 Peyach fi wrth g Arethe Scottish Free Commission Bill by taking evidence from a second panel of witnesses, Dr Henry Pete, Dr gedrangos Amstrong, Chris Stuart and Professor McGregor. Members have received papers submitted by each of our witnesses so we will go straight to questions from the committee. The first one will be to yourself. I am strong. What I like about your submissions is they're always nice and provocative. There's always lots of good stuff to chew over. I hope that your leg gets better soon, which is why you have the attire that you have on today as you've just explained it with a brief break. Let's go straight into it then. You comment in your paper that, in our view, the SFC's role in preparing reports of the assessment of reasonableness is too narrow. A preferred arrangement would be for the Scottish Fiscal Commission to prepare its own independent forecast, both from revenue and expenditure side, using inputs from the Scottish Government and civil servants and evaluating the fiscal consequences of their policy proposals. We'll dig into that in a wee second. I have to say what's interesting isn't the last panel. None of the people who made any submissions thought that there should be independent forecasting, whereas all four of you are of the view that there should be independent forecasting. Perhaps we should have mixed and matched that wee bit, but certainly we'll go into that. I wonder if you can then talk us through that a wee bit, Dr Armstrong. Yes, thank you. First of all, very good morning to the committee and it's a pleasure to be back here. The paper, I just want to say at the beginning, is a joint paper with Dr Catriona Lysenkerva, my colleague, so it's a shared work. We do think that it's important that the SFC has the capacity to make a forecast for a number of reasons. Here I would refer to the committee's Scotland's fiscal framework paper, which I thought was an excellent paper, and I note that there's a substantial section on moral hazard, which is very important. Moral hazard arises because of a lack of equal information, because of imperfect information, and that is why having a separate forecast is so important. Now, why is the forecast more important than a test of reasonableness? Reasonableness is, in my view, as someone who is involved in forecasts, it's actually quite a low threshold, because there's a lot of uncertainty. To be unreasonable, you've got to push somebody to be unreasonable, and yet, year after year, you can actually build up quite substantial errors. Rather than test this bounds of unreasonableness, I think it's much better to say, well, have your own forecast and we'll start from there. That is the first reason, because the threshold of reasonableness is inappropriate. It's too high to say that something is unreasonable. That's quite a strong demand. That would be the first point. The second point is that the rigour of going through a forecast is very important, because you realise that the more you look at data, the more uncertain it is that, until you play with the data, it's very hard to understand it. In terms of a discipline, I think that that's very important. Thirdly, for the overall framework of what's in the Smith commission, that the Scottish fiscal framework has to sit within the UK fiscal framework, and all that we know about devolving fiscal powers to sub-central government institutions, is that the issue of moral hazard has tended to come up, and therefore, to mitigate that, to provide some confidence, I think that the more independence and scrutiny and transparency that can be given, the better. The best way for that, in my view, is for this independent committee to be able to make the forecast and to do the policy costings. Thank you very much for that opening, and moving to you, Jeremy. You say in the very start of your paper, in the summary, that in line with the finance committee's recommendation, the Royal Society of Edinburgh is firmly with the view that the Scottish Fiscal Commission should be capable of producing its own independent fiscal forecasts and assessing the Scottish Government's fiscal projections. In doing so, the SFC should have the freedom to develop its own framework of analysis data sources and methodology. You go on to talk about the SFE not currently having access to data from HM revenue and customs, and the UK and Scottish Governments will need to address it so that the SFC has direct access to HMRC data. I wonder if you can take us forward in that regard, Jeremy. You are responsible. Good morning to everybody, and let me just know that I am speaking on behalf of the Royal Society of Edinburgh today and not as an individual, but as a representative of a group that worked on this topic. I very much agree, first of all, that it is necessary to have the ability to undertake forecasting within the Scottish Fiscal Commission. I agree with the reasons that Angus has just set out, but I think, and primarily amongst those, is the fact that until you undertake your own forecasting, you really do not get a feel of how the data is working, and then the key parameters that matter most within the forecast that is being derived. For example, as we note in paragraph 7 of our paper, it is suggested that the SFC will have the power to assess the impact on projective revenues of changing the value of the underlying assumptions. It is really difficult to see how that could be undertaken without having your own forecast that you can play with and test the sensitivity to variations, so I think that that is critical. In an ideal world, we would have within Scotland several independent forecasts that were around and could be used. We do not have that proliferation. I hope that there will be more going forward, and I am delighted that the Scottish Government is doing work in that area. I am delighted that Strath Clyde, where Peter comes from, where I have an association, is involved. However, there are not enough independent forecasts to rely on, and I think that that makes every sense for the SFC to develop its own capability. The point that is made about data is important. There are two elements to this. One is that it is necessary that MAUs be established with those that provide data at the UK level, as well as those that provide data at the Scottish level, and that a close working relationship will be established with the OBR as part of that. However, the other point is that there may be a number of items of data that simply are not available within Scotland in the right form and at the right time at the moment. I think that the one role of the SFC would be to point out the limitations on data and to work with the Scottish Government on improving data sources, so forecasting could be better embedded. There are a number of reasons why independent forecasting does make sense. I was surprised to read the other submissions from the panel that you spoke with earlier. I know that they are all accountants rather than economists, or taxpayers. Maybe you could argue that economists are just looking to the self-interest of their profession. I hope not. I think that it is necessary, but I think that the data point is equally critical that we have to have the right data to do the job properly. Last point, no forecast is ever going to be right. Every forecast is subject to error and uncertainties. Unless you have your own model to play with, you cannot explore those play with them and you will never get a good feeling of what really matters to make the forecast within the right bounds. Professor McGregor, what I-CAS actually said is that the ability of the commission to assess the Scottish Government's forecast projections is that the ability of the commission to assess the Scottish Government's forecast projections will be significantly enhanced if it has the capacity to generate its own forecasts for comparative purposes. You are in agreement with the panel there. Again, you talk about data limitations in the following paragraph. However, what I-CAS said in paragraph 12 of its submission, which I referred to, is that I-CAS does not consider it a need for the SFC to prepare independent forecasts as it would create duplication of work and not provide value for taxpayers, which is not the committee's view but certainly the Scottish Government's view. Can you discuss why you think that I-CAS is in error in terms of its assessment? I think that it would be foolish to deny that there are costs associated with providing independent forecasts. Those ought to be proportionate to the degree of sophistication that is used in generating the forecast. There is clearly going to be a learning process here. It is not clear exactly what models the Scottish Government is going to be using at this point in time. You do not want to have a sledgehammer to crack a nut. You need something that is proportional to the activity that is actually undertaken in the forecasting process itself. However, there is no doubt in my mind that—I agree with the other speakers for the same reasons that they put forward—the provision of those forecasts and the process of generating those forecasts is very important. I would expect, however, that those independent forecasts would, of course, draw liberally on the expertise that exists within the Scottish Government. We know that the Scottish Government has been seeking to develop its own expertise in terms of economic modelling. I will not list all the possibilities here, but we know that they have been looking at macroeconometric modelling. We know that they have been looking at computable general equilibrium modelling. We know that they have done some micro simulation modelling. We know that they have got a range of energy models. This is all good news. We need this to be able to properly understand the potential impact of policy. Equally, we need some external body to assess what has been done with these models and the quality of the analysis. The value added there will vary depending on the extent of fiscal autonomy that the Government has, but the greater the degree of that autonomy, the greater the case, the stronger the case for analytical ability and for independent forecasting ability. Given what we know now about the future extent of fiscal autonomy, it is such that there really is a need for that kind of analysis. I would have accepted it that for much lower levels of fiscal autonomy it would not be required, but it would now be about to go into a situation where that is essential. That is very important. As I said, there are roles for others as well in the independent evaluation. I doubt that it will be the case that the independent fiscal commission will be given a very broad-ranging remit that might encompass assessing contribution to the ultimate objectives of Government policy. Sustainable growth contains within it a number of objectives that are potentially conflicting. We might want to understand the impact of the policies that we are pursuing on that range of goals. I doubt very much that the fiscal commission will be given those powers, but I would like it to be. If not, the fewer powers that the fiscal commission has, the more important it is to have some other external independent sources, including, of course, the academic institutions that are partially represented around the table. It does depend on circumstances, on the degree of sophistication, on the degree of fiscal autonomy, but my judgment would be that, given what is likely to happen here in the near future, it is absolutely essential. Mr Stewart, in response to the response from the Scottish Property Federation, paragraph 6, supports what has been said already. In paragraph 3, you say that, with powers being devolved in income tax, it is important for the economic competitiveness of Scotland that effective tax and finance scrutiny of the Scottish ministers decisions is assured for the Parliament and the wider public. It is effectively a building on a lot of what Professor McGregor is saying there. We do not agree that that is assured merely by commenting on the reasonableness of Scottish Government forecasting, which, again, is much in line with what Professor Armstrong has said, so I wonder if you can expand on your thoughts a wee bit there. Sure. Thank you very much for the opportunity to speak. I suppose that we represent the property industry in Scotland and our concern is about that. We rely heavily on international markets for our capital, which is one of the key issues behind our industry. Going back to the point about reasonableness, we deal with that word all the time in our contractual responsibilities as property developers. It does not necessarily have the level of teeth that you would expect. From a capital markets perspective, the need for independent assessment is critical because that benchmarking against Government policy is really required. It needs to be there to give the capital markets the assurance that policy has been made with benchmarking stress testing with a robust challenge in place. That is probably one of our critical points. The closer you get to the bond markets, the more you have the ability to raise finance from those markets, the more they want to see stronger disciplines in place. I think that that is necessary. Jeremy, you have said—we have again touched on this already—assessing the long-term sustainability of the public finances of the RSE is frequently concerned that the focus is on short-term priorities to drive the Scottish budget process. You have said that you agree with the Finance and Committee recommendation that the functions of the SFC should include assessing the performance of the Scottish Government against its fiscal targets and assessment of the long-term sustainability of the public finances. How crucial do you think that that is in terms of legislation? In my view, that is becoming increasingly important. Peter MacGregor referred to the extent of fiscal devolution. I think that we can all accept that this is likely to increase rather than reduce over the years ahead. The more that you go down the fiscal devolution route, the more important it is that the Scottish Government looks at not just the short-term but the medium and longer-term perspective. That is why we very much agree with the Finance Committee that looking at performance against fiscal targets and critically assessing the long-term sustainability of the public finances should be a substantial element of the SFC's role. We think that that would be assisted if the Scottish Government itself prepared a medium-term budget framework while it produces its annual budget bill. Alongside the annual budget should be this medium-term framework and then the SFC should look at that, look at long-term sustainability and provide comments. As we note here, nothing is perfect in this world, but at least it would be able to point out critical issues and see matters that need to be looked at from a policy perspective or from a budgetary planning perspective. We think that that would be an extremely important element. The more that we go down the fiscal devolution route, the more that this is going to be critical. When we get to the stage of having fiscal rules that need to be adhered to at different times, I think that looking at the question of adherence to fiscal rules for the Scottish Government and also to assist in Scottish Government contacts with the UK Government would be a very valuable role for an independent or truly independent fiscal commission. You say truly independent as if you suspect that it is not truly independent or might not be. Do you have concerns on that front? I think that it is absolutely critical that everything is done to ensure that this is truly independent. I think that significant progress has been made over the last couple of years, a lot of it due to what this committee has been saying and others who have been giving evidence to this committee have been saying. I do not think that we can yet be 100 per cent certain that the independence is entrenched and will necessarily be there to be sustained throughout. I think that there are issues that need to be addressed and one of those would be to make sure that as much as possible as to the role of the commission is within legislation and can therefore be relied upon. That is one of the points that we make later on in our paper. I was not casting doubts on it, but I am absolutely underlining the need for utter independence and the utter perception of full independence. Dr Armstrong, you also say that the SFC should award a wider role in assessing the sustainability of Scotland's public finances. Do you suggest that we should work out modalities of the creation of a forecast liaison group as a pair of budget responsibilities that I have understood headed by the Scottish Fiscal Commission by involving civil servants working on the forecasts? I concur with the view that the SFC would ideally prepare a long-term fiscal sustainability report or whatever the new name would be, but that would be the idea. For three reasons. First of all, in the long term is where a lot of the challenges that Governments face lie. In particular, the demographics challenge is a longer-term challenge and that needs to be considered in short-term decision making because that is where a number of the difficulties are going to be coming. Second, we are particularly bad, notoriously bad at discounting the future properly. The climate change work has had a lot of discussion in this area. Our way of taking adequate account of future possible events where there is uncertainty about them is not historically poor. In other words, we just do not put enough weight on what is going to happen in the future. That is not a criticism of Scottish policy making at all. It is around the world. It is just the way that human beings tend to operate. I think that mitigating against that by assigning an independent body with that role. Finally, this is a bit of a public good. The SFC does not do it. Who is going to do it? This is really a public good. With the greatest respect to academic colleagues, this is a public good that has to be communicated to the public in ways that we can understand. I do not understand half the papers that are written, so it needs to be made. It is something that can be readily accepted by the public, so it can influence policy properly. That is why I think that it resides with the SFC. In terms of the fiscal liaison committee, this has been successful in the OBR model. It is as much about having the powers as using them, if that is clear. The fact that you have the powers to call on civil servants and people to provide evidence and to give you information and help you in the preparation of it. You have the responsibility to do that, if the fiscal liaison group asks for it. It actually puts the burden the other way around that you really do have to co-operate with the OBR now. It also gives the OBR access to a wide number of people. The independent report on the OBR pointed out that anyone might have 100 people working on the OBR's work because you have a broad network of people that have a responsibility. The reporting from the chair of it, making it the head of the SFC, being the chair of it, and reporting, in my view, to your committee, will be the most transbound democratic way to do it. I think that that is very important as a modality for making this work. There are two separate answers. One is the fiscal and the other is the liaison committee. Let's switch the emphasis a wee bit. Professor McGregor, you referred to borrowing and deficit bias in terms of your own paper. Someone, if you can talk us through that a wee bit. If you can comment first of all on the response from SEPFRA and their submission, they say that, in relation to borrowing, we would advocate that the statutory powers here should be expanded from projections of borrowing requirements to assessments of the sustainability affordability and prunes of the Scottish Government borrowing. The deficit bias refers to the notion that Governments might be tempted for a variety of possible reasons to spend in excess of the revenues. There might be a whole range of possible sources of that. One would be over optimistic forecasts of tax revenues, another might be a political cycle explanation of what is going on. However, for whatever reason, there has been a tendency towards the creation of deficits that are often judged to be excessive. Part of the role of the fiscal councils around the world typically has been to—that has been a major motivation in the formation of these councils as I understand it—to try to tackle that deficit bias and to provide some independent monitoring and checking of the policy position that has been adopted. Part of that implies that you have the ability to project forward for Government expenditures and for Government tax revenues and be able to assess the difference between those, depending on the precise fiscal rules that are in place, and monitor that and react to it if there is an anticipated problem further down the line. That means that you try to anticipate what possible problems might develop. That is an important role of the Fiscal Commission and would imply as well that the Fiscal Commission has the ability to do all that or require others to do it. The implication would be that it ought to be tight responsibility for monitoring progress on adherence to fiscal rules. That might involve advice on when it might be sensible to depart from those fiscal rules. If there are particular economic circumstances that suggest that, even though we have a set of rules, there should be some deviation in those particular circumstances, it would be appropriate for the Fiscal Commission to comment on that as well and provide some kind of assistance. Panelists want to comment on that particular issue. I am just going to ask one further question and then I am going to open it up to the rest of the committee because a number of people want to ask questions. To do with yourself, Mr Stewart, one of the things that Mr Swinney says when he comes to committee when we talk about how the SFC has had an influence in terms of the Government and forced it to make changes is about the non-domestic rates income, for example, when last year they said that the Scottish projections were over-optimistic and that the Scottish Government changed it. However, you are suggesting that paragraph 11 of your report that there is a significant role for the SFC to play in verifying Scottish Government forecasts that will be aligned to the forthcoming revaluation. We welcome strongly the inclusion of that as a specific part of the bill. You go on to say that you feel that it is important that it forms a core part of the SFC's consideration, so I am just wondering if you can tell us a bit more about your views on that. Yes, I think that this is just a general point about alignment with markets and in the principal towns and cities of Edinburgh, Edinburgh, Glasgow, Aberdeen, there have been fluctuations in valuation clearly over the course of the last eight, nine years, but in some instances there are rents, for example, of Haft. There are fairly major divisions between how the market is moved in different areas, so linking the way in which forecasting happens, especially around non-domestic rates, is critical to understand the impact on those markets and ultimately the tax take that will come from them. It is such a big part of property's contribution, I suppose. It is the nuances of those markets that we do not think are reflected, necessarily, that need to be reflected by the new SFC. That theme of market alignment really goes through the whole thing, because the impact on markets is critical. I have any further comments from anyone on the panel on that particular issue. Right, let us open up to the committee. First person to ask questions will be Richard to be followed by John. Thank you, convener. My first question is for Professor McGregor. Some of the concerns that certain witnesses have given around the production of independent forecasts has been the resources issue about how much it would cost the commission to do, but your paper suggests that, in fact, the commission should be able to access work that is already being undergone in the Scottish Government in terms of modelling, in terms of staff resource. Could the commission do that, therefore, minimising any additional cost but also maintaining independence of its own forecasting? It is a challenge, clearly, and you need to be very careful about how that is done, but I do not really see why not. I do not really see why the Scottish Government could not be given a set—or the Scottish Government of Commerce could not be given a set, for example—a set of alternative assumptions that the commission would like to explore. One that has already been mentioned that is going to be key is what is going to happen to net migration. This is not easy to judge. It would be perfectly reasonable for a number of groups to take different views of that future, and it may be very important to understand what the fiscal consequences would be of those alternative futures. That is a comparatively straightforward example of variation in assumptions, but if you think about all the assumptions that have to go into those projections of future revenues and costs, it is very difficult to do, very challenging, and each of the individual components is difficult when you are talking about forecasting deficits, and this is really challenging. It is inevitable that there will be alternative views about the likely futures, and I think that it would be great if the fiscal commission were in a position, either to do it themselves or to require others to do it. You are talking about using the same models, assuming that they were satisfied with the methodology and the modelling, using the same models to do alternative simulations. That is a way of minimising the cost. It does not eliminate the cost. I am not pretending that it is not costless, because it is part of the burden that is shifted to the Government economists, and they will require the capacity to do that, so I think that we need to be careful. I just wanted to comment on that, if I may. I agree with Peter MacGregor that there is no problem in making use of models that exist where they exist within the Scottish Government or elsewhere, provided those who make use of them A have a total understanding, and B have full access to the models and can use them as they wish. My preference would be not for the commission to ask the Scottish Government to test some sensitivities through their model, but actually for them to have the model in their own hands and then to use it themselves to have the skills and expertise in computing capability to use it, and to undertake their own calculations, rather than asking Scottish Government officials to do it. That would work much better, but we are talking about a proliferation of different models that are required to forecast different elements of that. Peter and Angus know that I am far more practitioner than I am, but that will be difficult. Making use of good modelling that exists wherever it be makes every sense, but you have got to know how it works, how it has got to be satisfied that it works well, satisfied with the data and preferably do the work yourselves so that you can then be satisfied the results of ones that you can place credence in. It is not starting from the scratch on all this work, and it is a key point in terms of the resources. There will be models there, and they will be used in a different way with their own expertise and their own views, but therefore coming up with independent forecasts for that necessity of creating a whole infrastructure themselves. Is that right? Let us be grateful that, rather than £20,000 a year, we are talking about £850,000. The £20,000 a year was an absolute nonsense, and so now one is talking about what looks to be a very reasonable level of resourcing, at least for the initial stages of the commission, but yes, it must be cost efficient in its use of resources and make use of what exists elsewhere that is of the quality and independence that is required. I completely agree with Jeremy Pee's point here, and I think that this point about transparency is really crucial and should characterise all aspects of the behaviour of the fiscal commission. My own personal preference, and I think that the Scottish Government economists are attending to this view, is that these models ought to be in the public domain, and they ought to be accessible not simply by the fiscal commission, but by others who have an interest in independent analysis, so I do not disagree at all. I also agree that, ideally, all of this work would be done within the commission itself, but I just do not see a necessary fundamental objection that part of this work would be done according to the requests of the fiscal commission, but by the models themselves. What reasons have you given for these not to be in the public domain already? If you have any comments on this, Dr Armstrong? First of all, in terms of the resources at the beginning of your first question, I think that that is why a fiscal liaison group with having the right to information and to an explanation is very important. You really have the power then. Knowing that you have that in your back pocket to call on makes the communication and the whole dialogue much easier. I think that that has been the experience with the OBR. In terms of the models, I would add a note of caution that we have all these wonderful macro models, but when it comes to forecasting fiscal revenue, they are pretty crude spreadsheets. I would love it to be a really fancy model that is all nicely coded up. They are not. They are big spreadsheets. They are 60 to 70-page Excel spreadsheets, which are just a nightmare to get your way through. That is what they look like, which are full of a lot of data, some of which are sensitive, some of which just happen to be market sensitive, non-published, HMRC data. I just do not see how that can be made public, but I think that that is exactly why the SFC needs to have these data sheets. I do not see how it could make a judgment on reasonableness without starting with that base of going through the numbers, because not until you have looked at them and thought, what are those elasticities and why are they that number and what happens if I put a different number in? Oh my word, look at all of that. A very competent user would take a couple of weeks just to figure your way around some of these models. It is interesting that you said that, therefore, there may be some problems accessing some of the data from a public point of view. In terms of the commission, everyone stated how important it is that it has access to data. Do you foresee any potential problems in it gathering the data that it needs from other agencies, particularly in terms of UK departments? Are you confident that robust memorandums of understanding will be able to be put in place to enable that to happen? Do you, Professor Pete, because you know what your previous experience of Government is? Do you know how open and transparent some of the agencies have worked? I was just thinking as a parallel on my experience as a member of the Competition Commission and then the Competition and Markets Authority. With that hat on, we run what are called data rooms, where data is put into a room and people having signed confidentiality agreements can access it, can run their own models, can work with it and can therefore make full use of the data, but then can't take the full data away. They can take the results and the evidence, but that way, if there is particularly sensitive data, as Angus Armstrong has referred to, then members of the Independent Fiscal Commission or their staff could be allowed to use them within that confidentiality context and then make use of the results without disclosing the highly confidential information to a wide spectrum of people. It can be done. Professor McGregor, do you want to make a comment on that? No, no, I agree with that. It is certainly doable. That is absolutely the key issue. The final question, convener, is on the issue of the independence of the SFC and ensuring that is guaranteed by legislation, because Professor Pete answered very carefully your question this issue earlier on. At the moment, my understanding is that the Fiscal Commission Council is actually consulted before or by civil servants in producing Government forecasts. In your submission, Professor Pete, you said that it would be alarming if the SFC would have a role in producing the same forecast that its task was assessing. Does that suggest that the commission's role needs to evolve quite quickly at the moment? Are you satisfied that the commission in terms of legislation is fully guaranteed in terms of legislation, or does the legislation have to go further or change anyway to ensure that the commission is as independent as it should be? On the second part of your question, we have said in our submission that one area where we do believe that there could be enhancement of the legislation is that the terms of appointment should be clearly set out in the legislation, as opposed to being left to the discretion of Scottish ministers. I think that that would make it easier for independence to be assured. On the first point, I am just trying to recall what the first point was. On the second point, there exists where the commission is consulted by the Scottish Government officials over their forecasts. You said in your paper that that is not a practice that you would encourage. I apologise for missing that point. I think that what we have said is absolutely the case, that if the SFC was there throughout the forecasting process, giving comments alongside the Scottish Government as the forecast was produced, it would be very difficult when it was then published to stand back and say, well, this is where we disagree with it. I think that there is a difference. There is no harm in being involved in technical aspects of how the forecasting is undertaken through the fiscal liaison group that Angus has referred to or whatever, but then when it comes to the actual forecasting itself of the making of assumptions and the working through the process, I believe that that should be undertaken by the Scottish Government alone and that the fiscal commission should then be in a position with its own forecasts, but also with a good understanding of how the Government forecast was derived, of making informed and objective comments that are then available to put into the public domain. If it was there within the camp as the forecast was made, it would be very strange to come out and give severe criticisms afterwards, so I think that the only way that it works is by being separate. I am continuing the whole area of forecasts. The opening paragraph, Professor Pete of your paper, talks about SFC should be a freedom to develop its own framework of analysis data sources. I am just wondering how that would work in practice. How could they develop their own data sources? Because, presumably, HMRC, for example, has to be the data source for income tax. The answer is that they would not be a data collection agency. What they would be doing is identifying which elements of data were inadequate or lacking and working with the agencies at the UK level and the Scottish level that were potential providers of the actual potential providers of that data and working with them to make sure that the data was enhanced and improved over a relatively short period, so that the work of the commission could be better founded. It is not a matter of going out and collecting data, but of working with the other agencies through MOUs to make sure that the data is all there that is required for this process. Can you tell us what we are going to ask next? Is it the MOUs or is it on the face of the bill that we need more about availability of data? That is a good question. Something should be within the legislation that it is part of the role of the SFC to ensure the availability of adequate data for the forecasting role. That gives them a locus that MOUs should be established as soon as possible and should be clear within the bill that MOUs are anticipated. However, if their role is set out within the legislation, they can develop it in the short term thereafter. Is it totally black and white that either the Scottish Fiscal Commission produces forecasts or they do not produce forecasts? Is there any grey area in the middle that they play with the model, do some of the forecasting—one of our previous witnesses said something like that—do some of the forecasting? Is there a halfway house or does it really have to be all or nothing? There is no halfway house as to what they are empowered to do. They have to be empowered to do as much of this as they see fit. What they then choose to do and how they do it, following what Peter McGregor said and elsewhere, fine. However, they have to have the full powers to do as much of this forecasting now and going forward as devolution is extended as is possible. I would like the legislation to be future proof to allow for extensions as devolution is extended. I think that I would completely endorse what Jeremy has just said. It is a bit like being half pregnant. It is either independent or it is not. Either you stand with total responsibility for this forecast that you have made or it is not yours. However, they would not need to necessarily—although they have given the power to do that, they would not need to do every single step themselves, because bits of it they would be comfortable with. They would sign off on the forecast, which implies that you are signing off on every step that you have taken part in that. You are fully entitled to use part of the forecast that the Government's civil service or the Scottish Government has given to you that you have requested and it has been provided for you. You are completely entitled to do that, but you are doing it accepting full responsibility. This gets to Jeremy's point that you have to have the capacity to be able to have full control over the forecast. How, in practice, you would like to use that power is your judgment alone. If you want to do all of the minutia, you can do it. The chances are that you are actually going to accept what the HMRC has largely provided. You perhaps tweak it, change it, and you will pick up that over time and experience. You need to have the capacity to say that if we do not accept the data that has been provided or the analysis that has been provided by somebody else, we have all the right to go back and say that we want to call for more data or do it ourselves. You have to have that capacity in order for it to be your forecast. Is it the danger that we end up with the Government to produce the forecast, the SFC to produce the forecast? They both dig in and defend their forecasts. We have got nobody kind of neutral to comment. You raised a good point. In our paper, we would just have the SFC providing the forecast that the Government then uses, because we do not think that it should be a beauty contest between two agencies. The problem with forecasting is that we will make a forecast next week. The week after, the whole world could change for some reason or another that was completely unforeseeable, but might make one of the forecasts, which was not reasonable, become exactly accurate by default. It gives a false impression of forecasting. It is very ambiguous and difficult to decide who is doing a better job or whatever. The first best from our point of view would be the SFC, which the Government would be obliged to take in terms of its setting up several old budgets. That would be our first best model. Assuming that that is not going to be where we are, and that we would strongly support the committee in the fiscal framework paper that you put forward, we think that the SFC has to have full power to do its own independent forecast. That has this unfortunate beauty contest, but since our first best option does not seem to be available, we would go for that. Again, I would endorse Jeremy's point that that means full capacity to do all the forecast if it needs to. I think that the fiscal liaison committee is important in order to request an explanation from either HMRC or Government forecasters for some work that they have done. That is an important part. That should be within the MOU. The MOU that works for the OBR and the British Government is very important. That is where the understanding and the co-operation are vital. That is an important part of that framework. I believe that that MOU should be referred to in the bill. Would the absolute ideal be that the Government produces a forecast, somebody else or somebody who produces a forecast, and the SFC stands back in comments on the MOU? The absolute perfect for what I would see is the OBR-type model, in which the SFC does the forecast, which becomes part of the Government's overall budget. That would be the first best. I do not think that this is a job to start calling it on to academicity, the best rule in the world that is paid by the Government. I think that that is a mistake. We should stop putting things on to other people. With setting the whole framework, if we believe in avoiding moral hazard, that means that we need transparency. Transparency requires that whoever is making those assessments can do it completely independently, and they have to be seen to be independent. I am not casting assertions that those people would not be independent, they have to be seen to be. Chris is quite right that when the Scottish Government eventually goes to credit markets to start borrowing, which I am assuming that they will have to do eventually, because with this much power, my personal view is that it would be necessary, that it is not only whether you are being seen to be independent, and I think that that is absolutely crucial. That is why I am afraid that I am slightly hardlined on this. I do think that this should be completely independent, not only independent, but with the full powers of explanation from other agencies. My next point was independent, so did somebody else want to come in on that? I think that it is central to the credibility of the organisation that it is independent. The perception is going to be everything, especially from capital markets. If it is not truly independent, if it is just analysing or using reasonableness, that is not going to be good enough. That is complete independence. Its responsibilities will clearly grow, and so its credibility needs to be strong. To be slightly cynical, can you ever have complete independence? We were over in Ireland. All the key people know each other. We heard that one of the people on the council was checking on somebody that he was pretty friendly with. You can never really have complete independence, can you? I think that everything has to be done to make sure that it is the perception. It is a small place, clearly, but I do not think that that is an excuse. Jeremy also made an important point on the terms of appointment. I take the issue that you are raising. How can you ever prove independence when you can never prove it? However, if you make the terms of appointment sufficiently clear and open to public scrutiny, you can minimise the risk of having something that people view with some doubt. I think that it is more about minimising the possibility of that. In terms of appointment, that is crucial. Professor Pete, you said earlier on that you are very much emphasising independence in the legislation. I was just wondering how much do you see it as we can force that into legislation? How much is it by appointing the right people who will be independently minded, who will stand up to John Swinney, or that kind of thing? I think that it is a mixture of both. I agree with what you were saying earlier that you cannot provide a 100 per cent guarantee of independence, but you can avoid obvious pitfalls. For example, you should not appoint people to the commission who are members of the Council of Economic Advisers. That was a clear inconsistency in setting up an independent body, which has been removed. There is not casting any aspersions on the individuals. It was just the wrong way to go. Is that the only body that you would not allow and overlap with? I think that I would be very careful about other areas, other agencies. I suspect that probably it is the one that matters most. However, any other body that was giving advice to the Scottish Government on matters that are related to the fiscal situation, I think that one would be very careful about appointing people with that role. I think that that could come out in the appointment process. One could actually have a note in the terms of appointment that absolute objectivity, as well as knowledge and understanding, was critical. However, I think that that is a matter of you touching and feeling it and seeing whether it works rather than absolutely prescribing who cannot be there. I think that it is so important that one gets this right, that the more one can build it into the legislation, the better. Can I just note in passing that I think the RSE agrees with what Angus was saying about the first best solution being this body undertaking the forecast and them being used. My hope is, I suppose, that over time, as A, we get experience of the IFS and the IFS proves its worth and gets comfortable in its working relationship with Government, B, as fiscal devolution extends, maybe the Government will feel comfortable within a relatively short period of saying that it will move to an OBR-type model and allow the IFS to produce the forecast that is then used within the Government budget. I hope that it will become comfortable with that. It has come a long way over the past couple of years in terms of enhancing the independent and enhancing the role, enhancing the resources, maybe moving at this stage to giving them the right to produce the forecast that is then used in budget is a step too far at this stage, but I hope that within time that might be achievable. So, I do apologise. Yes, I meant the SFC. Yes, that's just a clarification. Let's keep the IFS up and this is another. Okay, just a couple more points. Dr Armstrong and your, towards the end of your evidence, you said that we believe that the SFC should take the OBR forecast as given. Now, I'm just wondering if you could unpack that a little bit. Are you meaning for whole UK things, or are you also meaning for Scotland? Our understanding is that the OBR does pretty rough and ready stuff for Scotland. They just say, oh well, 8 per cent and that's it. Whereas we're actually expecting the SFC to dig in a bit more there. How are we seeing that? Forgive me if that wasn't specific enough. I was dealing with the macroeconomic forecast that the OBR produced. When the OBR says that they expect UK growth to be whatever, 2.5 per cent next year, I think that that 2.5 per cent should be given as a UK macroeconomic forecast. An interesting problem then becomes, is suppose the SFC were to say, well, we think Scotland is going to grow by 10 per cent. Well, there's an implication there that the rest of the UK is going to grow by one point something. That becomes tricky because that could actually clash with what the OBR has been saying about the rest of the UK. It's a remote possibility. Because of the size, one is 10 times bigger than the other, you've got to get a big divergence to make it statistically meaningful that you can actually identify this. I think because of that, that the 2.5, the OBR's UK forecast, should be taken as a given and that any Scottish GDP should operate within that. I think that what would be ideal is if the OBR could, in conjunction with the SFC, come up with an either indicative Scottish macroeconomic forecast, but it would have to be indicative because there isn't the data there at the moment to be certain on that. In my view, the weakest part of the data is the external accounts for Scotland. How would you have a current account for Scotland when most of it is with regard to the rest of the UK? That's really messy because then you've got to work out who owns what company north and south of the board. You're opening up a whole kind of worms in order to be able to do that properly, which is why I think the neatest thing is just to take the OBR headline macro number as a given until such time we have the right Scottish data which we will say, well, let's have a rethink about whether that is the best way to go about it. I agree up to a point. However, I think that it's very important that we appreciate the degree of interdependence between the Scottish and our UK economies and the extent to which spillovers are likely to occur depending on policy initiatives in either region. We've done some preliminary work looking at this and looking at the scale of these spillovers. In fact, they can be quite surprising and quite substantial. Given the existing data on interregional trade, that's an important qualification, it looks as if, for example, when you conduct an expansion in Scotland by stimulating exports, assuming that that was successfully achieved, the impact on the rest of the UK is almost the same scale as the impact on Scotland. That's because of the trade data that are there. We really need to understand the extent of interdependence of these economies if we're going to understand the impacts of policies appropriately and if we're ever going to aspire to any kind of no-detriment principle on a wider scale. I agree that what that does to us is to begin to raise questions about the nature of the interregional trade data themselves, but they are fundamental to understanding the extent of policy spillovers between Scotland and the rest of the UK. I would use that as a reason for suggesting that, yes, in the short term, but rather, in the medium to longer term, what we ought to be doing is really trying to correct that data problem and really look at this in some detail. The solution to this in the short term is that, as part of the arrangement with the OBR, that the OBR and the SFC should liaise as the OBR is finalising its forecast, and any input that the SFC can make as it sees how the OBR is moving to take account of particular issues that pertain in Scotland that may be relevant to the macro forecast at the UK level and vice versa are taken into account. So, it's part of the process of OBR producing its forecast would be to talk to the SFC and to take account of their expertise as it reaches its final decisions at the UK level. The final point was for you, Mr Stewart, in case you have left out or anything. Towards the end of your report, you talked about LBTT. Of course, that is interesting, because that is something that is already happening, albeit in the early days. You said that an independent forecast of the consequences, etc., would have helped with current LBTT forecasts. In the following paragraph, you make the point that it appears at the moment at least that residential revenue is below what was expected, although I think we'd have to put some caveats in there, like it's the first year, it's the first half of the first year and all that kind of thing. Let's assume for a minute that the forecast was wrong, that an independent forecaster would have given us a lower forecast of incoming taxes. That would have encouraged the Government to raise the tax level. Would you have been happy with that? I think that it's more about the behavioural analysis behind the forecast and looking at the stress tests, because ultimately what's happened is that there's a segment of the market, the top end of the market, which has really stopped moving. That can be completely attributed to the change in LBTT rates. Clearly, that has several implications. It means that the market isn't functioning as well as it should, because some people want to move down, and if you've got a part of the market that's sticking, that affects the whole market. I suppose that our comment goes to how much analysis was done in the banding to understand the behaviour that has occurred since the change was made. If the SFC did its own forecasting, stress testing and that could be analysed, that should inform or help inform the decision clearly. I don't think that it's because it's the first year. I think that it has just stopped dead a part of the market and other parts of the market are performing very well, but that's why there's a clear differentiator. Just that some things, especially new things, are almost impossible to forecast? I would have thought that the whole point of the SFC in an independent forecast is that it can use that data to be more informed about your thresholds. It's about getting it as accurate as you possibly can. I think that what has happened with the market, from a market perspective, has been pretty blunt at the top end, and it is affecting other parts of the market. I'm sure that that wasn't the intent at the point at which the changes were decided. There must be a better way to do that, and clearly the SFC and an independent forecast would assist that process. I take your point. I suppose that my one feeling is that it's still early days, but we'll see what happens. Thanks. That was a brief supplementary to John's question. Just to explore that a little bit further, OBR has been held up as a standard to which perhaps we should be looking in terms of the SFC, but OBR also did forecasting around LBTT. Therefore, it was that the income from LBTT would be even higher than that forecast by Scottish Government. There are no guarantees that, if an alternative forecast is produced to that from the Scottish Government, it will necessarily be accurate based on what the OBR was predicting in relation to LBTT. If what you say transpires to be the case in terms of the income that is derived from LBTT, is it? I'm not going to comment on the OBR. My comment is about what has happened to the market and how we can better inform the decision making so that the market is that behavioural analysis. The top rate of LBTT is now pretty overwhelming and is stopping transactional activity. Whether that was introducing different banding, whether that was more interim or whatever, it is just about better information at the point of decision making. I am sure that that is what we are looking for from an independent FSC forecast. I know that the Scottish Government forecast is for all transactions to bring in £381 million. The OBR, in March 2015, estimated £431 million. It revised that in July 2015 to £540 million, so it lifted its estimates after LBTT had been introduced. My comment is not about OBR and its forecasting. It is about how you collect the appropriate amount of tax from the real estate market and how you make sure that your forecasting delivers the take that you are predicting. With the residential market, a lot of that is behavioural. It should be easier to try and make that prediction. Whatever the level, and it just seems that that ability to predict, in the case of LBTT, maybe was not there. It has caused very clearly part of the market to stall. That is not a different question about the amount that you recover. It is making the market not work as it should be, which is not good for the whole of the market. I am aware that we have an evidence session with Revenue Scotland coming, so I think that that might leach into that evidence session. I suspect that most of the questions that I have have been asked and answered already, but can I tease out just a couple of things? I think that it was in Professor McGregor's submission that you talked about the scrutiny of the block grant mechanism. Whilst we are all clear about testing Scottish Government assumptions, you are very clear about testing a UK Government assumption with the block grant mechanism. I wonder whether you would expand on that, and I wonder whether the rest of the panel agrees with you. I do not think that the commission could be charged with coming up with an alternative block grant mechanism, but I think that it is exploring the consequences of the block grant mechanism that is eventually agreed. Perhaps I am exploring the possible consequences of alternative mechanisms as well. There has been quite a lot of discussion in this committee as well about the precise nature of the adjustment mechanism. It can be very important in governing the revenue that is available to the Scottish Government for that reason. Put it this way, some independent analysis of it would be very welcome, and if the fiscal commission is in a position to require others even to do that, I think that that would be very helpful and useful. I think that, longer term, it would be useful to monitor the behaviour of that mechanism, and if necessary, to think about whether there are grounds for renegotiating that if it is believed to be prejudicial in some way, or appealing to the no detriment principle if we are still doing that. I do not know whether I can work or not, but I certainly think that we should be informed about the consequences of this, given its potential importance. I think that this is extremely political. The formula and any adjustments, I think, are a matter for the UK Parliament overall, for all parts of the UK. It is not just a Scotland issue, and that is the place to discuss it. I think that it would be bad. I think that it would be negative to give this to the SFC, because it is such a political question that you are throwing the body into the politics straight away. You need the body to co-operate very closely with the Government, and you cannot legislate every item in this. It depends on having confidence between both sides, and making something so overtly political would make that difficult. I would not do that. I think that, in the committee's report where you talk about the intergovernmental relations, you have a very good point that that needs to be put on a whole better standing. It is weak, informal and ad hoc. That is not the right way to look at the application of the formula and the deciding what expenditure is what. You have the rule, the Barnett formula, the design of it. Once that is decided by Westminster, the application of it, which is in itself quite an opaque process, there is something like the Joint Exchequer Committee, which has all the constituent parts of the United Kingdom together to discuss that and see the application. I think that that could be much more wisely put there for the scrutiny. If there needed to be an independent assessor on that, then I do not see why there would be a problem with that. However, I would like to think that the constituent parts of the United Kingdom on a public committee open to public scrutiny could make a reasonable judgment on how to apply the formula. I do not think that there is an SFC issue that is too political. Professor Pete? It is a complicated one, is it not? As you have just heard from the different views, in our submission we do suggest that if the SFC is monitoring adherence to the Scottish Government's fiscal rules, then it could have a role in assisting the Scottish Government and Scottish Parliament in considering what the implications would be to those fiscal rules and, indeed, to the overall fiscal position of prospective adjustments to the Barnett formula and or further devolution of taxes. I think that, certainly, the SFC should not get engaged in the politics of what should take place, but just in analysing potential changes or proposed changes and helping to advise the Parliament in particular on what that meant could be a value, but sensitively handled. Mr Stewart, I do not know whether you have any observations. The second thing that I am picking up from you all is that there is an ideal scenario where SFC does the forecasting and there is a less than ideal one where the Scottish Government does the forecasting and SFC comment on that. How do we ensure that there are clear Chinese walls in both scenarios? That, I suspect, comes down to timing and choreography of some of the different inputs. For example, we understand that the Scottish Government would require to see the publication of the Scottish Fiscal Commission's forecasting in advance of them actually producing their own and producing a budget. There is an Audit Scotland model where they see that a couple of weeks in advance and they are able to amend some of the report by discussion and agreement. There is a suggestion that 24 hours in advance would be much more effective and ensure independence. How would you do this? How would you arrange the choreography? Should they see it before, during the budget proposals or after? If we are talking about the SFC producing a forecast, you have three scenarios. One is the SFC producing the forecast that is used by the Scottish Government. In that case, there would need to be an iteration on an informal co-operative basis. I do not think that it can legislate that. It is where the confidence of both parties is important to make this work and the respect between both parties. I would expect, as far as at least a month before the budget, that there is already informal discussions. Informal does not mean that it is not serious, but both sides are sitting down in private, saying that these are the questions that we have and these are the things that we think are difficulties. The number of meetings would take place would be governed by what the Government proposals are, what the past proposals are, what the difficulty of the forecast is. Sometimes they are easier than others. I do not think that you can legislate anything like that beforehand. To make these things work, it really requires co-operation and confidence in both sides. That has made it much easier. I think that this is an important experience from the OBR if, in advance, when you are creating those bodies, the statutory elements and so on, you give the SFC enough protection as much independence that will never have to use it. That is the point. It does not actually have to use half those powers because it has got them there if it needs them. That is why I say that it is the most important. If you then go from the first best to our second best, which is actually competing forecasts, but the SFC produces an independent forecast, again, I think that it will be useful to have a number of meetings. I do not think that one should try and prescribe what they should be. It should depend on the issues that are involved. As long as this committee has the power to make its independent judgment, I think that you can have confidence that both sides of that discussion would work in the best interest in knowing what their parameters are. I think that getting these parameters sorted out at the outset is really very, very important. But there are a number of meetings that happen. With the OBR on the Government, without explaining any secrets, there are quite a few meetings. I would agree that the model of how the OBR works with the UK Government is a very valuable precedent because it seems to have settled down very well. We have Professor Lenard's report, which again is very valuable looking at alternatives. Although there are not that many examples of an independent commission that is providing the forecast, I think that we have evidence there. I bow to Angus's experience of how this works in practice. One view that I would make on this question of timing is that I think that it would be dangerous for the Government to receive the report two weeks in advance. I think that one day sounds plenty of time to be prepared so that they know how they can react initially. But if there has been the right informal contact so that they understand the issues that matter without being given absolute numbers but understand the issues and the scope for potential differences, then just having the actual numbers a day in advance should be more than enough, and then one would be assured that the report that came from the SFC was its report, and the report that came from the Scottish Government was its report, which had not been done jointly with the SFC. I think that that is what is critical. Twenty-four hours seems plenty. Thank you. Just a couple of questions. In relation to forecasting, the Scottish Government certainly up until now has not been enthusiastic about the SFC having forecasting powers or capabilities. When the bill team gave evidence a couple of weeks ago, they changed slightly to say that they would not do anything illegally to prevent the SFC from forecasting, but it certainly did not sound like a ringing endorsement. With that in mind, do we need to have something about forecasting on the face of the bill? I open that question to any of the panellists. It has to be clear in the legislation that the commission has the right to develop and undertake its own forecasts in such a way as it sees fit as it develops its role. I think that that is important. I read the evidence from the bill team and others, and I did see movement, which was excellent to see, and I think that including that in the legislation will be very important. My view is that this would actually be helpful to achieving what I think we all agree needs to be achieved, which is that more power needs to be devolved to the Scottish Parliament. With devolving power comes this issue of moral hazard, which we have observed in just about every other country that has devolved powers to sub-central governments. I find it difficult to see how you have less independent scrutiny at sub-central government than you do at central government and make that work well. That is the invitation for moral hazard. Your threshold should be whatever independent scrutiny you have at central government. You at least match it at sub-central government, and therefore I think that it should be in the front of the bill that it has this independence for forecasting. Okay, wonderful panellists agree or wish to comment? Do you agree? Yeah, we agree. Okay. The only other issue that I wanted to explore—I do not think that it was in any of the papers, so feel free not to share a view at this stage and to reflect upon it if you wish, but, in my personal view, there is an anomaly in the current set-up and indeed in the draft bill in that business rates are not treated like other taxes at business rates. Instead of judging the forecast of business rates, the SFC are to look at some of the buoyancy estimates that underpin the forecast, which strikes me as a bit convoluted. Then council tax, which is obviously controlled directly by the Scottish Government, is controlled at a council level, which affects the overall public spending within Scotland. My personal view is that I cannot see why both of those are not explicitly included in the same way that the OBR will give a forecast on council tax and a forecast on business rates. Again, I wonder if any panellists have thought about that issue and would wish to share a view. But until you've just raised it and your position seems to be entirely reasonable, but I'm not an expert on either of these. I don't know for particular technical reasons why this wouldn't be so. I can't think of any. It seems entirely reasonable to treat them in the same way, symmetrically. I think it needs to be included. As I said earlier about LBTT and non-domestic rates, such volatility in what underpins that. I don't think that anybody completely understands this buoyancy model, or certainly I don't, so getting to the bottom of all of that is critical when it should be part of it. I have no expertise in the area, but the logic of what you say rings true to me. Peter, we have no further questions for the committee, so I've just got a couple of others to finish off with. The commission has endorsed the Scottish Government's initial forecast for residential LBTT without any behavioural analysis. Do you think that it's reasonable to produce such a forecast without looking at behavioural analysis? Do you know what to comment on that? Very briefly, I'm assuming that this is a false step, that what we're engaged in here is a developing process. I'm assuming that what's happened here is that there isn't a behavioural model available at the minute that they can readily use, and so are resorting to mechanistic methods for forecasting revenue on a very pragmatic basis. To answer the question, I think that, longer term, it would be unsatisfactory to be relying on mechanistic methods, since they might be okay on a very short term basis, but typically get it very wrong on a longer term, and maybe sensitive to policy changes, which you can't possibly capture with a mechanistic model other than judgmentally. I would have thought that, longer term, you would expect to see behavioural models being developed for important sources of taxation. Is that the general view of the panel? It's my view. I think that that's right. Property-related taxes are quite notoriously difficult because these things have long, big cycles in them. However, a degree of behavioural inclusion, to a certain extent there's mean reversion in many of these series, I think it'd be surprising if a well-thought-through forecast didn't have some sort of behavioural assumptions inside that forecast. I think that it would be easy to do it purely mechanistically. I'm not an expert, but I clearly understand property. Some of the assumptions don't have to be... We know that the whole of the market needs to function as one. That's clear. So, if there is an overly weighted banding against one part of the market and it has the effect of stopping it, that goes to affect the whole market. It's demonstrable in what has happened on the back of LBTT that some kind of understanding of how the market operates is critical. The closer that the forecasting can get to the market and can be informed by very capable agencies throughout Scotland who are directly involved in the market and who know day by day what is trading, what isn't, how things are working. I don't know how you build that connection, but it's critical. If this issue was on my desk, the first thing I'd want to examine with great care is why the Scottish Government and particularly OBR forecasts had gone so wrong. What were the assumptions that were incorrect? Were they due to failing to bring in behavioural assumptions, were the data poor and misleading? What was wrong with the forecast that led to such significant diversions from what actually occurred? I'd want to look at that in detail and then consider whether the best way of improving forecasting going forward was to insert behavioural forecasts or some other changes to the data or the models that might lead to a greater probability of being closer to actuality. That's an important point, because you're probably aware that OBR forecasts in terms of LBTT varied enormously. There were some 20 per cent changes to initial forecasts to the most recent ones that actually were just on LBTT income. Even on a bi-annual basis, they were making quite dramatic changes. Initially, as Mark pointed out, a lot of it was because of original, they just extrapolated Scottish figures from UK data, which obviously was skewed by London in particular. One of the misconceptions about the property market is that it has recovered. It hasn't recovered, it is recovering. What you will see from Aberdeen, for example, over the previous quarter or when the market was buoyant, the commercial transactions that have been a big part of the commercial property transactions in Scotland have happened in Aberdeen, that now will stop. There are fairly major shifts in the market because of its fragility that can be severely affected by subtle changes in LBTT. For a part of the market to stop operating as Aberdeen has done, that will have a consequence in most recent quarters numbers. That link to the market is critical. I don't think that LBTT is considered to be the major component in the property market in Aberdeen at this time. The key point is that, as a market, we are funded by international capital. I would say that 70 per cent of our market is not from domestic banks or domestic equity, it is international. The way in which our market operates is not consistent with the rest of the UK market because its funding for a lot of its transactions is based within its own market. The frequency with which buyers come into the commercial property market, for example, and the consequence on LBTT depends largely on which assets are for sale. I know that that sounds, it is not them investing in the market as a whole, so you can predict the market, they are investing in what asset comes up. Atria, beside the conference centre for sale, it is a good asset that will attract attention, but other secondary assets will not. Depending on a year, what comes up for sale will entirely depend on the LBTT take. One final point, which is that the commission has described its own role as one of challenge and inquiry with the aim of improving the Scottish Government's forecasting methods. What level of transparency should there be in relation to that particular work? Ideally, I think that it should be as transparent as possible, including the model. Of course, you cannot, as was mentioned earlier, release details of confidential data, but, subject to that restriction, I do not see a reason why there should not be complete transparency about this interaction. As I said, I think that it would be great if the core models, if not confidential data, obviously, were made widely available for others to use as well. I agree entirely that it should be as transparent as possible. That should include evidence to this committee and wherever else is appropriate, but it should also involve putting forecasts and comments into the public domain in an accessible manner. I am not talking about complex econometric modelling being included, or even Angus's 50 pages of spreadsheets. I am talking about documentation that can be read by the interested public and they can then form a view of the issues that matter and make their own judgments as to where it is going. Accessibility is critical, as well as transparency. Transparency in terms of the findings and the forecasts of the SFC are critical. Some elements that have been discussed are sensitive information, and some of the models are too cumbersome. Therefore, the second best to opacity, which is the first, is that whoever is charged with this has complete impartiality. We just have to accept that some things are confidential and sensitive information, that is our life. However, if you cannot have transparency and in the models that they use it to a certain extent, you need impartiality. Then impartiality becomes super important, and that is where some of the governance issues and the independence of the SFC become so important. Thank you very much for your evidence this morning. Is there any final point that you want to make to the committee before we wind up this session? Can I make one final point? Of course. There is not a discussion in a number of the papers about how the Scottish Government will be doing its borrowing. I have given evidence a number of times saying that if we want to have responsibility, this really has to be from independent parties, for example external debt. It struck me as who is going to make an assessment of the cost efficiency of some of that debt finance and that fundraising. It seems to slip through a number of the nets on the papers. I raise it as perhaps that needs to be thought about in the overall framework that you are setting out. Anything else, Henry? Thank you very much to everyone. That has been an excellent session. Thank you over at your contributions. I would just like to say to Angus that I do think that the Sasha Dystell look does suit you. I will call a brief recess to allow a change over the witnesses and that will break for members. We are all present and correct to our third item of business to take evidence for revenue Scotland as part of our draft budget scrutiny. I therefore like to welcome to the meeting Keith Nicholson, Eleanor Emerson and Robert Bucking. Members have received copies of written update from our witnesses. Before we go to questions, Keith will give us a brief opening statement. This is the first time any representative of revenue Scotland has appeared in front of the committee after our go live date of 1 April. I felt that it was appropriate that I, as chair, addressed the committee and highlighted some of the work that has brought us to this point before Eleanor takes your questions as chief executive and accountable officer. Revenue Scotland has now been collecting and managing the two devolved taxes, land and buildings transaction tax and Scottish landfill tax for more than seven months. I wanted briefly just to reflect how we got here. It has been a historic journey since the passing of the Scotland Act 2012, the three bills passed by the Scottish Parliament through to the creation and establishment of the tax administration programme and then the birth of revenue Scotland and the collection of the first devolved taxes. As you know, an important theme throughout this journey has been that of collaboration and engagement with key partners, a key aspect of the Scottish approach to taxation. The Scottish Government consulted publicly on each of the three bills and support nut legislation. It held workshops and meetings with interested parties such as the Law Society and the Chartered Institute of Taxation from a very early stage and established working groups to discuss the main policy issues. The knowledge and expertise of SEPA and ROS will carry out key functions delegated to them from Revenue Scotland, has been both vital in the establishment of Revenue Scotland and since its launch on 1 April. Continuity engagement is vital to our ongoing success and initiatives such as the regular meetings of both LBTT and SLFT forums, following on from the devolved tax collaborative meetings and a new programme of roadshows, which will be rolled out in the new year or provide stakeholders with the opportunity to feedback and contribute to the efficient management of the devolved taxes. Getting Revenue Scotland to where we are now has really been a team effort and Revenue Scotland board is now an important part of that team. We want to continue to build relationships with our key delivery partners to ensure Revenue Scotland has a successful future. The team of staff who have set up and run Revenue Scotland deserve enormous credit for getting us to where we are now and I don't want to steal any of Eleanor's thunder, but it's worth reflecting on the fact that in the first six months we collected more than 220 million in tax revenues with 97% of revenues made online. These are achievements which demonstrate the value of collaborative work done at every stage and show that we at Revenue Scotland can face any future challenges with confidence. Thank you very much for that opening statement. I'll ask initial questions as usual and then we'll go on to questions from the rest of the panel. I have a lot of initial questions, I have to say, because I think that it's an excellent, very detailed report and a lot of the questions I would wish to ask have really been answered by the report itself. But let's kick off with the first one, which is a total set of costs for Revenue Scotland, a collection of LBTT. As you say in paragraph 8, it is 5.5 million. That compares to the estimate of 6.3 million in February 2015, and it's always good when these things come in under rather than over budget. You've given it as a detailed breakdown of this on Annex C, and I'm just wondering if Eleanor can maybe talk us through that just a wee bit. Certainly. There are two main things that changed, one rather more significant than the other. When I gave you the estimate back in October of last year and then updated it again in February, I'd taken a fairly cautious view about the set-up costs that would run on into this year, knowing that there would be staff who would still have work to finish off on the set-up in this financial year, and then they would need to find other jobs. In fact, we were very efficient at finishing off the extra work, and staff found other jobs very readily, having the experience of setting up Revenue Scotland. We underspent on the staff costs just because people moved on more quickly than I had anticipated. The other element that you'll see that's slightly different is the IT procurement cost. That's really an accounting issue. It's to do with how much of the cost is capitalised and how much is spread over the lifetime of the contract. We're working on one treatment on an assumption of one treatment, but the final decision on the treatment of it is that the costs come in as part of the operating costs rather than in the upper procurement, so that's what's driven the change in the IT costs. Does that mean that the cost for this year will be higher than anticipated because of that change? We're absorbing that within the budget that we had. That's fine. On paragraph 12, you say that we remain very alert to the possibility of tax avoidance, but as yet, I've not had to use the powers available under the general anti-avoidance rule. On that particular issue, you say that you've not had to use the powers, and you're obviously alert. Where are we with that? Is everyone who's been submitting tax returns all doing everything in a hunky dory kind of way? Is the legislation proving robust? One would always suspect, because I know that you always thought that there was going to be a wee gap in terms of what was going to be collected and what perhaps should be collected, which indicates that there would be some kind of element of avoidance. Is that the gap not happened at all? Have you had some discussions about the size of that gap, if you recall, in previous committees? We can't really estimate the gap. All I can tell you is that right now, we are going after all the money that is due to be paid. Our job is to bring in the correct amount of tax, and we are using all the powers that the Parliament has given us to go after the correct amount of tax. Right now, if I were aware of any unpaid tax, I would be using the powers that we have given us to go after it. The sorts of things where people talk about tax gap, it tends to be rather a theoretical estimate and it tends to be done retrospectively based on looking at economic indicators and what that might mean about the size of various markets compared to the amount of tax paid. It's generally done for past years, not for a current or a future year, and it is notoriously difficult to do it with any accuracy. I wouldn't read anything into the fact that we haven't used the general anti-avoidance rule and that having anything to do with tax gap. We haven't used it yet because we haven't had to. We're using all the other powers that you've given us. We will use the GAR if we have to. We're at the moment entirely in the business of just bringing in all the tax that's due will come to trying to make retrospective estimates at some point, but it's really quite hard to do accurate. I don't have to use the general anti-avoidance rule, but you say in paragraph 18 that we may generate between £100,000 and £1.5 million of compliance yield in total within the financial year and £3 million to £4.5 million if we're able to retain specialist compliance posts, which were initially included in our staffing plans for one year. You've not got avoidance, but there's issues in terms of compliance. That's what the tax authority does, if you'll excuse me. We ask questions, we challenge things and returns that people have submitted. If we need to, we open formal inquiries and we've now done that in some cases. We do debt recovery work if people have declared and not paid. We do all the things that you would expect us to do to make sure that the money comes in. Just one other thing before I open up to the committee. I wonder if you can talk us through about the application of penalties that we should touch on in paragraph 17. We've given the taxes an initial period to bed in to let people get used to new arrangements and get used to the new taxes. The revenue Scotland board took the decision that we would start imposing penalties and interest for transactions from 1 October. That's in place now. Returns that are associated with those transactions have obviously been coming in. We've issued quite a large number—I think more than 300—pre-penalty letters, what we call pre-penalty letters. We ask people what was going on and why the return was late. We will get to the actual application of penalties quite soon, I suspect. Thank you very much for that. Okay, a number of colleagues want to ask questions. First, there'll be Jackie to follow by, John. Thank you very much. Rina, can I congratulate you on the work done? It is a smoother transition than I think anybody imagined, and I've no doubt that that's partly down to yourself and the staff team as a whole. I do want to focus on residential transactions for LBTT for a moment, if you wouldn't mind. My maths is shaky, but judging from the amount that was raised and the amount anticipated to be raised, even if I was generous and doubled the figure that's there, we are looking at quite a substantial shortfall of potentially over £40 million. I'm curious to know two things on the back of that. One—I don't know whether you're party to this, but there have obviously been discussions with the UK Government about forestalling. I don't know if a figure has been agreed, because there was obviously some effect of forestalling at the very start of the implementation of LBTT. Do we have a figure for that? Anecdotally, I would have thought that house sales December, January, February are pretty small. Have you done any modelling or estimates as to what's likely to be expected? We're not party to the discussions on forestalling between the Scottish Government and UK Government, so I'm afraid that I can't give you any update on that. There's something that I should perhaps draw your attention to other than the forestalling that's important when you look at our numbers and try to compare them to the forecast. Our numbers are cash flow effectively. They're tax-declared and paid within a given month. Forecasts are really based on—and accruals basis effectively—their transactions within a year. Obviously, in the first year of a new tax, that effect comes into play far more strongly than it will in subsequent years, because during the month of April, and to some extent May, we are not collecting—HMRC would still be collecting—some stamp duty land tax during April relating to transactions in March. We were collecting LBTT for transactions in April, but some of that won't have come in until May, because people have 30 days after the date of the transaction to submit the return and make their payment. There's a cash flow effect going on here as well as any other effects that people are trying to model. We have looked at seasonality. Obviously, again, brand new tax. We don't have any seasonal pattern to go on LBTT, but we have looked at stamp duty land tax, and yes, there is a seasonal pattern where you would expect more sales in the summer months, and it would drop off a bit in the winter and then pick up again, but we're not forecasting. We note that there is such a pattern. We'll see whether it comes out in the same way on LBTT or not, but it will take us two or three years to build up a really good seasonal model for LBTT, I think. Do you have an order of magnitude as to how much sales drop off by, based on your information from last year over stamp duty? I have no figure with me, I'm afraid. No, I understand, but in terms of planning for what you require to do, if there's obviously a huge number of transactions that have staffing implications, if there are far fewer, then it's not something that you need to take additional measures to account for. In fact, our workload is steadier than you might think. Because we get well over 97 per cent of our returns online, the flow of returns and the flow of payments are coming through fairly easily, and what we're doing is managing our compliance work and the reactive work when people raise queries. The work when people raise queries is largely around the highly technical, the more complex transactions, and that's driven by a whole range of factors, but they're not really driven by the volume of residential transactions that are fairly straightforward. For work planning purposes, we're not having to estimate transaction numbers in quite the way that I can understand you might think. Even though yield may go down, typically December, January, February, there's lots of other things going on that people turn their attention to. That's helpful to know. In the figures that you provided to the committee, you had the landfill tax for the first quarter. Do you have the figures now that you suggested Friday the 13th of November is when we'd get the second quarter? That was the date by which returns were due. We're just looking at what's come in now, so we publish the statistics usually a few weeks after the end of the quarter, so we don't have figures ready to publish yet. We will do our release sometime in December, and I'm obviously happy to provide a copy to the committee when it comes in. Thank you very much, convener. Thank you, John. Following on from Jackie Baillie's point, I realise that some part of the papers that we've been given are prepared by herself, and partly are prepared by committee clerks, and they may be drawing on other material that I'm not entirely sure, but when I'm comparing what we've received in for tax or anything else with an estimate or a forecast, it's good to have them on the same basis. I don't know if I'm saying this, but I'm looking at you, but it may be going to somebody else. To have a full year on an accruals basis and six months on a cash basis is kind of difficult to compare, so I think that going forward I'd like to see them on the same basis. Just for further clarification, would I be right and think that I thought that how sales would be low in the summer as well as the round of Christmas, is that right? Spring and autumn would be the high times, or is it not quite that way? I'm not the expert on this, but no, I think that we do see a bit of a summer surge, but the tax surge lags a little behind the housing market for the same reason that I was explaining that people have up to 30 days to submit the tax return, so you might see the surge run about a month behind the house sale surge or up to a month behind. The convener mentioned paragraph 8 and some of the things in there. I don't think that it was mentioned about the staff costs for 15-16 or projected to be around £227,000 higher than you estimated. What can you give us a little bit of explanation on that one? Just the difference between a theoretical estimate and the actual costs when you have a staff in place. We've had one or two extra temporary people in to help us out on a couple of things, so there's no particular one thing that I could point to that would... It's mostly one-off. It's mostly one-off. I said that staff costs are turning out to be higher than originally estimated. I don't know if, presumably in a competitive market, you have to offer what people are willing to work for. It's not really... It hasn't been difficult. It hasn't proved difficult to recruit good staff. We've got staff who've come in and we're on standard Scottish Government terms and conditions, so we're paying at the same rates as Scottish Government would. As things settle down, we can expect... We should have more access to staff costs. I mean, I was interested to, in paragraph 9, the figure mentioned of operating costs for 15, 16 or 0.76 per cent of the £498 million tax revenue forecast. That's an interesting figure. I'd be interested to know how it compares with do we know what other countries do or anything like that? The OECD published figures on this. Their latest published figures are, I think, for at least a couple of financial years ago or a couple of years ago. Again, it's all retrospective. There's quite a broad range, as you would expect, but I think I could say that broadly comparable jurisdictions might be somewhere between 0.8 per cent or 1.2 per cent, but obviously that's for tax authorities that may be doing a very different mix of taxes, and so you have to approach the comparison with a bit of care, but it looks perfectly okay internationally compared to some of the other things that we've seen. Great. That was me. Thank you. Okay, thank you very much. Mark, to be followed by Gavin. Thank you, convener. A couple of the points that I was going to raise have been mentioned. I just noted in paragraph 13 that there is mention of applications to defer payment on LBTT. I just wondered on that what would be the reasons that people would seek deferment, how much of those requests have they been received, and also if a deferment is granted, how long generally is it granted for? For example, if the consideration for a transaction was contingent on a certain event occurring, then that would be one of the reasons that we would allow deferment until that contingency occurred in the future. We would then take payment at the date that the contingency occurred, rather than the date of the transaction. Just to contextualise that example, what kind of circumstances would we be talking about there? If I were to buy a property from you and in the unlikely event of planning permission being granted on that property in 20 years' time for me to build a wind farm on it, my lawyer might advise me to put a contingency in the contract that would say in that event a significant sum would also be payable. That event is unlikely and no money is actually being paid, so although that is part of the consideration technically, we would not seek tax on that at that date. We would wait until the contingency occurred. On that basis, have there been many requests for deferment since LBTT came into effect? I think that there has been a total of 10. Of those, how many have been granted? We can perhaps, it would be better if we could do that. Yes, sure, and also in terms of the length of time of the deferment, because the example that you have given there is entirely fictional in a 20-year period, but it would be interesting to know where deferments have been granted and for how long those deferments have been granted. We will obviously have to be careful about protected taxpayer information, but what we are doing here is applying the LBTT legislation as set out or doing our level best to do so. I am very happy to provide that, but it may not always be possible to give you the exact length of time, because we do not know when the contingency will occur. I am also aware of the difficulties around, if you are dealing with very small numbers, the ability for individuals to be identified, et cetera. Obviously, if it is possible, we can receive those, but I understand there may be difficulties. Excuse me, are you okay? Do you want us to stop for a minute? In relation to the tax collected so far, Jackie Baillie has asked a couple of questions by that. If you do a really crude analysis of it, then we would appear to be halfway through the year. We would appear to have collected more than half of the target for non-residential and we would appear to have collected less than half for residential. You added an important caveat that this is done on a cash basis, not accruals, and I accept that entirely. If it was done on accruals, because I guess, since you have submitted this, we probably have some idea of what happened in October. Are you able to give some kind of idea of the magnitude of how big is the difference between cash flow and accruals? If we are looking at those figures now, would there be a couple of million added, for example, or a couple of hundred grand added? Have you got some idea of the magnitude? We honestly have not done the bit of analysis that would answer your question. We have, however, provided a lot of data to the Scottish Fiscal Commission, so I imagine that it will be looking into all of that when it reports, which I believe is due to duty very shortly. On the seasonality point, you are absolutely right with LBTT. We have only got six months, but I wonder if you were to go back five or ten years with stamp duty while it is a volatile tax and it depends a lot on the economy. Are there not some patterns that can be drawn on a kind of monthly—even in a bad year, presumably, more houses are sold in September than December, for example? Is there not some work that can be done on that so that we can have some idea of where we actually are? I agree that, if you just look at it on a cash basis, it may be that the second six months are naturally going to be better than the first six months. Has there been no analysis done on that? The issue is more that we have not done it. Revenue Scotland has not done it. We are trying not to confuse the landscape on forecasting. It probably would not help the committee or anyone else, really, if there was a Revenue Scotland forecast and a Scottish Government forecast. We are not trying to do that. We are concentrating on trying to make sure that the money comes in and we will provide the data to the Scottish Government and the Fiscal Commission who will look at the forecasting. I am sure that they have looked at STLT seasonality, as well, as a model. A question that is better put to ministers is your operational point. What has been collected and you are going out there trying to make sure that—fair enough—it is just a kind of slightly random question, but it was just intriguing me. There may be a good explanation. On table 3, which is on page 9 of your report, August 2015, total non-residential tax liabilities—15.5 million, total collected—15.8 million, so that your peers have collected more than the tax. Is there an obvious explanation for that? It is partly the cash flow thing and it is partly the legislation, as you may well remember, that allows taxpayers to amend their tax returns for up to a year afterwards. When they amend them, the liabilities and the cash payments start to get out slightly at the step. Obviously, as you will see, we have repaid the money now. I am just remembering the various discussions that we have had in the committee over the past few years as we were working up to this. Two things I remember were discussed at length. One was the new law, the new legislation, closing loopholes, particularly around some of the land transactions. It was a great discussion about simplifying what was already fairly complicated legislation. The other was, from your particular point of view, being able to answer queries, dealing with the public. A lot of that was going to be online. I remember that when we had quite a chat. I just wondered if—it is not here, but just a quick update on both of those issues—that they are streamlining as it appears to be. Is that the case? There were real efforts, as you know, to streamline the legislation. I suspect that it is simpler and, as you know, it has been aligned more to Scots law. However, to be honest, LBTT remains very technical. There are property transactions that are quite complex, just in their nature, in the real world. It is not always completely straightforward for anyone to arrive at what the correct tax treatment is for LBTT. The second half of your question was just answering. We did have quite a lot of discussion around inquiries and being able to supply relevant information or the changes that will be made to any taxpayer or property developer or so on who would be in touch with you in that regard. We have given you some statistics about the volumes of calls that we receive and correspondence that we have received in response times. The more complex queries that we look to do in a more—if they relate to a particular transaction—we look to do in a more secure way than just through plain email. However, we are answering our phones and we are answering our correspondence. In practice, most of our inquiries come from solicitors or from landfill operators or those who are acting for landfill operators. We do not get a very large amount from general members of the public because most people have a solicitor when they are dealing with their property transaction, so the solicitor does it for them. Are there any final points that you want to make before we wind up the session? I welcome the opportunity and I am grateful for you to give me the opportunity to speak to the committee. It is something that I felt it was important to do since it is the first time we have been in front of you since 1 April. Thank you very much and I am glad that everything is going as Jackie said very smoothly. Congratulations on the excellent work that you are doing. I will call a one-two-minute recess at the end of our public session in order for our witnesses, official report and the public to leave.