 I welcome everyone to the sixth meeting of the Public Audit Committee in 2024. Before I turn to the agenda items, could I just place on the record my thanks for the contribution that has been made to this committee by Sharon Dowie, who stepped down last week as the deputy convener of the Public Audit Committee? She was certainly a valued member of the team and I certainly valued her greatly as deputy of this committee. The first item on the official agenda is to welcome and invite Jamie Greene to declare any relevant interests. Thank you, convener. Good morning, colleagues. Good morning. In that respect, I have no relevant declarations of interest to declare. Thank you very much. The second item of business is for the committee to agree a new deputy convener. Parliament has agreed that only members of the Scottish Conservative and Unionist party are eligible to be chosen as deputy convener of the Public Audit Committee, and I understand that the party has nominated Jamie Greene for this role. Do members agree to the nomination of Jamie Greene as deputy convener? Can I congratulate Jamie Greene on your appointment and look forward to working with you? The next item on our agenda is to agree to take agenda items 5, 6 and 8 in private. Are members of the committee agreeable to that? We are agreed. Thank you. The next item on our agenda, agenda item 4, is consideration by the committee of reports prepared by the national audit office and the Auditor General for Scotland into the administration of Scottish income tax for the tax year 2022-23. I welcome our witnesses this morning, Stephen Boyle, Auditor General for Scotland, who is joined by Mark Taylor, who is an Audit Director at Audit Scotland. I am also very pleased to welcome Gareth Davies, the Comptroller and Auditor General, and also from the National Audit Office, Rebecca Maven, who is a senior analyst at the NAO. We have quite a number of questions to put to you. I am going to ask Mr Davies to give us a statement, but before I get to that, can I invite the Auditor General to address the committee? Many thanks, convener. Good morning, committee. Scottish income tax remains a key part of the package of financial powers implemented as a result of both the 2012 and 2016 Scotland Acts. The purpose of today's session is to look at the administration of Scottish income tax up to 2022-23. This was the sixth year in which the full amount of non-savings, non-dividend income tax collected by HMRC is payable to the Scottish Government. It is also the fifth year that HMRC has published Scottish income tax outturns in its accounts. Outturn figures are for the year 2021-22. Differences between actual UK and Scottish tax outturns and the amounts previously forecast will be adjusted now through the 2024-25 Scottish budget. This is known as a budget reconciliation. There will be a negative reconciliation of £390 million to the 2024-25 budget. This is the largest budget adjustment to date and is a tribute to the unexpected extension of the furlough scheme and relatively weaker economic performance in Scotland compared to the rest of the UK following the emergence from the pandemic. HMRC's annual accounts also include an estimate of Scottish income tax for the 2022-23 year. HMRC collects and administers Scottish income tax as part of the UK's overall income tax system. The NAO audits HMRC's accounts and the CNAG is responsible for reporting to the Scottish Parliament on HMRC's administration of Scottish income tax. Given that I reported to the committee to provide additional assurance on the NAO's audit work in line with the recommendation from the predecessor public audit committee in 2014, I also explained what the findings mean for the Scottish budget. I am satisfied that the NAO's approach was reasonable and covered the key audit risks. I am also satisfied that the findings and conclusions of the CNAG's report are reasonably based. The CNAG has concluded that the outturn of Scottish income tax was fairly stated. That provides a Scottish Parliament with valuable assurance over this aspect of the Scottish budget. In previous years, the contents of my report followed the requirements set out in the legislation and covers three things. Firstly, the outturn for 2021-22 and HMRC's estimate of Scottish income tax revenue for 2022-23. Secondly, the rules and procedures that are in place to administer the system. Thirdly, the costs recharged by HMRC to the Scottish Government under the service level agreement between the two. In terms of the outturn and estimate, the methodologies have remained broadly the same as last year. I have concluded that both are reasonable. External factors such as high interest rates in inflation, wage growth and high and volatile energy prices have created additional uncertainty in HMRC's estimate of tax revenue for 2022-23. Nonetheless, I consider that the approach that HMRC has adopted in reaching this estimate is reasonable. HMRC's compliance activity, that's its activity to detect underpayment of tax, began to recover in 2022-23 after being essentially stopped for period during the pandemic. This is obviously for the UK as a whole, but still remains lower than before the pandemic. Across the UK, the number of compliance cases that HMRC opened in 2022-23 was higher than the year before, but still 17% lower than in 2019-20, the last year before the pandemic started. The total tax debt for the UK, so this is tax that is due from taxpayers but not yet paid, was £43.9 billion at March 2023 for the UK as a whole, and this was £4.7 billion higher than a year before. That remains at a level roughly three times the average annual figure before the pandemic, so tax debt is a significantly bigger problem for the UK as a whole than it was before the pandemic. HMRC continues to assess the risk of non-compliance as a result of divergence between the Scottish income tax system and the rest of the UK in terms of the marginal rates charged. It continues to assess that risk as low, but there is a risk that this position may not be sustainable in the longer term as further divergence of tax policy occurs. My team and I work closely with the Auditor General for Scotland and colleagues at Audit Scotland throughout our audit, and I'm grateful for their co-operation and contribution to the work, and I'm looking forward to answering your questions. Thank you. Well, look, thank you very much indeed. Some of those areas that are methodological approaches, the implications of a greater divergence between the income tax systems of Scotland and the rest of the UK, I think, are areas that the committee will be particularly interested in looking at. If I could perhaps begin by putting what I think is an important question to you, so on the one hand you say that you think that HMR estimates are reasonable, but then you go on to say, I mean, if I look at paragraph 1.21 in your report, Auditor General, you identify the areas where there's a degree of, I think we would call, woolliness in the methodology that's used. So you say that HMRC does not fully understand the causes of the 2021-22 overestimates and socioeconomic factors contributing to the 2022-23 overestimates. There's a big reliance on sample data into the revenue estimates. PAYE and self-assessment amounts of portion to Scotland do not reflect the differing proportions of each type of taxpayer between Scotland and the rest of the UK, and I know I've written to you, I think, last year about that, and I may return to that. For example, PAYE assessments include areas that are not subject to Scottish income tax variation, such as dividend payments and savings interests and so on. Assumptions are the basis of HMRC's estimates of PAYE liabilities. So these are all holes in the methodology that you have identified. How does that then provide you with the comfort of saying what we've got before us is reasonable? Yes, I suppose the first point to make is, as we say in the report, is these are not new issues for this year. These have been features of the system essentially since it was designed, and whilst there are marginal improvements and new challenges arising each year, these points we have made in previous similar reports. So this isn't a deterioration, this is a continuation. Although, as I said in my opening comments, the environment for 2022-23 is unusually volatile. Remember that year we were dealing with the energy price spikes or sorts of unusual economic activity. So the impact of these uncertainties could be higher in 2022-23 than in previous years, even though it's the same list as in previous years. So I think that's an important point. So our conclusion that the methodology is reasonable is not the same as saying we are giving assurance on the number itself. It's a really important distinction in audit terms. Clearly this is an estimate for a particular purpose. This is not an accounting statement with all the technical requirements to be materially accurate. So the one thing we can be sure is that this will have an error in it, this estimate one way or the other. The question is, is it useful enough for the purpose as it was designed? Our conclusion is that you can take the same level of assurance on this as in previous years, although being aware that the volatility in the economy for 2022-23 means that estimation error is likely to be higher. I think we've made the point in previous years that for us as Parliamentaries it's important to understand what's going on out there because that, in the end, guides policy decisions and decisions around the Scottish budget, decisions around where income tax is set. So that's really why we are especially interested in it. I suppose it strikes me that I would be interested to hear from you if you were in the shoes of HMRC. What would you do differently? I think the challenge there is that most of the improvements that you could attempt to make to this estimate would come with a cost because it would involve more granular data collection, new systems implementation, more staff activity to determine higher levels of testing on things like accuracy of the tax base and so on. Whilst it would be possible to construct quite a detailed action plan for making improvements in all of these areas, they would all come with a cost. That's why my answer to those questions every year is essentially that's a matter between the Scottish Government and HMRC because only the Scottish Government can quantify the value of such improvements to it compared to the alternative use of that money. That has to be a policy decision for the Government rather than an auditor. It's absolutely possible to set out where technical improvements might be possible, but they would all come with additional costs. I'll maybe ask the Auditor General for Scotland on this. The committee's perspective on this is that the Scottish Government was last year in front of us and saying at that point that the service level agreement between the Scottish Government and HMRC was being renegotiated. I think that we had been asking them to look at how that renegotiation can release more of the data. We think that it should be available. Some of it is fairly basic information about the tax gap and the extent to which fiscal drag is having an impact on tax returns and so on. I think that we were led to believe that that was within the scope of those renegotiations, but yet when we look at the service level agreement it looks broadly similar to the previous service level agreement. When we look at a fairly useful indicator to Mr Davies' point, the fee that is paid to HMRC by the Scottish Government, and that is to collect data on £13 billion, £14 billion worth of tax, is £600,000. Maybe I'm looking at this from the wrong end of the telescope, but it seems to me that it's a question of if you paid a bit more, you'd get more of that qualitative data. I think that the view of this committee would be that that would be money well spent. I think that it's at the heart of both the NAO's judgment and those of Audit Scotland, convener, that the risk is increasing in terms of compliance risk as we project into future years. Primarily, building on already a set of circumstances that the committee has seen in previous reports, but with more recent developments of the divergence of tax rates between Scotland and the rest of the UK in terms of income tax, what that might mean in terms of taxpayer behaviour, all the more reason therefore for the Scottish Government—I think it's a judgment for the Scottish Government to make about what additional comfort and assurance it wants from the estimates that it has before them, given how central those are to its overall spending ambitions that it sets through the budget. Undoubtedly, the pandemic has played an impact on the volatility of the estimates. As I mentioned in my opening remarks, £390 million is the highest budget reconciliation to date, but it's an important point that you make, convener, £600,000 relative to £13.9 billion doesn't feel entirely commensurate with what's at stake here. Probably there's a limit to what much more I can say on that. I think what we would say is that we've seen progress over the course of the past 12 months with the availability of data sets and that there is a contend, a step in the right direction, but there's more to do both in terms of transparency and that the Scottish Government, together with HMRC, keeps us under really close review as they see the future data and outcomes coming into place, the extent of which they can rely upon these estimates in future years. I'll make your point. Mark might want to say a bit more about the circumstance of the data set and what comes next. Just on the question around the provisional outturn, the estimate itself, that this information sits alongside the information that the Scottish Fiscal Commission provides and regularly updates its forecasts. That information is prepared on a different basis, more looking at the granular data that's available and the real-time information that's available from the tax system. One of the things that we've always been clear about in relation to this estimate is that it's based on a kind of share of the overall UK position and as more data is available from what happens here in Scotland, we think that there's opportunities for a different approach to that. As others have said, there's then a cost-benefit judgment in relation to this particular estimate about whether that investment in that different approach is good value alongside the work that the Fiscal Commission does, the reality of it, and I think that we would all recognise that, as the estimate that's prepared here is kind of put to one side and government and parliament focuses on the Fiscal Commission forecasts and the update of those forecasts, we think that there's a bit more value that can be brought from this particular estimate, given it's genesis from HMRC. Before I turn to the deputy convener, I paraphrase them a little bit, but the sense I got from the evidence that we took last year from HMRC and the Scottish Government was that maybe that data that we are looking for is not available, so no matter what the price, it couldn't be extracted from the system, the data that's collected, it couldn't be disaggregated in the way that we think it ought to be. Do you have a view on that? Maybe I'll ask Mr Davies first and then come to you. Well, there is certainly existing data that is not being separately analysed and could be. Could you give us an example of that? Well, a good example. It takes the tax debt information that I mentioned in my introduction. Clearly, it is possible to attribute that tax debt to individual taxpayers and possible to know which tax code applies to those taxpayers, so that's one example where that's currently done on the apportionment basis that was just described, but could be more accurately analysed with essentially more resource devoted to it. There are systems issues that make it less simple than I've just made it sound, but even so, they're not something that can't be solved. I don't have much to add to the convener. I think that this is a live issue. To give the Scottish Government credit, I think that we've seen some progress over the course of the past 12 months and it is a cost-benefit decision, but one that should be informed by what's clear is that there's increasing risk in this environment for the Scottish Government. Together with also fair to acknowledge increasing transparency around some of the governance, I've seen publication of minutes, which was a long-standing issue of the Scottish Income Tax Board. But for that board and the Government more widely to take a view about where to go next in terms of their ask of HMRC by way of compliance. Can I draw your attention to paragraph 1.13 of our report, which is titled Estimated Further Liabilities? I wanted to draw your attention to this because it's an example of where improvements in systems have reduced the estimation inaccuracy very significantly. It's useful to have a concrete example of what we're talking about. I think that this is a fairly obscure bit of the out-term calculation, but there are PAYE liabilities that are not straightforward to calculate, often determined just once a year, where employers have agreements with HMRC for particular kinds of employee expenses and so on. As that paragraph says, there used to be an estimate uncertainty of £10 million a year in this part of the figure. That's been reduced almost nothing now because of an improvement in the system. At first, you might have assumed that £2,731 figure was a typo, but it is actually the resulting uncertainty, the remaining uncertainty in that figure. It's a really good example of where it's not material to the overall estimate, but where an improvement in the system has allowed an area of uncertainty to be essentially eliminated. Okay, thanks. That's very helpful. I'm now going to turn to Jamie Greene, his first meeting of the committee. Jamie, over to you. Throwing in at the deep end. Thank you, convener, and good morning to our panel. Before I move on to my line of questioning, I may just conclude the previous session. Obviously, we've done a bit of discussion around the variance in outturn versus the forecast. That sort of flags a concern to me in that the 2223 estimations from HMRC for £14.9 billion, what we might see when that outturn finally produces in terms of variance. Effectively, surely there can only be one data set that's at the core of all of this, and no matter who you speak to, whether it's the fiscal commission or analysts or policy makers or HMRC, they're not using different data sources, presumably. Therefore, the core of it must sit a very strong and robust set of data that can help to produce those forecasts. Is there confidence that the data set and the analysis of it thereof is done in a way that allows policy makers and analysts to produce the sort of forecast that then affect decisions around tax rates, for example? Is it robust, given what is quite a huge variance in outturn this year? Yes. I think it is, as you say, the same data sets that are being analysed here, but crucially at different stages of maturity. So, the reason, it might seem like very old news to be talking about the outturn for 2122, given that we're in February 24. But that's because, as you know, the tax timetable means that self-assessments, tax returns have to be in by the January after the end of the tax year. They then have to be processed and calculated. So it is really only in late 2023 that the accurate picture for 2122 can be calculated. As we say in the report, you could be very confident about the outturn figure because it's based on tax actually calculated for real taxpayers. It's based on their tax dues. Therefore, 99.3% of that figure is actual rather than estimated. So it's only at that point that you could be as confident as that in the accuracy of the number. For the estimate, which is the following year, clearly that's before HMRC has received the tax returns for most taxpayers, certainly before they've calculated, tested whether those are accurate and calculated the tax due. So it's the difference in maturity of these figures that's the heart of this. And as we were discussing in the first conversation, how much more accurate can you make that second estimate? So in this case, the estimate for 2223, because at the moment it's largely based on an apportionment of estimated liabilities for the whole of the UK. Whilst there's all sorts of checks and procedures that HMRC goes through to make sure that they've eliminated as much of the estimation uncertainty as possible, as we were saying at the start, you're still left with a fairly significant amount. That position is more uncertain when there's the volatility in the economy that we saw in that year. So it's essentially how much more effort and cost you want to put into improving the accuracy of that more recent estimate, knowing that there'll be another year before you get the accurate figure. Is that volatility perhaps that I'll come on to? Perhaps I can ask the auditor general in your opening comments and repeated in the opening biographs of your report. It gives a quite a broad brush overview in that using phrases like continuing economic uncertainty, tax policy divergence and weaker economic performance, et cetera. Now, they're obviously quite generic terms, but what work has Audit Scotland specifically done to look at the root causes of why that £390 million was so vast? I can happily start and bring Mark in on a moment to you. I think what we're not challenging is the reasonableness of methodologies here. For all the reasons that Gareth sets out in his paper and also from our own consideration of the fiscal commission's report, there are different sources and there are timing differences layered on top of that, the volatility in economic circumstances for when some of these forecasts and estimates had to be made as we were in the midst of and then emerging from the pandemic. We've set out paragraphs 64, 65 and then exhibit two of our own paper. Some of the circumstances that were specific to the scale of the budget reconciliation that is going to be deducted from the Scottish budget for £24.25 of £390 million. Part of the factors that the deputy convener mentioned in my opening remarks are very relevant here. When the fiscal commission's forecasts were made, circumstances then transpired in respect of extension of the furlough scheme. Then I mentioned also about the extent to which economic performance in different parts of the UK rebounded after the pandemic. There's very clearly, as the CNAG sets out in his report, significant increases in tax revenue in 2021-22 compared to the previous year, as we would have expected. I know that the committee is aware of this, but the requirements of the fiscal framework, as they relate to the spending power for the Scottish budget, are based on the relative performance of the Scottish economy compared to other parts of the UK. While both have rebounded significantly, it is important to note that the Scottish income tax rates rose by 14.9 per cent compared to 15.4 per cent in other parts of the UK. It is an element of success, but it is still an impact on the overall scale of spending provision through the Scottish budget. Those are the factors that are at a high level, deputy convener, but Mark might want to elaborate on where our work has taken us and where we might go next in terms of our activity on economic performance. Specifically in comparison, is it possible to quantify that under performance in numerical terms? The starting point for this from our perspective is that we are not the economists in the room and the fiscal commission are the economists in the room. A lot of detailed work that the fiscal commission does in analysing its own forecasts and what has actually happened through time is publicly reported at the heart of their work and they engage with the Parliament's Finance and Public Administration Committee on that. We draw a lot of our understanding of what is happening in the economy from their work. From our perspective, we then use that to understand what that means for the public finances and the risks of the public finances. At the heart of that has been, and the Auditor General has pointed to some of this in his introduction and earlier answers, baked into the system is a high degree of uncertainty and a high degree of volatility. That is a consequence of the system that is set up and agreed. Underpinning that is that budgets need to be set four years in advance of when the actual tax is known, what was collected. That is the heart of the system. Our work over a number of years continues to be focused on how does Government in Scotland respond to that risk and how does it manage that risk. That will be the continuing area of focus of our work. As Auditor General says, post-pandemic and in a high-inflation environment, the fiscal commission has been clear that the level of uncertainty and the level of risk in here is higher. It has been quite positive in its outlook for the tax position going forward, but it also makes the point that the risk to that is probably mostly on the downside. That is going to be the position, but given the factors that sit behind that around earnings and the levels of inflation and how they impact on the different systems, if those expectations do not come true, there is a bit of a downside risk going forward. We continue to stay alert to that in our engagement and our audit of Scottish Government about how it responds to that. I suspect that one of the biggest areas of interest and perhaps concern, not just of the committee but probably of the Government themselves, would be the effect of tax divergence and differential tax rates in Scotland versus the rest of the UK. You state that there has been little change in divergence historically. However, this tax year and perhaps future tax year has increased or will increase. The known unknown is what effect that will have on the tax base behaviour and the resulting tax intake. So, what work has been done by either Audit Scotland or the NEO in understanding how this behavioural impact will affect the final outturn numbers? What are the risks, essentially, and how high or low is that risk? Shall I start on that? If you paragraph 2.16 of our report onwards, that section covers the issue that you've raised there. So, HMRC is steadily building its work in this area to understand any emerging evidence of taxpayer behaviour on this question of, is the divergence that already exists? Is there any evidence that's driving different behaviour? For example, high levels of non-compliance by wrongly declaring an address on the other side of the border and so on. The conclusion so far is there is very limited evidence of that, if any, but I suppose the point we're making is that the year that we're looking at here, the estimate for 2022-23, the level of divergence is fairly low at that point. It's obviously increasing steadily for 23-24 and now 24-25. So, just as an expression of audit risk, this is an area that obviously, in our view, requires increasing attention from HMRC. In response to this, HMRC is building a couple of databases. One is a longitudinal survey of taxpayer behaviour over time, so this is building up several years of data on cross-border movements. So, in other words, recording all cross-border tax movements in both directions and looking at any trends in that data over several years to see whether particular changes in tax policy lead to any changes in behaviour. I think that's a very sensible set of data to be studying over time and it's good that they've been collecting it for a while in anticipation of this. So, it's too early for that data to shed any light on the question at this point, but I think having that in place is going to be very important for the future. So, I think that's one to watch and to seek evidence from HMRC on. There's also their regular cleansing of their data on the system, so regular matching of data across different systems, although that's getting increasingly difficult with stronger data protection rules. The amount of data batching that HMRC is able to do is reducing over time, but that's still an important component, and also routinely looking for missing or inaccurate postcodes. So, there's a mixture of standard controls you would expect in a system like this, plus also this kind of standing back to look at long-term trends in data, which are all designed to shed more light on the question you've raised. From our perspective, deputy convener, we support the CNAG and AOS judgment in this about the increasing risk that there is to both compliance and forecasting in respect of the divergence of tax policies, and therefore for the Scottish Government together with the HMRC to have a clear view about what activity it wants to undertake so that it's satisfied that it understands the scale of the risk and how it is being managed on its behalf. That probably goes back to the discussion with the convener about what the extent of compliance activity that is being undertaken, so that it has as reasonable forecasting to support future Scottish budgets as possible. Perhaps I draw the committee's attention to paragraph 53 of our paper, in order to set out some of the behavioural responses that taxpayers can take potentially as a consequence of some of the divergence of tax rates. Some of those, and the committee will be familiar with, we've explored some of the approaches. I would also highlight paragraph 56, and it goes back to the dataset that the CNAG referred to in terms of tracking over a longer term how taxpayers are responding in Scotland to the divergence of tax rates. That matters not just to the HMRC and the Scottish Government, but it's an important transparency point here, so that both forecasters and auditors, but perhaps policy makers and scrutineers such as yourself and the Parliament, have a clearer picture of taxpayer behaviour when setting budgets supporting income tax decisions in Parliament that this information is more readily available than it has been to date. So we'd say that there are signs of progress being made in terms of the publication, but there is some work still to do and scope to be more transparent and future deputy convener. We know what some of those risks are. They are well documented and talked about often, but quantifying them is going to be terribly difficult until we see what people start actually doing with their money. For example, if you look at the scenario of someone earning £50,000 in Scotland, they'll pay 20 per cent more tax in Scotland than elsewhere in the UK, and that's a substantial difference. Therefore, the volume of people that may that affects first of all must be easy to quantify, but what they do thereafter may be more difficult. I guess that's part of the problem because one presumes that if there is a huge shift in behaviour, that will affect future budgets and what's available to the Government. Is that fair to say? Is that a risk to how we forecast future budgets? These are the factors that the Scottish Government needs to be across to satisfy itself about. The extent to which they have good-quality robust data upon which to make policy decisions in future, and as I mentioned, that is as widely shared and published as is possible. I think that both CNAG and ourselves are satisfied that the risk is increasing, that taxpayers' behaviour will change as a consequence of the divergence of rates. The Scottish Government clearly needs to be on top of that risk as it relates to its forecasting, given how central the forecast will be to future budget setting for the Parliament. There is no figure for our estimate for tax evasion or tax avoidance in Scotland. The HMRC produces a UK-wide tax gap estimate, which includes the difference between the tax if everyone paid exactly what they should under the law set-out compared to what's brought in. That's of the order of £30 to £40 billion for the UK as a whole. The only estimate available for Scotland is an apportionment of that total figure from HMRC. It's another example of one of those bits of data that isn't separately calculated for Scotland, but just a portion from the UK figure. I'm going to try and move things along. Graham Simpson joins us remotely this morning, but I'm going to bring Graham in at this point. Graham Simpson, thank you very much convener. I just want to follow up on a couple of things. I think Mr Davis, you mentioned, in fact you've just mentioned and you mentioned previously, this figure for the UK tax gap, which I think he said was £43.9 billion. Obviously, we don't have a breakdown of where that debt falls in Scotland, England and Wales yet, but would it be possible to get that? Is that something that would be desirable? There are so many large numbers flying around in this report. I think you're referring there to the tax debt figure, aren't you, rather than the tax gap. These all sound very similar, but the tax debt is essentially the amount owed by taxpayers to HMRC and for the UK as a whole. This is tax that's been certified as due, but not yet paid. That is another example of the, for Scotland's figure, that's another apportionment of the total. HMRC at the moment doesn't calculate, build that up from individual taxpayers in Scotland. As I said in my opening remarks, that figure as a whole is roughly three times the level it was before the pandemic. This is a much bigger problem, if you like, for all countries in the UK receiving the cash that's due. That figure for Scotland is an apportionment of the total rather than an accurate build-up. But surely we know who owes what. Yes, and as I was saying earlier on, I used that as an example of where a greater investment in resources at HMRC, if the Scottish Government decided it wanted that figure on a built-up, accurate basis, it would have to agree with HMRC how that would be achieved. What we do know is that when you think about HMRC, you imagine one big system that calculates the tax due from each taxpayer, monitors the amount being paid, looks at any debt level and what's being done about it. That isn't actually how it's set up. There's a separate system for every bit of the tax system in practice. There's a PAYE system, a self-assessment system, two completely separate IT systems. There's a tax debt management system, and so that explains why this isn't as simple as reading off a single database. All the data you might want to know about taxpayers, it's much more complex. Okay, so if we wanted to know what the tax debt is in Scotland, that would have to be a request from the Scottish Government. Yes, it would have to be a variation in the agreed approach, because at the moment the basis for all of the figures set out in our report is covered by an agreement between the Scottish Government and HMRC. So our audit assessment that HMRC is applying the rules reasonably is in the context of that agreement, and that agreement would have to be varied, have to be improved in the way you describe if that system was to change. Right, I'm going to ask what I might describe as a daffladi question, and somebody can help me out here. So if I live in Carlyle, but I work for a company in Dumfries, and I travel to Dumfries every day to work, which tax rate do I pay? Do I pay the English tax rate or the Scottish tax rate? So it's our understanding that where you live, it forms the basis of your tax affairs, Mr Simpson, so you would pay UK or English-based tax rates for that. That's a definite. So my colleagues who work in Dumfries will pay a different rate of tax to me. That's correct. Depending on your postcode, if you have a Scottish postcode, you'll have an S code attached to your tax number, and that will dictate that you'll pay Scottish rates of tax set by the Scottish Parliament rather than English or Welsh tax rates, and that will again be dictated by your postcode. I think that Mark wants to come in as well. Yeah, just a slight clarification of that. So you would pay the Scottish tax rate, Mr Simpson, irrespective of where you live, because there are certain rules for MSPs and MPs as to which tax rates are paid. So the slightly less flippant point is that there are very specific rules that derive where people pay tax, largely it's due to residency, but there are specific exceptions and MSPs would be one of those. So don't think you can just jump on the M74 and pay less tax, Mr Simpson. Listen, I want to assure Mr Taylor this wasn't me trying to wriggle out merely using this as an example of somebody else, not me, not an MSP. So that takes me on to the point of, you know, in all the papers we've got here, it's quite clear that HMRC doesn't actually know where a lot of people live. I mean, why? Why is that? Well, I mean, the basic fact is it's not a legal requirement to tell them. So, I mean, that would be a measure that the UK and Scottish Governments could particularly obviously the UK as the host of HMRC could legislate to make it a legal requirement to inform each of MRC of any change of address, but it's not actually. And it's one of the reasons why HMRC has to invest so much time and effort in investigating the accuracy of the postcode of taxpayers. So it's a simple explanation for why that's not straightforward at the moment. So you think that, you know, it would be a good idea to perhaps change the law so that if you do it, for example, in my other example, I worked in Dumfries and lived in Carlisle, or moved to Carlisle, I would have to tell HMRC I'd moved to Carlisle or vice versa. Well, thankfully, auditors aren't legislators. So what appears to be a very natural audit recommendation for an orderly system like this, lots of other factors to take into account such as privacy and so on, which are for politicians, not for auditors. But it is important, though, that in order to apply the right tax rate, now that we have these different rates of tax, then surely HMRC needs to know where people live. Absolutely. It's not a legal requirement. So, I mean, going back to my other example, I'm sorry to hark on about this, but if I work for a company, let's say I'm on PAYE, a lot of people are. Is it my response? I know it's not a legal requirement, but whose responsibility is it to tell HMRC where somebody lives? Is it the employer or is it the employee? If you're on PAYE, it's the employer, and so HMRC does a lot of work with employers to ensure that they're accurately coding because it's the employer that ensures that the tax code for each employee is accurate. So this X prefix that is used for Scottish taxpayers is the responsibility of the employer to get that right. HMRC does have regular communication with employers through the PAYE system to do that and regular checks. So, do you think then the level of accuracy of this information is greater for people who are on PAYE than for those who are not? Well, there's an extra level of checking involved because the employer is involved as the intermediary between the taxpayer and HMRC. Obviously, for self-assessment, by definition, it's a direct tax return from the taxpayer to HMRC. Some of them may have tax advisers, some of them may not. However, there is an extra step and control in place for PAYE through the employer, that's for sure. I'll leave it there for now, convener. I'm going to move things swiftly along and invite Willie Coffey to put some questions to you. Willie? Thanks very much, convener, and good morning again. I want to pick up that issue of employers who do not and habitually do not apply the Scottish tax code to their employees. In the paper, we read that there are 37,000 cases, but we ask this question every year. Why is this continuing to be a recurring issue where employers are failing to apply the correct tax code to their employees in Scotland? I think it continues to be, essentially, because employers vary in their diligence around this responsibility. I think that's the straightforward answer for that. I think HMRC does have good arrangements in place for checking and identifying where this is happening. Actually, the bulk of those are not, when investigated in more detail, it's usually not a significant issue in terms of the level of tax involved. I think that's some reassurance to the committee on that. As auditors, our experience of systems like this you can never relax and assume that you have created a secure and reliable system. You always have to put energy into checking compliance. You may have new employers, for example, who are not used to the system and need to understand it and be introduced to how it works. You may have changes in ownership that mean that a previously reliable employer becomes unreliable. I say bitter experiences in order to suggest you have to keep putting energy into checking compliance, communicating what's required, and not assuming that once you've done that for a few years it will be understood and implemented because it simply won't. I think that constant vigilance is needed. We spoke with HMRC and we were told that the number of employers who persistently get tax codes incorrect is very low. Their activity to rectify that with employers is effective. It's not as though the same employers are making the same mistake year on year. I was going to ask that. Is it the same employer? It seems to be the same number of cases around the £30,000 mark that we hear. This is the 50th year of this. Truly, a member of the public would ask, why is it the same employer that is doing this? How many employers are doing this? Is it a small number or is it hundreds? Is there any further information for the committee to help us to understand this? We don't have the data on the number of employers that you may have. I don't think that I would add, Mr Coffey. It's welcome that the Scottish Government is taking an interest in this and that the Scottish Government has requested data from HMRC on the number of staff employed by each employer using the incorrect tax code. You are right in terms of the overall number of employers. I accept absolutely what Rebecca says about looking at the patterns that sit below this. I think that this is the correct next step for the Scottish Government so that it is absolutely clear on the scale of the issue that we are talking about. Are those very large employers or are they small? Is there any discernible pattern that then influences the overall compliance risk for the collection of Scottish income tax? I think that we asked a previous committee for more information about the employers, the numbers of them, perhaps even the names of them, who are habitually perhaps doing this. Is that possible for the committee to get sight of that kind of information? It would have to be a request to HMRC. We certainly don't have that. Okay. In order to try to give the public some reassurance on it, is the balance of this different each year, would you say, or are we dealing with the same offenders every year? I think that we can't be precise for the reasons that we've said, but the evidence that we have seen from HMRC suggests that this is essentially an issue of churn, that they spot a problem with a particular employer, they secure corrective action for that employer, but then there's another one that crops up somewhere else, and the number they solve and the number of new problems balances itself out each year, so that's why you see this apparently stable number, but it isn't the same companies every year. That's our impression, but the extra information that the Scottish Government has asked for would be helpful in getting into that topic. Is the money recoverable for those companies who have not applied the Scottish code, and who should do, is that tax recoverable, ultimately, when they correct the code? Yes, where errors are identified and the employer is responsible for the incorrect coding, then the employer is responsible for making that good. Okay, thank you for that for the moment. Okay, thank you. Just a very quick one from me, and it relates to this thing about third party data checks to get assurance. I mean, when I read the report, it suggests that 1.4 million, 27 per cent of records analysed could not be corroborated with third party data. I mean, that seems to me like quite a high number. Do you have any concerns about that, Mr Davis? Yes, I think, I mean, this is again not massively different from the proportions from previous years, so and the most recent exercise was only just available when we produced this report, so we didn't actually have all the detailed analysis from HMRC. So what we'll do for our next report is follow up on that and give you the final analysis as it will be available by then. But yes, on the face of it, it does seem worrying. It's a very large number that you can't get third party confirmation for, but this is an area where, for example, data protection rules that I was describing earlier on do kick in, so it isn't straightforward to make comparisons between different data sources, particularly if the data you're trying to compare it to was not collected for this purpose. Quite rightly, data protection rules are very tough on that, you know, if you didn't state the purpose for which the data was to be used, it can't then be used for that. So that's one reason. Secondly, the third party comparison could, the inaccuracy could be on either side of that comparison. So the fact that the residency status doesn't match for a particular taxpayer doesn't mean that the tax code is wrong. It might mean that the other data set is out of date and inaccurate, so it's important to be clear that doesn't necessarily mean that this is an error in the tax figure, but Rebecca might be able to add some more detail if that would be useful. I can add some limited detail since the publication of the report, we have had some preliminary analysis from HMRC on this year's results, and of that 27% of records that were unmatched, there were various reasons, as Gareth has said, that that was the case, but HMRC were able to match 83.5% of the unmatched records to Scottish taxpayers who sort of didn't have a tax liability either because they were abroad or under 18 or it was a duplicate record, and of the remainders they've told us that almost all of those were temporary reference numbers, so temporary national insurance numbers that were given to someone while they were awaiting a permanent one, and they've said they were confident that they've now matched over 99% of all records using other data sources. So this is something that we'll have to interrogate in more detail in next year's reports, but that initial number of unmatched records, it doesn't necessarily correspond to to an unmatched record, it's just that it doesn't match initially with the third party data. Thanks, that's helpful, and it may be that in due course we will invite HMRC in and perhaps get into some more of the detail of that. Auditor General, just a word from you on this. Thank you. On the topic of third party data exercises, you'll see from the reports that those are carried out every two years at present. In our own report, I think we're asking the Scottish Government to consider the appropriateness of that frequency of these third party exercises, given all the volatility that we've discussed already this morning, as to whether or not undertaking these more regularly might provide more certainty through to the Scottish Government its information-resultant forecasting decisions that it makes around budget decisions. We think that that's an important topic for them to be satisfied about how regularly these are being undertaken. You mentioned earlier on about the Scottish Income Tax Board, and one of the things that resulted from our work last year was that their minutes are now produced. I have to say that they are not the most illuminating and comprehensive of minutes I've ever read. It's basically an outline of their agenda and a couple of bullet points beneath them. Nonetheless, one of the things that is described in one of those minutes is the fact that they are planning as a board to be more strategic. Do you get any sense that that's been the case? I'm probably not in a position to comment with any degree of authority, convener, on whether that is or isn't the case. I think that we welcome the increased transparency and the activity of the Income Tax Board, but we've not done any specific audit work about whether they are adopting a strategic or operational position. It's something that we can give consideration to as part of our future programme. Some of the areas that we've been discussing for the last hour presumably would fall within the remit of a Scottish Income Tax Board that we're seeking to be more strategic in its approach. No question. I'm resisting to put a label on whether they are or aren't being strategic instead of focusing on the specifics that we have in front of them. As it relates to their satisfaction with compliance, third party data checks and so forth, all of these will be the decisions that the Income Tax Board and the Scottish Government more widely should have under their consideration how they choose to identify that as strategic or operational. I think that it's a matter for them, but we'll keep that under review. We may not be able to rely on the minutes to help us to get to the bottom of that. I'm sorry to bang on about some of the stuff that's maybe been partially discussed already. I have a big concern about the uncertainties of the tax revenue and the way that it's calculated. In 2021, I took out a list of all the anomalies in the NAO report, where there were estimates, guestimates, potionments and goodness knows what else. I did the same in 2021-22, and it's almost a carbon copy. We've got your report for 2022-23. Although I haven't gone to the trouble of taking out line by line, it's pretty much the same again. As you yourself said, Mr Davis, there's only been some marginal improvements. Is HMRC in breach of the service level agreement that it hasn't placed with the Scottish Government? No, we don't think so. We think that it's reasonably applying the requirements of the service level agreement. As I said before, that's a long way from saying that the estimate that it reduces is accurate to any particular percentage level, but just that, given what's required in that agreement, the approach that HMRC is taking is reasonable. What our audit provides is assurance that the processes in HMRC systems to deliver the estimate in that agreement are working as intended, and our audit gives you that assurance. It's really important that it's not the same as a normal audit of a set of financial statements, which is a much higher level of assurance on accuracy. This is essentially that the processes that HMRC has said it will apply in agreement with the Scottish Government, it has applied. Well, I'm not a legal person, but if you look at the SLA, page 5, clause 23, first bullet point in it, HMRC will identify the Scottish taxpayer population and collect from it the correct rates of SIT to ensure that the Scottish Government receives the correct amount of income tax revenue each year. Patently, it doesn't. Carrying on, there's another five bullet points here about, for example, identifying Minterian accurate and robust record of the SIT taxpayer population. You can't say they're doing that. What we can say is the outturn figure, albeit two years after the end of the year, is fairly stated. The figure that you get for 21-22 in this report, for example, is not full of estimates. That is an accurate figure based on actual tax due from Scottish taxpayers. It's subject to the same level of uncertainty as the whole UK tax take on levels of non-compliance and so on, which is the tax gap I was referring to earlier on, but that's no different for Scotland than for the rest of the UK. But, with respect, you've already said that that's an apportionment of the UK. There's no specific figures for Scotland. But not for the tax gap, no. But the point I was making is that HMRC would be able to say for 21-22 in this case, it has accurately identified the amount due for Scottish taxpayers to the Scottish Government. What is much more full of uncertainty in estimates is the estimate for 22-23. But this time next year, obviously, we'll be reporting to the accurate figure for 22-23. As I said, I'm not a lawyer, but going down these clauses are very simple and straightforward and quite clearly from your report, they're not complying with it. It's okay to say, well, in a couple of years' time we'll have a real figure. That doesn't say that it's timely information, that it's relevant information and that it's received by the Scottish Government in such a way and in such a speed to enable it to discharge its duties in respect of rate setting, forecasting for SIT and all these other things. I mean, this is a fairly solid SLA, I would say, but it's not being complied with. Well, we think that you can take reasonable assurance that it is being complied with. What it doesn't generate is an accurate estimate for the financial year that's just ended. You have to wait another year to get that. That's the point here. I think it's not right to say that the HMRC is not delivering an accurate record of tax due and paid to the Scottish Government. It is, but it takes two cycles for that to be accurate. The estimate is necessarily at the moment full of estimates and will therefore be out by an amount in either direction. So, I think it's really important that we clear what we're talking about. If the SLA said deliver an estimate for the most recent financial year that is accurate to within 0.1%, you would be absolutely right that HMRC is not delivering that at the moment. I would need much more accurate basis and expensive basis of estimation to deliver that. But what it is delivering is an accurate list of tax due from Scottish taxpayers and collected, which is, you can take the same level of assurance on the accuracy of that as the UK as a whole can take for HMRC's performance. Again, if you just look at this literally, the SLA says HMRC will identify the Scottish taxpayer population and it says again that they will identify and maintain an accurate and robust record of the SIT taxpayer population. You can't say they're doing that. Well, we think they're taking a reasonable approach to that because, of course, that reasonable approach doesn't mean 100% accuracy with no errors and that wouldn't be achievable in any conceivable tax system. So, the question is what's reasonable and we think the approach that's being taken at the moment, given the state of the legislation and so on, as we said earlier on, there's no legal requirement to declare your change of address to HMRC at the time it happens. So, in that context, we think the approach that's being taken at the moment is in accordance with the agreement with the Scottish Government and therefore reasonable. There is plenty of scope for improving the degree of assurance you can place on the various controls in the system and we've discussed lots of those this morning, but that's not the same as saying that HMRC isn't complying with its SLA. To me, there doesn't seem a lot of rigour room here. I mean, it says HMRC had to provide the Scottish Government with sufficient relative and timely information and data for assurance purposes and to budget effectively. Can you categorically say they do? Well, we've talked about lots of the areas where it takes two years to get accurate information, but again that, I think this basis for how the estimate is compiled versus the out turn for the year before, that is a shared understanding between the Scottish Government and the HMRC. That's agreed in some detail each year, how that's going to be approached. So, I think that's transparent and it's not as if the Scottish Government thinks it's agreed one thing and HMRC is delivering a completely different service. This is the service that it's been commissioned to deliver with all of the uncertainties in the most recent year that we've described. And where is the agreement on all these uncertainties? Is it an agreement that says that a portionment will be done in this way? Estimates will be done in this way? Broadly, yes. Each year there is a discussion about the approach taken to build up the estimate and that is a transparent process and it's understood, obviously there's a report from us every year confirming that as well, so it's understood on the basis that that estimate is produced and as we've described it would be a policy decision essentially for the Scottish Government to ask for a more granular approach to some of those estimates to get a more accurate figure and as we said a strong case to be made for that but it would come with a cost. Do you actually measure performance against the SLE? Not in the sense of routine monitoring of every line of it. We've got three clear responsibilities as we set out at the start of this report but one of them is that as HMRC applied the rules and procedures that apply to Scottish income tax correctly and so we test that in our audit work but that's not the same as every line of the SLA. Arthur General, do you look at the service level agreement and performance against it? Compliance with the clauses of it that are being read? It may be helpful to provide some clarity on what our role is here Mr Beattie, so going back 10 years now following the Scotland Acts, Audit Scotland doesn't audit HMRC that's very clearly a role for the CNAG and his colleagues in the national audit office. As I mentioned or alluded to in my opening remarks, my role is one that I've inherited from my predecessors at the request of predecessor committees that as the Auditor General for Scotland I provide an additional assurance report to the Scottish Parliament through this committee. What my role is to consider the NAO's approach to identifying the key risks to the successful administration of Scottish income tax to look at on a sample basis some of the national audit office's audit files, the audit evidence that they have accumulated in arriving at the judgments in their report and to take a wider view about the quality arrangements within the national audit office. As I mentioned, I'm satisfied that all of those factors have been appropriately discharged by the national audit office, so know what I don't do and what I very clearly don't have the powers to do is to have any direct engagement with HMRC and their compliance with the service level agreement that you specifically asked about. Given that this is an agreement between the Scottish Government and HMRC, wouldn't you have a role from the point of view of looking at what the Scottish Government's getting? I think that I probably have to repeat in a different way my previous position, Mr Beattie, that I don't have a role. This is a matter for Gareth and his colleagues to audit HMRC as it relates to the engagement that they have with the Scottish Government through the service level agreement. My role, as I discharged through my reporting, points out from the Scottish Government's perspective options and considerations where they can take further steps to satisfy themselves about the compliance, the accuracy and the quality of the data, Mr Beattie. As a layman, I would still say that the service level agreement is being breached. Moving on to something a little bit different, although it's something that we've touched on before. We've talked about Scottish income tax behaviour and what I would say to the door general is that is there enough published data and research on Scottish income taxpayer behaviour to assist the forecasting by the SFC and policy development by the Scottish Government and scrutiny by the Scottish Parliament? There is more to do here. There is more scope to publish data, building on some of the activity that the Scottish Government has agreed with HMRC this year in terms of the data sets and echoing Gareth's earlier point that we stated in our own report. There is increasing risk in this field, Mr Beattie, in terms of the divergence of tax rates between Scotland and the rest of the UK and how that might influence taxpayer behaviour in Scotland. Both the NAO's report and the judgments that we make in our own paper say that this is an area of risk, increasing publication, more data, more transparency to inform the decisions that the Scottish Parliament takes ultimately through the Scottish Budget. Some progress but there is more to do and there needs to be a... The Scottish Government needs to take a call on how close an eye it wants to take on some of the impact of behaviours. What's also at play here is that there is a time lag about the availability of data. We were operating in a, as Gareth described earlier, a couple of years behind in terms of cycle. Some of that will be through necessity, but again, this is an important area, perhaps, as the convener mentioned, for the Scottish Income Tax Board to take a view on about how robust that data is, how frequent the checks, how transparent the Scottish Government wants to be for its own purposes, but also for parliamentarians more widely. As you said, it's going to be a couple of years before we really understand the full impact of what's happening now. That's correct. Okay, just moving on a little bit there. Do the... Or at Scotland and CNDG believe that the strategic picture of risk could be more bespoke for Scotland because we keep coming across data that... Where it's not available for Scotland as such, and then there's apportionment from the UK, UK-wide figures. To be honest, the UK has a different make-up of taxpayer. I mean, it's all distorted by London, and that goes for any place outside London, but nevertheless London dominates, and any apportionment is going to be somewhat distorted. Can we do this better? Can we get better figures for Scotland? I think... The strategic assessment of risk is the HMRC process for looking at essentially the risks to collecting all of the tax due in line with tax law. So it's looking at non-compliance, it's looking at tax evasion, it's looking at economic trends and so on. So it's a... As you say, it's a very important bit of analysis. As the UK diverges both economically and tax policy, then I think you're right. I think that that analysis needs to become more granular, needs to take this increasing divergence into account. If it's going to remain as a useful guide to the allocation of HMRC resources, for example, on how to maximise the tax yield and so on and minimise evasion and all of those other objectives, so yes, I think this needs to be kept under careful review. It will need to improve its accuracy certainly by the nations in the UK and as you say potentially regionally within countries in the UK as well. And of course, how much do that is a policy question for ministers rather than one for the NAO, but we will continue to report on that and we'll continue to, in our role in the UK, hold the UK Government to account for how useful and accurate its strategic assessment of tax risk proves to be. Just continuing on risk, cross-border migration. There's a trend analysis referred to in paragraph 2.33 of the report. Has that been, has that report on cross-border migration trend been published actually? A CHMRC expected the analysis to be complete by January 2024. It's maybe putting them under a bit of pressure, but do we know where that is? Yes, Rebecca, do you want to come in on this one? Yes, so that refers to analysis or cross-referencing with information from the land registry about cross-border moves. That has not been completed when we asked recently post-publication, so we're still awaiting that data. Is there a revised date? We weren't provided with one. We were just told that it wasn't yet ready. It has previously been done in previous years, but for this year we don't have any information on that. So do we know if it's being done this year? That would be a question for HMRC, I'm afraid. We just inquired about the delivery of it. We don't know yet. Okay, and just leading on from that, so there's now be able to update the committee on the joint-funded work by the UK, Scottish and Welsh Governments on cross-border migration, obviously trends over a period. It's referred to in paragraph 2.34 and the HMRC work on retirement and migration referred to in paragraph 2.35. Where are we on these? It's the work that I was describing earlier on, which I thought was a welcome development of building up this longitudinal assessment of taxpayer behaviour across border. Because this is the only way we're going to be able to form a robust view on the impact of divergent tax policy on behaviour. So I think this is a good investment. It's going to be some time before we see any impact of the increasing divergence, obviously because that's happening in the current year and next year. So to show through in this data, it's at least two or three years away. But if you don't start collecting this, you haven't got the baseline you need to be able to then measure changes. So I think this is a good development. But you're monitoring this going forward. Okay. Thank you, convener. Okay, thank you. We're actually over our time allocation and I think there are some questions that maybe some members of the committee have still got. But I just wonder whether we can perhaps follow that up in writing with you if you're willing to answer questions on that basis. If that's accessible to the committee members, I know who have got some questions still to put to you, including Graham Simpson. What I will do then is draw the public part of this morning's committee to a close. Thank you for the evidence that you've given us this morning. It's been illuminating and we know that you've been challenged at points by the committee. And thank you for your honesty in your responses to that. We've got other work that we may need to do in previous years. We've in turn invited in the Scottish Government and HMRC to give evidence, and the committee will consider how useful that will be on this occasion. But can I thank Gareth Davies and Rebecca Maven for your attendance and your evidence this morning? Auditor General Mark Taylor, can I thank you as well? And with that, I'll draw the public part of this morning's proceedings to a close.