Under item 3 we will hear from members of the Scottish Fiscal Commission on a separate topic, which is our consideration of the Scottish Fiscal Commission Bill at stage 1. Members have received a written submission from Lady Rice; I invite her to make her second opening statement of the day.
Thank you, convener. You are obviously gluttons for punishment.
We are very grateful for the chance to come to the committee and discuss what is, clearly, an extremely important piece of legislation for us. We have read the written responses to the consultation received by the committee, and we are interested in how broad the range of views and perspectives is. There is clearly not just one right way to set up such a commission, but it is important to say that we must choose the right way to shape the commission for Scotland.
We have given a lot of thought to the issues, particularly forecasting—an issue that has been raised a number of times by the committee and others—and the fact that we need not only to act independently, as we believe that we are doing, but to be seen to be independent. Some of the propositions in our response aim to address that particular issue.
I have one point on forecasting. As a banker, I used to despair at the fact that everyone calls everything a bank, whereas some things are banks and some are not; the word “bank” is used widely. Forecasting is a little bit like that. There are official forecasts, which are done by the Scottish Government, but there is also forecasting as a tool. Richard Baker was beginning to get at that point in relation to our table on forestalling. Along with analysis and research, forecasting is a tool that can help to inform debate and judgment. It is almost as though there is forecasting with a capital F and forecasting with a little f, and the distinction must be clear.
Some of the topics raised in the call for evidence relate to matters that will arise from the fiscal framework. We have tried to address them, although only briefly, because we do not know what the final outcome in the development of the fiscal framework will be—no one does. When we know whether there is a charter or fiscal rules, and what the borrowing arrangements might be, we can expand our perspective on those points. For now, we have given the committee just some general responses.
Overall, our views are set out in the submission and we are happy to take your questions.
I will have to enforce a self-defying ordinance, because there are so many things that I could ask—I see Gavin Brown shaking his head. I shall try to limit myself.
May we exercise a self-denying ordinance in our responses?
Certainly not. I have a good mind to keep you here until tea time for that cheeky comment. [Laughter.]
I will ask about something that comes from the end of your submission. I thought that I would start with this, rather than delve into the other areas that will no doubt be the subject of a considerable number of questions. You say:
“The term of office should be stated in the Bill and the SFC believes, on the whole, that a single non-renewable term as currently pertains is effective, and that individual terms should be staggered.”
I was quite surprised by that because many of our witnesses agreed with staggering terms but suggested that members should serve two terms for reasons of continuity. Has it been that onerous so far?
We discussed the question, and we could be persuaded in different directions. In setting up the Scottish Fiscal Commission, we subscribed to the strong view that we should take on the role without any feeling in our own minds that we might like another term, which could colour our responses or the work that we do, even subconsciously. There is real value in having just one term.
The term can be shorter or longer. Staggered terms are important. Some of the respondents who have said that there should be two terms perhaps believe that there may be a dearth of other qualified and able people. I personally take the view that, given Scotland’s 13, 14 or 15 universities and the great deal of talent in the UK, we should be able to find people of talent here or more widely.
We discussed the possibility of having two terms as long as there was a process for renewal, or possibly a break of two years and then a return that was not necessarily guaranteed. We would not stake our feet to the ground on the issue of one term, but we see the advantage of it, and we have no debate with it for ourselves.
So you would consider the President Putin approach of having a term of office—
For ever?
A term of office followed by a gap, and then a return to the SFC by popular demand.
I will again move on to discuss something different: additional functions. You say on page 3 of your submission:
“The SFC believes it should have responsibility for assessing the Scottish Government’s forecasts on the sustainability of Scotland’s public finances, such as adherence to fiscal rules as an example, and it would welcome the Bill being amended now to anticipate this additional responsibility when it arises.”
I am sure that that paragraph will be welcomed by a number of our witnesses who have given evidence so far, and by some committee members, too. Can you talk us through a wee bit more of your thinking on the issue?
I have been asked to give evidence to the committee several times, including prior to the Fiscal Commission being created. One of the main objectives of creating a fiscal commission is to ensure fiscal sustainability. The raison d’être of a fiscal commission is exactly that.
I could go on at length about this—I have in the past.
If we take the view about the commission’s remit as it is expressed in the bill as introduced—that it should examine and comment on information and revenues that the Scottish Government may use to make up its budget—we have to worry about fiscal sustainability as a minimum. At the moment, I am not quite sure what that means. It may mean sustainability against the rules that are currently in place, under which borrowing is minimal. If that is the case, it is not a big operation, but it might be different in some other context. The work is one part of the minimum; it is not explicitly in the bill, and it might help if it were.
A second issue relates to sensitivity analysis. It would be rather difficult to say that we think that a given forecast is reasonable without having poked it a little to see whether it stands up against shocks of different sorts or slightly different outcomes in the rest of the economy. Sensitivity ought to be covered, and it would help if it were nominated, so that everybody knows that it will be one of the things that the commission considers.
What you have just touched on—the issue of reasonableness—has come up before: we had some discussions about that in taking evidence. What is your view of what reasonableness is? How do you define whether a forecast is reasonable? That is something that we have tried to get to grips with, and a lot of our witnesses had some issues with the question at previous evidence sessions, as you will be aware.
I wonder whether you will get any reasonable responses.
Let us see.
12:00
I do not like trying to define reasonableness because it is one of those things like the prime rate. What is the prime rate? It is what we say it is. It is like something from “Alice in Wonderland”. Defining it precisely is rather difficult.
The next thing that people tend to do is have a checklist to see whether a forecast satisfies this or that, and the obvious question is whether it is accurate. That has a place in this discussion. I caution you not to be too prescriptive. In attempting to make something objective instead of subjective, you will find it difficult to introduce any judgment and, of course, the forecasts have to judge what is likely to happen in future. I would therefore not be too prescriptive.
I note by way of illustration, with the indulgence of Lady Rice, that the American SAT scores for getting into university make an attempt to show what a reasonable performance is, and the work is done with checklists. What we observe now is that most universities do not use the SAT system because it is too inflexible.
Having said that much, I will give you a different answer about what we actually do. In the case of LBTT, I was trying to assess the forecasts not by producing official forecasts but by doing our own calculations to see whether we would do something similar.
To start off with, we look at specific scenarios around the tax, which is how we go about establishing reasonableness. We check whether the methodology is okay, whether things have been left out, whether specific economic factors that should have been taken into account have not been, and whether the assumptions about the forecasts are sensible. People often pick on last year’s example of the non-domestic rates being overoptimistic. We did not change the forecast; we changed what we thought was an unreasonable assumption up front. We do those sorts of things at the first stage.
We can also check against other available forecasts. It is difficult in this context but not impossible. Of course, the OBR forecasts are available all the time so we can check against them, although I am beginning to get the feeling that there is not a huge amount of competition because the OBR does work on a UK basis and, if Scotland is a bit different, something else will happen.
The third thing is to do the simple calculation—I emphasise that it is a simple calculation and not a forecast—to back up the position. I ask myself: if I did it myself, would I get a similar number? It would not necessarily have to be the same number but a similar number. This is the point: if the number is different, we can dive in to find out why it is different and learn something from what is wrong. That helps us to decide whether what the Government has put out is reasonable. If not, we do not then say that the Government should change the number but that we think that a different process or extra information would be a good idea.
That is how we decide whether a forecast is reasonable. I am sorry for the rather waffly answer.
I found it very helpful. If you think that the Scottish Government’s figures are not really as accurate as they should be, when do they become unreasonable? Is it at a 1 per cent, 5 per cent or 10 per cent differential?
They should be better than they were last year.
So you would not pin down a figure.
No, because circumstances might change. If we get into a recession, the 1 per cent would go out of the window.
I am not suggesting that the differential should be 1 per cent. I am asking what the Scottish Fiscal Commission thinks would be reasonable.
That is where judgment comes in.
It is important to emphasise Professor Hughes Hallett’s point. It is not simply a matter of metrics: it is a question of judgment and numbers, and judgment is really important. If we went through the rational or intellectual exercise of saying that, for one particular case or one particular tax, we would expect to look at A, B, C, D, E and F—whatever those things are—and we found that the Scottish Government team had looked at only one or two of those factors, we might find it difficult to say that the Scottish Government’s forecast is rational. That is not just because we are judging the numbers that it comes out with but because of the approach taken. That is all about judgment and it is the kind of thing that comes out in our challenge meetings.
On page 2 of your paper, you talk about transparency and you say:
“In order to ensure transparency at every stage, the SFC would publish Minutes of its meetings with the Scottish Government”.
Transparency takes a number of forms. For example, we talked about forestalling at the earlier session but we did not have any detailed example from you that shows the work that you did to come up with the numbers.
Andrew Hughes Hallett touched on non-domestic rates, but the figures for the original forecast were not provided, so it is not really possible for us to identify how much it was changed, and it is also not clear why the recommendation led to a change in the official forecast when other Scottish Fiscal Commission recommendations did not. For example, the SFC recommended the need to account for behavioural responses in producing forecasts for residential LBTT, but the report does not mention whether that was raised with the Scottish Government as part of the process of inquiry and challenge, and what response, if any, was provided. There is no mention of that in the Scottish Government’s own methodology paper.
Surely transparency goes a lot deeper than just producing minutes of meetings and decisions made.
The production of the minutes, now that we have a bit of support, is one aspect of transparency. We publish our work and comments on our activities on our website, all of which is expanding over time.
We are also learning what it is that you as a committee consider transparency to be, so it is a learning process for us as well. If you want a lot of the detail behind a table, we can provide that. We are new to this work and to your expectations, so we are happy to take guidance and advice about what you want.
Overall, transparency matters greatly. There is no point in having any of the debates or discussions without being quite open about what we are basing things on. We might be wrong about things too, and there may be others who come to us and point out that we are actually off base in looking at certain things. That can happen only if we are transparent, so we are looking for ways to increase that transparency and to increase our overall communications. We are happy to take any pointers you can give us.
Looking to the medium to longer term, we have already begun a number of what could be called research projects that will examine more deeply various aspects of the modelling of some taxes. We asked for provisions in the bill to allow us to publish technical working papers, which would allow us, on an ad hoc basis, to publish the details of the research. That will also contribute to the transparency of what we do.
That is helpful. In your paper, you say on the right of access to information that memorandums of understanding
“are the normal way of working across public sector areas; if the SFC’s right to establish relevant MOUs would be better safeguarded with more explicit wording on the face of the Bill, then the SFC would support doing so.”
I take it that you would be quite keen for that to happen. Your wording is a wee bit tentative.
MOUs appear to be the way to establish good working relationships, but the ultimate good working relationship depends on the people and the good will involved. MOUs alone—perhaps this is reflected in the tentativeness of the wording—do not ensure what we want to achieve, but they certainly help. They set out the boundaries and the fencing around what we might do.
Just so that you know, I should tell you that we developed an MOU with Revenue Scotland specifically for the data that we needed to analyse in relation to the outturns paper that we did. Revenue Scotland was extremely helpful, and we are now in conversation with that organisation to develop what I would call—although this is probably not the language that you would use—a perpetual MOU, so that we do not have to do one each time that we go to it with a request for information. It is helpful to have such memoranda in place.
I take it that you hope to develop MOUs with HMRC and the OBR. Your paper states:
“Over time and in order to do its job properly, the SFC’s goal would be to have a deeper relationship with the OBR and HMRC.”
Yes, and the import of that statement is that, once the new powers on taxes and related matters are finally agreed and come to Scotland, it will be more and more important over time to have those connections formalised.
The OBR’s doors have been open from the very beginning. Its staff have been co-operative and supportive, and they have invited us into their challenge meetings when they have been looking at Scottish taxes over last year and this summer. We have joined those meetings and have had access to their working and thinking along the way, and we have provided some challenges back as well.
In a way, some of the work is already happening. An MOU does not make it happen—or, at least, does not make it happen perfectly—but we think that it is a useful tool, which is why we made that statement.
The last point that I want to touch on before I open the questioning out to colleagues is, obviously, forecasting. You say in your submission that you are looking for better interaction, “direct engagement” and “contact with the forecasters” and suggest that
“The SFC could have direct engagement with the forecasters in the early part of the year when the forecasters are developing or refining their models”
and that its
“role would be to challenge and influence the efficacy of Scottish Government processes and approaches”.
Moreover, you say:
“In a rational world, there should be one official forecast of tax revenues and there should be an independent assessment of that forecast. ... These activities should lead to a check on the official forecast. They should not become the basis in themselves for informing the Draft Budget.”
Finally—and your position on this has been quite consistent—you say:
“The SFC, as the ‘assessor’, believes it should develop the latter type of forecasts alongside research, modeling and analysis.”
However—and, as you will expect, I am going to quote some comments here—you will know that most of the witnesses that we have heard from are not of the same view. Instead, they take the view that the SFC should produce its own forecasts. For example, the Royal Society of Edinburgh is
“firmly of the view that the SFC should be able to originate its own independent forecasts of the future fiscal revenues ... Indeed, to fulfil its functions the SFC will need to be able to produce independent forecasts.”
Professor MacGregor and Professor Swales have pointed out that “international practice” suggests that where
“fiscal bodies do not provide their own forecasts they typically have access to sufficiently detailed independent forecasts”,
and the Canadian parliamentary budget officer has stated, if the SFC produces its own forecasts, it will allow Parliament
“to challenge the assumptions underlying the forecasts of the Scottish Government”.
There are loads of similar comments, which I will not read out, but I just want to finish with Dr Armstrong and Dr Lisenkova, who argue that the SFC should prepare its own independent forecasts on the basis that
“Producing a forecast greatly adds to discipline.”
Of course, not all commentators hold the same view—I note the views of the Institute of Chartered Accountants of Scotland and others—but the general thrust among your academic colleagues is that the SFC should produce its own forecasts. Why are you continuing to take the view that you should just comment on the Scottish Government’s forecasts?
I think that there are a number of topics that pertain in response to that. I will start and then turn to both my colleagues.
We have talked about this issue a great deal, and it brings us back to the comment that I made in my opening statement about the official forecast and other forecasts. I wonder whether some of the comments that you have quoted relate to the latter kind of forecast. One of the tools in our toolbox should be the ability to produce forecasts, possibly for some taxes at certain points in time and for some other elements. That would be our choice, but we would need to see. You have already referred to the work on forestalling as being a sort of mini-forecast. We think that that should be in our toolkit, along with the ability to carry out analysis and, as Campbell Leith has suggested, do some technical papers and research.
However, that is different from producing the official forecast. We think that there needs to be one producer and one assessor of the official forecast. If you were to turn to the Fiscal Commission and say, “Actually, we’ve changed our minds. You do the official forecast,” we would need to be assessed by someone. If we were to do the entire forecast ourselves, we would just be duplicating what the Government was doing, which does not seem to make a whole lot of sense in value-for-money terms. If we did our own forecast and then needed to defend it, we, too, might be biased. We have had a conversation about a lot of these elements, but we believe that there should be the official forecast, which, right now, is done by the Scottish Government, and that that forecast needs to be assessed, which is our role. We should not be doing an official forecast as well.
I turn to either or both of my colleagues for their comments.
My view on this goes back to some of the issues that we have already discussed with regard to types of economic forecasting and analysis. At the moment, the Scottish Government has to produce the forecast, and it uses statistical models based on historical data to do short-term projections of that data. If we were to produce an alternative or indeed the official forecast on a day-to-day basis, we would probably have to replicate that style of modelling, and I do not think that there is a great deal of added value in having the Fiscal Commission replicate that kind of modelling.
I think that there would be added value if we were to do deeper, more fundamental projects that delve more deeply into behavioural aspects and results of policy change, and bits of analysis of the economy that are not currently featured in the forecasting methods. The need to do those things would depend on the economic circumstances, because sometimes some issues will be more relevant than others. That is the point at which we would need to do such a piece of analysis and bring it to the fore in order, in effect, to evaluate the quality of the short-term forecast.
Therefore, I would give us the role of doing that portfolio of research. We are not just sitting there twiddling our thumbs when we are not doing an official forecast; we are doing lots of bits of research that, fundamentally, go to the heart of what drives the official forecast.
12:15
Andrew, do you want to come in?
Yes—at length.
What Campbell Leith said is right. That is one aspect of the issue; I will not comment on it further. What he said is absolutely right.
I suggest that my reading of what is going on is slightly different from yours, convener. There are some people who say that we should be doing forecasts, and there are plenty of people who say that we should not, one of whom is the International Monetary Fund. I think that it is well worth listening to the IMF, which has oversight of such commissions everywhere else. There are not many such commissions at the—what should I say?—regional level as opposed to the sovereign level.
If we look at the numbers, we find that two—I think—out of 23 such bodies in the Organisation for Economic Co-operation and Development are doing forecasts as the official forecasts and nothing else. Then there is the question of second opinions. We are doing a kind of second-opinion exercise—the Government will produce its forecasts and we will provide a second opinion. I go back to a very old experience of mine in the Netherlands when the central planning bureau was doing the forecasts and it was the only body that was doing them. The constant complaint was, “We don’t know whether to believe the forecast; we’d like to see some second or third opinions.” That aspect is important. Doing it this way round gives us the freedom to compare the Government’s forecast with other forecasts, of which there are not many. There is the OBR’s forecast, and one would assume that, when Smith comes in, there will probably be a new industry in doing that sort of thing.
If I am to depart from what my colleagues have said—I do not know that I am departing from what they have said; it is just that they have not said this—I think that being asked to do the official forecasts would seriously compromise our independence. To do the forecasts, it is necessary to get data. We do not have access to all the data in the world, so we would have to talk to other people and share some knowledge of what we were trying to do. If we were to put forward one forecast as the official forecast and the one that was favoured over other possibilities—as has been said, there are many alternatives at any one point—that would leave the perception in the world that the policies that underlay that forecast were the ones that were favoured. Therefore, in some sense, we would be prejudicing the discussion.
An important point that I did not realise but which I discovered when I talked to OBR people a month ago is that, when the OBR does the official forecasts, the Treasury often comes along with a proposition—it says that it is thinking of making a particular policy change, it presents its projections and it asks the OBR what it thinks of them. It is natural enough for the Treasury to want to do that, because it would not want to be slapped down by the OBR saying, “Total rubbish,” at a later stage, but that shows that the Chinese walls have been broken down and the influence of the Treasury is seeping in.
Therefore, having outsiders having no influence over independent forecasts made by us does not guarantee that outsiders will not try to influence the commentaries that we produce. It is one thing to have an independent forecast; it is quite another thing to have an independent commentary attached to it. The OBR has problems with that.
Those are various different possibilities that suggest that our independence would be compromised if we were to produce the forecasts. What we are doing is the other way round, and that is a different matter altogether. We take what we are given and we say, “Is this reasonable?” We give the reasons for our view, albeit that we might not have published enough yet. At that point, we can make a judgment and people can take it or dislike it as they wish.
I have a final comment to add. This is a personal view that is based on informal and other conversations that I have had. I think that, over time, a bit of an industry will develop in Scotland of think tanks, university bodies or whatever doing their own versions of forecasts or delving into some of these areas. A comment that the committee has had from others is that there is nothing else to refer to—what else is out there? I think that that will develop over time. It is not surprising that it has not developed yet, because the devolved taxes are so new. It is almost a market thing and I think that it will happen.
This is an enduring conversation, but I would like to add an extra point. I had not realised that, when the OBR was set up and given an exclusive, official forecasting-only mandate—give or take a little sensitivity analysis—that was done secure in the knowledge that the IFS would do the further analysis. That does not exist in Scotland. If you did not allow us to poke about and think of other factors that should be important, and we had to make the only official forecast, we would be in a worse place.
I am keen to probe further, but I am under a self-denying ordinance and I know that a lot of colleagues want to come in, so I invite Richard Baker to ask the next question.
The Scottish Government has informed the committee that no restriction will be placed on the commission to engage in forecasting if it wishes to do so, and there is nothing in legislation that would stop the commission from engaging in forecasting, but is it the case that, as it stands, you have no intention of engaging in any forecasting and that any forecasting you do is for yourselves as part of the practice of assessing the reasonableness of the Scottish Government’s forecasts?
If you do not mind, I would like to go back to what could be described as forecasting with a lower-case f. We have started doing that kind of thing, so we are engaging in forecasting as an exercise, and that is absolutely something that we will do, but what we ought not to do is develop an alternate official forecast. That does not make sense, and would compromise our ability to assess the official forecast.
We are doing lots of pieces of economic analysis and we have lots of projects starting up, but we do not plan to replicate item by item the entire set of Scottish Government forecasts.
Of course, there will be a learning process for the Parliament in scrutinising the work that you are carrying out to assess reasonableness. In making international comparisons, we have looked at the Swedish example, which I know you have studied. In Sweden, the Swedish Fiscal Policy Council does not produce its own forecasts, but the National Institute of Economic Research does, and there are many bodies producing their own forecast figures. That means that there can be a lively public debate about how accurate the Government’s forecasts are and how close they are to the National Institute of Economic Research’s figures. Is it beneficial to have such comparisons out there? Given that, as things stand—it may or may not change in future—there is a dearth of separate forecasts from Scottish Government figures, would it be beneficial for the commission to consider taking on a greater forecasting role than is currently envisaged?
That is a fair point. I can understand that you want not just a second opinion but a third opinion and a fourth, so that you can make some comparison. Our comparative advantage would be in making the comparisons between them, and you are right to say that there is a dearth of such things at the moment. There may be forecasts of particular taxes or particular bits of the market from professional agencies, but there is not much else available apart from the OBR. The OBR forecasts everything that we want to talk about but on a different basis, so we already have a second opinion, and I am sure that there will be more in the future.
The deeper economic analysis in the rest of the UK is pushed by the IFS and there is not such a body here. If there were one here, that is likely to be where that would happen. If you look at Belgium and the Netherlands, you will see that, in those countries, the functions have been split between separate agencies, so they have obviously recognised the point. It is just a question of whether resources are available and which is the most efficient way of doing it. It would be nice if other forecasts came from somewhere.
If we had alternative forecasts, that would provide some additional information. It may be slightly overstated how much additional information that would provide, though. If there were other bodies providing additional forecasts using techniques similar to those that were used for the official forecasts, they might make different assumptions at different points and there might be different numbers, but they would be within the bounds to which such forecasts vary. What you really want to see is additional information on some aspect of behaviour or some aspect of development of the economy that is not being captured in the official forecasts or is not being addressed in the methods that underpin the official forecasts. That is the bit that gives you genuinely new information and allows for more fundamental scrutiny of the official forecast.
Bearing in mind the duplication of effort in producing a full, separate forecast, is the primary concern duplication, or is there a resource issue? If you chose to undertake such an exercise, would you have the resources to carry it out, as things stand?
If we chose to or were required to undertake such an exercise, we would resource up in order to do it, but we could not do it with today’s resources.
We already have the ability, under the financial memorandum, to increase our staff to some extent. What we would do with that increase in staff would reflect what Campbell Leith has just said—it would be to look into some elements more deeply to understand what is really happening, so that we could make a better, more refined judgment about the reasonableness of the official forecast. If we had to do forecasting, we would simply say to you that we needed to resource up to do that. We would have no hesitation, and it would be our choice, to the extent that we would need more resource.
The experience in Ireland was that its Fiscal Advisory Council decided to undertake its own forecasting later on, as it developed its own work. If it transpired that there was not the additional forecast that you are expecting, you could take on that role to a greater extent later if you thought it appropriate.
You make an interesting point. My sense of a number of the independent financial institutions is that they have evolved over time. I would be surprised if we did not evolve, but I do not know whether we will evolve in that direction.
Before I ask my final question, I thank the panel members very much for their answers so far.
We have spoken about how important memorandums of understanding are, and you have an initial one with Revenue Scotland. You have had a lot of meetings with different agencies. Are there any problem areas concerning access to information? Are there any departments at Scottish or UK Government level that you think need to be a bit more receptive to your requests for information? Have you found those relationships to be broadly quite constructive?
Where we have requested information, those requests have broadly been met. I do not know whether “concerned” is the right word, but we are certainly interested to see how the relationship with HMRC develops. The Scottish rate of income tax is not yet in place, so all that we have is anecdotal evidence about what that agency may or may not be able to deliver to us. Our understanding is that HMRC is very conscious of the need to deliver data to us in a way that keeps safe what has to be kept safe under privacy regulations and law, just as Revenue Scotland has to do. There are some areas that are not tested, but we have not had major issues to date.
The subject of forecasting has been explored with you, so I will move on to the question of appointments and reports. I think that your independence is critical in moving forward. Would you accept that you cannot really both be an adviser to Government and scrutinise what it does?
Yes—“adviser” may need definition, so that we are saying the same thing, but we do not conceive of ourselves as advisers to Government on policy or whatever. We see ourselves as assessors and as the body that scrutinises what Government produces. That requires interaction with the Government teams.
I entirely accept that. I was referring, in particular, to the fact that members of the commission had a role on the Council of Economic Advisers. In that case, there could have been a perception that would have been unfortunate. However, I gather that the position there has now changed.
That position changed, and it was raised at our appointment hearings—I am sorry, but I do not know what the correct term for that is. We took action at that time to end our terms with the Council of Economic Advisers. I can say to you that there would not have been a conflict, but that is a case where one might be perceived, and we wanted to address that.
12:30
Perception is all, unfortunately.
Some witnesses have suggested that the powers that ministers wish to have over appointments and all of that area might be excessive and be considered to be—dare I say it?—interfering. Do you think that the balance of powers is right, or would you again, to secure your perceived independence, prefer a more direct relationship with Parliament?
I will look to both my colleagues to give a view; we have talked about the issues. At the end of the day, we are answerable to Parliament; Parliament approved our three appointments. We were recommended through the ministerial line, but we were debated—I read the debate in the Official Report; we were actively debated—in Parliament and then our appointments were approved. We are also answerable here; the committee represents the wider Parliament. That is ultimately how we perceive ourselves. My colleagues may have another view.
No—I essentially agree with that.
I will slightly rephrase Jackie Baillie’s question, if I may. I will do it anyway. It is not so much about the power of ministers, but the suggestion that an alternative appointments committee or something should sit to make nominations, rather than ministers. I would not want to intervene on that. I do not think that it would make much difference in reality, but the perception might be important.
We will hire individuals externally as we build a team to help us in our work. We have started that in recent months with two part-time research assistants, and we will probably look for some full-time economist strength. We will advertise and have an open process to hire such individuals. Perhaps that is helpful to your thinking.
Again, perception is all. I do not think that the fact that ministers control that degree of the appointment process and take powers to themselves to do so is helpful to that perception.
Let me see whether I can manage my next question without it being rephrased. It concerns submission of the SFC report. We obviously have experience of public bodies submitting reports to Government in advance of publication. Technically, that is done for checking accuracy, but in practice issues of presentation have been covered as well. Again, there is in some cases a perception that there is Government interference in that process. Some witnesses have suggested that you should not give the Government any notice of your report, or that it should, if you are being generous, have 24 hours to respond. I am interested to hear your view on what would be appropriate, given that the OECD thinks that providing a report in advance favours the Government.
We have given a lot of thought to that, and we have heard views from the Finance Committee in the past as well. We have produced one report, which Jackie Baillie very kindly said that she had read, being new to the committee. In that case we submitted it in advance for a fact check, because we were told last year that that was the proper thing to do. That was very helpful to us and we continue to think that it is helpful to us to make sure that we have our facts correct. There is something in that. We did not change our findings—we were not concerned that we were offending anyone by doing what we did, and no-one asked us to change anything. It was purely a fact check.
That kind of engagement can be helpful, but one can play with the timings. The forecasters could have the forecast essentially complete, not to be changed, and they could see our report and correct any facts or suggest that the facts are different, or the report could be submitted before the forecast is finalised. We could look at those details, but we think that the process is valuable. It is hugely valuable to have a series of challenge meetings with the forecasters, because that is the value that we add, as Campbell Leith has said.
What we have tried to lay out in our submission to the committee is a way to balance what is a bit of a conundrum: how do we get that right? It may or may not be right—we would happily accept guidance—but we propose having the challenge meetings early in the process.
We would go back to the previous year’s budget and discuss suggestions or criticisms, and we would find out how the forecasters were changing or adapting their approach to a forecast and whether they were taking on a new, devolved tax—we will face new taxes relatively soon. In the early days, we would give the forecasters our challenges or whatever, but that would not be advice. We do that in OBR teleconferences that we join; it seems to be normal practice. We would then let the forecasters get on with running the numbers—we would not be part of that, because we do not look at numbers and outputs; we look at their thought processes going into that.
We could produce our report on the budget right at the end of the process. However, quite honestly, if we are to add value we need to be involved at the right points instead of simply sitting aside, as a think tank might do, and analysing the budget once it comes out. Forgive me for speaking in slang, but if the draft budget turned out to be rubbish—I do not think that it would—that would surely do no one in Parliament any good, and it would not do the commission any good, because people would ask, “Where were you?” We need to engage in the right way, and our thinking just now is that that is how we should do it.
I accept that you look at the processes and the assumptions that are made, but from our perspective, you gave advice on non-domestic rates, the cabinet secretary reflected on the assumptions and the modelling and made a change. We have no idea what he changed the forecast from or what was wrong with the underlying assumptions. Equally, with LBTT, you made a recommendation about behaviour and we do not know what has happened since then—we do not know whether the Scottish Government has taken that on board. It feels as though there is something not transparent about what has happened since you said what you said on those issues. My fear is that the interaction that you describe—which might be quite reasonable among reasonable people—is not transparent and the perception is that there is something not quite right.
My approach to this is not necessarily to make things difficult, but how can we ensure that your independence is guarded? That is the most precious thing.
Our independence is the most precious thing, so we continue to ponder how best to engage. One thought that occurs to me is that when we write our report on the draft budget, we should give the committee—or indeed, anyone who reads it—more detail of the “He said,” “She said,” type of thing. It may be helpful to detail such movements.
On the change that the cabinet secretary made regarding the economic determinants of non-domestic rates, we did not advise what that change should be, nor did we give a parameter; it was simply our judgment that the range looked somewhat optimistic. In such a case, we would probably have to have the Government spell out what the forecast had changed from, or you would have to ask the Government that question.
Keep challenging us, because we will improve the process as we hear those perspectives.
Thank you.
I like your use of the word “conundrum” in your written submission and in your answers to Jackie Baillie. That seems fair. More interaction might get us a better result; less interaction might be perceived as evidencing more independence but might not get us the best result. Do we need to have some of this in legislation or can we leave the legislation as it is? Is there going to be a memorandum of understanding or just practice? How can we put all that into practice?
I am not an expert on legislation. I will give my two colleagues a minute to think about how to respond to your questions.
My first thought is that, in all my professional experience—I am thinking of regulation, which is what I have dealt with a lot in my business life—I have learned that the more restrictive those statements are in trying to anticipate something, the more they omit things and do not anticipate the full story. More general expectations can be enshrined in legislation, but instinctively I say that we should not be very specific in legislation because, ultimately, that will have unintended consequences or lead to omissions.
Is your key point that all minutes would be published and that that would be how we answer the questions, in a sense?
Yes. People would be able to see iteratively what was being discussed, what had come to the table and the challenges that we raised. We started taking minutes of those challenge meetings once we had a bit of assistance, which was the beginning of the summer, and we have already put summaries of them on our website. In essence, we say what topics we discussed in a challenge meeting. We have the full minutes and will append them to our draft budget report so that they are completely public.
I agree with Susan Rice. It is an evolutionary process. We will post the minutes and add commentary in our report. If the committee then finds that that is insufficient demonstration of independence or that the method is still not quite convincing enough, then, in an iterative process, we will—
Find something else.
We will find something else until it works to everyone’s satisfaction. We are here to make it work.
Our goal is the same as the committee’s: that we are perceived as being independent.
I back that up. When the minutes come out, the committee will see some thoughts and comments about what the Scottish Government might do next about a particular problem. They are not just, “So and so said such and such”; they are a bit more detailed than that. What we are talking about is exactly what the monetary policy committee in London does: it puts up transcripts of who said what. Our minutes are likely to be a bit briefer than those are. For transparency, the MPC does exactly the same thing: it works on the minutes.
As I understand it, that model is similar to what happens in Ireland. Jean Urquhart and I were in Dublin to meet the Irish Fiscal Advisory Council and others. The council came up against a problem this year. It was also involved early on with many of the macroeconomic forecasts but not in the detail of the budget. However, when the budget was announced, there was such a big increase in spending—because the Irish are doing so well, apparently—that it had a knock-on effect on the macroeconomic forecasts. There was a bit of tension between the Government and the council as to exactly what was, in their words, being endorsed: what was reasonable, in our words. Is there a potential problem if you are not involved at the latter stages of the budget’s preparation?
That would depend on the forecasting methods that are employed. If there is a big regime change—a big change in tax policy—that no one has discussed before that point, there could be behavioural consequences and, if the modelling does not take account of that, it is late in the day for us to intervene to factor in the effect.
The model should be robust enough to cover the different scenarios.
That is the ideal situation.
IFAC also moved from a non-statutory to a statutory basis and was somewhat taken aback at the amount of administration, human resources policy, health and safety procedures and all the things that fall in once a body becomes statutory: although it is small statutory body, it has many of the same responsibilities as a big one. Are you aware of that and have you taken it into account?
We are very much aware of that. We have been thinking about it and are discussing and giving some shape to a transition team or board that would be supported through the civil service. That team would take us through all the stages.
We will go on to a statutory footing on 1 April 2017 so, from now, we have 16 or 17 months. However, we need to get ourselves ready well before that. We understand from the keeper of the records of Scotland that we have to keep records of all of our exchanges, emails and notes—everything. As a banker, I am used to that, and we have put in place processes to maintain those records. We want to get ourselves into those good habits well ahead of time.
We will need a risk register and we will need to have our finances audited. Over this year, we have started to have quarterly governance meetings—although we discuss such matters more frequently—specifically to focus on governance issues. We are trying to anticipate: we are planning what we need to do over the next year or so and are putting a number of things in place now, so that we are ready.
12:45We also understand that we have to have an accountable officer. We have talked about hiring a senior officer—I have suggested the title “chief of staff”, but that may change. We would advertise that post. I would look to that individual to keep us right. I do not underestimate the amount of time that all that would take. We appreciate that such things are important and we will certainly undertake them.
In the final paragraph of your letter you talk about resources and say that if your remit expands in the future you would have to commit more time. On the commission, you go on to say:
“The resources available to it will need to grow prudently to support the additional work.”
There has been some suggestion that we are funding the Scottish Fiscal Commission quite generously if we compare ourselves to IFAC, which has a full range of responsibilities, or to Sweden’s equivalent. Do you anticipate that your budget might not be sufficient?
The budget is partly built on an assumption of the Scottish Government’s pay rate for however many full-time economists; the rate might be different in other countries. The proposed budget also anticipates the potential need for us to use fully commercial office space. At the moment, as you know, we are being hosted in our non-statutory phase by the University of Glasgow—very generously—and that change will mean some additional cost.
I read the notes of the committee’s visit to IFAC, which points out that they have half a dozen or eight secondees. The notes do not say whether the secondees were carried in IFAC’s budget or came from Government. Robert Chote of the OBR has been very open with the Finance Committee in talking about the number of people in the UK Government who provide data and information and support the work of the OBR. I would not want to say that the proposed budget is right, wrong, generous or not, because that will depend on what it encompasses.
I assure members—my colleagues and I have talked and we are in absolute agreement on this—that we have no intention of spending up to budget. The budget is not there to be spent; it is there to allow us to grow and develop as we see fit in this incredibly important moment as we move from a non-statutory to a statutory footing. We would not go out and hire four full-time economists just because that was allowed by the budget; rather, we would hire one. We might in the future, if we felt that we needed another or as new taxes come in, decide to add to that.
To answer an earlier question, if we found that we had much-expanded responsibilities, we might have to come back and say that the budget is not enough. However, today we would not say that. It is a very fair budget that gives us scope to make independent decisions about how and in what ways to expand.
As John Mason said, considering the Irish experience was very interesting. Two things come to mind. Ireland is a smaller country, but it is in the eurozone and is independent. You are right to say that the staff were seconded—from memory, I think that they were all seconded for a certain period of time, including one from the central bank. The staff came with different experiences and they changed.
Of course, IFAC has five members and I wonder whether you have thought about that, given that you only have three. Is that under discussion? You mentioned the pool of expertise being very small. That was also a matter for discussion in Ireland and four of the five commissioners live abroad, although they are Irish. That was a revelation to me. Suddenly, we could look internationally. The Scottish diaspora is even bigger than the Irish diaspora, so there might be expertise around the globe that we could encourage.
You said that, given that we have 13 or 14 universities in Scotland, there should be plenty of people with the expertise to step into your shoes. Are there really people in Scotland with a very high profile and so on—again, perception being everything? Is that pool available here?
My colleagues may have anecdotal views, as they are academics and know their peers. We have not sat down as a commission and asked who those people are likely to be, but the governance of succession and succession planning should become part of our conversation. In a way, your question is a good trigger for us to move in that direction.
On whether the commission should have more than three members, it seems to me that two would be too few, and there is always the number 27 bus risk. Increasing our number might be a thought. All of us are quite prudent in saying, “Let’s grow in anticipation.” Our view is, “Let’s grow as we need to, as we become bigger, as we go into statute, if we need to grow.” I would not be surprised if we spoke to the committee at some point and said that we needed another commissioner. Their term would then be staggered, and they would gain the experience and the knowledge that was needed to keep all of it going.
You asked whether the pool is out there. I think that the committee occasionally talks to academics who might well be considered candidates, if they were interested in something like the commission. We have not determined who those people might be or how big the pool is.
I do not disagree at all. I would rather fix on particular expertise and skills than fix on a particular number—on having four commissioners instead of three, for example. There is a comparative advantage within the commission at the moment and it is working fine. We will keep an eye on that. We have certainly talked about it.
On the diaspora point, it is not just Ireland that has gone international. A lot of commissions do that. From my perspective, that is not an issue; it would be fine.
On the point about whether the pool is too small, there are plenty of people out there but the question is whether you actually want them. It is a question of what kind of people are out there. The answer might make the pool a bit more restricted; it depends.
It also depends on whether they would want to take on the role.
Yes; they might not want to do it. I know quite a lot about these things. In the Swedish example, the original chairman walked out in a huff because it was taking too much time. He wanted to publish a few academic papers at the same time; I sympathise enormously. Some people may not want to do it. It is a bit tricky.
Finally—time is moving on—you see yourselves not as duplicating work that has been done in a full forecast but as selecting particular pieces of work. Would you say that the change from stamp duty to LBTT, and the figures that we have on that at the moment, merit that being the kind of work that you would home in on? It is quite a small part of the budget overall, but it is important.
I can see that Government policies come through into such types of forecast. How would you deal with drilling down into, or looking more comprehensively at, any one particular part of the forecasting, while avoiding or ignoring Government policy that will actually be a determinant of the outcome?
In our initial report, there are several instances in which we suggest that it would be best if the behavioural effects of policy could somehow be incorporated into the method of producing the forecast. We could analyse those behavioural effects, build up a model or piece of analysis and feed into it whatever the policy happened to be. In that way, it would be ready to do the policy analysis that you needed to do.
There might be factors other than Government policy changes to consider. For example, regulation over the past year has made the mortgage application process much more rigorous than ever before, particularly for first-time home buyers. There is a lot of debate going on as to whether it is too much, but it will affect the number of people who can qualify for a mortgage and is completely outside Government policy. It is therefore important that we look at all those factors, including the external ones.
If there is any kind of break in the economic conditions that underpin an extrapolation of historical data, it is behavioural responses to those changes that you will be starting to look at.
You are quite right about the need to set priorities with regard to which taxes to look at and so on. They also come in different sizes, and I would put more weight on the bigger ones and insist that the errors—if and when we quote them—are set out not only in percentages but in pounds so that we can see, as it were, what bang they will have on the budget. That would help with prioritisation quite considerably, and I hope that we do that in future.
I also think it important to avoid commenting implicitly on the underlying policies. In other words, you have to take them as is for now. If you are told, “We are going to put this tax up for sure”, you might be able to make such a comment, because you would be commenting on a public announcement. Otherwise, we do not want to be put in the position that everyone has referred to of saying, “This forecast’s rubbish, because everyone knows that this tax rate’s going up or down or whatever.” That is the reason for the slightly delayed effect with regard to forestalling; indeed, that is a case in point, because it is the sort of issue in which you are able to say what effect it will have in terms of pounds only when it actually comes in—and even then things are somewhat uncertain. It is important to do that work at that point, but we should not second-guess ahead of time.
Thank you.
I want to ask a very brief question about forecasting just to get my head around where we are. Lady Rice has tried very helpfully to disaggregate forecasting with a capital F and forecasting with a lower-case f, and I think that the analogy is a very helpful one. I presume that your work involves a degree of modelling that essentially replicates elements of the forecasting method in order to test the robustness of forecasts without having to begin at the beginning and end at the end of the whole forecasting process. Is that a fair assessment of your work?
It involves a mixture of things. We go through the nitty-gritty of the Scottish Government’s methods but, as I have said, we are also undertaking a series of projects on things that are not contained in the methods that are being used. We are not just replicating an element of the forecast but looking at something that, as far as the Scottish Government is concerned, is off-model to find out whether that is an influence or factor that should be taken account of but which currently is not. It is a mixture of a step-by-step analysis of what is actually being done and looking at other off-forecast issues that probably should be taken account of but which are not at the moment.
On the separation between you and the Scottish Government, there is a perception that you are looking over the Government’s shoulder and asking, “Are you sure about that?” How involved or otherwise do you think you need to be to ensure that the challenges that you make are taken seriously and that the information that is laid before Parliament is as robust as it needs to be?
It is not really our job to ensure that information laid before Parliament is as robust as it could be—that is the Government’s job, because the Government lays the information before Parliament. We are not saying, “You must do this”; we are suggesting that the Government considers certain factors because they might have a big impact. The Scottish Government can take that on board or not. At the end of the day, that is the Government’s choice.
If the Government produced a forecast that, in our judgment, was not particularly robust, we would say so. In that case, the Government would have laid a non-robust forecast before Parliament. It is not our job to ensure that the Government does its job. The specification for our role would be different if that were the case.
13:00
I appreciate that. Ian Lienert, who has done some work around comparisons, suggests that the exertion of influence over forecasts changes the definition of the role from independent assessor to adviser. In your perception, is your role as independent assessors maintained or do you become advisers? To me, an advisory role is much more heavily involved in directing what happens, rather than challenging the Government.
On whether we would expect the teams to be answerable to us in some way as to why they did or did not take advice that we gave them, we are not in that kind of relationship with them.
As I said before, this is how the OBR operates. It has a series of challenge meetings, produces tables and data and explains its thinking so far; it speaks to key individuals in the meetings, who raise questions and challenges and then go off and do more work. We are doing the same thing in Scotland.
As an academic economist, I am used to academic seminars in which someone presents their work and is interrogated as they do that. That is not always comfortable for the person who is presenting their work. People go through the work line by line and make very critical comments. That toughens up what the person does. That is the model that I have in mind.
When we do our independent pieces of research, the intention is that we would release them as technical working papers. We would present them to interested audiences and go through the same process as the academic seminar—come and get it, see what we have done and kick holes in it if you can.
As I said some time back, our approach is motivated by the question whether we would do something the same or differently. At that point, you can suggest that someone considers a variable or other factor that has been left out. It is up to them to decide what to do. They might come back and say, “We tried that and it did not work”, although if we have an extended purdah in which we are not talking, that might be more difficult. However, it is entirely up to them whether they try it and come back or not. The motivation from our side is whether we would do the same, which is the same for a seminar.
If you go beyond that, you get more heavily involved and, as I said in the general peroration on forecasting, you would then find that your independence was compromised. That is why I do not think it a good idea to do that. I suspect that most commissions have reached the same conclusion and so produce back-up calculations, rather than the official forecasts.
That concludes questions from committee members, although I have three or four still to ask—I am only joking.
It has been a long session. We appreciate your involvement, concentration and the clarity of your answers.
It has been very useful, as all such conversations are.
Thank you very much. No doubt we will see you at the committee again before too long.
13:03 Meeting continued in private until 13:05.Previous
Devolved Taxes