerable we are and what could be done to ameliorate any potential dangers.
Would you be keen for a Scottish IFI to have risk assessment at the heart of its remit?
Yes, absolutely.
Professor Leith, would you like to comment?
I am a little bit less negative about the role of the central forecast in the OBR’s work. It is true that it makes a number of assumptions in order to produce that forecast, but the vast majority are assumptions about asset prices, which are notoriously difficult to forecast as they involve forward-looking behaviour. Share prices today depend on expectations of share prices tomorrow. If I buy a share today, the price that I am prepared to pay depends on what I expect to sell it for tomorrow, so we can get into all kinds of technical issues to do with expectations, and the assumptions that the OBR makes are largely to sidestep those technical issues.
The big assumption that the OBR makes, which is perhaps less innocuous, is the assumption about monetary policy stabilising inflation within five years. However, given the usual time lags that we assume for monetary policy, in normal circumstances that is not too heroic an assumption. It may be that, in the current period, with interests rates at the zero lower bound, it becomes a stronger assumption, but it is one that the OBR would acknowledge that one may wish to deviate from, and it reserves the right to deviate from that assumption if appropriate.
The assumptions that the OBR is making when it produces the central forecast are not too unreasonable. The central forecast has thrown up instances when the OBR has said that the Chancellor of the Exchequer will not meet his fiscal plans, and the chancellor has sometimes responded by extending the duration of the austerity measures. Sometimes he has acknowledged what the OBR has said but has not adjusted policy. The central forecast is, to some extent, doing what it is supposed to do and is highlighting when fiscal policy might not allow the Government to fulfil its fiscal plans.
The OBR also undertakes some scenario analysis that is, as Jim Cuthbert says, quite limited in scope. However, it gives a feeling for the size of the multipliers that are involved, such as how much a negative shock of a given magnitude would affect the fiscal outcome. If you think that that shock is inappropriate or that bigger shocks are likely, you just scale up using the multipliers that the analysis has revealed. That sort of analysis provides information. It does not account for every possible scenario, but it gives a feel for the sensitivity of the central forecast to shocks, and that information can be useful.
I veer more towards what Professor Leith has just been saying than towards Jim Cuthbert’s suggestion, which is to take away the central forecast. We must remember that the basic function of existing institutions such as the OBR is to forecast and evaluate likely outcomes on the basis of current policies, and then to comment on their sustainability. There is therefore a requirement to indicate clearly to the Government the extent to which its policies are sustainable and the extent to which the outcomes are likely to be as the Government anticipated or would wish.
I accept that no point forecast can be taken as accurate and likely to happen. It might be that one has to have a range that expands around that forecast over time. However, it is still necessary to have a central view that enables comments to be made on the sustainability of policies and their implications for the economy and policy decisions.
In Scotland, what an IFI would need to do would be different, and it would also be contingent on the result of next September’s referendum. If Scotland were to go forward to independence, an IFI would need a much broader remit than if we were to remain in a devolved—albeit potentially increasingly devolved—state. It is highly desirable to set up an IFI for Scotland now, but one should consider a somewhat narrower remit that fits the present circumstances while thinking through what would be required in the event of a yes vote. At that stage, the question of exactly how the full forecasting role would work would become wholly pertinent. There is time to think that through, and we need time to develop the appropriate macroeconomic model. There is not a fully tested and ready model available in Scotland, so we should take it one step at a time; we must not, however, move away entirely from having some view of the future, because such a view is sometimes necessary to constrain Governments, and can certainly provide them with guidance.
I just want to correct Jeremy Peat. He implied that I said that I want to take away the central forecast, but I was not saying that. There is an important role for projecting or forecasting, which should be done, but its status could be changed slightly.
To give a good example, I think that the OBR central forecast is fine, but it depends on a number of assumptions. There is a likelihood that we will reach a turning point within the forecast period. I give the example in my submission of the growth of the external assets and liabilities of the UK as a whole—they are growing at about 8 per cent per annum compound as a percentage of gross domestic product. In a sense, that cannot go on for 20 years, and it is unlikely to go on for nine years, because that would mean a doubling, from 700 per cent of GDP to 1,400 per cent of GDP, in nine years. There is therefore a good chance that there will be a turning point in the series, certainly within the next nine years and possibly within the next five years.
We should bring out and discuss such issues rather than just present a central forecast that implies that, over the next five years, things will go on sweetly on that course. By all means, we should do the central forecast, because it is important to see where we are heading at any particular point in time, but it is also important to think more constructively and positively about the factors that might lead us to depart from that path within the timescales that we are talking about.
We are well into the meeting and I have a lot of questions that I want to ask, but my colleagues are also keen to come in, so I will ask one more question. I can always ask more at the end if colleagues do not cover the issues.
My question, which is again for Jim Cuthbert, is on feedback. The fourth page of his report states:
“For certain important variables”—
for example, world interest rates and exchange rates—
“the OBR is relying upon the view which the markets are currently taking about what will happen in the future. This raises the potential for feedback effects between OBR forecasts and market views”.
That is a fascinating aspect of your report. Can you give us any evidence of feedback that you know has taken place as a result of forecasts?
I cannot give specific examples from the UK context, but an example that I have mentioned before is that when, in America, there was a suggestion that policy would change on quantitative easing, it immediately had effects on the market. That is the sort of thing that can happen. Robert Chote’s response to my report, which the committee will see, significantly, and with some rather false modesty, underplayed the likelihood of that. In relation to the feedback effect, he said:
“The fact that no-one should expect us to be better macro forecasters than anyone else—and that we are not ourselves policymakers—means I would not worry too much about the potential feedback effects you mention. You suggest that our forecasts are an important influence on the market’s view of the future—while that may be true of the fiscal forecasts, I doubt it is true of the macro forecasts.”
I just disagree with that. If the OBR were to say that GDP will grow at only 1 per cent, that would be regarded as an authoritative independent view that there was going to be a change in the UK’s economic prospects, which would be bound to have an effect on the markets. A body such as the OBR has the power and credibility that can lead to important feedback effects.
Therefore, for example, if the OBR said that the economy was going to grow by a spectacular 5 per cent next year, a lot of companies would think that their business was going to improve and so they would invest in new plant, machinery, staff and training. Alternatively, if the OBR said that the economy was going to shrink or stagnate, that would have the obvious opposite effect. Professor Leith, what is your view on that?
For me, the major feedback effect for such policy questions is the monetary policy response to fiscal policy. When an austerity package is implemented, what happens to the economy depends on how the monetary authorities respond to that package. If they loosen monetary policy, reduce interest rates and do quantitative easing, those policies will offset some of the negative consequences for GDP growth. If the monetary authorities do not do that, GDP growth will be that much weaker. The central forecasts take market projections for interest rates, but the scenario analysis that the OBR does introduces a Taylor-type interest rate rule to its forecasting model, which accounts for such feedback effects.
The monetary implications are the most important. If we had in the UK a forecast of very rapid growth such as the convener mentioned, manufacturers might tend to gear up their investment and recruitment plans, but I suspect that the markets, particularly in an era when we have forward guidance from the Government and the Bank of England, might say, “Whoa! That means that interest rates are going to go up much earlier than we had anticipated.” The shape of the forward yield curve would change substantially, as happened when there was talk about the quantitative easing programme in the US, which Jim Cuthbert rightly mentioned. That could have severe implications for the whole economy through a monetary feedback mechanism.
I think that people, including the markets, would listen to the OBR. They might not necessarily believe that it was right, but they might believe that it would affect decision making by the Government and the Bank of England, among others.
In other words, feedback effects can have an impact, depending on the prediction.
My colleagues have been very patient in waiting to come in with their questions. The next question is from Jamie Hepburn.
10:00
Professor Leith said that the OBR’s main assumptions are around asset prices, but I note that in your paper, Dr Cuthbert—you touched on this in your exchange with the convener—you say that the
“key assumptions”
are
“about the success of government policy”,
which makes
“the central forecast of relatively little interest in itself.”
You make it pretty clear that this is not a case of the OBR failing to act independently, but say that
“this is an almost inevitable feature of forecasts produced in a heavily policy influenced environment.”
My question, therefore, is in two parts. First, is it not inescapable that the forecasts of a forecasting body that interacts with the political sphere will be
“produced in a heavily policy influenced environment”?
Secondly, if that is considered to be a problem, what lessons can be learned from it in establishing an independent financial scrutiny body in Scotland? Is the solution as simple as giving the body the wider risk-assessment role that you mentioned?
I certainly agree—indeed, the main part of my paper argues—that that is an almost inherent feature of forecasting in a policy-influenced environment. The response would be twofold. First, the status of the forecasts should be lowered to make it clear that they are actually projections rather than forecasts and that they are conditional statements about what is going to happen, given certain assumptions. Those assumptions might be regarded as the most likely, but that should be made explicit to allow people to judge for themselves how likely they feel them to be.
Secondly, there should be much more emphasis on risk. Earlier, I gave the example of turning points. You might know that there is likely to be a turning point but you will not be able to forecast it. It is simply impossible to say when such discrete events are going to occur. You can make a forecast, but you need to make the assumptions explicit, make it clear that the forecast is conditional on there being no potential risk event during the period in question and provide an assessment of the risks of a turning point that would invalidate the whole forecast.
It would be quite easy to do that; the difficulty in doing so is essentially a political one. If an independent body to which one attaches great status tends to corroborate Government policy, that is a very convenient shelter for politicians. In order to move the OBR, say, into the kind of role that I have suggested, you would need to rewrite the remit that it got from politicians, and in setting up a Scottish body, you have a good opportunity not to make the original mistake and not formulate a poor remit in the first place.
Thank you for that.
Professor Leith, the rationale for establishing what you describe as fiscal councils in the introduction to your paper is very interesting. Essentially, you suggest that the case for establishing such a body in a devolved context is less clear because the Scottish Government has to present a balanced budget. Does the move towards the introduction of borrowing powers, the Scottish rate of income tax and so on change things slightly? Are you saying that there should be no body at all or that, as Professor Peat has suggested, the remit for any such body should be narrowly defined under the Parliament’s current powers?
Internationally, the basic argument that has been used to justify the establishment of fiscal councils is that Governments are subject to what is called the deficit bias. In other words, they do not save for rainy days during good times and debt tends to accumulate over time. To the extent that a policy-making body cannot issue debt, it cannot suffer from that bias so it is less necessary to create a body to correct for it.
However, given the direction of travel, with increased devolution of powers and the possibility of independence following the referendum, there is a case to be made for capacity building and building experience in this area. To the extent that the Scottish Parliament has borrowing powers, the case for a fiscal council can be made.
So the case for creating a fiscal council is stronger now than it might previously have been.
The more borrowing powers the Parliament has, the stronger the case.
May I comment on a point made by Jim Cuthbert?
By all means.
I have to say that I disagree with the idea that the OBR will inherently support Government policy. Its assumptions are transparently laid out and, in general, quite reasonable; indeed, it has said on three separate occasions that the Government was going to break its fiscal rules, and it would not say that if it were inherently biased towards saying that Government policy is always right. In November 2011, for example, it told the chancellor that he was not going to meet his primary fiscal rule and, as a result of that prediction, George Osborne extended fiscal austerity for another year. Whether that was a good policy is another discussion, but the OBR has tried to hold the Government to account.
Just to make a potential correction, I am not saying that the OBR is inherently biased—far from it. It acts honestly and independently and I would not like it to be suggested otherwise. As an analogy, if you can see that your ship is heading towards rocks a mile away, you will assume that the captain, too, will see them and take corrective action. As a result, your forecast will be that you will miss the rocks. You will only say, “We’re liable to hit those rocks,” when you are 50 yards away and the captain has no chance of correcting the ship’s course. I am not saying that the OBR is biased in any sense—it will blow the whistle and is perfectly willing to do so—but it is in the nature of things that it is liable to do that only when it is clear that policy cannot avoid the looming danger. Where there is a long lead time and policy has a chance to take effect, it must assume that policy steps will be put in place that might be successful. I repeat that I am certainly not suggesting any inherent bias in the OBR.
I suppose that another danger might be if it, too, does not see the iceberg coming, but I will leave that for the moment.
Professor Leith, you have already said a little bit about the implications of an independent Scottish body coming up with a different forecast from the OBR but I am not quite clear what the problem might be in such circumstances. You have mentioned the media bunfight that could take place but, beyond that, is it a problem in and of itself if the two bodies come up with different forecasts? After all, Jim Cuthbert made the point that it would be for policy makers to make their assessment on the basis of the full spectrum of evidence and one might argue that these are only two—perhaps competing—pieces of evidence.
At the moment, the Government can pick and choose from lots of competing economic forecasts to justify what it does. However, the idea behind the fiscal council is that a policy maker funds it, gives it special authority to make forecasts and must respond to those forecasts. Of course the policy maker can choose to ignore the fiscal council’s forecasts and implement policies that contradict its findings, but it still has to respond to them. If multiple forecasts from multiple bodies impinge on the same policy, the means of holding the policy maker to account will be less effective.
Even when different policy makers fund different bodies?
Yes, if they are forecasting the same thing.
And that is notwithstanding your point that there are already any number of independent bodies making such forecasts.
But we are talking about the two bodies that the policy makers are jointly funding to produce the forecasts that are used to hold them to account.
Okay.
Can you explain the statement in your submission that
“the limitations of available data for Scotland also place constraints on the level of sophistication that can be achieved by any forecasting body”?
I am not an expert on the availability of Scottish data but, from reading the OBR’s memorandum on how it forecasts, it seems to me that it is engaged in quite a mechanistic exercise. In that memorandum, it claims that one reason for that approach is the lack of available data. Other data that could improve the sophistication of that forecast may be coming on stream, but I would not know about that. The Scottish Government is engaged in trying to improve its macroeconomic modelling capabilities at the moment but, again, that work is not quite on stream yet. There is still a lot of capacity building to be done to enable us to produce solid, sophisticated forecasts.
That is a useful point for my next question to Professor Peat. In the Royal Society of Edinburgh’s paper, he suggests that the committee should examine whether the powers of the body should be invested in the OBR. Given what Professor Leith just mentioned and that, at our last meeting, Robert Chote himself said that the OBR would be unlikely to give the level of detail that the Scottish Government needs, does that rule out our giving it that role?
The present arrangements under the Scotland Act 2012 certainly merit reconsideration. If an independent financial institute of some sort was appointed in Scotland, it would be right and proper for some of the role that the OBR fulfils at the moment in relation to devolved fiscal powers to be transferred across to the Scottish institute. That would lead to a greater probability of the right data being analysed in the right way within Scotland.
The Scottish institute would have to work closely with the OBR. I worry about different forecasts from different bodies, particularly when a parameter comes out of them—such as the level of expected borrowing—that would be important to decisions that the two Governments make jointly. If somebody else’s independent forecasters came with different forecasts and the politicians then got into a debate, it would be somewhat unhealthy. I would much rather that the two independent forecasters ironed things out, worked out exactly why they were coming to different views and converged as much as possible.
I tend to believe that, if an independent body is established in Scotland, some of the powers should be transferred to it rather than remain with the OBR.
In essence, you are not saying that we should not establish such a body and just let the OBR do the forecasts.
Establishing the body in Scotland would be valuable. It is exactly the right thing to do, starting with more limited responsibilities and then being ready to expand as required following the referendum next year. We should be getting it in place with the macroeconomic modelling work to which Professor Leith has referred. We should ensure that that work is firmly understood. We should get a transparent, well-respected model in place, look at the data requirements and make sure that any gaps in data are filled or that there are plans to fill them as best they can be filled.
A lot of preparatory work could be got going in addition to work on the Scotland Act 2012 and other areas. As part of that work, it would be of value to establish the capacity to analyse particular issues. It is the right thing to do, and the right time to do it is in the first part of next year, but I would not like to have an all-singing, all-dancing body too early. There should be a step-by-step introduction.
That is helpful.
One of the issues that we have been considering is how any body should be appointed. You helpfully say that it should be accountable to the Parliament. We explored the issue with Robert Chote, so it would be useful to know what the other witnesses think.
You also suggest that it could be
“a consolidated body that includes”
the Scottish Parliament information centre. SPICe is an integral part of the Parliament and has other roles, such as providing responses to individual requests from MSPs on matters that have nothing to do with fiscal forecasting. Would that not be quite a muddled relationship that would be too close to the Parliament?
I will split the two issues that you raised.
As far as the body’s appointment is concerned, we were worried about how to avoid potential political influence in setting up an arm’s-length and objective body. We thought that the right way to achieve that would be through appointment independently by the Scottish Parliament, to which the body would be directly accountable.
10:15We also thought that that would provide the Scottish Parliament and its committees with access to a relationship with a body that could help them in their work of scrutinising policies and decision making. That would be a valuable way forward. We talked about that for some time with some very informed folk and we felt that that was the best way forward. It would help to avoid the risk of political influence.
The OBR has done remarkably well to maintain its independence, as my colleagues have said. For me, the appointment of Robert Chote was critical, because he came with a reputation for being independent and no one could suggest that he had any parti pris, given his track record. That was a very good appointment. However, if appointments to the body were made by the Scottish Parliament it is probable that there would be at its head a similar person of repute, independence and strength—and the same would apply to the other members.
As far as the positioning of SPICe is concerned, we are aware that an awful lot of things are happening and there are an awful lot of suggestions of bodies to be established. We did a separate submission on the national performance framework, which we discussed with the Finance Committee in another context a while back. There are aspects in there of a group that is needed to do this or that, so we were worried about a proliferation of independent groups.
We thought that, as a starting point, the IFI could be hosted in a body that included SPICe. It would not be part of SPICe, but could sit alongside it for pay, rations and organisational purposes. That would get it going. As and when it needed to grow and be a more substantial body, it could then be freestanding and separate.
There is an awful lot going on that needs a lot of work and a lot of people, and we are trying to think of how to minimise the institutional costs of establishing such things.
I am sure that the Cabinet Secretary for Finance, Employment and Sustainable Growth will appreciate that.
We hope so.
Do the other two witnesses have a perspective on how such a body should be appointed? Do you think that the question whether it should be appointed by the Government or the Parliament is particularly important?
There are international examples. The Swedish Fiscal Policy Council recently asked to be brought under parliamentary control in the hope that that would strengthen its independence, but there are other fiscal councils that are under parliamentary control and have been subject to political interference.
What is more important is that the head of the organisation and, if it is a council, the council’s members are strong, independent figures; that the budget is guaranteed for a reasonable timeframe; that the length of appointment is for a reasonable timeframe; and that there is an advisory board attached to the body that can act as a buffer between the body and policy makers who do not like the implications of particular forecasts. Such devices to strengthen an organisation’s independence are critically important.
Does it matter whether the Parliament or the Government appoints the body?
There are examples of councils that are organised in one way and of councils that are organised in another, and it is not obvious that one is better than the other.
I absolutely agree that the body should be accountable to the Parliament, but I do not think that that is a panacea. In the Scottish Parliament system, the committees were meant to be independent and in some respects there has been disappointment when, on occasion, committees have acted politically. The Parliament is not immune from political interference, but clearly the body would have a much better chance of being impartial if it was accountable to the Parliament rather than the Government of the day.
Two other factors are important. One is resources: unless the body is adequately resourced—in relation to not just staff numbers but the quality of staff—it will struggle and not make an impact. Another important factor is the mechanism for appointing its head. Consider the recent appointment of the governor of Bank of England. The Bank of England is an independent body, but the appointment of the governor was seen as a very important step that some would say was unduly politically influenced. It is extremely important that you consider how the appointment of the head can be independent of political interference.
I add transparency to that list. It is utterly important that the work of the body is as transparent as the body and its head wish it to be.
Much of what has been said chimes with evidence that we have taken about forecasting, assessing and projecting. Another aspect, which has been hinted at but which we have not quite got to, is that some fiscal councils recommend courses of action. A criticism of that approach is that it brings the body into the political sphere. Do the witnesses have a perspective on that? If I may develop Dr Cuthbert’s analogy, do they foresee rocks ahead, and do they recommend putting the ship into reverse or starting out on a different course entirely?
I would like the fiscal council to recommend courses of action.
Let me give you an example. In November 2011, the OBR said that, following the financial crisis, it thought that the level of output would be permanently reduced, which affected its forecasts for the sustainability of public finances. As a result, George Osborne felt forced to extend his austerity measures for an extra year, a decision which could have negative effects on the economy. In that situation, it would have been ideal if the OBR could have said whether it thought that extending the austerity measures was a good thing to do. Given the economic situation at the time, I imagine that it might well have advised that that was one instance in which it would be better to delay the fiscal correction. It is important that a fiscal council is able to recommend action.
I slightly demur from that. As Jim Cuthbert stressed, it is important that the body should point out the risks and, in particular, pick up risks that it thinks are increasing in scale and are of real substance. The body might then want to suggest that action be taken to minimise or reduce the risks, but it should avoid specifying in detail what that action should be. That would be to go beyond its remit and risk its becoming seen as a policy maker or policy driver, with all the political repercussions in that regard.
The body should identify and crystallise the risks and stress where they are becoming important, and it should suggest that action is required and then step back. That is my preference.
I thoroughly agree with Jeremy Peat. The body’s role should be to identify risks rather than to suggest what action should be taken.
Jamie Hepburn talked about the sensitivities around politicians becoming involved in appointing a fiscal council. Where would the potential dangers be if politicians were involved in appointing people who were allowed to make recommendations? Perhaps Professor Leith will comment.
The politicians would be just one step removed at that point. The way the OBR works is that the chancellor makes the suggestion but the Treasury Select Committee has the right of veto on the appointment of the head—that is a power that the committee does not have in the case of the Bank of England. In that sense, there are some checks and balances.
For the body to work, the head must be seen as independent. It would negate the whole purpose if the people who made the appointments made the process blatantly political.
I agree. We can never totally remove the risk of someone getting through who is not 100 per cent independent, but the more transparency there is, the better, as I said.
Requiring the person to present himself before an appropriate high-level committee of the Parliament and to be seen to be asked questions by all MSPs from their different perspectives—who would then agree, preferably unanimously, that that is the right person to go forward—is as good a basis for getting the right person as I can think of. We can never dispose of the risk entirely but, if we go through due process transparently and in an even-handed manner, we should get the right person.
We have talked a little about why it is a bad idea to have a range of forecasts. The suggestion is that it is better that the Scottish body and the OBR reach agreement on areas of common interest and produce one forecast.
That flies against my gut feeling. For example, there are three of you here—we did not ask you to come here with one voice; we wanted to hear your three voices. When we put money into a pension fund, the fund managers spread it around; some are more optimistic and others are less optimistic. It is helpful to have a range of forecasts. Can you convince me why it is not good to have a range of forecasts in this situation?
There are a range of forecasts out there, so you already have a diversified portfolio of forecasts. By appointing a fiscal council, you are giving primacy to the forecast of the body that you have funded, made the appointments to and attached importance to.
From what Dr Cuthbert quoted, it seems that the OBR does not believe that it has that level of authority and does not want it. It has been given an almost mythical authority.
In what sense?
It does not believe that its forecasts are having the effect that Dr Cuthbert feels they are having.
People in the OBR do feel that their forecasts have an impact. There were the incidents that I mentioned in late 2011, when policy was changed as a result of the OBR’s forecast. That is the fiscal council working.
I believe in a range of forecasts; I do not believe in a spot forecast and I never have done in my work as a professional economist or in commentating. However, you must have one central view and then look at the potential diversion from that.
It is important for anybody in the public or private sector to look at the what-ifs around the potential diversion from the central expectation. You may plan your policy primarily on the basis of the central forecast, but you need to consider the risks around the forecast and ask, “What if this happened? What would be the implications? How would we respond to that? Are we ready to do so?”
The other certainty is increase in diversion over time. The forecast for the 12 or 18 months ahead is likely to be more reliable than the forecast for what will happen in 24 or 36 months. You should have a range of forecasts, be prepared to respond to circumstances if they differ from the general expectation, and take that into account in your planning, but you should always have a central view from which you start.
That is similar to what the Bank of England does with its fan charts, which is quite a helpful way of working. You will find that most private sector organisations also follow something like that process. There is a requirement for an independent body that provides forecasts to Governments to have a central view, but all the risks, uncertainties and scenarios should be carefully set out alongside that.
Dr Cuthbert, you cited quantitative easing as one of the big factors. If I understood you correctly, it seems that the OBR is making quite a big assumption. Would it not be beneficial to have another body that was not making that assumption?
I do not think that you need the other body. I was a bit unsure whether your question was whether we need one forecast between the OBR and the Scottish forecasting body or whether you were talking just about an individual forecasting body.
I believe in having a range of forecasts. By all means, you can have one central forecast—in a sense, you need that for policy purposes—but it is important to understand what has gone into that forecast and what the assumptions are. It should be made much clearer what the assumptions are in the OBR’s central forecast, particularly in relation to quantitative easing, the output gap and when we will get back on to the trend rate of growth of the economy.
As long as the assumptions are clearly displayed, you will know more about your central forecast and the mythical status that tends to be given to central forecasts will be eroded—as, in my view, it should be. People will have a better understanding of what is likely to happen and how the forecast may be wrong.
I very much believe in having a range of projections and making clear what different variables and assumptions are involved and what effect they could have on what will happen.
10:30
Thank you—that is helpful.
Professor Leith, in your submission you question the cost of such a body and you mention that it would take a bit of time to build it up. We have not talked about how big it would need to be. The OBR has 17 staff; if we split that according to the size of our population, we would end up with 8.5 per cent of that—in other words, 1.5 people, which does not sound very many.
Professor Peat’s idea is that we could combine the body with something else. That appeals to me, because it would be simpler. We should be able to do things more simply in a smaller country. Even if the body has limited powers, how big will it need to be? What will the cost be?
On average, fiscal councils internationally have between 15 and 40 full-time members of staff. The size depends on what remit the body is given and how detailed the forecasts have to be. If it has to cost individual policy proposals or the policies of the Opposition, extra people will be required. The Dutch Central Planning Bureau has more than 200 members of staff, as does the American body, but, typically, such bodies have been 15 and 40 members of staff.
Is that what you would recommend for Scotland?
It depends on what the degree of devolution is, what powers the Scottish Government has and what needs to be scrutinised. The size will reflect that.
In our submission, we suggested that a working group could work with this committee or another committee to think about the initial body that might be established. I agree that we would not want to start with a body of the scale that Professor Leith has just identified; five to 10 people should be perfectly adequate to get the body up and running for its first month or year of activity. However, when it comes to the type of institution that would be required in the run-up to and following independence, a body of substantially larger scale would need to be planned for. The answer to the question is therefore: it depends.
I was interested in your suggestion about setting up what your submission calls “an independent expert group”. Is it necessary to set up yet another group?
It would be a short-term group. Once the committee has taken evidence from us and from other people, it will have a host of ideas; the committee’s new adviser and Scottish civil servants will undoubtedly have ideas, too.
I love transparency, and I think that such a group could provide the committee with advice from a range of different perspectives and experience, which it could use to determine the way forward. That would add an extra dimension to the process of deciding what the initial independent fiscal institution should be and where it should be based. It would give a degree of validity to the process.
It will be necessary to be careful about what the body is asked to do. A valuable function that such a body could perform would be to independently cost policy proposals that are put forward by Government, but—done well—that is an extremely resource-intensive process. Therefore, at least initially, until the body had found its feet and was properly resourced, it would be necessary to ask it not to indulge in the work of costing policy proposals because, if done properly, that would suck in a huge amount of time and resource.
I would like to make an additional point. When the OBR was created, it was lucky enough to be able to take the Treasury model off the shelf and use it. A Scottish OBR would not have that luxury; it would be involved in model development as well, which is extremely resource intensive.
My final question should perhaps have been my first.
Professor Leith, if I understand your submission correctly, you are saying that the starting point is to try to keep control of debt and not to borrow more and more. To turn that on its head, I presume that, if we had a system like the one in Norway, where money is put aside and the Government is sensible about that, we would not even need such an organisation. I do not know whether Norway has one.
It has in place rules and a sovereign wealth fund, which I guess is like a reverse fiscal council.
Are we saying that we need such a body because we do not have self-control?
Many reasons are provided in the economic literature about why the political process may give rise to a tendency to have deficits. If we look at the Organisation for Economic Co-operation and Development economies over several decades, we can see that there has been a general tendency for debt levels to rise, even before the current financial crisis. That is why Governments are adopting fiscal councils—to correct that bias.
I want to ask about the timescales. The cabinet secretary has stated that it is his intention to establish the new body prior to the implementation of the newly devolved taxes in April 2015. That seems to me to be very soon. All the work that you are speculating might be done with advisory bodies answerable to Parliament would need to start now, I guess.
Yes, it depends on the scale of the body that you want at that initial point. If it is limited solely to the devolved tax powers under the Scotland Act 2012, it could be a relatively small body, improving on the method used by the OBR. However, if you are anticipating there being more devolved powers or independence in the future, with greater borrowing powers and a more elaborate fiscal council to deal with the broader forecasting issues that emerge because of that, the sooner you build capacity, the better.
Everybody agrees that the remit of the OBR is quite restricting—I think that the OBR would say that, too—and you have also stated that it would be important in the short term to come to some agreement on Scotland’s figures and forecasts with the OBR. Is it important that the wider remit, or the remit that we would want in an independent Scotland, is the remit that is worked on now, given that the IFI would acknowledge that there is a reputational issue at stake, to do with what Scotland will look like and how it will handle those issues? If we are looking to improve things here and not to pick something off the shelf that is found wanting, how do we square that circle and allow it to happen? We can get excited about the prospect of the new body, which I would be keen to see, but what would happen in the short term?
I think that you are asking whether the Scottish OBR having a different remit from the rest-of-the-UK OBR could cause some kind of tension. As long as the Scottish OBR is giving advice on the policies that the Scottish policy makers have control over, it can run alternative scenarios based on the instruments that the policy makers have at their disposal, and that work would not be restricted by the fact that the OBR is not doing the same for the rest of the UK. I do not see an obvious conflict there.
I take a slightly different slant on the question. There is a fairly narrow time window between now and the time when that body would have to be up and running and fully functioning, which is why it would be desirable to have a more detailed look at exactly what the body would look like, how it would operate and how it would be appointed, both in the context of the Scotland Act 2012 and looking forward beyond that in the event of a yes vote.
That would mean that two critical steps would be needed during the first part of next year. One would be a clear timetable on the macroeconomic modelling work that Professor Leith referred to. I do not know how far that has gone or when it will be ready, but it will be needed, so that needs to be assessed. The other thing is appointing a head of the body or a head designate. It would be valuable to have someone in place at an early stage so that they can oversee the work and be in regular contact with the OBR, because I would like to think that before the Scotland Act 2012 is implemented in 2015 there will be a period in which the Scottish body can shadow the work of the OBR from a Scottish perspective while looking at different modelling work and different assumptions and getting as close together as they can on the models, the data and how the model will run so that they can come out with results. That way, when the Scottish body takes over, it will be as seamless as possible and it will minimise the risk of there being different views from different bodies.
I would get going very early on macroeconomic modelling, get a person in place to run it, perhaps with people on contract in the short term, and get working on the elements that relate to the Scotland Act 2012 requirements as soon as possible in close consultation with the OBR, but I would also think through what would be required post a yes vote in the referendum next year so that, if there is a yes vote, work can accelerate towards being ready by the date of independence to have a much more fully fledged and sophisticated body that is ready to run, with agreement in Parliament on its functions and understanding of the resource requirements and everything else that goes with it. There are a number of steps on the way, but there is no scope for dilly-dallying along that route.
I think that I agree with Jeremy Peat. An incremental approach would be desirable.
I presume that the immediate requirement is a fairly short-term one to understand what factors bear on variations in tax receipts in the fairly short term. The body should focus on that initially. In the longer term, one would think about the demographic and other economic factors that would bear on long-term trends in tax receipts and the evolution of the economy. There is no reason why one should not start off by concentrating on a fairly short-term, narrow remit, which would be valuable, and then build out from that. The way to build confidence is to master one particular area first and then build out rather than promise everything and not have the resources to deliver.
To go back to the ship and the rocks analogy, would the Westminster Government or the chancellor, George Osborne, have known the OBR position? Would that have been discussed and then not accepted by him? Did the OBR then go public and say, “We’ve given this advice, and it has not been taken”? I cannot remember the details of that. That defines the relationship, in a sense. As has been said, the Government can, of course, decide not to accept the advice.
The Treasury is integrated with the OBR in the budget-setting process. I think that the Treasury indicates to the OBR the kind of measures that it is thinking about, and the OBR will bounce back to it that it needs information on X,Y and Z in order to be able to scrutinise them. Issues bounce back and forth between the Treasury and the OBR so that the policies that the Government is thinking of implementing are fully costed by the OBR, that feeds into the forecasts, and both sides know what the forecasts are. As part of that process, the chancellor may choose to respond, as he did in November 2011, or not to respond, as he did in December 2012, but he will know the OBR’s position when he gives his statement.
I think that you have all said that you think that there is sufficient distance between the work of the OBR and that of the Treasury, but it is not always seen like that, and there has been criticism. Some economic journalists, for example, see them as being too close. Do you have an opinion on that and on what clarity is needed?
I would be very reluctant to come to the view that they are too close. I think that Robert Chote and the people who work with him will be determined to maintain their independence. Of course, they had to take the Treasury model and they had some civil servants on secondment initially, but there is no way that those people and those around them would limit their independence. That is critical to their role, and the people who have been chosen are the right people. They will have words in the ears of Treasury officials from time to time, and if they see the boat approaching the rocks, they may whisper in a crew member’s ear first of all before shouting to the captain, if I can put it that way, but I see them as scrupulous in their desire to remain independent, and we must make sure that, if there is a similar body here, that is maintained as an absolute requirement.
10:45
Yes. One can make an interesting distinction between what one might call political and institutional independence. That topic came up at the meeting of the interparliamentary finance network—IPFIN is a body of international financial scrutineers—that I addressed in the Parliament. We had an interesting discussion.
A body such as the OBR is undoubtedly politically independent—I am quite sure that it has no political bias. However, any body in that situation finds it very difficult to be what one might call institutionally independent. If a body is in the position in which, if it makes an adverse comment, it could move the markets and perhaps cause a sterling crisis or whatever, then it is very difficult for that body to speak out with completely unvarnished frankness. That is an inescapable part of life. Whatever body you set up will inevitably owe a large degree of loyalty to—and even without recognising it will have invested a large degree of faith and loyalty in—the very political system that set it up and in which it is embedded. Therefore, it will find it very difficult to make calls that may impact adversely on the whole structure of that system. That is an inherent difficulty that one has to live with.
A difficulty that the OBR faces is that, around the time of the budget, it is necessarily liaising with the Treasury behind closed doors because it cannot reveal the contents of the budget in advance. That interaction is still valuable because it allows the OBR to influence policy timeously and ensure that policy avoids the rocks.
At our previous meeting, Robert Chote spoke of the importance of the OBR getting information from Her Majesty's Revenue and Customs. He made it pretty clear that any body established in Scotland would also need access to that information. The Royal Society of Edinburgh’s paper says such a body
“should have the right to develop its own framework of analysis, data sources and methodology”—
and, vitally—
“should have full access to data and any related analysis or evaluations made elsewhere.”
I presume that all the witnesses accept that the body would need access to HMRC information, but please correct me if that is a wrong assumption. How would it secure that information? There is a memorandum of understanding between the Scottish Government and HMRC on the Scottish income tax rate. Would that suffice as a mechanism, with the Scottish Government acting as a bridge to the information, or would that impinge on the operational independence of any body? Should it have its own direct relationship with HMRC? If so, how could we make that happen, given that we do not have provisions to do that with HMRC?
I am not an expert on that topic, but I envisage that it would be much more straightforward for the new body to have direct access to HMRC in its own right because it could then ask the questions, make sure that it gets the data in the way that it finds most valuable and return with more questions. Working through an intermediary would be inefficient, with the potential to get things wrong in minor ways. Data are complex, and dealing directly with the people who own the data is always by far the most effective way of working.
If the body is established appropriately, I do not see any difficulty—I cannot see anybody objecting to it having a statutory right to access and use that information on some prescribed and fully set out basis.
I agree.
I, too, agree, but what happens if HMRC has to spend money on redesigning some of its data sources to meet the demands of the new body? HMRC data is not 100 per cent data—a lot of the important information is collected from various samples, which may need to be redesigned to give the required accuracy for Scotland. In that case, there will need to be a mechanism for influencing HMRC. It is not necessarily the case that this body, by simply talking to HMRC, will be able to achieve that. We also need to think about what mechanism will put pressure on HMRC so that it comes up with data of adequate quality.
That is an interesting point. This Parliament is unlikely to be able to force any such mechanism, because we do not have that legislative competence. I am not looking for there to be a problem, obviously, but do you think that that could be an issue?
As long as we are in a devolutionary context, that is really a question about the functioning of devolution. It is the sort of thing that the Secretary of State for Scotland and the Cabinet should be getting interested in—whether the thing is working properly and the different parts of the system are providing the quality of information that is needed.
That is helpful. Thank you.
I have a couple of brief questions. Professor Peat, in your paper you talked about there being an interaction between this independent fiscal body and Audit Scotland. Can you expand on that?
That would come essentially in the context of the possible remit to look at what we call
“strategically important forward-looking issues and commitments.”
We are aware that Audit Scotland does a huge amount of valuable work, mainly retrospective but with forward-looking implications. We would not want this body to set out down that path—which, incidentally, I would see as a role that came later in its remit rather than earlier—without ensuring that it is in close contact with Audit Scotland. That would get the best out of the work of both and avoid wasting resources.
There is broad agreement that this body has to be independent. We have talked about how it would be appointed, and whether it might be by Parliament rather than by Government. We have talked about who would head it up and the importance of their independence. Can panel members suggest anything else that we could implement to ensure that the body is genuinely independent and, just as important, is seen to be independent?
The only other comment I would make is to echo what Professor Leith referred to earlier, which is some form of advisory board to act as a ringmaster and ensure that the body’s independence is maintained and not challenged, and that there are people with sufficient seniority and clout to be able to stand up to any attempt to influence it. That would be one added means of minimising the risk of intervention.
Does Professor Leith or Dr Cuthbert have anything to add to that?
I would just add that it would be very important that the head of the body—or the body—has a clearly established statutory right to speak out. They cannot be muzzled.
I agree with those points. A strong advisory board is very effective as a buffer between policy makers and institutions. There are also the details of protecting the budgets and terms of appointment and so on, which are all necessary to ensure independence. It is those little details that start to matter.
Thank you. That appears to have exhausted the questions from the committee. The session has run on longer than anticipated, so I do not intend to ask any more questions. I just wonder whether each of our witnesses has any final comments.
I think that you have exhausted most of the comments that I had to make. Thank you for the very interesting questions.
No mad rush for any final comments, then. Thank you. That brings the session to an end. We agreed earlier to take items 3 and 4 in private.
10:53 Meeting continued in private until 11:32.