 Fygoeddo. Welcome to the sixth meeting in 2019 of the Finance and Contitution Committee. The usual story of my mobile phones please, make sure that they are not in a process that will interfere with proceedings. The first item on agenda this morning is to take evidence on the Round Table format. Today's round table will focus on Scottish VAT Assignment and we will take evidence from Mark Taylor, who is the assistant director of Audit Scotland. Charlotte Barber, director of taxation at ICAS. John Cullinan, tax policy director at the Institute of Taxation. Professor Graham Roy, director of Fraser of Anders Institute. John Irland, chief executive of the Scottish Fiscal Commission. Dr Paul Matthews, senior analyst at the Office of Budget Responsibility. I warmly welcome all our witnesses this morning to say that I am very grateful for your coming along to this morning to help us with this task, which is on the face of it technical, but when you start digging under a bit, it's quite a complicated area, so I'm sure you're going to be able to help us significantly as part of our discussions this morning. It's intended that this discussion will be as free-flowing as we can make it. We have structured it around three themes, so if you want to contribute, just catch my eye or indeed Jim's eye, and we'll make sure that you can get in to say your piece. We need everybody to contribute as successfully as possible. The two areas will be around how VAT can be assigned to Scotland effectively. There are busness and transparency of VAT assignment methodology and issues around VAT forecasting and risk to the Scottish budget. Now, it's inevitable, we're going to give it a flow across these areas, so if I feel one area is already exhausted as we go through this, then we'll just move on to the next. With all that being said, firstly we will consider the theme on how VAT can be assigned effectively to Scotland, and I invite Willie Coffey MSP to start that discussion. Thank you, Willie. Thank you very much, Bruce. Good morning, everyone. We know that the first 10p of the standard rate in the first 2.5p of the reduced rate are to be assigned to Scotland using a fairly simple estimation of the total VAT take across the UK. A couple of questions is to hopefully start us off. Could the assignment of VAT to Scotland be done more effectively or even more accurately? How much of an influence do you think it could have on policy development by the Scottish Government? If the VAT receipts fluctuate, is it fair to hold the Scottish Government to account when we have no levers to influence any of it? Right. Is there a start for 10? Who'd let you kick off that very simple question? Graham, thank you very much. I was going to break it down a bit if I needed to, but Graham, go for it. Okay, so there's a couple of things in there. Firstly, one of the big challenges we know with assigning the VAT is being able to come up with a robust estimate, and the papers that have been published so far set out one way of doing it, and you're right that it's a relatively simplistic way to do it by basing it on essentially our share of consumption across the UK. The challenge with that is that even if you do that, the potential for variability within those estimates is significant. If you look at the GERS numbers, you're talking about it being plus or minus a couple hundred million either side, just based on statistical uncertainty that goes into that. There's a big question there about the accuracy or ability to get a pinpoint accurate estimate of VAT revenues in Scotland. That comes back to a couple of things. One is the complexity of the VAT system, so it's much more complex than income tax. Secondly, it's much more complex to estimate VAT revenues just given the fact that you're having to worry about what people are spending on a day-to-day basis across a wide variety of products, which therefore means that you can't have the same level of certainty or the same accuracy of estimation that you could do with some income tax. There's a big issue there about the accuracy. There's a question for Parliament to think about taking the level of risk that it wants to take on, given that level of uncertainty that exists purely from a statistical point of view, irrespective of where you think that there's volatility in the economy, but purely on that statistical point, there's a level of risk there. I think that your second question was generally about the ability to be accountable for VAT revenues or to grow VAT revenues. The principle behind assigning VAT was because it is correlated with the strength of your economy, and therefore, when the economy does well, you gain the benefits from doing that and your economy does worse than you are accountable for that happening. I guess that there's two questions. One is about the ability of the levers that the Scottish Government has to control the economy. Secondly, the levers that it has to control changes in VAT revenues specifically. Do we have any levers that can say when trying to encourage people to spend more on certain types of products that are vatable? Do you want to do that? Do you want to try and encourage people to spend more rather than save less? Is that the type of things that you want to do? I think that there's a whole host of different issues in there about the ability to use your levers in a way to try and move VAT revenues. As I said, that then links back into the complexity of how you estimate it. I might add to that that ICAS has said from the outset that you need to be careful about the flow on consequences as to whether it would then lend itself to the Scottish Government looking towards the kind of industries that generate VAT. Food industry tends to be zero rated, so you might not go there. There are those kinds of issues, but I'm not sure that the correlation is that strong between your individual policies and this VAT assignment model. I'd have thought that it's much more broad and it's just the general strength of the economy, assuming that the economy equals spending. I would have thought that there are two aspects to it. One is, if you like, the fluctuations and the volatility that are due to the policy intention. If there is more VAT receipts—let's ignore the estimating problem for a moment—if there are more VAT receipts in Scotland due to the strength of the Scottish economy, that's the kind of intended part that you would get if you had a fully devolved tax. Whether it's fair or not is one thing, but I guess it's more of a political decision. Is that what you want to happen? As far as that is concerned, one thing that I haven't seen anything about—I think that Charlotte just alluded to it—is the distributional side of things. There's a point made in discussion of Scottish income tax about the progressivity of the system. Generally speaking, the belief is that if incomes were evenly spread, there would be more consumption. On the other hand, maybe some of that consumption is things like food that are zero-rated for VAT. I don't think that I've seen an awful lot of analysis as to how those policies might fit together. Yes, if policies are adopted that boost the Scottish economy and therefore spending in a general sense, you would generally expect to get more VAT receipts, but I haven't seen much analysis of it broken down distributionally or otherwise. The other side of it is the whole thing is if you like the unintended volatility, which goes down to the fact that we're having to use a lot of estimates. Rather simplistically, I tend to have the feeling that it doesn't matter so much if it's inaccurate on day one, if it then remains constantly inaccurate in the same way, because you've given up the same amount of block grants and nothing changes. So from one point of view, what does it matter? To be honest, this is not a tax specialist area, this is a statistical issue. I didn't see much in the official papers about the volatility of those estimates, but some of the comments about them seem to suggest that it could be quite volatile indeed. I get the impression that that is an area of risk for the Scottish Parliament. As a committee well-known knows and has spoken about it on a number of occasions, of course, what matters ultimately is the interaction between the UK tax take and the Scottish tax take, however, assessed. There's something about, from a policy perspective, how do those UK policies play in Scotland and how can Scottish Governments and Parliaments match or diver from what the policy perspective is at a UK level? It's the interaction of those two. Clearly, there's no ability to vary tax rates or to vary the tax model, but there are other things that the Governments can do around encouraging economic growth. What there is is a step removed from that to consumption and how that flows through to the VAT model, just to reiterate, and I know that we'll come back to it around the estimation process. That is essentially something new that's brought into how the fiscal framework operates. There's been small degrees of it elsewhere, but essentially this whole adjustment is based ultimately on an estimate, and that is fundamentally different from what the position was before. That raises questions that I know we want to get into about people's trust of that estimate process and the ability of all parties to rely on something that they know as an estimate. The last thing to say is just to emphasise that point about if it's consistently wrong, and that's less of an issue, but it's knowing that it's consistently wrong is the challenge and the extent to which the volatility sits on both sides of the equation, the block grant adjustment and the tax take in Scotland, and how much is that driven just simply to pick up on Professor Roy's point about the statistical estimation process and to the extent to which those two pieces of volatility are correlated or cancel each other out or potentially build on one another. Because the tax calculation basically works out what the UK position is, works out what the Scottish position is and takes away the Scottish position from the UK position, there's a danger of that amplification in this process more so than elsewhere. John, given that the SFC is in a situation where you're going to have to forecast on the back of estimates, if I've got that right in terms of the future and the way this works, how concerned should we be in those circumstances if we'll never have any real outturn figures from a Scottish perspective on VAT? I think there are two dimensions to that question. I think there's first a question which is not for the commission, which is in a sense the political acceptability of that, and people's trust in the assignment data and model. We can talk about some of the technical aspects of that later, but I think there's that political thing, but I think from our perspective as a commission, it's more of a grey continuum. The difficulty that the assignment process assigns for us in forecasting is that you can imagine that you have this underlying VAT revenue for Scotland. That's the conceptual thing you're trying to work with. But because you have a statistical model in its estimation, you introduce noise. That noise is hopefully random, but perhaps not random, so it moves around. The job of the commission will be not just to estimate that underlying trend, but it will be to estimate or forecast what that random error is in the assignment model. That's very hard. The simplistic answer is if you've got random noise and something moving randomly, the best estimate is to just take the average. But on a year-by-year basis, guessing or forecasting what the noise in the model is will be difficult. Given that already the conversation is identifying problems with the estimated model, and you mentioned basing an alternative approach and out-turn data, I'm just wondering from a practical point of view what would be involved in that, because clearly it would have been creating separate Scottish VAT points for businesses. It would have been a major burden on businesses. I haven't done any thinking about what the impact of that would be, what the impact on business would be and what the cost would be of doing that alternative approach. Our membership has always adamantly been against that. You see the kind of problems disentangling with Brexit, pulling Britain out of a European market. We'd have similar problems here if we'd have to set up a separate Scottish VAT network. The whole point of VAT is designed for a single big market, because it goes in and out on a production, adding values as it goes along. In broad principles, members in business and practice would be against the imposition of having to go and do all that administrative work. Even then, I'm not convinced that we should truly nail it down as to what is Scottish VAT. In immediate terms, businesses are under pressure around other elements, whether it be Brexit or making tax digital for VAT. I don't think that they would welcome it at all, if that's any help. I'm sorry. That's the response that I think would be expected from business in these circumstances. My question was not to suggest that we should do that. Should we be concerned that the system that we're going to use, because we can't do it, is it going to be robust enough? Yes. I was toying with that very problem coming down the road this morning, because we wouldn't want to account specifically for Scottish VAT. We see the problems in this model, but I think that the assignment model is better, ultimately, than trying to identify proper true out-turn VAT. I think that we would be understandable to look at it. It would be very burdensome, but you might get a more accurate result. In one respect, I don't know that you would even get an accurate result, because you would need to design rules to say what Scottish VAT is. If you look at where the place where a supply for VAT purposes is deemed to take place, that's something that, even within the European Union, they've redesignated from time to time. When we have the Scottish income tax come in, ultimately, an approach to defining Scottish residents was adopted, which is different from what defines UK residents, if a mixture of practical and other reasons. I don't think that you would end all discussion and say, oh, there's a simple answer, that's a Scottish receipt. You'd have to consider and design a whole set of rules to define what it was and then make businesses do it. It would be burdensome, but it wouldn't end all the argument about whether we got it right. Patrick, I guess that leads into some of the area that you are interested in. Maybe it's not the right time, but if it isn't, forgive me. Is there a right time? I suppose what I'm interested in is why we are doing this. As I was saying to colleagues before the meeting started, I was on the Smith commission. I didn't argue against the proposal to assign VAT, but I was never a wild enthusiast for the idea that simply assigning a proportion of VAT would achieve any actual objectives. I can understand the desire that we had to have a big number in terms of the total proportion of the Scottish budget that came from taxes that were in some way under the control of Scottish ministers or the Scottish Parliament. However, in the absence of the ability to use policy levers to change the rates or bans or to redesign VAT, for example, to incentivise or disincentivise different kinds of consumption, to achieve social or sustainability objectives, in the absence of a great deal of control over things such as how much of the money that people spend goes through shops on high streets in Scotland, as opposed to Amazon accounts, what are we going to achieve by doing this? If it has ended up being more complicated and less convincing than we thought it might have done, should we step back and think again at this point? There is going to be a review of the fiscal framework. When that happens, should we review it as well and decide in the light of what happens over the next few weeks in UK politics, for example, should we decide whether a different approach is desirable or achievable? I think that looking at what the Smith commission was trying to do—I was not so, I will second guess it—was trying to build in accountability for the performance of the Scottish economy having impact on the budget. That is fine in principle, but when you look at the methodology about the practicalities of how VAT will be able to do that, there is so much uncertainty within the statistical measures. There is a big question mark about whether, even if the Scottish economy was changing relative to the rest of the UK, that would be reflected in the VAT revenues that you receive here in Scotland. Given the uncertainty about the calculations, it could well be the case that that does not happen. There is a fundamental question about whether that link on accountability is achievable with the asset nation given the complexities of the process of assigning VAT. The principle about assigning VAT is solid, but the practicalities of doing it, given the costs and the risks involved. There is a second question that the committee might want to reflect on about the transparency of the system. We have already seen the challenges around that with people understanding how income tax works, what happens when the statistical estimates change around income tax, but that is based on a much bigger sample, a much bigger, more robust than VAT. There is a question about transparency around the system for that. The final point about the Smith commission is about the ability to use Scottish Government leavers or Scottish Parliament leavers in order to influence the budget. That would not happen with VAT, not even talking about the ability to change VAT rates, but to be able to trace through changes in government policy having an impact on VAT numbers. There would be question marks around that as well. Does it not also rest on a deeper assumption that at an overall scale, a country scale, more consumption of VAT-chargeable goods and products is by definition a good thing and that what government is going to be accountable for is just maximising that, however? Yes and no. In the sense that if your economy is growing more quickly in the UK, so that you are not changing the levels of consumption or encouraging people to consume more but your economy is growing more quickly, that should show up in terms of your relative performance and VAT receipts. However, you are right. You could be a situation where, for the same level of growth, if you encourage people to spend more and to spend more on certain types of products, save less, then that would boost our relative performance and VAT. Your comment that you make is right in that context. You would be encouraging people to spend more and there is a question about whether that is right. There are various incentives in that in terms of whether there is economic value in getting people to save less and spend more in ill circumstances. In that context, yes. Tom, did you have a comment? Yes, just to pick up, Patrick made reference to the review of the fiscal framework that is due in the next couple of years. In one of the areas that we have discussed previously is whether any consideration should be given to greater flexibility around the Scotland reserve given the volatility of income tax receipts and many discussions regarding forecasts. From correct, Professor Royce, there is a confidence interval around VAT, perhaps a couple of hundred million. If we look at some of the other taxes, we will be discussing landfill tax. I think that the forecast for next year is about 104 million. I think that the cumulative forecast for LBDT next year is about £646 million, but £200 million either way in VAT. There is £180 million committed towards the attainment challenge, £120 million to high scale. That is a significant element of public spending that is up in the air depending upon which way back we seek our forecast to be. If we are to take on that assignment as do we have to re-look at the fiscal framework and the provisions around the Scotland reserve to give greater flexibility, is there enough flexibility in the current arrangements to manage that volatility? I will come in on that. As Tom Arthur explains, it is how all those things interact together that provides the challenges to the extent of the management job. Where all the bets go against you, that could be a sizable number, where things cancel one another out is less of an issue. If all the bets go for you, there is effectively a real challenge as to how you can spend that windfall in a way that does not commit spending that you do not necessarily have the funding for in future years. That is inherent in the system. On the review of the fiscal framework, at this stage we have not had for the big taxis, certainly for income tax or for VAT assignment any sense of the actual numbers and the actual reconcilations that will be taking place. We can talk about it in conceptual terms. We have seen what has happened in the devolved taxis, but as social security plays through, as income tax plays through and as VAT assignments plays through, that is when we begin to get a much more of a greater handle on what is the size of that volatility. It is important that we have a degree of track record to help support that review of the fiscal framework, where we can get more of a sense of the numbers. A challenge for us all at the moment is to get a sense of how much we are talking about. We have some numbers, we have some revised forecasts that give us some sense of that, but until we have more of a track record, that will be a hard thing to do. Particularly in relation to VAT assignment, we have a conceptual model, we do not have the detail behind that model yet, and what we have so far is no sense of how that would play out historically and how much that level of volatility is contributing to that. It is a big challenge. Given the forecasting, with the caveat that there are forecast reconciliations, I think, £145 million, I think, £472 million, would it be prudent perhaps to delay the implementation of VAT assignment until we have actually seen what the outturn actually transfers to be for the previous two, well, this and the previous tax year regarding income tax? We would not have any outturn for VAT, but to get a sense of how the outturn actually matches up to the forecast, to get a sense of what volatility exists with regards to income tax. Will we begin to get a picture of that in July? I think it is from the first figures, whether it is £185 million, I think it was? £145 million. £145 million. We will find out in July how real that was. Sorry, Dr Paul Matthews. Thanks. Just to come in on the point of reconciliations and volatility, there are, of course, two moving parts. You have got the SFC's forecast, which is going to be moving because they are, quite rightly, completely independent of the OBR. Their forecast can move in totally different ways to how we choose to forecast, and then we will be going at different times as well, which can be particularly tricky. Our UK VAT forecast is actually really quite stable. It ticks along quite nicely. The previous year, plus a little bit of growth, is a fairly good predictor of what is going to happen next year at the UK level, but that is mostly because we have not really had any major policy changes. Who knows if there will be policy changes coming up in the future? I think you might have quite a risk if, say, at a UK budget there was a policy event and then it would be required for the next forecast, the fiscal commission to react and feed through. What would that mean for the Scottish assignments, and then those differences could be set at different points in time feeding through so that reconciliation, ultimately, when it happens, when the outturn occurs, could be done on two different bases. The OBR would look at one set of policy positions in the fiscal commission another. If I could just ask a question, and I appreciate how we would only be able to answer this in a technical, from an OBR perspective, but where in the UK to leave the European Union without a deal has been suggested that that would be on a part of what happened in 2008? Now, I believe that the policy response was to be just VAT to 15 per cent. If such a policy layer was pulled, would that make it almost impossible for this year to be an effective year to transition year from butter? Would it make no difference? I think it would be very difficult to know what is going to happen. Just give that as one potential example. Yeah, consumption, but the point is that we have actually had a really, you are completely right, we had a really stable period of VAT policy compared to things like income tax and then stamp duty LBTT, which are chopping and changing a lot the policy design. It has looked very stable and then, as you say, if there was a desire to boost consumption in the short term and VAT was used as a policy lever, things could move quite chromatically. I think that that has been very helpful kick-off in the starter, but I think that we need to start getting to some of the nitty gritties around methodologies. So our next topic for discussion on the robustness of transparency of VAT aside methodology, I invite James Kelly to kick that off. Okay, thanks a lot, convener. We have started to drift into elements of this discussion in the previous section, but I am supposed to put this in context. What is trying to be achieved here is to look at the UK VAT calculation, to calculate the Scottish share and to assign an element of that Scottish share in the Scottish budget. In bearing mind, the budget is important in terms of expenditure, the decisions taken by the Scottish Parliament. It is important that that assignment is accurate. So to try and achieve that, the Governments jointly have produced a paper with setting out a methodology. For that to work, the methodology has got to be robust so that it produces accurate calculations and also has to be transparent so that people viewing the calculations and ultimately what is assigned into the Scottish budget have confidence in that. So what we are interested in in this section is looking specifically at the methodology that has been outlined by both Governments in this paper. How does it serve that purpose? Are they being robust in delivering accurate VAT assignment and giving transparency to try and inspire some confidence in the calculations? Do you want to kick that off again? Okay, great. Thank you. So a few things on that. I think it would be great to see a lot more detail from both Governments about the methodology. You are talking about a 16-page report with no numbers and no estimates of sensitivity about how those numbers are changing or moving around and it is just £5 billion worth of revenues. I think that there is a lot more information about the robustness that the methodology is going into. It is all well talking about a methodology, but what does that mean in practice in terms of the estimates? To give you an example of that, the methodology will work for about 70 per cent of that. That will yield consumers expenditure on VAT and it relies on one survey called the living cost and food survey. To give you an assumption of that, in GERS for the last few years you are talking about maybe 500 households maximum in Scotland being surveyed for part of that. The Governments have boosted it as part of this new methodology. We are probably talking, I do not know exactly, but we are probably talking about maybe 1,000 households max on the living cost and food survey. To put that in context, you have 2 per cent of taxpayers for the survey of personal income. There are about 50,000 versus a few hundreds in the context of the living cost and food survey, so the robustness of that estimate is always going to be much less than it would be for other taxies. That creates problems. Martha mentioned that it means that the spread of confidence and errors are quite large, so a couple of £100 million on either side. To give you another example of that, in the most recent GERS numbers, when the Government boosted the living cost and food survey, in looking at the latest number that it had—that was £1617—that took £300 million out of Scottish VET estimates, simply by boosting the sample. It has nothing to do with difference in performance, but just because you increased the sample size of that survey, there was a vision down of £300 million from the Scottish VET share. I had a vision up of £300 million on the UK share, and that illustrates the potential sensitivity of the methodology to revisions in the analysis that goes into it. I think that there is much more transparency about just the potential for that to be an embedded feature of the VET assignment that would be really helpful. Thank you, convener. Picking up in Graham Roy's points on some of the issues around surveys, the living cost and food survey in particular, and how robust and reliable that information is, given that the transition period is one year only, I am wondering whether survey information based on one year only is the best way. Given that 2019-20 will coincide with other potentially major events around Brexit, why one year would normally be okay, but 2019-20 might be an unusual year? Why not three years? Is there any way for it to be reflective or a kind of rolling survey? The way that it works in GERS at the moment—I do not know if this is the same for this new methodology, but the way that it works in GERS is to use a three-year average for the survey, but that is not for any structural reason other than simply because the quality of the survey is relatively weak in the context of the number, so you have to use a three-year average to give you some degree of confidence in it. I think that you are right. That then opens up some interesting questions. If the whole purpose of this is to improve accountability on a year-by-year basis, if you are using an average over a number of years, then you break that link between accountability, policy changes and actual outcomes, because you are smoothing the actual numbers over a period of time. I think that your general point would be really helpful to see perhaps in the next iteration of material that we see from the Government, just about how sensitive the estimates are to whether you use one year, two year or three year spreads, what happens when you use different numbers of people being surveyed. If you add an extra hundred people or 200 people into the estimation, does that radically change the results? The results from last year suggested that it does. If you add in more households, it will potentially need to lead to fluctuations in the numbers that come out. I think that much more information about just how sensitive this methodology is to the data that you put into it would be really helpful. Perhaps convener that is something that we need to pursue with both Governments the size of surveys but also the time frames as well. I think that the fundamental point that Graham is making is really important that there is a real need for much greater transparency in the details of the assignment model. We have a slight advantage because we had to produce a forecast in December, so we have seen some preliminary versions of the assignment model and the data. In fact, if you look at our December report, you can see that we, in effect, published the 2016 outcome for the assignment model in its very preliminary form, which hadn't been agreed by the JEC. I have a slightly different take on the sampling issue to Graham, and I think that basically this just underlies the difficulty of two people working in the same sort of field getting hands on information and the sense the way we desperately need this published. In our September statement of data needs, which we published, we asked very clearly that the two Governments published the assignment model and the underlying data, and they did this through an official statistics publication which all the users were consulted on, and as far as that hasn't happened yet, and we still don't have published data, although we have some access to some data. My understanding, just to go back onto the technical issue about the survey, my understanding is that the preliminary assignment model just uses one year as opposed to three years, and also that the sample side has been boosted, so it's been doubled, and for the next iteration to be published, the number of households responding will go up to 720. In the current version, which is published in 2016, data is 360 households, which is 0.03 per cent of the number of households in Scotland. The issue, I think, in the 2016 data wasn't so much the sample size has increased, but they have been in sample variation, so because the sample size is relatively small, there had been an estimated switch between zero-rate and exempt goods, and reduced rate and the standard rate, and that caused that drop in the A2 receipts. All that underlies the volatility issue, as we think that that was a rogue drawing of the sample, and also the need for much more transparency and a publication of the underlying details of the assignment model. Did you say that the boosted number was 720 respondents? You get about a 40 per cent response rate in these sort of surveys. Are we basing our outcomes on what VAT estimates are on 700? No, not quite. What you're doing is using the information from those 720 households to estimate a split between exempt, zero-rated, reduced, and standard-rated VAT. Once you've got that weighting, you can then use the consumption data to estimate VAT. It's a detail, but it's an important detail. It's a small number, but these sort of survey, and again, this is the need for transparency, don't be thrown by the small sample size necessarily. We don't know. What we need to know and what I think we need to have published are the confidence intervals around that sample size. Can I just... We are all thrown, I think, by that small sample. So why shouldn't we be thrown by... Why should we feel reassured? That's a big enough number. Okay. Why shouldn't you feel reassured? Because the statistical properties of samples don't vary in size with the sample size. They vary in the size of the square root of... the inverse of the square root of the sample size. Which means... Which means crudely, as you increase the sample size, you get diminishing returns to scale in the increased accuracy. And as you reduce the sample size, you have a smaller size, you get... You clear? That's me going. Sorry. Mark, I've got a couple of politicians you want to come in as well. Just on the transparency point, and you'd expect me to say... The transparency is really important there, and there's two elements to that. One, the one we've talked about, which is effectively publishing workings and the detail so that people can understand the detail, but there is something, and it's apparent already for the methodology, there are lots of details. And we've been talking about one of them, and there are many, many more which are built on ONS estimates, Scottish Government economic estimates, which themselves are built on lots and lots of details. So transparency, as far as all that workings is really important, and it absolutely subscribes to that, but also transparency in terms of interpretation and assessment of things like what's the extent of risk in there and a professional assessment from those that are putting the estimates together about the extent of that. Now, we've got some of that at the detailed level for the things that this is built on that are already published by ONS and Scottish Government. It's how does that get aggregated up so that transparency includes a sense of the things that we were talking about today. Neil? The document talks about unregistered traders, and obviously there could be a variance between Scotland and the UK on that. I wanted to ask about the impact of the cash-only economy, and whether that is higher or lower in Scotland than the rest of the UK, and could that have an impact on the VAT figures? Sorry, Willie. I just wanted to ask, I mean, businesses are VAT registered in Scotland. Is it not a safer methodology to just pick out the actual location data that's coming in much in the same way as we did with the income tax thing, which we all know was a great success in estimating the numbers of Scottish taxpayers? Shall we not do that? Charlotte might give you a reason why that wouldn't work, or maybe I'm wrong, but... It's more difficult than just individual traders, because you could have large businesses here, but there are businesses everywhere. So what's Scottish? You could have lots of consumers buying stuff online. Do you know where are they buying it from? What are you trying to measure? The whole concept of VAT is that it kind of sits around a business, so any business that you have will have input tax coming into the business and output tax coming out, and you marry one against the other, and in a production line, say, you have half a dozen processes in, out, in, out, in, out. Of course, you know how much business goes across the UK. We've had discussions here before. It's quite difficult to pin down what is Scottish VAT, and I think it's partly that we were talking about it earlier, you were as well, John. It's actually quite difficult to pin down that model as well. Do you know that both models have their difficulties? Can I just bring those back to an area that I've got concerns about? It's just the very fundamental point about if we put rubbish into the system, we're going to get rubbish out. So, from what I understand, in broad terms, this model for VAT assignment methodology works by applying VAT rates to the estimates of expenditure in Scotland through something called the total theoretical liability model. Now, I haven't seen anywhere in any of the descriptions from the Scottish Government or the UK Government about how successful that model actually is. If that's the model that's being calculated for the amount of the VAT, and maybe Paul can help us here, how is the output from that model compared with the actual outturn itself? Does it work? That's what I'm saying, because there's no evidence at all in any of the papers that we've got about how robust that model is. Yeah, so I'm very happy to comment on that. So it works in that our forecast for VAT, for the purposes of the way that we're using it, which is forecasting, it's okay, but— Wait a minute, wait a minute, it's okay. What does that mean? None of our forecasts are 100% accurate, and our VAT forecast at the UK level has lower errors than our average tax forecasts. So our UK VAT forecast is doing okay. So that's a forecast compared with outturn? Yes, exactly. So the key thing is— So tell me what is the variable in the other one? What is the difference in that okay? I'm going to answer a slightly different question, which is, we know the money that's come in to HMRC, the receipts. That is a definite known outcome. The VAT theoretical liability model is estimated, built up from a lot of ONS estimates for consumption, then there's various judgments made about what is standard rated and zero rated, similar to the assignment methodology, till you get this total theoretical liability. Now this number is a lot higher than the actual VAT receipts. The difference between the two is known as the tax gap, the VAT gap, it's a residual. There's lots of things, we don't know what is going on in there. It could be genuine aspects of the tax system which are just not captured in the theoretical liability model. So for example businesses under the threshold, but it could also be things like avoidance and evasion that's taking place. This VAT gap is quite large, we're looking at sort of around 8%. For the forecast, this is not such a problem because, as we were discussing earlier, if the error is constant going across the forecast, and this is what we assume with the VAT gap, other than some measures, policies that the UK Government have introduced, generally it's doing an okay job holding it fairly flat. But it is a large difference. I guess the real question for us is, because this is a good tax based on relativity between the two jurisdictions of what will happen to the Scottish tax gap and how do we know what that level is going to be compared with the rest of the UK? We will never know, because we will never have Scottish receipts. The difference is between theoretical liabilities and receipts, which is known at the UK level, and we'll just have theoretical liabilities in Scotland. Could I just help and just by giving some numbers perhaps, numbers can tie things down. Just in, we've been looking at the OBR's forecast and record of this because it obviously gives us a sense of how difficult this is going to be forecasting Scotland. If you go from the March 2012 forecast onwards, and they're two year ahead forecasting error has been about 2.4 per cent on that bridge, which Paul is absolutely right, is lower than a similar sort of thing across all their fiscal forecasts, but it's higher than the income tax forecast error, which is about 1.7 per cent. So you can get a sort of sense of the relative magnitudes there. Just on that vagap point, Paul, my understanding is similar to Paul's, and certainly when we're doing our forecast model for the assignment model, we use the UK data on the tax gap to move from the Scottish theoretical tax liabilities down to our forecast of Scottish receipts. I just want to say a bit about the difficulty of working out even conceptually, where that belongs. Last night, I paid a hotel bill and there would have been that on it, so that hotel consumption took place in Scotland, but I'm not a Scottish resident. If I'd come up here for a kind of ordinary run-of-the-mill business and that that was quite properly reclaimed, it would depend on the outputs of that business, what took place where, and if you just took the receipts from the hotel chain. First of all, if they'd got hotels in England, they'd have an extra job of work to break down what took base in Scotland, but even then you wouldn't get to, in quotes, the right answer. Here we've got forecasts, which possibly stand to be corrected by outturns, but we've also got estimates for things where it's not even totally clear what the conceptually right accurate answer is. That's the difficulty, I think. The question is about the tax gap issue that we're talking about here. In terms of the methodology that's going to be used for Scotland, frankly, I just don't know, will it just be assumed that the tax gap is the same in Scotland? I think that because it's going to be based on the theoretical liabilities in Scotland and theoretical liabilities in the UK, it effectively ignores the tax gap. There are many reasons why the tax gap could change in lots of odd and strange ways, but I don't think that that would be a particular problem for the assignment methodology. Okay. Any other questions before we go in this area before I ask another one? Just before we move on from methodology, I noticed that, in the paper produced by the Governments that said that the Scottish VAT assignment model is similar to the VAT revenue sharing arrangements that the Canadian Government uses with some Canadian provinces. I just wondered if any of our guests knew which Canadian provinces and how that was working out, and whether the affected Canadian provinces were happy with arrangements. How robust is that in Canada? If it's going to affect what we do here, we need to know. I should say that we are just a bit of context. We are publishing a new forecast this afternoon, so we have been very busy. I read this on the train up, and that exact point on the Canadian provinces was very interesting. I made a note to look into it, because clearly looking at international context is really useful for these things. This is what we've tried to do on income tax to understand behaviour changes, looking at different states in the US, different parts of Switzerland, and so actually looking at some international comparators, that would, I think, be a potential good step forward for this work. I just wondered if anybody here today knew anything about the international comparators. Graham, how we go? A couple of things on the Canadian example. One thing that we have to be slightly careful is trying to read direct comparisons between models that are contextually quite different. So in Canada, for example, yes, there is a common, in most provinces, there's a common GST sales tax that goes across all of Canada, but in many cases it's on top of an existing domestic sales tax that is levied within the province. So they actually have outturned data, which is helpful as a benchmark that you can compare against what's happening at the federal level. I think that two other points about the Canadian example. One is that the quality of Canadian data is actually really quite high, so they produce lots of economic accounts at a province level, which we don't have in the Scottish context, so they produce, effectively, national accounts in many cases at a state or a province level, which then gives you much more accurate information that you can then draw on and also compare against in these instances. Again, the third thing, which again gets back to maybe one or more principles about on the Smith, where Smith was heading in this, in Canada, my understanding of what the system in Canada is really designed to do is to try and be used as a way of helpful equalisation or at least providing a mechanism to increase the share of Canadian tax revenues that flow amongst the states. Whereas it's less about giving accountability to states to try and grow their economy in that context, it's more just about assigning just a portioning revenues across the Canadian provinces that is something that's set at a national level. Whereas the context here is quite different, it's about injecting accountability into the Scottish budget and not the other parts of the UK, whereas in Canada, it's quite a different system. Would it be quite good to look at the data that is produced elsewhere and how that's configured and how the evidence is found, if you like? I appreciate entirely the point about context, politics and systems are different, but in terms of the range of information and how they gather, information might be helpful. Where the whole fiscal framework is getting into is the UK system for the public finances and statistics, robust enough to cope with devolution and transfer significant public finance powers to devolved Governments. When you look at places such as Canada, they have much more information and much more accessibility to data than we do. It's more a question about whether the pace of change in the statistics and the data that's available in the UK is able to cope with what's happening in the wider fiscal framework context. Is it appropriate to use that Canadian model and apply it here? Is it safe for us to be doing that? I think that my understanding of what they're talking about in the paper is that the idea that you're portioning the national tax based on consumption within provinces is the general principle of what happens in Canada. I guess that the question then comes into no one. From this context, that would seem to be the sensible way of doing VET. The question is, do you have sufficient robust data and information that lets you do that in a way that can minimise the risk and give you the confidence? I have one more question before we move on. I just don't believe the numbers, so I just want to make sure that I'm not... I might be wrong. I'm quite often wrong, so let's just test it. One of the things that will obviously apply to this system is domestic tourism in terms of the VAT that we're taking from it. According to page 436 on the annex to the Scottish VAT Assignment Summary of Assignment model, according to the 2014 tourism figures, English and Wales visitors spend about £1.7 billion in Scotland and Scottish visitors spend about £1.2 billion in England and Wales. Now, intrinsically, that just seems to me cannot be right, with a population of £50 million spending £1.7 billion as a population of Scotland of £5 million spending £1.2 billion. I have my sense that that figure needs to be challenged, right, or is it... Is anybody got any thoughts? If nobody's got any thoughts, I'm going to challenge the figure, because it just seems to me intrinsically that it can't be right, but there's nobody to respond into it. On you, Paul. I'm not going to comment specifically on that figure, but there are a lot of other data sources beyond the living cost and food survey that are mentioned in here. I think that one of the things that you probably want to be concerned about is how constant... Maybe that's right, maybe it's wrong, but actually if the error is constant, then it shouldn't be a problem, but if this is going to chop and change and the 2014 tourism figures are replaced with a new survey and they use a different methodology and suddenly those estimates change dramatically, then obviously that will feed through the assignment model and be quite different. There are other aspects where the methodology is using things like the annual survey on hours and earnings, which is an okay data set, but we're looking into a much better one, the real-time information from HMRC, and if you are moving the data sources, then you're potentially moving the error and thus the assignment. I would agree with what Paul has said, and I think for what it brings down to me is, again, we need much more transparency about what the assignment model has, the data sources that it uses and also an understanding of the variability and the methodology of those data sources. In a sense, the sooner that the two Governments can publish something, I think that everyone's interests will be well served. We'll move on. Our final theme on this is issues around VAT forecasting and risk to the Scottish budget. I guess we're into that area anyway, Murdo, but do you want to take it on, please? Thank you, convener. We've obviously already touched this to an extent, but maybe I could start off by asking about VAT forecasting and maybe come to you first, John, saying that you've got that happy job in your course. I'll try and address. My two starter questions, which overlap, are how does forecasting VAT compare in difficulty to forecasting income tax receipts, for example, which is something else that the Fiscal Commission is doing? Secondly, what is the short-term risk from forecasting error and how might that be managed within the context of the Scottish budget? Is it more difficult than income tax? It's just different. Mechanically, it's quite straightforward because we have an economic forecast. We take the components of that economic forecast, particularly consumption. We put them into a relatively simple model and that comes the result. Whereas the income tax model is much harder because it has much more data, it has a very large sample size in relative terms and we have the individual micro data. In terms of technical difficulty, forecasting VAT is relatively straightforward. The question probably wasn't about that, though. It was more about how difficult in conceptual terms and how difficult it is to get an accurate forecast. The experience of the OBR is interesting. Even with a richer dataset and out-turn data, the forecast area around VAT is larger than the forecast area around income tax. As I said before, the real conceptual problem for us is not so much estimating what VAT assign levels are. It's because the surveys that are used in the assignment model are volatile and we have to predict the volatility in those surveys. That's really hard. The example that we already have to hand is the December forecast that we published, where we think that there's a rogue drawing from the Living Costs and Food Survey for 2016, which suppressed consumption. That caused the forecast to be lower. However, how do you make that judgment? Because the reconciliations will work around that rogue drawing, how do you, in advance, guess what that rogue drawing will be or forecast what that rogue drawing will be, and that's tremendously hard. That's a forecast difficulty. What impact will it have on the Scottish Government budget? As yet, because the Joint Executive Committee hasn't made a decision, we don't know what the arrangements are, but let's just assume that it's like income tax. If you think about the data that goes into the assignment model, it's the UK Blue Book is particularly important because of the outcomes at the REUK level, the rest of the UK level. The lag suggests that it's a similar sort of lag to income tax, so 18 months to two years. If you think about income tax and the sort of reconciliations issues with that, then I think you've got a good idea of the sort of mechanics of the issue, or the likely mechanics of the issue. In terms of the numbers, we took the OBR average forecast error of 2.4%. If you applied that to the magnitude in Scotland, you're talking about an average error of the low hundreds, somewhere between an average error of approximately between say 100 million and 200 million, but that's just an average error. That gives you a sense of the potential issues. Now, as everyone keeps saying, the reconciliation depends upon the combination of forecast errors for the Scottish VAT and also for the rest of the UK, so it's not just that number, but you're talking about something that is not as potentially large as income tax in sort of magnitudes, but pretty chunky. Paul, do you want to reflect on some of that discussion? I think I would just bring in the point, when we say forecast error, this isn't necessarily a bad thing that we would be, you know, we got something completely wrong. It can actually be just policies that we weren't aware of at the time, so if there is a policy change, then our forecast for the year ahead is obviously going to be different because the policy has changed. VAT has been relatively stable as a policy area compared to, say, income tax, which has been changing all the time, and if we have sort of these policy errors and trying to understand policy costings, as we know from income tax, that's pretty hard understanding what the behaviours are going to be around this, that could add another slice of variance on top of your existing economy forecast error plus the sort of methodology error of trying to get to your survey, and then you've got policy costing error on top of that, we could have quite wide confidence in that force. Getty, about Dangella's point that she made earlier, if we're now talking about risk to the Scottish budget, to reduce that risk to the Scottish budget, would it be advisable, the right thing to do to spread this over a three-year period instead of just taking an encapsulation of what's going on in one year to make sure that we're reducing some of that risk? I don't know if that would be the right thing to do or not, but I'd like to get some reflection. The division of labour between the commission and the government series is very clear. It's our job to do forecasting, their job to do the assignment model, but as a statistical thing, if you pool sample sizes by averaging over years, you increase reliability. Increase reliability, so that we reduce the risk. Just picking up on that point, two things. What would be really helpful to see is within the numbers what happens when you start to do that, so does it reduce the risk? What happens when you boost the sample size, do those numbers move around a lot, and then you get an idea back to whether the Scotland reserve mechanisms are the borrowing power sufficient to cope with the level of risk that exists within that? It would be really helpful to see in the next iteration of the papers that the Government has produced just how robust the numbers are to that, but that gets back to a more fundamental question that you again may want to reflect on, that if you're having to rely on three-year moving averages for the VET revenues, if you're having to worry about smoothing them over time, how does that get back to the point about accountability for why the policy is being assigned in the first place? The idea is that you should be held accountable for movements in the budget. That happens with income tax, so you can then see if your outturn is lower, therefore you have to pay back the forecast, but if you're smoothing out the VET revenues for three, four or five years, all you're really doing is replacing Barnett with something that's very similar to that, but it's just smoothed over a number of years, so you break that link between the accountability and the policy implications. On the point of the sample and how it plays out, something that I would quite like to look into, or potentially the Governments to look into, is that outliers in the sample and how much they matter. Because we've got 720 households, let's remove the most extreme spending one. How does that change the share? How would it feed through? What would that be the end result in terms of Scottish VAT? We're talking about £5 billion here. Would the change be tens of millions, potentially from individual households who happen to be at the extreme of the sample? Just so that we can sense in future years if the sample does change, you get more of these outlier households coming in, how much effect would that have? On Murdo Fraser's second part of his question about how manageable is all of this, I think that for me there's two aspects for this. One is something that we've said in Audit Scotland for a while, is what's the Government's policy particularly around reserves but reserves and borrowing and how does it set up that policy in a way that enables it to cope with whatever the uncertainty that's around at the corner and is something important in there. There's then the question of that, when adjustments are made, how do spending plans cope with that volatility, which inevitably will be there? To what extent does everything get baselined and then you're storing up problems for the future? Or to what extent are you able to modify your spending plans to cope with that volatility in a way that's more of a short-term response? Key to that, and this is where the real challenge is, is knowing what's a permanent and recurring effect because it's due to the economic performance and what is just to use John Ireland's term noise in the system and that's a real challenge for Government and for Parliament to get a sense of those two components but it's something about what, how does financial planning, how does financial management get ready for the inevitable volatility that's around the corner? Mark, you also talked about transparency and one of the things that we got from David Eisner in his paper to us was around that issue. I could just pick out some of the words there, just something else that I think we need to reflect on. The paper states that the model calculates that the VAT occurred in each of the sectors at UK level is done by using confidential HMRC data and intermediate consumption figures but gives no further information on this and then says that the Scottish share of VAT in UK exempt sector is calculated using market data. It's all complicated but effectively a lot of this is being done on confidential HMRC data which we do not get access to. Do you think that this is appropriate in the way that we're trying to design a new model for Scotland? I think that what I'd say is that absolutely I'd recognise that there's likely to be an appetite from Parliament, from the Scottish Parliament, about a degree of independent assurance over some of that and some of those calculations. The questions then rise about how auditable is that and how, given all the problems that we've talked about or the challenges that we've talked about today, to what extent could you have an audit of that information in any case that would give you that degree of assurance and then in the context of the recent agreement between the two Governments about how Audit and Accountability works with the recent Audit and Accountability framework, to what extent our Audit Scotland and National Audit Office are able to sit down and design and agree an approach that both gives the assurance that you would be looking for and is workable given the complexes that we've talked about. I think that where we start from is we recognise that as a legitimate question and one that we need to explore but I think that we also recognise the challenges in trying to give you, I think, the assurance that is inherent in your question there, convener, but it's something that we're setting off to try and explore and try and understand. Having more information about how it works and where the numbers come from is a necessary part of that. I just wanted to say that on behalf of HMRC, they have a statute that governs them that basically makes it a criminal offence to reveal anything to do with particular taxpayers affairs but also anything that could be reduced to that and quite understandably, therefore, they have an institutional aversion to risking that. In fairness to them, I can totally see that the issue doesn't make for transparency or your accountability or for being capable of explaining very easily to the public, particularly if there are adverse changes that are going to cost them more taxes. To solve that problem at the root, one would need to look back to that criminal statute. It's not just a question of HMRC behaviour, as it were. That's a good balance. Thank you. I mean, just to say on that point, we regularly bump up against that taxpayer confidentiality issue when we're trying to understand why forecasts have changed and HMRC are very robust in defending taxpayer confidentiality. But I think there could be something which is quite tricky here, say if outturn changed quite dramatically and then the answer was, well, something's happened in some confidential data and we can't tell you what it is but you've got less money or more money. That might be quite a hard message to explain or sell. Having again, because we've had access to some of this data to do our forecast, I think that once the Government has published the details of the assignment model, you'll get a very clear sense of where the confidential data is and how important that is. I would just delay answering that question really until you see the details of the assignment model. So we're going to get more than just the summary that we've been provided with already. So my understanding is that the two Governments, well, I know because I've been invited to it and I'm going, and there's a workshop in the end of March where the Governments are going to discuss the assignment model in more detail and I understand that they're hoping to publish something in the spring. Is that a workshop just involving Governments and the agencies around about them? No, no. I think that it's an open workshop. Murdo, do you have any more questions on that, Eric? One final question just from around the risk issues just to finish off this session then. What can the Scottish Government do to increase the growth of Scottish VAT relative to the rest of the UK VAT revenue? What model of things can we do that we can increase the amount of VAT in Scotland? That brings us back to the very beginning, really, because what's the point if I don't know when you're leaving us? Any suggestions? Graham? I think that this gets back to the conversation with Patrick Harvie earlier. The whole purpose of this from the Smith commission was the idea that you can better link the tax base in Scotland and therefore the Scottish budget to the performance of the Scottish economy. All else remaining equal, if the Scottish economy can grow more quickly than the rest of the UK, then our VAT base will grow more quickly than the rest of the UK, all else remaining equal. That therefore building that accountability in Scotland would gain the benefits from that and of course if it grew less quickly then we would take the risks. I think that there's then questions about how you do that and I guess the debate about how you grow your economy more quickly and all the issues that go with that. There is also the subtle point that for a certain level of growth one way you could would be to try and get people to spend more money on vatable goods or to increase the proportion of income of people who spend a lot of money on vatable goods. Again, a bit like where are the income tax, where there's incentive to grow higher-rate taxpayers because they pay more income tax, you would have the incentive to encourage faster growth in people who would spend more money essentially if this mechanism was to work. That then gets into some interesting questions about whether you want to do that and whether that's consistent with ideas of inclusive growth, whether that's consistent with ideas about shifting money away from consumption to savings and vice versa, but I think it's quite hard to see unique levers where the Scottish Government could do something that would boost vat in a particular way other than trying to just grow your economy more quickly. I just want to contribute at this stage. That's the case. I want to thank you for your contributions. I think that it was a very useful beginning to our discussions because I get the sense that this is only the beginning and hopefully we'll inform our sessions that we have with Government in the future. I think that it certainly helped us greatly to understand some of the challenges that exist. There's a lot of work to be done. That's pretty obvious. Again, thank you, though, but I suspend at this stage to allow a change over our witnesses. Thank you very much. Colleagues, our next piece of business is to consider subordinate legislation relating to the Budget Scotland Act 2018, amendment regulations 2019 draft. We're joined for this item by Kate Forbes, who's the Minister for Public Finance and Digital Economy and the Scottish Mackay, head of finance co-ordination in the Scottish Government. Before we come to the formal consideration of the minister's motion, we'll take evidence on the Budget Scotland Act 2008 amendment regulations 2019 draft. I welcome our witnesses to the meeting and invite Kate Forbes, the minister, if she wishes, to make an opening statement. I will, if that's okay. Thank you so much, convener. The spring budget revision provides a final opportunity to formally amend the Scottish budget for 2018-19. This year's budget revision deals with four different types of amendments to the budget. Firstly, a few funding changes. Secondly, a significant number of technical adjustments that have no impact on spending power. Thirdly, some whitehall transfers and finally, some budget neutral transfers of resources between portfolio budgets, including a modest budget redirection to ensure that we maximise our available budget. The net impact of all those changes is an increase in the approved budget of £3,576.2 million from £40.505 billion to £44. You've got the numbers in front of you. I'm not going to read them out. Table 1.1 on page 5 of the supporting document shows the approved budgets following the autumn budget revision and the changes sought in the spring budget revision and the supporting document to the spring budget revision and the brief guide prepared by officials to provide background on those changes. The first set of changes reduced the budget slightly by £3.3 million in comprises funding, which has been allocated over a number of lines as detailed in the brief guide offset by the repayment of £175 million of farmers' loans FTs. The second and most significant set of changes comprise a number of large technical adjustments to the budget. Those technical adjustments are mainly non-cash and therefore budget neutral, as they cannot be redeployed to support discretionary spend elsewhere and have a net positive impact of £3,303.8 million on the overall aggregate position. It's necessary to reflect those adjustments to ensure that the budget is consistent with the accounting requirements and with the final outturn that will be reported in the annual accounts. By far, the largest of those adjustments relates to an increase to the AMI provision for future NHS and teachers pension costs. That flows from the outcome of appeal court rulings on the judicial pension scheme and firefighters pension scheme discrimination claims. The ruling has significant implications for future costs of unfunded schemes and we've had to adjust the non-cash AMI budget by £2.3 billion to meet the potential future costs of remedy. The UK Government is expected to appeal this case further, and while the final position is not likely to be resolved for some time, that adjustment is made on the basis of legal opinion on the probability of a successful appeal. With regard to Whitehall transfers and allocations, there is a net positive impact on the budget from a number of transfers of £275.7 million, the most significant of which are the transfers of £157.3 million from the DWP to fund the devolved responsibility for carers allowance and £78 million from Treasury in respect of the agenda for change health pay award. The final part of the budget revision concerns the transfer funds within and between portfolios, which committee members will be familiar with, to better align the budgets with profiled spend. The main transfer between portfolios are noted in the supporting document and the guide. As we approach the financial year end, we will continue in line with our normal practice to monitor forecast, outturn against budget and wherever possible we will seek to utilise any emerging underspend to ensure that we make optimum use of the resources available this year and manage the necessary carry forward to meet additional spending commitments, as disclosed in our draft spending plan. Finally, as I draw to a close in line with the recommendations of the budget process review group's recommendations, my officials have included in the brief guide sent to the committee an indication of the forecast outturn position at 31 January. That is the latest position available when the brief guide was prepared and, hopefully, has given the committee the best view of the emerging position. Provisional outturn figures will be announced by the cabinet secretary in early June. I hope that colleagues have found the guide helpful, although, as always, we are open to suggestions for ways to make it even more helpful. Maybe I could set off there then, because it was helpful that that was introduced as part of the budget process review group and information is now available. I appreciate, minister, that it is still a challenge for the committee in whether it is an area of underspends or Scotland reserve, given that that is a continually moving picture. We appreciate that. Therefore, in order for us to carry out our scrutiny rule more effectively, when the final balance of the reserve, for instance, is published in June, we will not be able to provide the committee with a table detailing all of the movements in the reserve throughout the year to make it more available to us in terms of our scrutiny role. Although I recognise that more movement yet may transpire, I am in a position at this stage to see how much reserve will be available to meet any of the potential shortfall that may come from the reconciliation process in relation to income tax and the fully devolved taxes once the outturned figures are available in July, because that will be an important moment. Yes, in principle, I am very happy to provide the committee with those final figures in terms of getting your head round the figures in front of us. Obviously, members will be aware that that position changes quite regularly, so we were keen to give you the most up-to-date position, but recognise that it means that reconciliation can be a challenge. I am very happy to commit to providing more information. You mentioned the point about devolved taxes and the need for reconciliation. Obviously, with the first reconciliation due next year, as part of the forecast for the dual balance, it states that devolved taxes income forms £136.2 million to ensure that we are prepared for reconciliation. So, it is £136.2 million in the reserve to put forward that purpose, and if it turns out that the estimate is correct. That is the forecast. That is the forecast of the surplus tax receipts that we are holding in the reserve. That is quite clear. Thank you. Thank you, minister. Just a follow-up to the convener's question, I am looking at the forecast residual balance for 2019-20, which is in the table on Annex C from the brief paper. There is a balance forecast of £300.2 million. Broken down, you just explained £136 million that is there for devolved taxes income. There is £85.5 balance set aside to fund spending commitments. Can you just explain what those spending commitments are? Yes. Those spending commitments could, for example, be used—members will be aware of recent coverage on things such as teacher's pay. That balance that is set aside for future spending commitments will likely be used to support known pressures, for example the teacher's pay agreement. There is not a specific ask that this is tied to at the moment. It is more of a kind of float that is there to deal with things that might arise in future. In the spirit of managing our finances prudently, yes. We recognise that over the course of the year there will be emerging challenges that need to be met, so that is reflected in the residual balance. Okay, thank you. Just one more question, convener. The figure for financial transactions is £78.5 million. Where does that arise from? Is that Treasury Barnett consequentials? So the financial transactions there are in relation to—well, that is after allowing for additional spending commitments of £47.5 million in 1920, and that is— Principally the repayments of farmers loans that are contributing to that balance. Okay, and presumably there are restrictions by what that money can be used for. Yeah. Activities in hand to explore how we can best deploy that in 1920. Okay, all right. Thank you. I was going to the questions. James? Yeah, thank you. Just a small question on the residual balance in the reserve before I go on to another question. This £300 million figure, does this include the £54 million that was transferred to the reserve as part of the information that was released by the Cabinet Secretary at stage 3 of the budget? So the £54 million, so in terms of the follow-on from— So that is the balance of the additional consequentials that were received very late on as a result of the UK supplementary estimate. So what we've tried to do in that table is show that net position so that £300 million does effectively include that remaining balance. Right, okay. That's clear. So whereabouts is that then? It's obviously included in one of the other lines. Well, you can see in the table that was provided there, we come down—when we look at 1920 spending commitments, we've got the expenditure commitments reflected in the budget bill of £313.5, then the stage 2 additional funding of £94 gives you £407.5, and then we've brought in there those late additional budget consequentials in offsetting those additional spending commitments, so that just leaves the pot at the bottom. So we've brought the full amount in there in arriving at the remaining balance that's available. Sorry, so I'm not trying to be awkward. I'm just trying to understand it. So where you're saying you've brought that amount, and where is the £54 million included in what table? Well, it's the difference between the £148 that's brought in there, and the £94 that's already allocated. So the balance of £54 is effectively supporting wider expenditure, but the way it's reflected is that it's just coming down to be reflected in the reserve. It's been deducted from that line. Right, okay. That's fine. In terms of the portfolio movement, so it was a £43 million underspend in our credit in relation to capital housing receipts, have you got the background to that? So that's income from repayments of FTs for the year, which are shown as negative expenditure, and so they look like a reduction in expenditure, but they're currently estimated at £43 million, which is made up of £29 million of shared equity receipts, £10 million of charitable bond receipts, and £4 million of other FT receipts. Estimates for the year have been made from the various trends to date and known scheduled repayments. Sorry, so why is that a negative then? Because it's income from repayments of FTs, so it's shown as negative expenditure, so it looks like a reduction in expenditure, but it's presentational in terms of being income from repayments of FTs. Right, sorry, I'm not trying to go forward, but if it's income, I'm still not quite understanding why. If it's income, why has it been shown as negative? That's just the way that income is shown. It's reflected as a minus in overall, so it supports the idea that the income offsets expenditure, so it allows further expenditure. And in terms of the enterprise budget, there's been a £56 million underspend there that's been released elsewhere. I'll go into detail about that. That's a mix of different transfers. If I go through it, I'll finish the aspect. The resource reductions are as follows. The £6 million underspend in relation to enterprise zones, which is a result of the full construction and operation contract not being expected to be in place between Scottish Enterprise and Strathclyde until after a stage 4 review in late January 2019, which was later than initially forecast. That's offset by £5 million, which was deployed to the innovation industries-level programme to cover the costs of fire recovery fund. In terms of the capital, which is again financial transactions, there are reductions of £55 million, which is a release of £28 million of emerging or planned underspend in FTs in relation to Scottish European growth co-investment project and the European investment fund. There's another £25 million of FTs, which is a transfer to communities and local government portfolio in relation to investment into the Scottish partnership for regeneration of urban centres, or SPRUCE, as it's more commonly known. That's a transfer of funds for the first building Scotland fund investment into SPRUCE, which is seeking to complete a number of commercial and industrial property investments using that investment this year. I just wanted to ask about the £3 million receipt from the UK Treasury for the Presidential visit on policing. Is that a fair and accurate estimate of the cost of the policing for that visit? Do we have to claim that? I remember a bit of a discussion about whether Police Scotland was able to reclaim that money or not. Obviously, a white hole in the UK Government Treasury transfer to cover the cost. You'll remember some of the noises that were at the time of the Presidential visit. I can't remember which cabinet secretary it was at the time, but I think it was who'd written to see that that scene was not on the Scottish Government's invitation that it should be covered by a Treasury, so that transfer recognises that commitment. There will be no other questions. We now move to item 3, which is consideration of the motion. I invite the minister to move S5M-16046 that the Finance and Constitution Committee recommends that the Budget Scotland Act 2018 amendment regulations 2019 draft be approved. Do members have any other comments? No other comments. I put the question. The question is that motion S5M-16046 be agreed. Are we all agreed? We are agreed. I thank our witnesses for their contribution. It was a very useful discussion. We'll now publish a report to the Parliament setting of a decision later today. I think that we'll just continue, but we'll allow a changeover of Government official. Very neat transfer. Our next piece of business is to consider the subordinate legislation relating to the Scottish landfill tax standard rate and lower rate bracket. We are joined for this item by Kate Forbes again, the Minister for Public Finance and Digital Economy and Ewan Cameron Nielsen, who is the finance director of the Scottish Government. Before we come to the formal consideration of the minister's motion, we will take evidence on the Scottish landfill tax standard rate, lower rate order 2019. Again, I welcome our witnesses to the meeting and invite Ms Forbes, the minister, to make an opening statement as she wishes. Just very briefly, the order specifies revised amounts for the standard rate and the lower rate, so standard rate, the figures will be in front of you. Those proposed rates would come into effect from 1 April 2019. The committee members will wish to note that they match the UK landfill tax rates from 1920 as set out in the UK Finance Act 2020, because we want to avoid any potential for a new phrase that I learned this week, which is waste tourism, to emerge as a result of material differences between tax rates north and south of the border. I am interested in some of the underlying policy and rationale in relation to the taxation rates, but I suppose that the qualifying materials, which I appreciate, are a sister regulation. There is a recycling business in my constituency called Brewster Brothers. They are an aggregate supply company, so they recycle and reuse things like waste soils in their waste. That attracts the lower rate of landfill taxation. I suppose to put it simply, how I understand it, is that dumping soils remains, in the words of this company, too cheap, and that the lower tax rate applied is contradictory to the circular economy, in that it is easier to dump soils, which I appreciate are less polluting than other materials. However, bearing in mind, we do not want to be wasteful of any resource, which is quite unusual for people to be arguing for increased taxation. I wondered whether the Government had any plans to review the number of bands and the qualifying materials that are slotted into the different taxation rates? We obviously keep it all under review, but that point touches on a key issue with this tax, which is that it is trying to achieve two objectives. It is perhaps the only tax that we want to see a reduction in revenue raised because it demonstrates that it is having its intended impact of reducing landfill. It is clear to see with the figures over the last decade that total landfill volume has fallen. The SFC revenue forecast over the course of the next five years demonstrates that it is having the intended impact of decreasing revenue from the tax. However, having said that, in light of the objectives of the tax, it will certainly keep the rates under review. One of the reasons for changing the rates this year is to ensure that we are achieving the objective—and it is not the case that materials are just being moved across a border, for example—but that we are reducing the overall amount of landfill. That example touches on the key objective of the tax, which is to bring in revenue, but secondly, to reduce landfill. We know that landfill is decreasing, which is good. We know that the revenue forecasts are decreasing, which, ironically, is good, but we need to keep an eye on unintended consequences, particularly around the behavioural impact on companies. I appreciate that the minister says that this is constantly kept under review, but I wondered whether there was a specific point in time, either in the short or medium-term future, where there will be a more granular look at unintended consequences. That has to be done in conjunction with the cabinet secretary for the environment. I know that she and Mr Mackay will be meeting again in the not-too-distant future to discuss how both portfolios are managing the tax to ensure that it reaches its objectives. Alex, I neglected to bring in the last discussion. Forgive me, but it is landfill time. My question in the last session was answered. On this landfill tax, we were supported by the landfill tax aims. However, one of the unfortunate consequences is the increase in flytipping. If you could demonstrate how the additional revenue is going to get to those, you have particular councils who are having to deal with this issue. Obviously, with all the sources of revenue, it is redistributed through to local authorities, but I recognise that, on an issue like this, it has got to be a partnership approach with local authorities, because, often, local authorities are managing and doing the collections. SIPA has a key role to play to when it comes to flytipping. They launched their campaign in the middle of last year to raise awareness of flytipping and make clear what the fines were for flytipping. There is also a role to close on different public bodies in Scotland working together well. Revenue Scotland has developed a reputation for better compliance, because it works so closely with SIPA. On landfill tax in particular, the levels of compliance are improving significantly because of that partnership. I think that that is probably the key way in which we will get a grip of flytipping as well as resourcing local authorities. Thank you very much. You mentioned the fiscal commission forecasts on landfill tax minister. Those forecasts are based on a central assumption that the ban on biodegradable waste going to landfill in 2021 will be implemented. You will be aware that, late last month, COSLA expressed serious doubts on that, saying that it is unlikely that that can be achieved on the timescale that is expected. If it is implemented, that will continue to reduce the revenue from landfill tax. If it is not, it may raise more revenue but fail to achieve the policy objective. What assumption the Scottish Government is working on is that the ban will be fully implemented in 2021 as planned. Moving on from that, if that happens, that means that revenue takes a big hit, but one of the consequences will be more recyclable waste going to incineration, which does not have a serious place in a truly circular economy. The UK Government has indicated that it intends to look at the option of introducing a tax on the incineration of waste. Is the Scottish Government looking at that? Has it discussed the potential for a necessary transfer of powers to implement that kind of policy? There are a lot of different questions in there, so I will try to take them one by one. First of all, I am very aware of the points that Patrick Harvie raised. The ban has been set in legislation since 2012, so it is disappointing that local authorities have not had significant time to prepare. Is there still uncertainty around the readiness of some councils that is disappointing? We know that 14 local authorities have long-term solutions in place, including the major authorities such as Glasgow, Edinburgh and Dundee, and the other authorities have interim solutions in place. Our focus now is trying to work with those authorities who do not have a solution in place to identify ways, such as using collaborative procurement and improved recycling, to comply with the ban as soon as possible. At the moment, although some of those questions are for the Cabinet Secretary for Environment, our intention is that the ban, which has been in legislation since 2012, would still be our objective, and that was reflected in the Scottish Fiscal Commission's forecast most recently. However, that work with COSLA and local authority waste management services continues to try to address that challenge, but our policy still prioritises waste reduction, and I think that that is reflected in the forecasts. In terms of additional taxes, as far as I know, we have a constant rolling programme of work looking at further devolution of taxes and the ways in which we can improve our current tax regime. Any changes in that area have to be taken forward in collaboration with the Cabinet Secretary for Environment, because it is one of those areas where we can achieve multiple policy objectives if we work together closely. If you are concerned about waste tourism as a possible threat, then presumably if the UK Government did look at a wider disposal tax, there would be a real need for Scotland to have the power to do the same. Absolutely. Again, there are unintended consequences of only having partial devolution of taxes, where we have full devolution over policies around the environment. However, if we were hamstrung by not having the tax powers to be able to control waste management—waste tourism, for example—that would significantly hinder our environmental policies and objectives. I do not understand the use of the language of being hamstrung. Ever since the Kalman commission in section 80B of the Scotland Act, this Parliament has the power to create new taxes. We already have the power to create new taxes. If there is a need for a new, we already have the power to create new taxes. The point that I just made was that there is a rolling work in the Scottish Government looking at how we can improve our current taxes and increase the devolution of taxes. The point that I was trying to make there is that, if there are two portfolios that are both trying to achieve two very similar objectives, they have to be working together, and the tax regime has got to reflect the environmental policy objectives. Just to be clear, that is not about the future devolution of powers that the Parliament does not yet have. It is about future exercise by the Government of Powers. It is about making sure that we use the powers that we have currently. To Patrick Harvie's point, if the UK Government were to introduce a new tax, for a good example at the moment, is the consideration of a digital services tax our priority would be to look at what impact that will have on Scotland and to respond appropriately. If there are no further questions, we now move to agenda item 5, which is consideration of the motion on the order. I invite the minister to move S5M16045, which the Finance and Constitution Committee recommends that the Scottish landfill tax bracket standard rate and lower rate bracket order 2019 be approved. Members, do you have any further comments? There have been no further comments, and I put the question. The question is that motion S5M16045 be agreed, are we all agreed? We are agreed. I thank our witnesses for their contribution to this discussion. We will publish a report to Parliament setting a decision on the order later today, and I now close this meeting at the Finance and Constitution Committee.