 Good morning and welcome to the 26th meeting in 2014 of the finance committee of the Scottish Parliament. Could I please remind everyone present to turn off any mobile phones or other electronic devices? Our first item of business this morning is to decide well to take items 4, 5, 6 and 7 in private. Are members agreed? Members have indicated their agreement. Our second item of business this morning is to take evidence from the chair and members of the Scottish Fiscal Commission in relation to its report on forecasts for the devolved taxes and for non-domestic rates. Therefore, I welcome to the meeting Lady Susan Rice CBE, Professor Campbell Leith and Professor Andrew Hughes Hallott. Committee members have copies of the commission's report, so we will go straight to questions from the committee. As is custom on the committee, I will ask them questions first. Members can come in subsequently if there is any questions left to ask. I will try not to steal all the good ones at the start. The commission describes its approach to consideration of the Government's forecast as one of, I quote, inquiry and challenge, followed by response, followed by further inquiry and suggested improvements. I am just wondering whether you intend to publish details of this inquiry and challenges recommended by the committee. Let me respond on the first instance, and I will turn to my colleagues. Are you happy if the three of us have a conversation with you? Yes, I should have made that clear at the start. I should have said that either one or more people can comment that it is not all directed at your self-race. You can answer that, and if your colleagues wish to answer that, that is great. It may be that you do not wish to answer at all certain questions, and other colleagues may be more appropriate to refer them to the professors on either side of you. Absolutely. Are you used to answering that? That is very helpful, but I will start off in response to the question. Those were my words. We three drafted our report, but we had different bits and different sections and then made it one voice, we hope. I was simply trying to reflect on the nature of how we did things. With some other physical bodies in other countries and other places, they actually look at forecasts after the fact, for instance. Our job is to look before the fact. It was really to point out that the nature of what we are doing, we started once the forecasters were ready in the latter part of the summer, having a presentation from them on their approach to forecasting their models and so forth. Inquiry and challenge was a matter of asking a lot of questions, saying, what do you mean by this? Have you considered this? Have you considered that? Out of all of that kind of two-ing and fro-ing, we ultimately drew some of our conclusions. For instance, one of the conclusions that is very clear in the report is that new taxes and not the decades of historic data and that's a challenge and data is an issue. Out of the inquiry and the challenge came some of our conclusions. In a summary sense, the nature of that was really encompassed in the report. If you're asking, would we publish something more specific, if you tell us what you'd like, we can do it. If you'd like, for instance, a summary of meetings or conversations we had with the economic forecasters, we can prepare something of that sort if you'd like that. We felt that we had reported overall on the nature of how we worked, which was to challenge beforehand. The results from all those conversations were the results that we put in our report. Did you forecast change the result of the inquiry and the challenge? We're not forecasting. No, but your assessment of the forecast, I should say. Quite clearly, there are two different forecasts. There's one that the Scottish Government has made, for example, on new taxes, etc. There's one that the OBR has made, and there's, as you know, a £51 million differential in that. Is it to say whether your assessment of those forecasts changed? That's the best way. I will ask my two colleagues, because they each took a tax primarily, although we shared all of that, which was very much a committee effort at the end of the day. However, our understanding of what was happening changed absolutely as we had the conversations. Some of the economists went back and came back to us to answer more questions, to give us more information as we dealt into it. However, in terms of our assessment, we started by finding out. We didn't start by saying, this is good or bad. We started by saying, help us, understand what you've done. You can't start in any other place. I would maybe not answer your question in exactly the words you've posed it, but let me turn to either of my colleagues there. The nature of inquiry took several forms. Not only were there written submissions from the Government to us and queries written in response, we also had meetings, which essentially took the form of an academic seminar. People were asked points of clarification during presentation of their forecasting methods, and then there was a long series of questions afterwards. We also received the spreadsheets containing the models from the Scottish Government. In a sense, that's all part of the inquiry process, even though we weren't specifically asking Scottish Government economists at that point, we were just looking through the models ourselves. There's a wide range of activities taking place in terms of the inquiries that went on. The initial part of that was not so much reviewing forecasts, because they weren't finalised at that point, but it was exploring methods. The large chunk of the initial part of our inquiry is looking at, how do you propose to forecast, in a general sense, what underlying methods, what techniques, what models have you built to enable you to do that? The bulk of the inquiry was focused on that. Relatively late in the process, the forecasts are finalised, and then we look at those in light of the discussion of our methods that have taken place. How things developed changed considerably during that process. Your view of how it was all came together, I wonder if you can give us a bit more detail on that, Lady Rice. I think that the quickness was that there wasn't a big window of opportunity for any of us to come together, understand that and produce our reports. It was a very intense period, but it wasn't over many months. Obviously, we were only formed in July when we had our letters of appointment. Help me when you ask that. I'm trying to remember the specific words that you used. I'm trying to find the depth of my short-term memory. You were basically saying that what you looked at at the beginning of the inquiry is not really the way you explored it, but it changed a bit and evolved over the period. Sorry, so let me make that clear if I haven't done so. We started out by needing to understand from the forecasters what models they were using, how they were approaching the forecast, just how they were coming at it to begin with. We couldn't really make a judgment about how reasonable their forecast would be once we saw them unless we understood their approach to it. That was the early part of our engagement. Then, as Campbell has just said, laterally, when they had more or less concluded their forecast, we then looked at those and needed, obviously, to make the judgment about whether they were reasonable or not. You'll have seen our conclusions in the report. In one area, we pointed out that the forecast was within the bounds of reasonableness, if you will, but quite optimistic, and that was responded to. That was the outcome of the second phase when we applied what we knew about their models and their approach to their forecast. Is that clear? Sorry, if I'm not making it clear. I think that it's me that's not so clear rather than yourself, but I'll move on a bit. Throughout your report, you've talked about data, and obviously there's a frustration. The committee shared this frustration for a long period of time about the kind of sources and the quality of data that's available, specifically to look at Scottish figures being able to drill down. I'm just wondering whether all the information that you considered necessary for your report was made available in good time by the Scottish Government? The point about data is a very important one, and I referred to it just a minute ago when I answered your first question. It is that these are new taxes or taxes with a new shape. There is not a long history of data that could feed into the models that the forecasters were using. My two colleagues will tell you that's what economists want. They want a long series of data, because that means that they are more comfortable with the forecast that they come out with. Given that fact, we and the economists had a lot of conversation about what data they had, what the sources were, was there anything better, were people cooperating with them when they went to acquire data? Was this the best that they had? Perhaps two other comments, and I might turn to maybe Andrew to add a little more on this. One comment is that I've also prompted a series of meetings with some of the agencies that provide data, and I've started providing data, to the Government economists. Those conversations are ongoing. For instance, SEPA, having a further meeting coming up, having had a couple already, Revenue Scotland registers of Scotland, and so on. It's for us to understand how willing and happy and responsive they have been. We make our own judgment about that, but certainly there has been absolute willingness that we could judge on all sides in terms of the sources of data. What you have from the work that the OBR has done in the past, and you can understand why this would be the case, they have, in some instances, gathered data and then made a proportional cut and applied that to Scotland. That is not the same as having Scottish-only data, and that's where we all want to be. It could not be the case in this first round, so I just want to make that point. The other point is that we had amongst ourselves a great deal of debate, some of it well into the night sometimes, about what do you do then as a forecaster and as an economic modeler when you don't have a perfect data set, because actually life is rarely perfect, and you're often in situations where you don't have and what are the best approaches and ways to deal with that. Andrew, do you want to say anything about that? I could do. Just going back to the beginning of that question, and the reason why you want long data series, if you can get it, is that you're trying to look for some regularity in the data, and if you've only got a short period, you can't tell whether it's just noise variation or whether there's something regular in there, so it's important to go back as far as you possibly can. It takes a while to discover sometimes whether the data does go back in time and you didn't know or it's been hidden somewhere. For example, we've discovered that the housing data from Register Scotland potentially goes back to 1617. It goes a long way back, but it's very patchy. It's apparently pretty good back to 2003, and then you come up and say, well, how good is that? How much regularity can you get out of that? In other cases, it's obvious that you want some, again in housing, you want some financial data or something of this kind, maybe of a sort that doesn't exist at all, or as Susan just said, it may exist but only at the UK level, and the problem with that is that you take out a Scottish cut that may be the best you can do up until now until you collect some of your own or find some other way of doing it. Of course, it's going to reflect the rest of UK conditions rather than Scottish conditions, so in housing, numbers may get blown apart because of the London effect. You're left saying, well, is that cut good enough or is there some other way we can make an assumption which we believe takes a London effect out, but we wouldn't actually be sure until we can look in a post-mortem sense to see how well it did. So it's a very long way of saying inevitably there's compromise when you do this. You know how you would like to do it ideally, then you discover you can't do it ideally, and the question is, how far are you prepared to compromise? I think that the methods that the Scottish Government has used up until now are reasonable. That's not to say that they couldn't be improved. I think that our job for the next year coming is to see what we can do to get them to improve, which is going to be, excuse the phrase, but a bit touchy-feely as to how you can sense what we go forward and whether it's reliable. I mentioned earlier that the difference in the forecast between the OBR and the Scottish Government is about the order of £51,558 million, as opposed to £609 million. I'm just wondering which of those is the most accurate. I can try to answer that one. Essentially there's a kind of continuum of ways of forecasting, and the Scottish Government and the OBR are really at opposite extremes of that continuum. The OBR has a large macroeconometric model with hundreds of equations describing all aspects of the economy, and that's largely because they have to forecast all macro variables plus the whole range of fiscal variables that they're interested in. That's the bulk of their work that's focused on that. In forecasting the devolved taxis, they then tend to look at, given the limited data that there is, a portion of some kind of share of UK projections to Scotland, which may be even just a historical average, but it may have some slight drift in it. The Scottish Government, on the other hand, looked at each tax in a case-by-case basis, and they've used what data they have for that tax, and they've built small, simple, statistical, sometimes or sometimes simply extrapolation based on historical data to project those variables forward. It's a completely different approach. It's not obvious which one is going to produce the best forecast. There's harsh races between those forecasting techniques that are conducted in the literature. It's not automatic that having a big hundred equation model is going to dominate a simple statistical analysis, so no one approach is better than the other. Essentially, it's fundamentally different ways of doing it. In terms of specific taxis, the land and buildings transactions tax are forecast to be slightly higher under the OBR's forecast than the Scottish Government's, but the OBR is using UK-wide projections, which has a more buoyant housing market than Scotland does, and they're shading those down a little bit for Scotland's share, but not to the extent that it appears that the Scottish Government is doing implicitly by building up from Scottish data. Is there some kind of fundamental differences in the way that they're doing it? Yeah, I know what's fundamental. You can answer almost how long is a piece of string, to be honest. One of the things that we're quite keen to know is which forecast you think would be more accurate. There's a difference of some 15 per cent in terms of residential transactions for the OBR. There's a 9 per cent difference in landfill tax, so those are quite substantial proportionate differences now. Those are fairly small taxes, but if we go forward in terms of other devolved taxes, such differences will be very significant. That's why we're quite keen to get the most accurate forecasting and to get your assessment of which forecast would be able to deliver the most accurate prediction of where we are. In order to do that, you need to attach what are called standard arrows to those forecasts, which neither forecaster currently does. It's very difficult to do with large-scale macro-econometric models. The Scottish Government is insufficient data to build statistical models that can assess the statistical accuracy of their forecasts either. However, if you look at something like the Bank of England's forecasts, where they do have sufficient data and a small-scale model, they can attach errors to their forecasts, which is how they produce their fan charts, which forecast inflation and the upper and lower bounds for that. My conjecture would be that if you drew similar fan charts for these, doing proper statistical analysis, the fan charts would get wired very quickly and would encompass both sets of forecasts. They're not statistically different. They look largely different, but statistically they're not different. In one instance, where there really isn't a lot in terms of bulk, a lot of Scottish data, the Scottish economists took a three-year average, which is a very sensible approach, so they were quite conscious of the limitations of the restricted data that they had. I just want to touch on what I think, because I want colleagues around the table to have an opportunity as well. That's on the issue of non-domestic rates. You talk about buoyancy. You suggest that they are on the optimistic side. Indeed, Mr Swinney has reduced his forecast by £83.5 million, and you also say that in noting that NDRI revenues are five to six times larger than land bills and transaction tax, the commission states that it would pay to make them as reliable as possible, as quickly as possible. To my mind, I'm not really sure why you're suggesting that, because what Mr Swinney said on 10 October, when he presented his draft budget, was that since 2008, some £13.1 billion has been collected in NDRI, and yet the variance has only been £40 million cumulative, which is 0.3 per cent. Surely that shows that you're saying, for example, Professor Leith, that we're in the margins of error when we're talking about a difference between the OBR and the Scottish Government of some 15 per cent and some LBTT predictions, and yet on NDRI, we're talking about only a 0.3 per cent variance in delivery. Why, as a group, are you so concerned about this particular issue? I can respond. It's essentially that the taxis are taxis on different types of activity, so the land and buildings transactions tax and the landfill tax are taxis on transactions, whereas the non-domestic rates are essentially a tax on the stock of rateable floor space. The thing that's essentially changing in the forecast are the changes in rateable value floor space, which is new business premises coming online and so on. That's why the forecast is essentially that the forecasters are having to forecast the change in domestic rates rather than the level of domestic rates. Perhaps our report is slightly misleading in suggesting that there's volatility in the aggregate level of that. Essentially, we should have compared the change in domestic rates tax revenues to the tax revenues from the other taxis in terms of looking at accuracy of forecasts. It may be 0.3 per cent, but it's quite a lot of pounds. It's a much larger tax, and that's why we're concerned about it. I'm picking up behind the times here slightly, but you said that it was about £40 billion. £40 million out of a £13.1 billion sum over six years now. At that time of a recession and all the rest of it and yet the accuracy of the figures, it's quite remarkable to be able to predict such figures within 0.3 per cent. That's over a number of years. I don't have the individual figures through the individual years, but they may cancel out. There might be quite a lot of variability year by year, and we're forecasting for one particular budget. That's why you can also get lucky. In this case, perhaps I'm lucky. I'm not sure which one it was, but that's why we're concerned with it. It's potentially quite a large number of pounds. Is there any year when it's been particularly way out of the predictions when Mr Swinney has predicted it being significantly higher than in actual fact the sum delivered, because I'm not aware that that has been the case? No, I don't know the numbers. We don't have those numbers. We don't have those numbers. But that will bear looking at and you're pointing us to some as we work on our work plan for the coming 12 months. That's a point that we'll definitely pick up. We certainly need to go back into, as I said during the post-mortem on all three taxes in fact, to see how well they performed. This took quite a long time when we had to do it in a month. So if you wanted to come and join us at midnight. No, I think we all appreciate what you're doing. We do appreciate that. I mean, you're putting your finger on the right thing. We need to look back and say how well they performed. That phrase that you picked out is mine, and I'm concerned with the size of the potential impact on the budget, and it is a lot larger, potentially. Well, thank you for that. I'm going to let your colleagues come in, and the first person to ask a question is going to be Jamie to be followed by John. Thank you very much for the work that you've done. As a commission, you have been able to endorse as reasonable a forecast made by the Scottish Government, and I basically just want to return just very briefly to the issue that the conveners explored in relation to the difference between the forecast of the Scottish Government and the forecast of the OBR. The conveners explored it quite thoroughly, but just for absolute clarity, and just to go back to the points that Professor Leith was making about the difference in terms of forecast methodology, is the difference explained simply by the fact that the OBR comes to a UK assessment and a proportionate share of that to Scotland? Largely, yes. That's the fundamental difference across the two approaches to forecasting. There are other slight differences in using a three-year moving average as the base year for forecasting non-residential transactions, whereas the OBR doesn't do that, so there's other subtle differences. The landfill tax is based on DEFRA data or a DEFRA model for the OBR, but it's built up from SIPA data for Scotland, so there's other small differences as well, but fundamentally the difference is that the OBR uses a big macro-econometric model and then scales a portion of that for Scotland. The Scottish Government economists start with a small set of data relating just to those taxes and forecast, roll that forward to forecast those taxes. When they're a proportionate share or they're just doing it in terms of Scotland equates to about just under 9 per cent of the UK population? No, they have some historical data for the amount of tax revenues raised in Scotland for that particular tax relative to the rest of the UK, so they use those historical averages. In recent forecasts they've been tweaking them slightly. For example, the UK housing market's more buoyant, so they've reduced the share of stamp duty land tax apportion to Scotland based on the fact that the Scottish housing market has not been as buoyant as the rest of the UK, given the London effects. Although they still think that it'll be more buoyant than the Scottish Government, it does, apparently. Yep, so they haven't scaled it back. Just turning again the issue that the convener touched on as well, because it does come through as a major theme of your report, the frustration about not having access to data. You've set out that this is something that should improve over time, but how long will this period of time be? That is how long is the piece of string, absolutely, that's the question. Obviously, 30 years from now there will be a series of data that covers 30 years, so that's how long, maybe none of us would be sitting around this table at that point. The real issue is to go in and do a kind of drains up just to see whether there is anything more that could have been done for this round to get more data, different data, get it differently, or whatever. Then it is to say how does one accommodate in a forecasting sense to the fact that we have not the ideal set of data and there are ways to do that. Forecasting isn't, well, you'll all know this. You cannot predict the future, you just make your best shot at with what you have, coming out with a picture of what the future might look like. The data will improve in a specific way. I've had some conversations with SIPA, it looks to me, and I'm going out to test this on the ground, as it were, in a couple of weeks, but they will be much closer to the landfill sites than obviously the OBR would ever have been able to be. In terms of the relationship of tax applied to what's going in, we believe that they will have very accurate and truly on-the-ground ability to reflect what is the real case. This isn't to fault one side or the other side about these differences, it's just the way it has been. I think that the data will improve over time, but it will be a gradual improvement. We won't get the 30 years until 30 years are up. Maybe this is a question that is unfair to ask of you and maybe you won't want to answer it so you don't feel free to dodge it. Feel free to answer it if you want. You've not heard the question yet, and it's not that difficult a one, but something that has been suggested previously, so there should be some sort of dedicated statistical agency for Scotland looking at data, not just in the area that you're looking at, but on a much wider basis. Is that something that you think would be useful for you as a commission that would be also useful for other public bodies, for the public at large, for us as Palantarians? Forecasters need data, so it doesn't matter where it comes from. If a new statistical agency would generate that data, then it would be a good thing. You have something like registers of Scotland in their job is to hold and create data and analyse that data and so forth, so you'd have to at least ask the question. I am avoiding giving you an answer yes or no, but you'd have to ask the question, would there be an overlap with what is already done and is what is already done sufficient? Is it good enough? Or would an agency be helpful in that regard? What we really need is that over time the new taxes generate revenues, we'll have the data for that, so new data will be generated, but there must exist, there were transactions in the Scottish housing market before, there was landfill being sent to landfill sites and taxes being paid before. Those transactions presumably are recorded somewhere that may not be easily accessible, but if that data still exists in some raw form within existing bodies then it would be good if those bodies could be encouraged to dig that data out and create it in a usable form. In my experience of this sort of thing, there's often more data there than you think, but of course it's not until you get into one of these forecasting exercises and this is for the first time it's being done that you know what questions you want to ask. I now know I want that kind of data as opposed to something else, so there'll be improvements in that sense and there'll be improvements when you go over it two or three times, so this is not 30 years, this is three years down the line. When you refine the ad hocory, which is obviously in there, it was some of the variables. All of those things will happen quicker than other things, they went with the perfect answer. On the statistical agency, I would have thought that it would be useful in one sense at least, whether you want to pay the money for whole agencies is another question, but it would be useful in one sense because it would allow the focus to be on the Scottish data and to the extent that this is being derived from somewhere else, which might have been UK data and you're trying to carve it up and that sort of thing. Giving you an explicit focus like that would be helpful. One last question, convener, on your report. You welcome feedback from any reader to the sport, and I turn my mind back. I think it was your point, Professor Layton. Maybe I've got the term already, but you talked about the whole idea of peer review, I think it was, and I'm just wondering if you've had any such feedback. We have done an exercise to look at other fiscal commissions or similar bodies in other countries, and they are all slightly different. There isn't anyone that is absolutely like ours, but to see what they've done in terms of reviews. The OBR is the one closest to our worlds, has had a review done by a Canadian with experience, and I've read that, I think we've all had a look at that. That review was done five years in, so that was a peer review, but obviously it gave the OBR the chance to establish itself, work out its methods, have a little bit of opportunity to see did they get it right or wrong, did they change, did they adapt, and so forth. We agree that it's absolutely appropriate to have a peer review. We've been two to three months in operation, so it wouldn't be happening in the next two or three months, I would guess, having said that. When the report was published, we didn't send it out widely. We put it on our website and put out a little press notice so anyone who was interested could find it, but I specifically sent it to the OBR just to get a review there. We will, over time, as we continue those conversations with the other bodies and agencies with whom we have started to have a relationship, actually request some feedback and some input from them over time. It's good to get that sooner, but a formal peer review with capital letters I think would be some period down the line. I don't know exactly when. I get maybe that the formal wasn't necessarily suggesting that you'd have a formal peer review, it was just maybe what on and you have. You've sent it on to the OBR and the rest, so you've sent the answer to the question. People are looking at this and you are getting some feedback then. Yes, yes. We've had meetings very recently with Revenue Scotland and with registers of Scotland. In those conversations, there was some feedback as well. We would welcome that. I mean, this is, as I said in it, it's a first for all of us. The more that people come into the conversation and tell us what they think and what works and what's clear and what's not, the better we'll do it next time. Thank you, John, to be followed by Michael. Thanks, convener. There's obviously been quite a lot of questioning, and I suspect there will be on the amount of data and availability of data and those kind of things, so continuing with that, your comments under LBTT, the non-residential model and your recommendations, you talk about creating a new Scotland-specific dataset may be the only reliable strategy, and then you say, which is the bit that worries me, that is neither a short-term nor cheap undertaking. My question is, how do you decide, there must be a time when more data doesn't actually get you a better forecast? So spending money on getting more data would be a waste in that case. How do you decide how much data you need and the points being made that some data may be there is just not available? I mean, how do we balance that up? I guess there are lots of ways to answer the question. Part of the answer, and then I'll turn to my expert colleagues for more of it, but part of it is the quality of the data you have. So it's partly the amount of data because models require volume, and that's very helpful, as Andrew has explained, but some of it is actually the quality of the data. How good, how specific is it? There we do see a focus really beginning to happen and has started already in Scotland to get Scottish-specific data. That's really important. That means that the data we do have, even if it's limited in terms of bulk and volume, is perhaps better data and that's important. In terms of your direct question, how much is enough and where does the cost benefit come in? I don't know if either Andrew Campbell wants to say something, Andrew. Sorry, a few things, I suppose. You're right to pick on that particular element, because I think that that was one of the weakest parts of the forecast being made at the moment. The question is, if we do this properly, we would go back and say what kind of model do we want, how would we in principle try and forecast this, and then see what data that would demand, and then discover no doubt that it didn't exist or we have to dig. I'm not sure that we're going to do it, but we can tell people what to expect. So it's a bit difficult to answer specifically ex-ante, because we haven't done that exercise and the Scottish Government, people haven't done that exercise. At that point you discover, yes, we do desperately need Scottish-specific data, so we need to set up a project to do that, which we might commission. That's how I would think about going about it. Starting at the model end, rather than saying of all the data that you can imagine that affects the non-residential sector, what do we have? The convener made the point earlier that the forecasts on non-domestic rates income had been quite accurate over six years. Professor Hallott made the point that there might be a lot of fluctuations within that. Is that important? There's clearly a borrowing limit, and therefore if the fluctuations were too great, it would affect the borrowing limit, but in one sense we are more interested in that there should be a balanced budget or whatever over six years or whatever. Is the one year so important? We are interested in a balanced budget over time. I'm not sure that the Scottish Government is in a position to be interested in that, because they have to balance as best they can every year, because they can't borrow effectively. The amount of borrowing that is allowed under current arrangements on current spending is scarcely visible to the naked eye, so half a per cent of GDP is tiny. They are constrained. That's why I said that. If they were to go to a model in which they were asked to, as a convention is, to try and balance over the cycle and forgetting the last financial crisis, which is a bit extreme, it would be a whole lot easier and we would be interested in the performance exactly the way you say. I need to go back and check how much fluctuation there was year by year. That's the general point, Mr Alonso. My other question is specifically on the landfill tax. One or two points jumped out at me. One was that you noted that SEPA is reporting data with a two-year time lag. That sounded a bit concerning. Can you comment on that? We just came out of our exploring the sources of data. SEPA are, as you'll know, they have been changing the way they're gathering the data on landfill. They have actually something that I think we all agree, I certainly think, is very good because they'll be looking, unlike the OBR, which just looks at landfill as reported and then says, you know, the tax is such and such, SEPA will be looking at illegal landfill, you know, where people have broken the rules and there will be, in essence, as I understand it, a bigger penalty, a bigger tax for doing that. So they're in the process of changing their approach. They are and have been developing an approach where they actually will go out and view and visit sites. I mean, all of this will lead to some difference. So where they have been, that was a comment about what has been, but going forward, this is already being refined and will continue to be refined. The data will come out more quickly in future. Yeah, I can't tell you about the time lag, but this isn't, actually, it's a very important, more general question, which is the timing of availability of data. So that is specifically with regard to SEPA and they, I believe, are thinking about the timing. But you also have data held by the registers of Scotland and they have a quarterly reporting pattern. And does that pattern suit, you know, in other words, what are the times when we all need data and what are the times when we need data? So we're in starting conversations, in the middle of conversations, to say, this is when it would be best to have those data, this is the timetable. I mean, is there a general problem that the data is there, it's just coming out too late? Yeah, there is or there could be and that's why we want to basically have a discussion with all of the data providers and come up with a view that says, this is when the data would be most useful and most effective in terms of future forecasts. So yeah, that's an issue. That's not to say that there is a problem that led us to say that any of the forecasts were totally unreasonable, because we haven't said that, as you know. But there's an issue here that we'd like to get under and perhaps influence some change going forward. So I'm sure there's a committee that would be interested to know if some, you know, if you're getting data quickly from one area and not from another, then we'd be happy to, you know, make comments on that, I'm sure. Could I say that this is the data as it is normally reported from these areas? And I hear what you're saying and thank you very much. Leave it with us to have these conversations about the regularity of data and the periodicity of the data that comes out. And if we do need some muscle, thank you for the offer. Thank you. I mean, another specific figure was the 16 per cent, where it appears to me, if I'm reading this correctly, that the actual landfill waste going in should be raising a certain amount of tax and it's 16 per cent less that's arriving at HMRC, which also concerns me a wee bit. Are we clear what that means? That's our understanding of it as well. There's stuff being dumped and the tax level should be higher, but the HMRC data is in conflict with that. Because that then makes forecasting almost impossible. Sorry, I didn't interrupt. No, I understand where you're coming from. There is a certain amount of waste tourism, which is the jargon term, and that is those purveyors of waste sometimes cross borders and it becomes slightly complicated for the OBR because they deal with firms, if you will, those who create and put the waste to landfill. If those firms are crossing the border, it's a little harder to distinguish, so that probably accounts for some of the difference. But as I say, I have some confidence anyway that the changes that SEPA are bringing in will be much more specific and more accurately reflect what's happening on the ground in Scotland. That's the place that we want to get to. Okay, and my final point, still on landfill tax, I think that the point has been made, if I'm correct, that the environmental folk and within government are really keen to push down the amount of landfill, so that's quite an aggressive target. Then the finance John Swinney and so on are basically reflecting that in the budget, so the two are exactly the same. That gives for consistent government, I would accept. I suppose that my question is, should the two be exactly the same, or should the environmental folk be aggressive and the finance folk should be cautious? We've had some conversations, that's a very good question as well. Aggressive and cautious, we spoke, Andrew, you were the one who prompted some of the early discussion about the fact that the model took a straight line approach, that by next year we should have only this amount going to landfill, and here's where we are now, and there has been some progress because of the escalation of the taxes that has had an impact, behavioural impact, and therefore what else do you do but show a straight line, and that's really what you're reflecting. We had a little bit of discussion, so this gets back to the question about challenge and inquiry, which is really discussion, so we had a bit of discussion, what if the tax was raised significantly, would that have a different impact? Would firms do something else with their waste? It might go to landfill, it might not, it might go down south, it might go somewhere else, and so we explored some of those issues. But in terms of is that a good way or not to do it, it's an acceptable way to do it. The only thing I would add to that is we need to do some backtracking on that one too, because if it's a straight line or something simple like that, you ought to be able to look at it and say, well, did they keep to it in the past, or are they slipping? I think the word slippage appears somewhere, and at the moment we don't really know. The difficulty is, of course, we don't know that those really are the targets. We know what the end target is in 25 years' time or something, so we know here and we know here, but, of course, in principle, it might do anything in between, and we don't really know. But if we put a bit of pressure on the saying, well, tell us, are you keeping up with this, then eventually it will come clearer. It's also, when Susan's right about the difficulties in knowing exactly what you're talking about, because OBR data is somewhat different from SIPA data. OBR, sorry, SIPA data is, as it were, more Scottish, and therefore presumably better, so we hope we're in the right place. There's also a twist, which I'm not sure we've, at least I don't fully understand. There's a difference in the mix of what's being dumped in the ground and the tax rates differ on that, and we don't have specific data on that sort of thing. Numerically, it probably doesn't make a huge difference, but it will be a little bit more in where the differences are coming. If I was out of private opinion, I would reckon that the SIPA approach is more appropriate for those forecasts, but I can't prove it yet. We may be on to later. It's very much on our agenda and has been discussed. What happens next from your point of view? You've obviously produced this report to sit alongside the draft budget. In terms of your work streamers, what happens next? Well, a number of things. We've already had a session ourselves as a bit of a drains up, I would call it, in terms of what we did. It really was a very intensive effort to get this out. My understanding from talking to the fiscal unit within the Government is that they are shortly to have their own drains up about the process. I think it was intensive for them as well. We then have agreed that we'll come together and just discuss this and sort of see what worked, what didn't work, how do we work together differently more effectively if we need to. I think our one primary assumption is we want to start much earlier for next year's budget because we'll have the ability to do that. We couldn't do that this year. That's one thing we're doing. We are continuing to create and develop these what I call relationships with relevant organisations and bodies. I've mentioned some of the Scottish agencies. We'll be going down. We had spoken to the chairman of the OBR at the beginning of our process in August, but we've been in touch. We will be talking to him more fully. We've had a conversation with the parliaments and I'll get the name of it wrong budget. It's Simon Wakefield's unit that looks at the budget, I think, the expenditure side of the budget. To make sure that we know what people are doing, to see if anyone has ways to help us or feed into our work, we've been in touch with a network called IPFIN, IPFIN, which is the UK nations of fiscal and budgetary bodies. They have a get-together for a day in November, and I believe I'm going to go to that. They invited me and said, well, let's go meet, talk and find out. We've been in touch with the OECD. They have a get-together in the spring, and again, they've invited us. We think that that will be very useful because they have fiscal commissions from a number of different countries and we can learn from them. Your own findings, I think, your February report from this year about, as you were doing, your own consultation on whether to set up a fiscal commission, you had taken some evidence, I believe, from the Swedish commission and from the Irish commission. We've been in touch with both of those. In fact, Andrew has been out and spoken to the Irish commission and we have some paperwork from the Swedish commission. Part of our work is to find out more about how others do it, how we can improve what we're doing, so it's building out these relationships. It's reaching more widely around these networks. It's understanding absolutely what more do we need and do we want and having the right timetable to go into next year's budget round. Then picking up on some of these questions that you're raising with us this morning, some of which come out of our report. That is the core of what we need to do, is to explore these questions and find some answers. How much data is enough data? What do we do when we don't have it? I mean, these are important questions for the future, so we will set out to do all of that. The other part of our work programme, and this may seem more on the side, but actually it's important, is, as I think you know, we are being hosted at Glasgow University. We've been very cooperative and are in the process of giving us a little office and process of helping us find a couple of research assistants, so we need to continue that process until we are actually functioning there. We have a website, which you may have noticed doesn't have very much on it, but we cut the report out at the right moment in time, which to me was what really mattered. I think our three names are on it, so we want to do a little bit more with the website. Thank you. Any suggestions from anyone about what you'd like us to do, please feed those in. A couple of members have asked about the differences between projections from the OBR and projections for the Scottish Government for devolved taxes, and some explanations or some theories have been put forward for the differences. Just looking at the tables in front of me, one of the OBRs when it comes from March of this year—and the Scottish Government presumably is from October—might be September—I guess September, October this year—and the OBR one is in relation to stamp duty land tax, and the Scottish Government one is specifically in relation to land and buildings transactions tax. Could either of those factors go some way to explaining some of the difference? A simple answer is absolutely yes. I'm not sure the Scottish data will not all be at one point in time, whether it's August 1 or September 1, or whatever. They will have done their best to get the best and most up-to-date data that they could. There will be some timing difference. Of course, that makes a difference, and it is a different tax in terms of what it will bring in. The next question is for the bottom of page 4 of your report, when you're talking about the residential model for LBTT. Right at the bottom of that report, Roman numerals 3, relatively high tax rates applicable to the upper band of the new LBTT may also induce an additional behavioural response, which has not been factored into the forecast. Obviously, behavioural response is something that we've talked about as a committee, and we've put others on. In terms of the behavioural response to that tax and to landfill tax, was there any sort of work being done on that that you were able to see, or are you basically saying that there's none being done, but we think that there ought to be some done going forward? The formal techniques that the Scottish Government is using don't include any behavioural responses, so it's not that kind of modelling. I think that in terms of the housing market side of things, they have looked at various models that are more behavioural or more structural that would include those kind of effects, but they haven't been particularly successful in managing to find workable versions of those. Those are typically very difficult models to successfully develop, so at the moment they don't have such models, we're encouraging them to look into the possibility of developing them. Add to that, I think that one of the reasons why this is not at the top of the agenda, it wasn't in that brief period, is because the amount of transactions, the value, I should say, the transactions at the top part of the tell is actually rather small, so they focused on other things. First, the only other thing I would say is, presumably if there's a behavioural response, one would imagine it was going to be driven by the difference in the tax rates in the top band and the next band down, which is relatively small. The one below that is rather larger, so I mean your question is the right one, but it might be actually a different bit in the distribution. It doesn't help you very much, but… Next question. You made the point that we want to see enhanced forecasting methods, and one of the main areas is data that has been talked about, so I want to ask about data, but other than data that will obviously take a degree of time, are there any other specific obvious areas where you think that the current forecasting methods could be improved for next year? Obviously, the data one that we've discussed and the behavioural aspects, if there are ways to bring in any of that, that would be very helpful, and that would apply to each of the taxes. There are different behavioural responses potentially to each of them, but I will again turn to my colleagues for fuller answer. In the non-domestic rates, we're aware, without knowing exactly how far it's gone, we're aware that the Scottish Government people are working on trying to get, some behavioural responses into that part of it, so we'll be watching that, and I think that perhaps we're going to sit down and actually go through it with them step by step before we come to the next report to see what we can say. The same should apply elsewhere, but I think probably it's not as advanced at the moment. If we move on to non-domestic rates, your view as a group was that the buoyancy increase for business rates growth, in your opinion, was on the optimistic side. The Government, I think, has obviously reduced slightly its forecast on the back of that. Can you just talk through what actually happened there? Did you say that we think that it's optimistic by x per cent? Did you simply use the term optimistic side, and they then did it? Just be useful just to know how that came about. We looked at their approach to forecasting, which is essentially, I think, it was eight years of data they have, and they have the historical average for growth and buoyancy over that period, and then they have a range of macroeconomic and microeconomic indicators, which they use to try to suggest whether the economy is going to be growing faster or slower than in previous years. They looked at those indicators and took those to be reasonably optimistic and then went to the upper level of the historical bands that we'd observed for buoyancy. Perhaps it was correct to take an optimistic view of how things were going to economic conditions going forward, but, in terms of the impact on buoyancy, it was putting it to the upper limit of what had been observed in the past. What we did was point out that this seemed to be at the upper limit of anything that we've seen in the limited data that we have. It appears optimistic. It may possibly be reasonable, but it does seem on the optimistic side. I think that Mr Swinney adjusted Scottish Government's forecasts on the basis of that, but we didn't tell them what adjustment to make. It's not our job to do that. Last question. Again, business rates. Obviously, there are previous years to look at, and it ties in with a previous question. Is it something that you would plan to do for next year, or is it something that you could do to look at the estimates of projections for business rates? Obviously, we've got, whatever it is, 10 or 15 years' worth of data, presumably for that. The original estimates versus the outturn and what the estimates were based on is a work that can be done on that to see how we can improve the model going forward. For all taxis, we would seek to look at forecasts and outcomes. The more information we have on that, the more we can inform improvements in modelling. It's difficult to do, because you want to look at the differences and then you need to ask the question, why did it happen? Cheap point to say, this tells you that the inquiry and challenge approach is actually effective, and then several of these points have happened. But when you get to that stage and then transfer it to how much would you want to make an allowance in the future for this sort of thing, that's much more judgmental. Part of our job is to make judgment, and we do that better and better the more data and evidence that we have, so it's a good suggestion. Does the classical commission have any input into our involvement with the Smith commission? I will explain exactly the involvement. I wrote to Lord Smith some little while after that appointment was announced and wished him well in a challenging task. I know him, so it's fairly easy to say it in these words. I said that I just want to remind you about the new Scottish Fiscal Commission that we exist. We did not intend to make a submission to the Smith commission, because that's not our role. We are, on the other hand, ready if they want to come to us at any point to share an idea or try something out or ask us questions, we would respond in that case only within our remit. We're very conscious of what our remit is and that we don't stray, and so that seemed to be the appropriate way to engage with the Smith commission. If, again, any guidance there would, you know, I'd appreciate that. But in terms of your other roles, will some of you be making submissions to the Smith commission? No. Oh gosh, I'm just trying to think. There is a non-incorporated, if you will, body called the 2020 climate change group. It's a group of volunteers from business and government and others who try to influence momentum on the climate change agenda, and they may be possibly making a statement to the Smith commission, but it's not done yet, and my thinking was possibly, you know, I could be excluded that it's not even a real body in the sense that it is a non-organisation, it's a group of volunteers. In terms of any of my own other engagements, the festivals forum, I have a very big involvement, I chair the board that looks after the kind of strategic positioning of the city of Edinburgh and all of the festivals, and they have not made a submission. If there was some discussion about that, I would think about any relationship to the fiscal commission and would not necessarily put my name to something. So I don't, I'm not aware of any other input that would cross over my worlds, and I just turned to Campbell and to Andrew to ask if you have any. No, I think my answer is exactly the same, and the fiscal commission doesn't do policy advocacy, so he couldn't suggest something as such. If they came to me or to us, we could say something. It's a bit difficult to say what it might be. No, I think if they approached us, we would see in what basis they approached us, they may not do it, because why would they in a sense, given what we're responsible for, but if they did, we would discuss whether and how we could respond to them. Campbell, is there any overlap with anything? I have colleagues at the University of Glasgow who are making written submissions to the Smith commission. I was asked to be involved but declined to give them my position on the fiscal commission. Okay, thank you. Just finally, and it's going back to the data, the missing data if you like, but when you looked at the systems that the OBR employ, was that really impressive? Did you get excited about the kind of detail in the data collection that they have in order to make their forecasts? The OBR have access to the full set of ons and treasury data needed to run multi-hundred-equation models, so they have a large set of data on which to base the forecast. Is forecasting them with that kind of data always been impressive? I mean, my experience on this committee is not great, but the OBR have often been fairly spectacularly wrong in their forecasting. The OBR's approach is to build this very large macrochonometric model. The academic literature has performed harsh races between alternative approaches to forecasting and puts that style of macroeconomic model up against relatively simple statistical models. Over a short-term horizon, the short-term statistical models often outperform large-scale macrochonometric models, but the OBR, on the other hand, needs to produce a set of forecasts for a huge number of variables, and it needs to be coherent and consistent, so you need a big model to be able to perform that task. Those are trade-offs. Sorry, Campbell. It may be worth noting that we had exactly that kind of debate amongst ourselves as well about whether one type, the big or the rather more just the simpler approach, was better. Actually, the simpler approach that we see being taken here out of necessity also felt quite comfortable and helped us to come to the conclusion that the results were reasonable. On the collection of that kind of data, do you foresee other institutions at universities and so on who, it seems to me, often do quite detailed research into some of the areas that will be relevant to you? Will you engage with them? I think that the answer is yes. We certainly spoke about that at the beginning, and I think that we probably know some of those universities that do gather up a fair amount of data. Again, we've reached out a little bit, and that's part of our follow-up program, those bodies that we might have relationships with. We might also, at some point, want to commission some research, and we would likely turn to a university to do that, so as being partial independent and having the integrity of academic research. So, I guess I could just add to that that I can think of several universities or institutes which do do some research. Usually, it's on one bit of the economy, so if we wanted to look at, and I can think of one case where it's focused on the labour market, so if you want to look at some of the factors that are important there and whether that affects the forecast we're looking at, we can do that. I'm not sure we're going to find very much which does the Scottish economy per se. Maybe it should, but well, I can think of one case, but I wouldn't use that model having been into it. So, it's patchy. But generally, I think what you're saying is that the fact that it is almost kind of microcosmic, perhaps, compared to the OBR in some of the areas that you're looking at currently, that the chances are that it could be much more exact. The smaller models. Yes, there is a chance that that could be the case, but part of that is possibly in the data collection and the integrity of the data as it's stored and reported. The biggest issue is actually getting genuinely Scottish-specific data. I mean that's really important and that will help hugely in terms of the quality and the assurance that we feel towards the forecast. So, we see the way forward that that is what is intended and that is what has started here and therefore that's an important place to be. I think that the other thing, and I talked about meeting some of the other fiscal commissions from elsewhere, that's the kind of question that we can chat with somebody completely outside our ambit and just get a little bit of their experience, what have they found has been useful. So, it's a good question. That is a concluded question from the committee. I'm just wondering if there's any fuller point that she would like to make. Anything from yourself? No, I think we're coming. No, I think we're coming. I haven't heard of that. Yeah, just thank you because you've obviously gone into the report in some detail. I'm a note taker and I've made a note of a number of these questions. I'll make sure that they all come up on our agendas going forward for discussions. So, we appreciate that. Thank you very much for the evidence that you've given to committee. This morning, I'm now going to call our recess until 10.45 to allow a change of witnesses and to give members a natural break. Okay, our next item of business is to take evidence as part of our inquiry into proposals for further fiscal devolution. I therefore like to welcome to the meeting Professor Ian McLean and members of copies of Professor McLean's written submission. So, we'll go straight to questions from the committee and as is fashion on this committee, I'll start off and once I've asked you some opening questions, then we'll expand the session to colleagues around the table. So, first of all, I have to say it's a short paper, a very interesting one actually Professor McLean. I'm sure there'll be plenty of food for thought for members of the committee. I think you've listed for example your view on what should be devolved, what should not etc, but there's one or two areas where you've left to weeb some caveats, so I'm going to try and explore those weeb but further from you to get your views, sorry. For example, you say that the rest of income tax could be devolved. So, I'm just wondering if you can tell us should it be. Thank you very much, convener. I think this issue makes less difference than most of the parties in the debate on both sides think because what is really important as I say in my notes is the marginal principle that the margin you and colleagues have to decide between taxing and spending and you already get that under Cullman under the Scotland Act 2012. I'm not myself clear how much extra you get by having the whole of income tax devolved. Nevertheless, I think it's perfectly possible to devolve the whole of income tax and we would need to be clear what we mean by devolve. If it means devolving the right to set the bans and even who qualifies for who is required to pay income tax then devolving the whole of income tax is a big deal. If it's a matter that the bans and exemptions and so on remain uniform at UK level then I think it's not a big deal. I hope that I'm making myself clear. Do you believe that it should be such that there's control over the size of bans who qualifies etc etc? I can see arguments both ways. All this is given the way the referendum vote went. Scotland remains in the UK. Therefore, although having the power to set rates and bans would give this Parliament a lot more autonomy and a lot more responsibility there would be pretty heavy transaction costs beginning with defining not only people but sources of income as either Scottish or non-Scottish. I'm torn here, convener, perhaps between head and heart. I'm not sure which is which but at any rate the principle of maximum autonomy does mean devolving control over rates and bans. Practical considerations suggest that that would be really quite difficult. Your view differs from that with the Foreign Prime Minister, Gordon Brown, in terms of the impact that it would have on MPs in the UK. You say here that MPs would remain responsible for the world UK tax structure and for macroeconomic management. Is that the case? The former Prime Minister will have to speak for himself. My view is, as I've set out in my note, that, since the Westminster Parliament would remain responsible as an absolute minimum for rates and bans in England, the rest of the UK, and for those taxes that can't be devolved, it seems to me that there is a role for Scottish MPs at Westminster in all circumstances, even where the whole of income tax in Scotland to be devolved. You've pointed out that, in your view, North Sea oil should be the first candidate for further tax devolution. I wonder if you can expand on your rationale for that. Because it's related, you talk about co-operation tax, which you don't think should be devolved, but do you think that the revenue raised from co-operation tax vis-à-vis North Sea oil should be the same? I think that the three issues here are, if you will, what economists call the economic rent derived from North Sea oil, which is captured possibly by some of the taxes other than co-operation tax. Then there's co-operation tax itself on North Sea oil, and then there's the rest of co-operation tax. In preparing my note, I think that members will have noticed that I am an avid disciple of Adam Smith, and I'm very honoured to be speaking a quarter of a mile from Panmure House, where he wrote The Wealth of Nations. It seems to me—I'm pleased that it also seems to the Cabinet Secretary—to be right to start from Adam Smith. Smith argues that rents are the most appropriate subject for taxation. In a modern context, I would say taxation at a devolved level. The pure principle of tax policy is laid out by Adam Smith and his successors, especially David Ricardo, that lead me to say that rents from the North Sea should be the next tax base that this Parliament should control. The reason I say that co-operation tax within a continuing union should not be devolved—well, a very easy reason, I can say three reasons, Amazon, Starbucks and Google—and all members will have seen the performance of these three companies before the Westminster Public Accounts Committee—given an opportunity to avoid tax, multinational corporations will, and given that they will, if a Parliament such as this one, or I know that Northern Ireland is asking for the same, were to take control over co-operation tax levels, there might be some gain to the revenue received by this Parliament, but there would, sure as eggs are eggs, be a loss to the revenue received in the UK from co-operation tax. As my note says, I don't think that within a continuing union co-operation tax in general should be devolved. Co-operation tax on North Sea operations is a slightly different matter, because North Sea oil is where it is. You cannot pretend that it is in Luxembourg, and therefore I think that devolving the whole of North Sea tax is perfectly feasible, as well as desirable, on the grounds that Adam Smith gave. Thank you very much. I'm really impressed by your clear, concise answers. I do like that, I must say, actually. I don't always agree with them, but I have to say it. I'm sure that members will have their own views, too. The other area where you talk about assignment—again, you say—I'm trying to tease out a direct answer on this one—is that you talk about the proceeds of VAT in Scotland that could be assigned again. Do you think that it should be assigned to Scotland VAT revenues? I've seen a reason why it shouldn't, under where the result of the Smith commission, which is the context in which we are all talking, to be some form of divo max, then max could involve the assignment of that revenue. It would give this Parliament some limited control, but only limited because you would not control the rate or the base of that, but some limited control say to, I don't know, encourage retail developments in Scotland so that people spent more, if that was what you and colleagues thought was the policy desideratum. But I think it's more a matter of giving the Parliament revenue consistent with its spending responsibilities in order to reduce the vertical fiscal imbalance that I open my notes by talking about. Yes, indeed. I know that you gave us some quite interesting—it's the school information at the back—looking at various different countries in terms of that. Just one final question before opening out to colleagues around the table. You talk about what taxes should be excluded. You talk about capital gains, income tax on savings and inheritance tax. Can you give us your thoughts on those three? The reasons are more practical than principled. Income tax on savings, members will know that there has been a lot of discussion between, in particular, within HM revenue and customs and the UK Government, and I'm sure the Scottish Government. The sheer physical difficulty of identifying a Scottish source of income seems to the revenue to be an insuperable rejection to devolving income tax from savings and investments. Whether it really is or not is not for me to say. I expect you have in mind to interview officials of HM revenue and customs, and I think that that would be a question to ask them rather than to ask me. On capital gains and inheritance tax, maybe the cases are different, maybe I should have written more carefully. Capital gains tax, my reasons for not recommending devolution, are that essentially, because it's a tax on capital and capital can move around, you would see all sorts of schemes as soon as the rate varied, companies would incorporate in Scotland, and the same issue that I raised with corporation tax would come up again, that you would get, to speak frankly, pretend incorporations in whichever jurisdiction had a lower tax rate. Inheritance tax, given that most of what is caught by inheritance tax is property, and property is where it is, is fixed, there might, on second thoughts, be a case for devolving inheritance tax. Of course, it could be a policy issue if this Parliament wanted to take a different line to the Parliament at Westminster as to the rate and base of inheritance tax, particularly whether the threshold for inheritance tax should be raised or not. On second thoughts and possibly contradicting what I say here, I might say go for it. Excellent, glad to hear it. So I should have, I was going to pass on, just one thing I did mean to ask you, what is your view on the, on devolving ear passenger duty? That's obviously something which has been in the news quite a lot in the last 24 hours. Yes, I'm aware of that. For the noise that has been made about an extremely small tax, see my table too, I'm surprised that there is so much fuss made about this in either direction. I would say, so my overriding comment is that APD is so small that people should be rather wary about listening to the vested interests and you and the public have been lobbied by airlines who would like to pay less tax. Well, that's hardly a surprise. Airports are where they are and therefore APD comes under the Smithian principle that you tax the least mobile base. I don't think, given that, given the configuration of airports in the UK, I don't think there are any huge policy issues. It has already been said, I think it may have been in the Scottish Government White Paper, that the Scottish Government would like to reduce the rate of APD paid at Highland and Island airports. Well, I can see obvious policy sense in that. Competition between Scotland and England, not really a huge issue, people are not going to dry, I mean it's 100 miles from Newcastle to Edinburgh airports. It is not really a matter of severe competition between adjacent airports where the rate of APD to differ. I think that, with all deference to the lobbyists who have been lobbying very hard on this issue, that it's not a particularly big issue either way. Okay, thank you very much for that. Colleagues around the table, the first to ask question will be Malcolm to be followed by Jimmy. In relation to assignment of VAT and possibly corporation tax, I mean obviously that would have the advantage of increasing the proportion of our budget that we raise ourselves, but would there be any economic advantages from assignment or indeed other advantages or disadvantages? It's hard to see any real advantages. You just get the revenue. In the case of VAT, it's a really big tax and so the revenue income is important to your budgeting. But when you think about it, do you have policy tools that would make you change behaviour in relation to people's expenditure? Well, you might indeed want to do that, but I think what you would want to do is to change the coverage of VAT and the rates, and you can't do that within an EU member state, as my note says. So I don't see any huge advantages from assignment other than the one that I mentioned in my previous answer, that it gives you control over a higher proportion of the tax revenue accruing in Scotland. I suppose the next question follows on from that in a way, because I suppose one of the problems about assigning VAT would, you know, what would happen in a recession. Moving on to oil, it's not, I suppose, just in a recession, but obviously there would be loss of revenue in a recession, but of course even under current circumstances, as we know this year, the price of oil fluctuates quite considerably. So how would you deal with that particular problem? I mean, this year might be a good example, because I think a lot of the predictions for oil revenues are quite a lot lower than anticipated by either the Scottish Government or the OBR. So how would you deal with that particular issue? Because obviously if that was part of your revenue, that would be quite a significant part of the Scottish budget. I start with what the people of Scotland want, and as my notes say, it's well established from surveys that the people of Scotland want this Parliament to control all domestic public spending. I don't think, for the reasons that I start out with on VFI, that it's responsible for this Parliament to control spending if it does not also control taxing. This leads me to say, for the reasons that I have given earlier, that North Sea and Vat—well, North Sea is a suitable subject for devolution—Vat is a suitable subject for assignment. If this brings the set of issues that you raised, that North Sea is volatile and that Vat is volatile in a downturn—it's one of the least volatile of taxes, but it is volatile in a downturn—then all I can say really is that the Scottish people should be aware of what they wish for. They wish for maximum devolution, and this is the consequence of maximum devolution. If they didn't want devolution, they could be protected from fluctuations in oil production more than given that they do want. It seems that they want devomax. In terms of powers for this Parliament, is the implication that you would have to have very significant borrowing powers to cover those kind of shortfalls? Are there some economic problems in relation to that, whether in terms of interest rates or in terms of what would be feasible for the UK macroeconomic management or whatever? I recall saying on the previous time when I gave evidence to this committee that, on borrowing powers, I was a borrowing powers fundamentalist, and I was maybe telling some tales out of school from the Kalman independent expert group, of which I was a member, and of which members of this committee will be familiar. My view, but it was not the view taken ultimately by the IEG nor by Kalman, was that market discipline is the control that really works and that this Parliament should have borrowing powers. Of course, borrowing powers to cover fluctuations are one thing. Borrowing in order to fund current expenditure is quite another thing and should not be done and where this Parliament to do it, the markets would notice very quickly and impose penal rates. Borrowing to cover fluctuations, I'm all in favour of. You've used the term divo max quite a lot and featured in our debate very prominently yesterday, including in my speech, but you obviously want to exclude certain taxes, so what do you mean by divo max? Sorry, I was just using the popular phrase and it has been very widely undefined, if I may put it that way. Practical divo max for me would be the devolution of the list of taxes I've set out in this note with the possible addition on second thoughts of inheritance tax. So divo max is what you define it as? No, if a policymaker came to me and said, I want divo max, I as an academic can only say, well, what do you mean by divo max? And then I would say, well, this is a good idea or a bad idea or a good idea up to a point because here is how the Adam Smith's principles of taxation applied to your proposal. Now, of course, I lot, my last question really, and because you've given such clear and concise answers, we have a problem coming back to you because you've already given your view so clearly and concisely. I mean, as you've suggested, perhaps this income tax issue is going to dominate to an unnecessary or undesirable extent, but the fact of the matter is partly because Gordon Brown has raised the stakes on the issue and partly because every party seems to have a different view. I suppose I'm probably asking you to repeat yourself, but I suppose I'm genuinely interested in this issue. I think Professor Ronald McDonald is going to have things to say about this, but we haven't seen his submission for next week. I mean, in terms of what Labour is proposing on income tax, I mean, I suppose Gordon Brown's looked at it from in terms of the implications at Westminster in a kind of way you've touched on that. But can you think of any other economic reasons why it might be desirable to withhold some income tax? I think Gavin McRown in his book suggests that there might, there could be quite good economic reasons for keeping some of the income tax. I can't remember his detailed arguments for that. I think that it was to do with emergencies arising. You need to have some income tax powers at UK level, but I may be paraphrasing them wrongly, but I wonder if, in fact, you think that there are any sound economic arguments for dividing income tax in that way, or whether you think that at the end of the day is down to a judgment about the consequences of it for the position of Scottish MPs at Westminster? I think that it's primarily the latter, and I didn't dwell on that point because it seemed to be not to be germane to my evidence to this committee. If I asked to give evidence to William Hague's committee, which I have not been to, I would talk about that at considerable length, or at whatever length seemed appropriate. As to economic advantages of holding back some proportion of income tax, I confess that I have not read Gavin McRown's new book, and I should have done, and I wish I had, but he also is very well able and very well qualified to speak for himself, and I hope that you will consider asking him if you haven't already. I would guess that the sorts of reasons that he and others who make this point are having in mind are reasons to do with sudden shocks, and, after all, a sudden plunge in the oil price might be the most likely shock that Scotland might be exposed to. Sudden shocks can be damped within a union by redistribution. Redistribution is very difficult if this Parliament controls almost every tax and every tax base, which takes me back to—I'm just repeating myself—that the Scottish people should be aware of what they wish for, but we know what the Scottish people wish for. Maybe this Parliament should say careful people of Scotland because devolving the maximum amount of tax possible means less room for cushioning shocks. So, last point, just to summarise what you were saying, but your own view on the specific issue of the effect at Westminster of devolving income tax entirely to this Parliament, your view is that that would not, in itself, whatever the pros and cons of different arguments about the position of Scottish MP at Westminster, you don't think that the devolution of income tax is really going to make any significant difference to that? For sure, it will make a difference. It might throw questions on the number of Scottish MPs at Westminster, and again, that's not a matter for me here nor for this Parliament, but I can see that that could be an issue. I think the nitty gritty is going to be in the area of does devolution of income tax mean devolving control over rates and bases, and that cuts both ways, because if the rates and bases are not devolved, then I think there really does have to be a Scottish presence at Westminster because you're controlling the level of the personal allowance or saying whether there should be a 50p top rate. There are lots of highly, highly political questions, which were I living in Scotland, I would want to be represented by an MP who could help to make my views on these matters known. If, on the other hand, the devolution was so extensive that this Parliament had the power to set different rates and bases to those applying in England, for sure this brings a classic West Lothian question, would Scottish MPs be allowed or advised to vote on the English rates and bases, but even in that case, if a given level of personal allowance is set in England, that quite severely constrains the freedom I think of this Parliament to set a different rate of personal allowance to avoid gaming behaviour and as we all know in matters of income tax, it's the rich who are in the best position to game. People will suddenly turn out to have a house in Scotland if it benefits their tax position or in England if that way works. So I think, I'm sure that you will be aware that I'm thinking on my feet here, but I think that in either case there remains a case for having Scottish MPs in Westminster. Okay, thanks. Okay, thank you. Jamie, to be followed by Michael. Thank you, convener. I want to turn back to the issue that you raised, convener, unless the part of your paper, Professor McLean, where you suggest that control of North Sea oil should be the first candidate for further tax dilution. That tallies with evidence given from the Charter Institute of Taxation, Charter of Accountants in England and Wales to the Treasury Select Committee yesterday, but—and this question isn't meant facitiously, it's meant quite generally. You obviously don't literally mean the North Sea, presumably you mean that we should control taxation of resources that are within Scottish jurisdiction. I did mean that, yes. Apologies for the shorthand. No, that's okay. But it does beget a more serious question. Control of tax is all well and good, but should we also control the licensing regime for the extraction of such resources? It would make sense that the one should go with the other as part of the overriding picture that tax and spend responsibility should go together. In practice, it would be hard to unwind any promises given by the UK government to companies exploring the North Sea, because—pardon? Or elsewhere. Or elsewhere, yes. West of Shetland, whatever, the same issues apply. Should this Parliament want to give new incentives to encourage exploration—I don't know if that was the question behind the question—I would be fine by that. Of course, new incentives mean less tax, at least in the short run, less tax receipts. Yes. I was also thinking that there may be areas that, thus far, the UK government has decided shouldn't be utilising the Scottish Parliament. The Scottish Government may take a perspective that they should continue not to be utilised or maybe they should be, I suppose, at the same point. I am thinking specifically on based the extraction of resources. There is currently going through a licensing arrangement, and perhaps the Scottish Parliament would say that we do not want to do that, or maybe we do. I am also thinking that the convener might have a perspective on that. There are apparently deposits in the Firth of Clyde that have not been utilised. Maybe we would decide that that is okay, or maybe we would decide that we would like to utilise them. I am out of my comfort zone here, colleagues, because this is into the nitty-gritty of oil taxation. Although I do not know the answer, I know a man who does. I think that Professor Kemp of Aberdeen, who is very well known to all of you, would be the right person to pose his questions to. I would be surprised if there is any source that there is a policy decision at any by any government not to have exploited, but I am outside my area of knowledge here. Okay, that is fair enough. That might be a moot point, but I was getting a sense that if there wasn't the restriction that VAT could not be set differential rates within a member state, would you be recommending that it should be devolved? If there wasn't a restriction, then the issues with VAT would become the same as the issues with excises, tax and tobacco and alcohol and petrol and so on. The issues with those, although I do not mention them in this paper, I have mentioned them in other stuff that I have written, is that, although it is mitigated by the thinly populated border between Scotland and England, nevertheless, if rates vary only by a penny or two, you will get people who drive 100 miles from Glasgow to Carlyle or 150 miles from Manchester to Gretna to profit from a lower rate. In one perspective, that is all very fine if this Parliament is the one that gains the revenue from, say, having a lower excise, then it might not worry too much about the fact that tax receipts in the current UK as a whole have gone down. Where it would be more difficult, and I accept that this is more about excises than about taxation, is where this Parliament decides to levy higher excise on behaviour that it would like to curtail, like smoking too much or drinking too much or driving too much, then instead of there being an enormous hypermarket at Gretna, there would be an enormous hypermarket at Carlyle, and people would drive from Glasgow to take advantage of the lower rates. Were VAT allowed to vary, you would get the same sort of thing times about five or ten, because VAT is five or ten times bigger than excises. Anyway, to take on the idea of assignment, you have already talked, I suppose, the rationale there is that the Parliament of the Scottish Government could try and influence behaviour that could increase revenue. How effectively equipped are we to do that? We have to be very pedantic here. This Parliament may increase rates, which may or may not increase revenue. I am aware of discussion about imposing a minimum unit price on alcohol or other devices to encourage pro-social behaviour. I just think it is a fact about the world that this Parliament's power to control behaviour by this sort of means is not unlimited, because people will go somewhere where the tax rate is less to get their spirits and cigarettes. That is the fact about the world that this Parliament cannot really do anything about. If VAT was, the revenue was assigned, it should then the Parliament, this Parliament and the Scottish Government, albeit it would not be responsible for setting the rate should it have some form of statutory rate to be consulted on the rate. That might be a very sensible thing to ask for, yes. That was a very concise answer, but thank you for it. My last question in terms of the implications of further financial powers for the role of revenue in Scotland, you suggest that it could make sense that other taxes should continue to be collected by HMRC, but there is not a slight concern there that HMRC is not in the legislative competence of this Parliament. If we had taxes that we were responsible for legislating for, but we could not legislate for HMRC, is there not a slight concern there? Would you suggest that HMRC should become a shared legislative competence, although it might be difficult to do that, or would it not be easier to go back to what I think has been what everyone has thought would happen is that the role of revenue in Scotland should be enhanced to deal with the further devolution of taxes? My answer on that was only driven by what academics call barefoot empiricism. I looked on revenue Scotland's website and I found that it is not there yet. If it has not even got a website yet and it has got a few hundred million pounds of taxes coming its way very soon, then I do not think it is quite ready to deal with 10 billion of receipts of income tax. It is a purely practical suggestion. Agency agreements between, as it might be, HMRC and Revenue Scotland would be very easy. I cannot imagine that I would not think that they would require parliamentary authority, but if they did require parliamentary authority, then the Sewell motion procedure would be very easy to apply. I am just thinking on terms. We, as a committee, have a scrutiny in our role, and we could probably find it easy to scrutinise revenue Scotland, which, after all, is a creature of statute that has been created out of this. Body was HMRC. It was interesting that your answer could be taken to say that the answer is then that you need to enhance the responsibility of revenue Scotland, so that is a possibility still then. That is up to this Parliament, yes. The normal presumption would be that the Parliament, which levies a tax, also collects it, but, obviously, without taking any position on the union or anything, there are economies of scale in tax collection, most especially for the big ones, income tax, national insurance and VAT. I wonder if Professor McLean could help me to understand a bit more some of the technicalities of the block grant adjustment. We have seen two taxis coming to Scotland under the Kalman proposal of the Scotland Act 2012. It is very uncontroversial. The limited taxis are not very big, but the devil has been in the detail. Getting the block grant adjustment, I agree, has been very difficult. The cabinet secretary has made us aware of his frustration around that. If we were to look at more taxis coming and more adjustments, how technical, how more problematic is that going to become? Would we have to have one standard adjustment for all those taxis or would each individual tax that was transferred to Scotland require a different set of adjustments? It speaks to one of the frustrations that I know a lot of people have, which is that how Barnett works, what Barnett consequentials and so on, is entirely in the hands of HM Treasury. It is not statutory. If this Parliament or this Government does not like what HM Treasury is doing, there are currently, to my knowledge, no mechanisms except perhaps the Joint Ministerial Committee to pursue that. I would rather have some neutral ring holder—I have written about this elsewhere, not in this evidence—that the example that I would commend to this committee would be the Commonwealth Grants Commission in Australia, which is not controlled by the what they call Commonwealth Government nor by the States, although it is licensed by all nine of those bodies to make allocations. If the future of block grant remains in question, I think that it would be desirable for it to be controlled by a public body of some sort, which was not an agency of one of the parties to the argument, brought under some sort of joint control of, say, Scottish, Northern Irish, Welsh and UK parliaments. Of course, I can see very quickly down the road there would be a question of, do they have equal votes or does the UK have more votes than the others? I do not have an immediate answer to that question. Obviously, we have had a look at the issue of the Welsh Assembly's perspective on Barnett. Do you have sympathy with their position? I have sympathy with their position as regards Wales, because their position as regards Wales is very clear, which is that Wales does disproportionately badly out of Barnett, given how poor it is. They are quite clear that they would prefer a needs-based assessment. Now, this question, as members will know, was also aired in the Carmen process in Scotland, and Carmen, and advising it, the independent expert group, did not say that it wanted a needs assessment to replace Barnett. As everybody in this room knows, on a plausible needs assessment, the block grant coming to Scotland would be less than under the present arrangement. If we are to continue down this road, the idea that somehow Barnett could remain unchanged, do you consider that to be a conceivable prospect? Can people who argue for Devo Max also ring alarm bells because it is going to have an implication for Barnett, with any justification? Yes, I think that there are two almost balanced fallacies here on the opposite side. There is the fallacy made by people who appear in Scotland very shortly before a referendum and make a vow that contains two inconsistent promises, one that the fiscal autonomy of this Parliament will be increased and two that Barnett will be protected. At some level, these promises are incompatible, and you would need to interrogate the people who made the vow to see which one of those they really meant. On the other hand, to be fair in my condemnation all round, it is inconsistent to demand maximum possible devolution of taxation, including taxation of, for instance, North Sea oil receipts, and at the same time, it is asked to be protected from the consequences of that devolution by having Barnett-type transfers. That is a fair answer, thank you very much. Thank you, John. One of the things that I do not think has been mentioned so far is that, in the bullet points in your executive summary, you mentioned unfavourable demographics for Scotland if we wanted to take full control of social protection. Can you expand what you mean by unfavourable demographics? I mean the unfavourable post-65 demographics that have been much discussed in the last six months, that the pension age population as a proportion of the population is somewhat higher and the projections going forward are somewhat less favourable for Scotland than for the UK as a whole. There is also a related question that morbidity, chronic disease, is worse in Scotland, and there are similar policy issues, but you can do something about disease in Scotland, you cannot do anything about the age structure of the Scottish population, not in the short term. If you take on responsibility for social protection, then you take on at the margin finding the money to pay for it. That is a policy choice, which is not for the likes of me to say whether this Parliament should take. Are we saying that, at the moment, Scotland spends less proportionally on social protection than the rest of the UK does, but that, in future, there is a risk that that would switch? No, I am not saying that exactly. I am saying that, at the moment, social protection, as you all know, is predominantly reserved matter. Therefore, the risk, or the shock, if you will, of Scotland having an older and worse health population is absorbed at UK level. If that function is devolved, then it is for Scotland to meet out of its own resources. Presumably, it could work both ways. Generally, my constituents are dying earlier, so there is a risk that they are going to live longer and that would cost us more, but on the other hand, we also have extreme ill health, so there is an opportunity that we improve the ill health and benefit from that. So there will be arguments on both sides, is that the case? Oh, yes, there could be arguments on both sides. I do not think that any politician is going to say, come to Scotland so that you can die earlier, but I realise that that is not what you just said. No, we actually only have to travel on a railway line through Glasgow and you die earlier, but switching on to North Sea Ireland and such, the whole question of—I mean, you actually were in favour of that because it is linked to rent, which I come back to again, but usually the argument against that is volatility, but again, presumably, the volatility can work both ways. If the projections for future oil revenues are lower and then there is more opportunity that we might gain from having the revenues, but if the projections agreed were higher, the risk is that we would lose. Again, it could work either way and it really depends on how the adjustment was made. Of course it can, yes. It can work either way. If receipts from North Sea oil are devolved and they go up, then this Parliament and the people of Scotland are in a good place. I do not think that there would be any question of adjusting—well, we would have to see what formula succeeded Barnett, but if any formula that succeeded Barnett was based on needs, then upsurge in oil revenue would be on the face of it irrelevant, but also I think formula funding would become a very small part of the finance available to this Parliament where there to be any form of divo max as defined not by me, but by the politicians who agreed what it meant at the end of the Smith process. I am interested in your argument that, based on Adam Smith, the rent side is the one that we should emphasise. We have tended to, in this paper and elsewhere, look at existing taxes and who controls them. We have not looked so much at possible new taxes. First, would you argue that we should move all taxes towards the rent end? Something like land value tax has been suggested over the years. Is that something that we could or we should be able to introduce in Scotland rather than just moving around existing taxes? I am laughing slightly because, to be in favour of land value tax in some quarters is taken as a sign of madness. In fact, the arguments for land value tax are perfectly sane and they were made by Adam Smith and they were made by David Ricardo. Land value tax has been taken over by some people who might be regarded as rather cranky, but the underlying arguments are good. Where this Parliament moved towards a land value tax, it would be, I assume, substituting that for council tax and business rates, which are the existing taxes on land and property. Speaking only as somebody interested in tax structure, I would be delighted if this Parliament did that because I think the underlying economic arguments for land value tax are sound. The Parliament already has come under immense pressure, I know, from interests and lobbyists in this area, but let me say that I am a huge fan of the work of Andy Whiteman in this area, who will be known to all of you. On that matter, I suggest that you ask him. You are giving us a lot of suggestions for future witnesses, I have to say. One of the things that you said was that if there is more devolution, there is less room for cushioning the shocks. I guess that that would be true if it is an external to the UK shock, but if it was an internal shock coming from Westminster, presumably devolution actually gives more room for cushioning? You are thinking of tax policies— Welfare reform, cuts to welfare reform. Yes, okay. Bedroom tax and the like, which is a currently controversial example. Yes, that is true. There would be more room for cushioning if this Parliament decided that it did not want to implement a tax, a welfare tax that it did not like. Of course, if that results in less revenue, it has to meet the revenue somewhere else or adjust spending policies, but that is the whole point of being a ffiscally irresponsible Parliament, which is what underlines my whole remarks. The final thing was that HMRC has already been mentioned and Revenue Scotland and so on. When we did put out the bids for the previous taxes, in fact, it seemed to be the case that we could do it more cheaply than HMR revenue and customs, which makes me wonder whether one of the options—although you said that there are economies of scale, would there be an option to simplify the tax system if we could have control of all income tax and national insurance? First of all, we could put the two together and have a simpler system that might cost less to run. That is possible because it is easier to run a tax system in a country of £5 million than in a country of £60 million. On the other hand, we are in the notorious territory of computer systems, government and big IT failures. My note of caution is that big IT failures are perfectly possible in a country of £5 million, as we all know they are possible in a country of £60 million. It is worth exploring, but I would not put my shirt on it. The suggestion is that, if it was income tax, we would be starting with where we were. Presumably one option, rather than just transferring everything to revenue Scotland, is to split HMRC and take a chunk of HMRC that would look after income tax, and that would become whatever we call it in Scotland. Do you think that that is an option? You should take advice from the revenue authorities. My understanding is that that already happens in the sense that the tax-raising function is organised tax by tax more than it is region by region. There are people who collect income tax and there are different sort of people who collect stamp digital land tax and so on. I do not want to detain the committee, but this actually goes all the way back to Robert Peale in 1842 when income tax was introduced and before him to William Pitt in 1799, I think, who wanted to ensure that there was a different administration for each tax on the principle in those days that a gentleman should not be interrogated by a tax collector as to the whole set of his income. I think, although you should take evidence from the tax authorities, that HMRC is already organised functionally rather than regionally. Professor McLean, can you just talk through briefly what is happening on table 1 of your submission? You are basically talking about the vertical fiscal imbalance, which I think is one of the priorities that you want to deal with. Just to try and get my head around the figures, if we take the UK, which is helpfully in red, or at least on my copy, 36.4 for central revenue and 4.2 for sub-national revenue, above that Spain is 22.7 and 12.2, yet the vertical fiscal imbalance at the end Spain and the UK are fairly similar. Can you just talk me through how those figures come out? Yes, with pleasure. There are different ways of measuring VFI, and for this purpose I just took the simplest possible, which is in effect to subtract column 2 from column 4. The 14.5 minus 4.2 is 10.3, and in the row above 22.6 minus 12.2 is 10.5, with a surrounding error there, but it rounds to 10.5. One problem is that the year for which these data were taken, and I took this year because this was work at the time in connection with Kalman, was the depth of the financial crisis in all OECD member states. In every single case, you will notice that spending exceeds revenue, or every country in this table was in deficit in that particular year, but that does not affect the general principle that one way to measure this is simply that difference. There are more complicated ways, but I went for the simplest. What drives the relatively high UK figure is that sub-national revenue is such a low proportion of revenue collection that, in the UK, it is only council tax and business rates that are levied by bodies lower than the UK Parliament. That is changing in relation to Scotland, but Scotland is only 10 per cent of the UK. The changes that we have been discussing in this session would not in themselves change that number all that much. That is helpful. I am grateful for that. You talked about in your paper some of the areas that you would not devolve and gave explanations for those, except for inheritance tax. You have said that now you might have a slightly different view, but what was your additional reason for saying that inheritance tax would not be a candidate for devolution? I was once again applying the pure gospel of Adam Smith and saying that inheritance tax is a tax on capital. My reason for my change of mind in real time while sitting in this committee is that when you think about the assets that people who die typically have, in most cases it is a house and a house is where it is, so it is not subject to evasion. That is for the moderately rich. I think that there is a whole separate issue of inheritance tax avoidance and evasion by the extremely rich, of which I have views as a citizen, but I did not come here really primed to give them, but it is an area where I would hope that this Parliament and the Westminster Parliament would look. You talked a little bit about the devolution of taxation for North Sea oil. If that were to happen, there would be a block grant adjustment, as we have seen with LBTT and landfill tax. If there were to be a block grant adjustment and oil taxes were to be devolved, do you think that the Scottish Government would stand by their oil revenue projections that they came out with a couple of months ago? I think that that is a question where I am downed if I say yes and I am downed if I say no. I think that the Scottish Government can make whatever projections it wishes, but the revenue will be whatever it is. Any forecasting body needs to make projections in order to do sensible planning of future expending commitments. I think that that is as far as I dare go, given that I am aware that opposite opinions on this question will be held round the table by members. I am grateful. Thank you. A couple of points arising from some of the things that you were talking about, Professor McLean. One was the—and it comes up often—about variations on taxes across the border and suggesting that if we tax cigarettes really highly or whatever that there will be a massive warehouse in Cumbria or in Scotland, depending—do you think that that is really a concern elsewhere, where every country in Europe the border is crossable at Northern Ireland to Southern Ireland and so on? Has it stopped? It did not stop tax on wine, for example, in England when people could nip across a channel and come back with crates of wine, and they still do. Is it seriously a consideration? For sure it is a consideration, but, as I said earlier, it is a consideration which this Parliament cannot do very much about. People will jump in their cars and drive 100 miles for a few pence off. Some people will, some people will not. Scotland is relatively lucky because you happen to have a thinly populated border. The issues are more difficult for Wales, much more difficult for Northern Ireland, but I would say that limits the freedom of this Parliament to set as it might be tobacco taxes, but, of course, it does not take it away. This Parliament would still have the freedom. It is interesting in that case that your data from GERS, which I recycled in my table 2, shows that tobacco duties are one—after aggregates levy, tobacco duties come second in the fourth column in Scotland's percentage of UK. That is saying that there is more revenue from tobacco duties in Scotland per head than there is in the rest of the UK. Where are this to vary? Some people, of course, would drive to Carlisle for their smokes, but not everybody would. Thank you. The other thing I wanted to ask about was the costs of devolving taxes, which, during the pre-referendum debate, were cited as something that was unthinkable. However, to my uncertain knowledge, it may not be mentioned at all when it came to the vow. Do you have a view about that? The costs of setting up a tax administration were an area of controversy before the referendum, and indeed an area in which I had some involvement. Some numbers, which I have to say I thought were both unrealistic and misleading, were produced by my academic colleague Patrick Dunleavy from the LSE. When you drilled down into his numbers, it turned out that his estimates of the cost of setting up an independent government were actually the same as everybody else's. It was just that he had put a label that he called transition costs. If the vote had been yes, those costs would have been inevitable. Given that the vote is no, the costs are, to some extent, optional. The Parliament can decide whether to expand the capacities of revenue of Scotland to cover every tax, or to have an agency arrangement with HMRC. I do not know which would be cheaper, but the opposite points have been made in this discussion. On the one hand, there are economies of scale, on the other hand, taxing 5 million people might be less expensive per head than taxing 60 million of people. I just do not know the answers to these. These are very quite weighty matters, which would have to be explored further. On your support for a land value tax, I hear what you say about the people being seen as mad wanting to raise the issue of a land value tax. Do you agree that it is something that really should be debated and given serious consideration in Scotland? It could, in fact, be quite a large revenue earner and, indeed, fairer than either of the council tax and the business rates that we collect at the moment. I do think that, but I must stress that that is a citizen rather than an academic. Mostly it is a citizen rather than an academic, but I think that the fairness point is that there is a very strong argument made to repeat myself by Smith and by Ricardo that land should be taxed in proportion to its value, and that is the least distorting of taxes. On a personal note, when I was about 15, I picked up Tom Johnson's Our Scots noble families, first published in 1909, which I think was the greatest radical manifesto of its era in Scotland, but when I commend that, which members will already know, I'm speaking as a citizen, not as an academic. Just so that I'm clear about the difference, if you were speaking as an academic. Speaking as a citizen, if I had a vote in Scotland, I would vote for whichever party made promises to introduce land value tax. Speaking as an academic, I have to say, well, the here are the advantages and here are the disadvantages. An obvious disadvantage from any tax change is that there would be losers, and you would for sure hear from the losers. I think you already are hearing from the losers. In any major tax change of this sort, you hear much more from the losers than from the winners. That would be, for what it's worth, my advice as an academic. I was just saying exactly the same thing to Jim Finlay for nothing, when you said that, you know, the issue is always, it's not just that you hear more from the losers, it depends on how many losers there are relative to the number of winners also, and the extent to which they do lose out. We've concluded questions from the committee, just one or two things, is to round up from myself. In terms of table two, you know, you can't revenue Scotland 2012-13, which includes geographic share of North Sea oil and descending order by tax. What's interesting, of course, is that the fourth highest and also the disproportionate share from Scotland is gross operating surplus. One thing can give us a wee bit of detail as to what that is specifically and why Scotland has a much higher share in the UK relative to its population. Yes, convener, I noticed that as well, and I'm simply recycling numbers which your Government produces, the JARS series, so I don't know, it would be easier for you to ask the JARS team to come and give evidence, but I assume, and I think this is the assumption behind your question as well, convener, that this arises because there are functions of government which are public in Scotland and private in England, water comes to mind. I don't know that that's the correct answer, and I would advise you to check with the JARS team. Yes, to be honest, it really wasn't so much the content, it was more the proportion, but that's a reasonably good answer. In terms of tax system, we've talked about costs of setting up an independent Scottish tax system or enhancing the tax system here through revenue Scotland, whatever, and the difference in costs. One of the things that the Institute of Fiscal Studies said before the referendum was obviously that if you start a new tax system, more or less from scratch, you can get rid of a lot of the horrendous anomalies that you've got in the UK system. I and I'm sure all MSPs got a thing through that said that some 34 billion remained uncollected from financial year, 1112, 6.8 per cent of the UK's tax take. We're not talking about money, which people, we said, the revenue have said, oh, you don't have to pay that because of some kind of avoidance, but that's money that should have been collected, so 34 billion uncollected, which of course is more or less equivalent to the entire Scottish budget. If such tax powers were devolved to Scotland, given that it would have almost a clean sheet of paper, not quite, but almost, would that not allow us to make a much more efficient tax collecting system? We hinted it for 5 million people, it might be easier than for 60 million. Yes, I do think that there are arguments on both sides. It might be easier for 5 million people to chase tax avoiders. On the other hand, the very fact that some new differences between the Scottish and the rest of the UK tax system would be created, unfortunately and inevitably, in itself creates more opportunities for tax avoidance. There's a notorious example, historical example, from Northern Ireland, where this is all on the public record, has been for a long time. A meat corporation, the Vestis, took advantage of a tiny provision of Northern Ireland tax law that nobody else had noticed to, in effect, almost exempt their company from tax law, although they had no domicile in Northern Ireland. You would be balancing the greater ease, if you will, of going after avoiders in a country of 5 million, rather than a country of 60 million, against the new opportunities for avoidance that any variation in tax rates automatically creates? There's an actual fact to be more opportunities, because if you had a principles-based system, as opposed to a rules-based system, surely that would reduce the likelihood of avoidance, because it's what you mean through the principle, rather than the specific wording of legislation that is tied to the UK in 300 different tax avoidance rules that have had to be specifically developed? I can only perhaps cynically refer to the title of one of the works of my colleague Andy Whiteman, whom I mentioned earlier, whose book is entitled The Poor Had No Lawyers. The converse of that is that the rich do. I think that I will just leave it there, convener. Thank you very much for that. I was going to say to you, given that we're now more or less winding up the session, if there's any filler points that you may wish to make in addition to anything that you've said, that you've if any light bulbs have come and switched on or anything like that, along with inheritance tax one over the last hour or so? I think that we've covered everything that I came here expecting and hoping to cover. Thank you. Well, thank you very much, actually. I think that we've certainly appreciate, as a committee, your very forthright answers to your questions. So thank you very much for that. Given that that's now the end of the public session, I'm going to call a short recess. Five minutes are so in order to allow Professor McLean and indeed the Fisher report in public to leave.