Finance and Constitution Committee

Meeting date: Wednesday, October 3, 2018

Official Report 778KB pdf

---

Pre-budget Scrutiny 2019-20

Our next piece of business is to take evidence from Her Majesty’s Revenue and Customs as part of our pre-budget scrutiny. We are joined for this agenda item by Jim Harra, who is the director general of customer strategy and tax design at HMRC, and Jackie McGeehan, who is the deputy director of income tax policy. I welcome both our witnesses to the meeting this morning. Members have received copies of HMRC’s annual report. Before we go to questions from the committee, does either of our witnesses want to make an opening statement?

Thank you, convener. I will make a few short points. First, I confirm to the committee that HMRC has provided the annual report on Scottish income tax to the Scottish Government as set out in our service-level agreement. The report has also been provided to the committee along with the relevant extract from the department’s accounts on Scottish income tax.

I also want to draw the committee’s attention to the successful delivery of the changes to Scottish income tax for 2018-19, which were announced in last year’s budget. We worked very closely with the Scottish Government at every stage of implementing those powers, including incorporating the 2018-19 changes, and we look forward to delivering on the budget announcement that we expect later this year.

Finally, I would like to apologise, convener, that a letter from me to you in the summer went astray somehow. I do not know why that happened. You should have a copy of it now.

09:30  

Okay. I will not ask any more about that. James Kelly would like to begin the questions this morning.

I want to focus on the issue of the income tax outcome for 2016-17. According to your report, the outcome was £550 million less than estimated. That was partly attributed to the movement in additional and higher-rate taxpayers. Initial estimates were based on the survey of personal incomes, and the movement from that resulted in 5,000 fewer additional-rate taxpayers and 43,000 fewer higher-rate taxpayers, which contributed to the reduction of £550 million in the estimated outcome. I am interested in an explanation of the methodology of the SPI, how that drove the number of taxpayers, and how you arrived at the actual outturn figures in your report.

The SPI is conducted on a sample of taxpayers. We gather the data about their incomes from their tax returns. The most recent survey was for 2015-16, which is what the 2016-17 forecast was based on.

There are some issues. The first is that projecting data from a previous year and making assumptions about wage growth and population changes builds uncertainty into the forecast. In addition, although the survey is regarded as representative at the UK level, it is less accurate at a sub-UK level, when you try to break it down by country or by region. We are working with the Office for National Statistics, the Scottish Fiscal Commission and the Scottish Government about what improvements can be made. Meanwhile, as the Scottish Fiscal Commission says, those would be the key reasons why the SPI overestimated the number of higher-rate and additional-rate Scottish taxpayers, which was a significant factor in the inaccuracy in the forecast.

In addition, for the 2015-16 SPI we did not have the identification of Scottish taxpayers so we had to try to identify them from that survey based on postcode information. When we produce the 2016-17 SPI, we will be able to do so based on the identifier of Scottish taxpayers. That will make SPI more accurate for forecasting purposes.

The outturn is based on the actual information from pay-as-you-earn and self-assessment returns for people who are identified on our database as Scottish taxpayers. Therefore, it is a much more accurate outturn, based on much more up-to-date data. It has been audited by the National Audit Office as part of their audit of our accounts and our trust statement.

Taking 2016-17 as a baseline position, because that is the first year for which you have captured all the Scottish taxpayers, how confident are you that the figures that you have captured provide an accurate baseline?

We are confident that that is an accurate outturn. It has been audited independently. It is based on much more up-to-date information. There is a table in our accounts that shows some further adjustments that we have had to make for events that had not yet happened at the point when that outturn was locked down. For example, there will be some people who pay late, and we have had to look at historical payment rates data to adjust that for non-payment by late payers. However, those are quite marginal adjustments, so I believe that that is an accurate outturn, which has been independently verified.

What is the size of the Scottish sample for SPI and what role does that play, moving forward to future forecasts?

I do not have the information for the size of the Scottish input to the SPI. It is an ONS-owned piece of statistics. I can certainly get that information for you.

Can I follow up on that question? Your service-level agreement with the Scottish Government requires HMRC to provide the Scottish Government with sufficient, relevant and timely information and data for rate setting and forecasting for the Scottish income tax. Jim Harra, you said in answer to James Kelly that there are issues about the Scottish data in relation to the UK data. You say in section 2 of the HMRC annual report:

“Although the SPI is considered representative of the UK taxpayer population, it is less reliable at a sub-UK level.”

As you might imagine, that gives me some concern, given that our process is very much based on forecasting. Given the significance of income tax forecasting to the size of the Scottish budget, do you think that it is reasonable that your own report describes as

“less reliable at a sub-UK level”

the data that you are providing to support the process? I think that we were all under the slight misapprehension that the data that is being looked at for the Scottish circumstances was the same as the data for the UK circumstances, but it is obviously not. It is the same information base, but it is not the same level of data.

Yes, you are right. It is exactly the same dataset and the Scottish Government and the Scottish Fiscal Commission gets data that is materially identical to the data that the OBR gets from that set for the UK. You are right that, when you drill down into the SPI to look at Scotland only, the information is less reliable than it is if you look at the whole-UK picture. That is the data that we had from the 2015-16 survey. Together with the Scottish Fiscal Commission, the Scottish Government and the Office for National Statistics, we are looking for what improvements could be made. The key thing is to get data that is more up to date and to have data that is based on actual Scottish taxpayer identification. Those are the two key steps for us.

Given that any forecast error is potentially significant for the Scottish budget, how pressing is the work that is going on to ensure that the Scottish Fiscal Commission has increased access to the data on Scottish income tax at the same level as is available for the UK? For example, why does the SFC have access only to the public-users tape data for the SPI?

The Scottish Fiscal Commission has exactly the same access to Scotland-level data as the OBR has to UK data. There is no question of reduced access. The only slight exception to that is at the very top of the income ranges. We have had to do a bit more aggregation in Scotland to avoid identifying individuals but that is immaterial in terms of the level of data access. The issue is the quality of the underlying data in the survey rather than restricted access.

I am sorry to labour this point but can you tell us a bit more about what is going on to improve the quality of the available data so that we can understand it more clearly?

The statisticians are working with the Scottish Fiscal Commission and the Scottish Government on how the base of the survey could be improved to provide data that gives a more accurate picture for Scotland and also to give it more timeously. This year we have pulled forward the timing of when we produce the data in accordance with our agreement to support the Scottish Fiscal Commission. We will be looking to do that again next year.

I am sorry, James, if I cut across any areas that you were going to go into.

I have one follow-up point. I reiterate the point that the sample size is important. I am sure that I read somewhere that it might be as low as 2 per cent. That is quite a low sample size and you are extrapolating from it, which obviously adds to the potential risk of error. You have this baseline outturn figure now and you have the on-going SPI survey. How do those two things interact to update your forecasts?

I am not a statistician, but I know that, as mentioned by an earlier witness, even what seem to be relatively small sample sizes can give very accurate levels of data compared with the full population. Obviously, the larger the sample size, the smaller the error rate. However, sometimes, increasing the sample size can result in only a marginal improvement. It is something that the statisticians in HMRC, the ONS and the Scottish Fiscal Commission are considering.

What was your second point?

How do the baseline outturn figures and the on-going SPI survey data interact? You have the survey data and you have these baseline numbers for the actual number of taxpayers in your report for 2016-17. How does it work to update forecasts?

The outturn is based on the full set of just over 2.5 million taxpayers. It is not survey-based at all. What that outturn tells us is that it was significantly lower than the forecast for 2016-17. However, having a baseline year now, if you compare that with the forecast for 2017-18 for example, that assumes a level of growth in Scottish income tax receipts that, while it is possible, I think is not likely, and therefore I would expect those who issue forecasts—the SFC and OBR—to take that into account in the next forecasting round.

Some other issues with that year that I think affected forecasting were to do with the devolved powers and related to non-savings and non-dividend income. As well as having to estimate the balance between Scottish and non-Scottish taxpayers, you also have to estimate the balance between savings and dividend income and non-savings and non- dividend income. That was a particular challenge in 2016-17 because the UK Parliament introduced changes to the taxation of dividends around that time. That had a behavioural effect on people who control their own companies for the timing of their dividends. I suspect that when we analyse the outturn for 2016-17, we will find that the proportion of the total UK income tax receipts that came from dividend income was different from the assumption in the forecast. That will wash itself through in further years.

I am sorry to labour this point. In the outturn data, you have 2.5 million taxpayers, whom you have now identified, and you have this on-going survey. How, for example, is that data used to update the 2017-18 forecast? How is the survey data used to interact with the 2.5 million base dataset that you have to update the forecasts?

HMRC does not issue the forecasts. The OBR and the Scottish Fiscal Commission do that. We provide them with the data that we have. When they made their forecasts in the last round, they did not have the 2016-17 outturn data. For the next round, they will have that, plus some further data around economic determinants, which may well change their forecast, but it is for them to produce the forecasts with the data that we have made available.

I want some clarity about the issue of the number of Scottish taxpayers. In table 2 in section 2 of your annual report, you have a figure for all Scottish taxpayers of 2.528 million. Does that include everyone who has an S code or does it include only people who are actually paying tax? I am trying to capture a figure for people who are earning but at a rate below the personal limits level.

The data is only for people who are paying tax. A larger number of people would potentially be Scottish taxpayers if they had non-savings, non-dividend income within the taxable range.

Do we have a figure for that larger number?

I am afraid that we do not have it here, but we can let you have that figure.

That would be helpful.

Table 1 in your report shows the outturn figures. I am interested in the breakdown between the Scottish figure and the figure for the rest of the UK. If you look at the self-assessed total there, it is 79.76 for the rest of the UK and 4.36 for Scotland. That represents a share for Scotland that is about 5.4 per cent of the figure for the rest of the UK, which seems a bit on the low side.

In the next line, on PAYE, the Scottish share is about 9 per cent of the figure for the rest of the UK. Do you have a narrative behind that? Remember that these are actual figures; they are not forecasts. Why would the Scottish share of the outturn be so low compared with what the population share might be, which would be around 9 per cent?

09:45  

That will be because of differences in the profile of Scottish and RUK taxpayers in terms of their sources and levels of income. It is not a legislative test that puts you into self-assessment. You are in self-assessment if HMRC feels that is necessary in order to administer your tax affairs. The difference will be down to the profile of taxpayers in Scotland compared with taxpayers in the rest of the UK and how many of them are in self-assessment. That balance will be different partly because of the profiles of employment versus self-employment and other income but also income levels, because that is one of the determinants of whether we feel that we need to put someone in self-assessment.

In your footnote there, just below the table, it says that the first line

“Includes an element of PAYE for SA customers”,

so the figure in the first line would be even smaller if that were not so.

Say, for example, that I am an employee of HMRC but I am also in self-assessment. When I complete my self-assessment return, I would put my employment income on my return. In this table, that would have come out in the SA figure rather than in the PAYE figure. The PAYE figure is really for people who are only in PAYE and do not have to complete a self-assessment return. If you have to complete a self-assessment return, your employment income and the tax on it is reflected in the SA.

Are you happy that it is completely accurate? There seems to be quite some discrepancy between the figures for Scotland and those for rest of UK.

If you like, we can go away and get you some proper analysis of that. I have described what I believe are the reasons for that discrepancy. We will see whether we can get some analysis to back up what I have told you.

These are two ways of administering tax as opposed to two different sets of tax liabilities. As you have mentioned, for example, people who are in PAYE are also in self-assessment so they are in a different line, but we can try to get you some more analysis that explains that.

Alex Burnett is interested in issues of consolidation.

My questions are on the theme of reconciliation and the issues with the three-year time lag. What can be done to have some sort of interim reconciliation? Your report mentions that HMRC intends to publish real-time figures, which would obviously improve the ability to do that.

Can you clarify whether we are just talking about PAYE data rather than self-assessment data? If so, would the figures be more accurate for Scotland because we have a higher percentage of PAYE taxpayers?

Lastly, is there any possibility of publishing real-time data on self-assessment? I appreciate that the profile of payment during the year will vary, which is maybe why that is not done, but you mentioned that you look at historical data on payment rates and what is paid late, to give you a profile. Could that be improved so that you can have interim data on self-assessment?

A couple of years ago, we introduced a new information technology system for pay as you earn and a method of collecting payroll data—it is called real-time information—which means that, for pay as you earn, we now have in-year data that we gather from employers, which we did not previously have. The data is supplied by employers every time that they run their payroll, so it is generally monthly but it can be more frequent than that. We have shared that data with the Scottish Fiscal Commission and the Scottish Government to support in-year forecasting.

There are a couple of issues with that. First, it does not include self-employed income, which is about 16 or 17 per cent of the income that you are interested in. Also, with our pay-as-you-earn codes, we sometimes adjust for things that are not relevant to Scottish income tax. For example, if someone has savings income, we may adjust their pay-as-you-earn code to collect tax on that, so the amount of tax that is being deducted through pay as you earn is not always precisely the same as what the Scottish income tax outturn will be.

I said that there were a couple of issues, but a third one is that the information is based on the S code each month. However, the test for being a Scottish taxpayer is not a monthly test; it is an annual test. It is possible that we could deduct tax from someone on the basis that they are a Scottish taxpayer in April, May and June, but they might move south of the border and, for that tax year, they will turn out not to be a Scottish taxpayer and therefore those first three months of collection would come out in the outturn, or vice versa.

Those are some issues with the quality of the data that the Scottish Fiscal Commission and Scottish Government need to take account of. Nevertheless, we think that it will give a very useful source of data for in-year forecasting.

As far as the self-employed are concerned, we do not have any in-year data. The self-employed give us their income data on their self-assessment returns by 31 January, after the end of the tax year, so there is quite a lag in collecting that data. We will potentially get more in-year data from the self-employed in future. We are rolling out a new policy called making tax digital for businesses, which involves receiving quarterly updates of data from the self-employed. That policy launches next April for VAT, but there is already a pilot out there that enables people to do that for income tax purposes as well. In future, we may well get in-year data on the self-employed, but currently we do not have any.

In summary, you said that you can get 80 per cent accuracy in real time.

Yes. The real-time information data based on pay as you earn will cover 83 to 84 per cent of the total income that is within the Scottish income tax, although that is subject to the issues with the data that I mentioned.

I think that Tom Arthur is also interested in that area.

My questions have largely been covered, but I have a brief supplementary to some of the points that Mr Kelly raised. How provisional is the provisional estimate of revenue for 2017-18, which I think is £11.9 billion?

I think that that was the OBR forecast in March. We did our own forecast and it was in our trust statement. That used data that was a little more up to date because it was done slightly later in the year, but it is not materially different. However, now that we have the outturn data for 2016-17, which was not available when the forecast was made, you can see that the forecast assumes a level of growth in Scottish income tax between 2016-17 and 2017-2018 that, although not impossible, is unlikely. I expect that the figure might come down in the next round of forecast.

Thank you for the clarification.

I believe that Emma Harper has concerns about data.

Are you confident that all Scottish-rate taxpayers have been fully captured and that the 2016-17 outturn data is accurate and is a current reflection of reality? You spoke about people who might spend the first three months of the year north of the border and then move south, which would create additional complexity.

I will let Jackie McGeehan respond to that.

We are confident that we have captured 98 to 99 per cent of Scottish taxpayers in our data. There is no definitive list of Scottish taxpayers. Obviously, the population moves around and people come in and out of it for various reasons. However, each year, we carry out a big data-matching exercise to identify Scottish taxpayers and improve our data. We compare all our UK records against third-party data to corroborate the details of individuals. About 47 million records were matched by name, national insurance number and so on. In that group, only about 5,000 individuals were identified as having a third-party Scottish address where our records showed a UK address. About 4,000 individuals had a third-party UK address where our records showed a Scottish address. That was a pretty good matching record.

We then wrote to all the individuals for whom there was some doubt over which address was correct. We had quite a good rate of response to those letters. More than a third of people responded, which is a high rate for something that people do not really feel that they need to do. In most cases, the individuals confirmed that the HMRC address was correct, or they told us what the correct address was and we updated our records.

We then looked at the Scottish unmatched records, of which there were around 1.1 million. We took out duplicates, those who had moved abroad and some that had temporary reference numbers. If someone takes on a new employee and does not have all the information about them, a temporary reference number might be allocated in the tax system, which does not accurately reflect where the person lives. We also took out the records where there had been no tax activity in the previous five years. We whittled down that population and, after removing all those groups, we got to about 490,000 records where more corroboration was needed.

For individuals who were employed, we looked at where their employer was and whether they had a pay-as-you-earn record with a Scottish employer, and we found that about 280,000 individuals had a Scottish employer. That left 208,000 uncorroborated records. Within that number, some individuals will be in Scotland but in self-assessment and we have not been able to check their address yet, or they will be employed by a large employer that has employees across the UK, so we would not be able to see from the pay-as-you-earn record that the employee is Scottish. We got to over 96 per cent firm corroboration plus an assumption that, within the remaining small number, some will also be Scottish, so the figure is 98 to 99 per cent. We repeat the exercise every year.

Another important aspect, which the committee might want to touch on later, is communication to ask people to tell us if they move. Most people tell us without any prompting, but we keep reinforcing the message that they need to tell us if they have changed their address.

It is a long process—I am sorry for all the detail—but we do a lot of work to check the addresses.

Thank you. The detail is helpful.

It is good to hear the detail of the work that is going on, because questions were raised about that at the beginning of the process. That leads into Murdo Fraser’s area.

Yes—it follows on nicely. I want to ask about compliance issues, which you address in your annual report, in relation to potential behaviour changes as a result of tax changes. You talk about potential cross-border migration, which in light of the level of tax change is probably unlikely, but it is perhaps more likely that high-net-worth individuals will shift income towards dividend payments. Are you keeping a close eye on that? How are you monitoring it?

Clearly a differential in tax rates creates incentives. The risk that people will change their behaviour in response to those incentives depends on the scale of the differential, the scale of effort involved in changing their behaviour, what the compliance rules are and, as you heard from a previous witness, the chances of being caught if you break the rules. In the case of Scottish income tax, if we assume that Scottish income tax is higher than UK income tax, which is the case for higher earners, that is an incentive for people to disguise their identity as a Scottish taxpayer and pretend not to be one—Jackie McGeehan has described the work that we do to validate identities—or to avoid or evade the tax on their income by either underdeclaring it or changing its nature from non-savings, non-dividend income.

Some people have scope to incorporate and change their income into dividends, and there are limited rules to prevent them from doing it. In the case of people whose working relationship is employment, there are quite strong rules that prevent them from shifting employment income to incorporation and dividend income. However, for people who are self-employed, there are no rules to prevent them from incorporating their business and drawing out income by way of dividend. Tax-motivated incorporation is an issue for the UK tax system as well as for the Scottish tax system because, even in the rest of the UK, the differential between corporate tax rates and income tax rates creates an incentive.

10:00  

On the other hand, the dividend taxation changes that I mentioned, which came in a couple of years ago, reduced to an extent the incentive to incorporate, because they increased the level of taxation on dividend income. It is a behavioural response that is entirely possible and I think that the Scottish Fiscal Commission has estimated what impact it will have on Scottish income tax receipts. There are no rules to prevent self-employed people from incorporating.

I assume that you will monitor on an on-going basis whether there is a trend in that direction. Will you be able to identify that?

As an administrative and compliance organisation in relation to the self-employed, we do not have a direct interest in that, because it is not non-compliant behaviour, it is something that they are at liberty to do. However, policy makers and forecasters have an interest in it and, therefore, we supply data that enables them to assess that. A couple of years ago, the OBR did quite a bit of work on the level of tax-motivated incorporation across the UK and its impact on receipts. In its most recent forecast, the Scottish Fiscal Commission estimated the impact on Scottish income tax receipts, which is between £5 million and £30 million a year.

You will collect the data but, ultimately, the Fiscal Commission will express a view.

Yes, as well as policy makers who keep an eye on the issue. The dividend taxation changes that came in a couple of years ago were in part a recognition of the fact that, as corporation tax rates come down, the incentive to incorporate grows, and those changes counteracted that to an extent.

That is useful, because there are suggestions that there might be a reduction in corporation tax. I am making assumptions here based on what I see in the media, but the chancellor may reduce corporation tax in his budget, which would potentially have an impact on the amount of income tax that is paid in Scotland, which obviously supports public services in Scotland. From the information that you have, how soon can you pick up what impact that is having? We would be concerned if the amount of income tax taken in Scotland was reduced as a result of a lever that is being used, for legitimate reasons, at another place.

The UK Government has set out its road map for corporation tax rates over the next couple of years—it is to go down to 17 per cent—and that has been factored into the forecasts. You are right that it is within the UK Government’s power to change that road map, and you would have to revise the forecasts accordingly, but at the moment the forecasts already have baked into them an assumption about where corporation tax rates are going, based on the road map that the British Government has announced.

If a new corporation tax rate was announced, the economists would forecast the behavioural effect of that, and the Scottish Fiscal Commission and the OBR would turn that into their forecasts of the impact of tax-motivated incorporations. Over time, you would then have to track whether the behaviour matched the forecasts. It would take some time for that to come through. People would not go out the day after an announcement and set up their company. It would probably be the next time they see their accountant, who will say, “Have you thought about doing this?” That would happen over a period of time. We can track it in the growth of company registrations and the growth of the corporate tax base, but it takes time to monitor.

You must have historical data on the shift from paying income tax to paying corporation tax. At this stage, that will be across the UK.

We do, and the OBR published a fairly comprehensive report on that in 2016, I think. I will happily send a copy to the committee.

Would the Scottish Fiscal Commission be able to get a data set for Scotland that would tell it what was happening in Scotland specifically that would be accurate enough for it to make forecasts?

I know that the Scottish Fiscal Commission has made a forecast. I would need to check what data we can provide as opposed to what assumptions it will have to make and test later on. I will check that.

It would be useful to know that. Angela Constance has questions on transparency.

Good morning. Given that anything to do with the fiscal framework is somewhat complex and that we would like to make more people engage with it and understand it, I have a few quick questions about transparency. There have been instances in the past where fairly fundamental information has been buried away in technical annexes and the fuller information has come out in dribs and drabs. For instance, the 2016-17 global figure—the figure of £10.7 billion for Scottish receipts—was buried away in an annex, and Mr Harra then wrote to the committee and supplied further information.

Looking ahead, will you publish the 2017-18 non-savings, non-dividend Scottish income tax outturn figures? How will you do that in a publicly accessible manner? Will you discuss with stakeholders—whether that is the committee, the Scottish Fiscal Commission or the Scottish Government—the timing of that information, how it is best put in context and the method by which it is communicated?

Under our service-level agreement with the Scottish Government, we have a commitment to produce the annual report and we are happy to discuss with people the format of that and how it is made more accessible to people. The purpose of the report is to ensure that information in a voluminous set of HMRC accounts and the trust statement is extracted and made more accessible, but I am happy to look at what more work can be done on that.

On timing, the report is part of our overall accounts and trust statement, which are published around June each year, which is as quickly as we can possibly do it after the end of the financial year on 31 March. We have pulled that forward in recent years and we continue to try to do it as soon as we can but, given the scale of what we have to produce, there is limited scope to bring it in much earlier. However, this year, as well as the figure for the Scottish income tax there was also an interest in the figure for the rest of the UK, which we did not produce at the same time—we produced it a few weeks later. Certainly for next year, we will aim to produce that at the same time, so that there will be a comprehensive set.

We are working with the Scottish Fiscal Commission on what further official statistics we can publish at that time on Scottish income tax. For example, I know that there has been a fiscal framework recommendation that the RTI data that we provide monthly to the Scottish Government and the Scottish Fiscal Commission be published every month, and that is our intention. We are working through some of the data issues that I described, which are not relevant to Scottish income tax but which make that data less than perfect, to ensure that, when we publish the data monthly, we can describe it effectively.

I thank our witnesses for their contribution. You have provided useful information, and I am very grateful to you.

I suspend the meeting to allow a changeover of witnesses.

10:08 Meeting suspended.  

10:11 On resuming—  

Our next piece of business is to continue our pre-budget scrutiny by taking evidence from the Cabinet Secretary for Finance, Economy and Fair Work. The cabinet secretary is accompanied by the Scottish Government officials Aidan Grisewood, who is deputy director, tax, and Daniel Hinze, who is deputy director, fiscal sustainability. We welcome our witnesses to the meeting. Members have received copies of the Scottish Government’s fiscal outturn report, which will help to inform today’s evidence session and our deliberations. Before we move to questions from the committee, I ask the cabinet secretary whether he wants to make an opening statement.

Yes, I do, but I will be very brief. The fiscal outturn report was requested following the work of the budget process review group. We have tried to make it as comprehensive as possible, but it is, of course, another first as part of the new arrangements, and we will be happy to take further suggestions on how we can progress the report to aid understanding of the outturn and any other workings.

I welcome the report’s publication and the engagement on it in relation to reconciliation, the fiscal framework and the transfer of powers. I recognise that the committee took evidence from the Chief Secretary to the Treasury last week; indeed, I met her the day before her evidence session.

I am happy to take questions on the report.

Thank you, cabinet secretary. There is something that I would like to get on the record. What effect do the lower-than-forecast income tax revenues for 2016 have on the Scottish budget for 2016-17 and 2017-18?

The earlier estimates were the result of a survey based on income tax. As we move closer to outturn, we get more exact numbers from HMRC, which the Scottish Fiscal Commission uses. In short, there is no impact on the Scottish budget as a consequence of that issue.

That said, all decision makers must be mindful of the full detail of the composition of the Scottish tax base and what that looks like in policy terms. However, the statistical issue that you raise has no impact on the budget.

But the fact that the number of additional-rate and higher-rate taxpayers is lower than estimated will have implications. What impact will that have on your policy?

Of course, we need to bear in mind the numbers that we can draw down from the Treasury and the numbers that we understand that we will raise in tax from any tax policy changes we make. In doing so, we must be cognisant of all the complexities of the fiscal framework. Fundamentally, however, we have a clearer understanding of how many additional-rate and higher-rate taxpayers we have, which will inform future decisions. That clearer understanding will inform not just decisions that the Government takes, but decisions that all parties in Parliament would take in arriving at any budget position or any future policy on tax rates.

The issue to which you refer was to do with baselining, so it has no budget impact, but in policy terms, we would certainly want to be mindful of it.

James Kelly has some supplementaries in that area.

Good morning. I have a follow-up question. The 2018-19 budget was agreed by Parliament in February. When the Fiscal Commission looked at the income tax position later in the year, it downgraded the income tax forecast by £208 million. In addition to that, as you noted and HMRC told us earlier, there are 5,000 fewer additional-rate taxpayers and 43,000 fewer higher-rate taxpayers. If the income tax forecast in the current budget has been downgraded and it stands to reason that there are fewer income tax payers, surely that will mean public spending cuts in the future budget.

10:15  

I am trying to be clear and helpful with the committee. There are separate issues here. We are talking about different points in time and different bits of the process. The early baselining issue was the issue that the convener asked me about, which has no budgetary impact, because we reconcile the numbers and we arrive at a main baseline position, to which the UK Government and the Scottish Government work. For that reason, there is no impact on the budget.

I understand that some members might have misunderstood what that means for the Scottish budget. There is no £0.5 billion hit on the Scottish budget. As the Fiscal Commission gets more data and more outturn figures, the numbers that we work with become more advanced, but the baselining issue has been addressed.

I turn to the more recent forecast, which Mr Kelly makes a fair point about. The closer we get to the fiscal events, the better, because we arrive at actual outturn numbers and clearer estimates. Even the Fiscal Commission’s most recent forecast for the medium-term financial strategy is also subject to change. For example, the gross domestic product statistics are different from what the Fiscal Commission had estimated, in Scotland’s favour. Taking the GDP element in isolation, that will enhance the position as regards the forecasts that the Fiscal Commission has for what might be raised from income tax. I respect the fact that the situation is complex and dynamic but, when we get to the Scottish budget, we will have the numbers from the OBR and we will have the latest forecast from the SFC, which will be informed by more up-to-date data. All of that is helpful.

To go back to the convener’s original point, we now know that we have fewer additional-rate and higher-rate taxpayers than was thought in earlier years. That has all been addressed in the baselining issue, but policy makers should think about that in determining their propositions for future years. That is about as clear as I can be on that matter.

If you have fewer taxpayers and less revenue coming through, particularly at the higher rates, when you cite your budget for 2019-20, surely you will have less in that tax envelope, and that will feed through to cuts in public services.

No.

That is the policy outcome if you maintain tax at its current level.

I am trying to be as helpful as possible, but if we understand the fullest possible workings of the fiscal framework, the block grant adjustment and all the elements that go along with it, it is not as simple as that. What we do is relative to what the UK Government does in its tax policy. I am trying to describe where we are in the composition of the tax base.

If, for the sake of argument, I had adopted the proposition that was put to me by the Labour Party during last year’s budget process, I would have been more heavily reliant on additional-rate and higher-rate taxpayers. It is true that the composition of the tax base is really important in arriving at the tax decisions that we take, but it is driven by the Scottish Fiscal Commission’s forecasts and its understanding of our policy.

I understand Mr Kelly’s point that, if we have fewer taxpayers in a particular tax category, that has implications for the outcome, but the historical issue—which was about the baselining to which we were all working—has been addressed. Of course, the starting position for the Scottish budget is the block grant; all the other adjustments come after that. I hope that that is more helpful.

What will your approach to tax be in the future, bearing in mind the challenges, some of which you have outlined? What is your view on tax policy? Are you minded to keep tax at the same level or are you looking at any changes?

I know that this is pre-budget scrutiny, but it is somewhat bold to ask me what my tax plans are in advance of the Chancellor of the Exchequer outlining his tax plans. It is a very fair question, and it is a nice attempt to get me to reveal the future tax position, but if I were to do that, it would bring pre-budget scrutiny to a whole new level.

It is a fair question, based on all the information that we have had.

I think that the committee is very well sighted on the process that I must undertake. The UK budget is a huge fiscal event, which impacts on the block grant and the block grant adjustment, as understood by the OBR and SFC forecasts. I will work my way through all of that before I present to Parliament what our tax plans are.

The baseline has now been established. For the record, can you confirm that, from here on in, it is the rates of growth relative to the baseline that will be the important issue, not the information that has already been provided—the historical stuff?

That is correct.

That is what matters.

Yes, it is the divergence that is the issue, not the starting baseline point.

Thank you for those answers, cabinet secretary, which were very clear. I want to move on to the issue of borrowing, which is dealt with in chapter 5 of your report. I hope that your answers will be just as clear on this as they were earlier. Talking about capital borrowing in particular, you say in your report:

“In total, the Scottish Government will have accumulated £1.459 million of capital debt by the end of 2018-19, well within its overall £3 billion limit.”

That is true. It is also true, as you report later in that chapter, that the Scottish Fiscal Commission has judged these levels of borrowing as being

“reasonable and ... compliant with the terms set out in the Fiscal Framework.”

However, is it not also the case, cabinet secretary, that if you continue with that level of capital borrowing for the next few years—and of course you have not said whether you propose to do that or not—if you continue to draw down your maximum annual allocation each year and continue to borrow on the basis of relatively long terms, such as 25 years, you will likely run up against your borrowing limit by 2022?

That analysis is correct.

Is that what the Scottish Government proposes to do?

That is not my proposition.

Could you say any more about what your proposition is likely to be?

I know that Mr Tomkins asked for clear and concise answers. That was a clear and concise answer and I can now go into greater detail. If I draw down at that level, that proposition is right: we would hit the cap. I was able to outline some of this in the medium-term financial strategy. I have not set out the longer-term capital plans.

In the programme for government announced by the First Minister, there was an indication that we want a national infrastructure mission. I think that everyone here supports investment in the infrastructure that will help to support our economy today and prepare us for the future. The Parliament as well as the Government will have choices to make about how we invest in our infrastructure and how we fund that. A mix of investment in our infrastructure can achieve our infrastructure needs as a country, but of course we will use our capital borrowing powers—as well as our resource borrowing powers—responsibly.

I am not clear on what the chancellor will do but, if there is a comprehensive spending review in spring next year, that might give us the ability to set out further multiyear budgets in terms of capital. The resource budget has been a one-year budget, as has the capital budget, but I think that it would be helpful for everyone if we could set out multiyear capital budgets.

I hope that I will be able to give more detail at the Scottish budget and beyond, but some of that is contingent upon a multiyear settlement from the UK Government so that we can understand what capital we have. We want to use the capital borrowing powers wisely in a way that gives stability and stimulates the economy. However, there can be a mix. To invest in infrastructure, it does not necessarily just have to be those capital borrowing powers that are used. There are other levers that can be used.

Thank you for that answer. You are clearly right that some significant choices confront both you as a Government and us as a Parliament. You have indicated that you do not want to run up against that £3 billion cap by 2022, which indicates that in future years, your priorities for capital borrowing will be different from the priorities in the past two or three years; we are on that trajectory at the moment.

I am not asking you to give details of what you will be proposing in the budget, but can you enlighten us a bit about the structure and nature of your thinking on the factors that you will take into account when deciding when and how to change course from the current trajectory?

I would not describe it as changing course as such, because I was clear in the medium-term financial strategy what borrowing powers should be used over a further period. I am genuinely reflecting on the point that we all want to continue with the infrastructure investment. There are huge capital demands—a huge call on resources—but we can use a mix of levers to fund infrastructure investment.

In this pre-budget scrutiny, I am being clear with you. We have used the borrowing powers, as outlined, to stimulate and support our economy. I hope to set out longer-term plans on capital. I have done it on some substantial commitments, for example for local government for childcare and early years, for housing investment and resource planning assumptions and for digital. There is a £600 million commitment for digital. I will be transparent about how that is funded and will return to Parliament on that over the course of the budget.

I have been clear that we want to use the borrowing powers responsibly. We return to the review of the fiscal framework and engagement with the UK Government on our arrangements, our limits, our borrowing powers and every other part of the fiscal framework as part of that review leading to 2021. I am trying to reassure Mr Tomkins that I am looking at the range of levers that can be used to invest in infrastructure while using our borrowing powers responsibly.

That is helpful to a degree, but in order for the Parliament to engage in effective pre-budget scrutiny, it would be helpful if we could get a little bit below the top line. We know that there is a range of levers, we know that there is a range of financial devices, and we know that capital borrowing is not the only tool that you have in your toolbox. However, with respect, you have not been any more specific today than you were the last time we talked about this, a few weeks ago, in relation to how you propose to use that different range of levers that you have available to you. In order for us to be able to do our job effectively on pre-budget scrutiny, it would be useful, if not this year then in future years, for us to get a bit below that top line.

There is one outstanding top-line question. Do you anticipate that when it comes to the fiscal framework review, you will be asking the UK Government to raise that cap above £3 billion?

It is a premature question but I would always, as Scottish finance secretary—as any finance secretary would—want as much flexibility as possible. We have self-imposed prudent limits—for example, the 5 per cent self-imposed rule on resource spending as part of the actual budget. We will always act prudently but the borrowing limits for the Scottish Government are set in that agreement so it is premature to ask what I will be asking from the UK Government in those discussions.

Incidentally, I would want to revert to the committee first to get the committee’s view on how we should engage with the UK Government in that review. I am sure that you would want to be part of that engagement process as we review the fiscal framework and the written agreement.

I am a Scottish nationalist and I want Scottish independence. I want as much normality as any other country in the world in terms of the borrowing arrangements that we can deploy, but what I have at the moment is an agreement with the UK Government, and we will return to that at some point in time.

I understand Mr Tomkins’s desire for more detail on the capital budget going forward but the budget is the most appropriate time for me to present that. That is the point at which I outline my plans but I was able to cover some of the parameters within which we are working in the medium-term financial strategy.

In our report, the pages that Mr Tomkins has referred to outline some more of the detail on borrowing beneath the headline commitment. I know that Mr Tomkins wants more on what the future looks like but I can only return to that comprehensively at the next big fiscal event.

The major shift since the last time I was before the committee is the national infrastructure mission, where we outlined that commitment to lever in more finance to take our infrastructure investment to internationally competitive levels. That will require a range of financial tools to be able to deliver it and of course that has to be transparent.

Emma Harper, we have reached the area of your question a bit sooner than expected. Rather than come back to the subject later, do you want to ask your question now?

10:30  

Yes. Good morning, cabinet secretary. Adam Tomkins has just talked about repayment of borrowing and resource repayment. The repayment of capital borrowing is over 10 years, and the repayment of resource is over three to five years. Is it reasonable that we should be required to balance the budget annually despite the increasing dependence on tax revenue forecasts to determine annual spending limits? The budget spending repayments are over three to five years or 10 years. Would it not make more sense to have the same flexibility as other Governments to balance the books over a longer period so that maybe statutory fiscal rules could be introduced?

That partly touches on the point that I was just making. As a devolved Administration we are working within the framework—the arrangements that we have signed up to—but sometimes it feels as though we are working in a somewhat constrained fashion. Some fiscal rules we have imposed on ourselves to show good governance in relation to fiscal restraint—the 5 per cent rule on resource spending, for example—and we are well within our borrowing limits. I think that we are showing a prudent approach, but independent Governments around the world can enhance their borrowing and should do so in a prudent and fiscally responsible way.

Right now, the UK Government has a degree of fiscal headroom. It is up to the UK Government how it chooses to use that. We have made the point to the UK Government that it should use that to turn the tap on for spending, which would help to stimulate the economy and support our public services; of course, devolved Administrations would be part of that as well.

I would appreciate more flexibility but, that said, I am working within the arrangements that we have agreed and I am trying to get on with it.

I was trying to follow a particular route earlier, but inevitably this is going down a slightly different route than I expected. I will try to keep the flow going by bringing in Tom Arthur and then Patrick Harvie because I think that it is all playing into the same area now.

This is a supplementary to Emma Harper’s question and it is about flexibility. The fiscal framework was a product of the pre-Brexit era. As things stand, it seems as though we are heading towards a hard Brexit, which Sir Anton Muscatelli described as representing

“the most unhinged example of national self-sabotage in living memory”.

Do you believe that there is enough flexibility within the fiscal framework to contend with the particular shocks and damaging effects that a hard Brexit could have on Scotland, given that, for example, the Fraser of Allander institute estimated between 30,000 and 80,000 job losses and, in the case of a no-deal Brexit, the governor of the Bank of England estimated that house prices could fall by up to a third?

That is an excellent question. No, I do not think that the fiscal framework agreement envisaged these circumstances and therefore I do not think that there is enough flexibility to deal with such a shock. That said, of course, the shock could be to the whole of the UK; it might have a disproportionate impact on Scotland. We are seeing some of the evidence around that, whether it is to do with the working-age population, impact on GDP or a whole host of other matters. As it stands right now, on the GDP statistics we are outperforming the rest of the United Kingdom, contrary to what the Fiscal Commission forecast—I will just very gently throw that in there; it is a material consideration.

The issue for us is how we could respond to those challenging economic circumstances. It is true to say that Brexit was a bit of a Tory gamble, then a guddle, and now it is an act of economic self-harm. We need to see where the negotiations get the UK to. I will not focus on Brexit today other than to say that I think the parameters of flexibility we have are somewhat constrained and limited in those circumstances. I hope that we can continue to grow our economy and enjoy success but that hope is somewhat challenged by the threat of Brexit right now.

I am working within the agreement but we need to look at some of the parameters that constrain us, for example, on the budget, what was budget exchange, the Scotland reserve, and how much I can draw down. Even if we have it in the reserve, I am constrained as to what can be drawn down, even if we have it in our resource, and that is a challenge.

These are matters that we will return to as part of the review that we have agreed to, but it is a very challenging environment at the moment with the threat of Brexit looming large over us.

I hope that we have not gone too much into the territory that Patrick Harvie wanted to cover.

No—it is fine. Thank you, convener.

The cabinet secretary touched briefly on the review of the fiscal framework in his discussion with Adam Tomkins. I am interested in how that review should take place and what the early thinking is on its nature—the way that it should be conducted. In the current plan, it is supposed to take place towards the end of this parliamentary session. It might be difficult to get consensus if it happens in the months immediately prior to an election, so earlier up-front engagement on the broad approach to it might be useful. The Chief Secretary to the Treasury did not seem very interested in getting into a discussion about that last week. Is the Scottish Government giving any consideration yet to how that review should happen? I hope that we all agree that it should not take the form of the UK Government reviewing and imposing its solution, but it will not simply be a matter of the Scottish Government saying what it thinks should happen. How can we get an open, collaborative and participative process that does not repeat the breakneck, behind-closed-doors process of the Smith commission?

I appreciate the question and the way in which it was asked.

I want a collaborative process that comes from the Scottish Parliament and the Scottish Government. The review is, of course, some time away, and the pressing issue for me is delivering budgets year to year and the circumstances therein. Members will know that we must have an independent review that informs the dialogue and decisions by the Government.

The Chief Secretary to the Treasury and I have not yet touched on the review in the joint exchequer committee, but in a meeting that I had with her, I underlined the point that, because of circumstances that we have touched on—the volatility and economic challenges that the UK faces and specific issues that we have in Scotland—she should continue to be mindful of being flexible when we can show evidence that a change of approach may be needed. She was open to that. I am saying that, short of the review, we should still have a consensual basis on which we can address issues, and she said to me that she was certainly open to that.

On the overall, fundamental review, my understanding was that there was to be a full session’s worth of implementation and then delivery of the review. Substantial issues are at stake, including the model that is used, indexation, the limits and the other arrangements relating to our powers. We will all have to give time to reflect on that, and I want to do that in a collaborative way. I am open to further discussion with the committee in that regard.

There ought to be a good agreement between the two Governments with, I hope, the input of the Parliaments by this time next year. I am not talking about the detail but about how things will be taken forward and the timescale for public involvement, transparency, open participation and other stakeholders to comment. If we go past 2019 and into 2020, it is inevitable that, in the pre-election atmosphere, more battle lines will be drawn and there will be less space for any kind of discussion about the matter.

I appreciate that point, and I will certainly take it up. That analysis of the politics and the heightened political sensitivities is fair. I am not sure that the issue will be at the top of the Treasury’s list of things to do at that point in time, so I am happy post the Brexit negotiations and pre the Scottish Parliament elections to ensure that we have the space, time and opportunity to do things right and in a transparent fashion.

I think that that plays into the area that Angela Constance was interested in.

Yes. I want to discuss relationships and working together. The cabinet secretary is perhaps aware that, in July, the Public Administration and Constitutional Affairs Committee at Westminster produced a report that concluded that, 20 years after devolution, Whitehall still has not really got with the devolution programme. That is evidenced in its structures and, crucially, in its culture of working. The report also concluded that a lack of appreciation and understanding of devolution does not just go against the principles of devolution but is contrary to good governance across the UK.

Last week, we heard from the Chief Secretary to the Treasury, who described some of the structures in which there is work with the Scottish Government. In particular, HMRC was mentioned. Obviously, she was positive about the relationships and the existing structures to assist communication. What is your perspective on the existing arrangements?

Ms Constance has asked a good question. It depends on the minister and the civil servant—maybe that is true of every Government. Structures such as the joint exchequer committee and the finance ministers’ quadrilateral are in place for the matters that I deal with. We might disagree over policy but, when it comes to the fundamental workings, we try to get on with what is agreed, because it is important that people have confidence in the competence of tax collection and the devolution of powers for financial arrangements or anything else.

As Angela Constance is well aware, getting everything that the Scottish Government requires, whether in respect of the transfer of social security powers or other powers, can sometimes be very difficult. There is a sense that the UK Government sometimes does not get devolution, but there are education processes in the civil service to try to make people aware of devolution issues. I have vehicles to raise ministerial issues, and officials have working arrangements, as well.

I watched the earlier evidence session, in which HMRC was probed on the release of data that the Scottish Fiscal Commission requires. Although the Scottish Fiscal Commission is independent of Government, we want to support it in getting the data that it needs from HMRC and others. Therefore, we apply pressure where we can and when that is required.

We try to get as much competent understanding of devolution as possible so that we can get on with what is agreed. Our politics can be radically different, but, to be totally frank, the understanding of the nature of devolution sometimes depends on the minister and the civil servants involved.

A greater appreciation of devolution would be helpful. I have clearly expressed to the UK Government what helps us and what does not help us in respect of the budget cycle, for example, because that is important for decisions that we take.

I do not know whether that is of some assistance.

I am interested to know what you, with your experience, would wish Whitehall to do differently and what you and the Scottish Government could do differently, bearing in mind that it takes two to tango.

I will be as concise as possible. The devolved Administrations should have mutual respect and be treated as other Governments in the United Kingdom rather than as other Whitehall departments or sub-departments. That is how the Scottish Parliament and the Scottish Government should be perceived. Scotland should be perceived as a nation and an equal in the United Kingdom’s arrangements and should therefore be given our place in that regard. Sometimes it feels as if the Administration is a Whitehall department or sub-department that has to be managed. All the devolved Administrations, whose finance ministers I am very close to, would share that view. We were meant to be partners in the United Kingdom arrangements, and I would like us to be treated as equals in them.

What can the Scottish Government do differently?

If we were independent, we would be certainly seen as an equal of the UK Government. Short of that, we want to try to progress the commitments that we have in the administerial arrangements and the UK Government’s understanding of what we are doing in Scotland. We all understand that for us to get on with the transfer of powers we need the operational arrangements to work clearly, whether that is providing data, providing access to the necessary officials or allowing us to get on with resolving some of the outstanding issues. That would enable us to maximise what we have agreed around devolution.

10:45  

As I said, the places where I would raise disputes, as finance minister, would be the finance ministers’ quadrilateral meeting, or the joint exchequer committee, or directly with the Chief Secretary to the Treasury, the Chancellor of the Exchequer or other ministers, as I have done. I would raise concerns directly with them. The other ministerial arrangements have not worked particularly well—for example, the joint ministerial committee on Brexit negotiations, to go wider than my brief. The Scottish Government most definitely has not felt well involved in that regard, but when it comes to finance I try to get on with it.

Can I press you, cabinet secretary, because I do not think that you answered the question? What would you do differently?

In terms of—

I mean in terms of how the Scottish Government operates to improve relationships. That was all about what others can do.

I do not think that the Scottish Government can improve on what we do. In all seriousness, I do not have any complaints from the UK Government other than that they disagree with our constitutional position. I am not aware of what I could do differently, or what could be recommended that I do differently with the UK Government to progress the matters that we have before us. I can, however, tell you what I think the UK Government should do differently. If the committee has suggestions as to what I should do differently to engage with the UK Government, I am interested in that.

I do not know whether the Chief Secretary to the Treasury highlighted in her evidence one element of change, which is the recognition that the joint ministerial committee arrangements have not worked well. The UK Government knows the displeasure of the devolved Administrations in that regard. We are considering a new forum for the finance ministers. There is the finance ministers’ quadrilateral, at the moment. Of course, the Northern Ireland Executive is not operational right now, but that aside, we are looking at the meeting of finance ministers, as part of the existing ministerial arrangements.

Again, convener, you are asking me to say what I could do differently and you are asking why I do not answer that question. You need to tell me what you think I should do differently.

I will send you the list.

And publish it.

We will get into some specifics now. Willie Coffey and Neil Bibby are interested in air departure tax and VAT, respectively.

Cabinet secretary, could you clear up the air departure tax issue that was raised last week at the committee? I think that it was Neil Bibby who raised it. Liz Truss said that there was no approach, that she was aware of, by the Scottish Government to the UK Government on the air departure tax and how it affects the Highlands and Islands exemption. You wrote to the committee yesterday. Could you clear up exactly what the position is?

The position is exactly as I have repeatedly reported it to the committee. Fortunately, I have the repeated written communication to prove it. I know that some committee members have accused me of misleading Parliament, which is unfortunate, but the reality is that I have been totally accurate on the matter, so those members will have to reflect on what they have said.

On whether I approached the UK Government about the need for notification to the European Union, I have correspondence that I can share with the committee. I have a letter here from Mel Stride MP, who at the time was Financial Secretary to the Treasury. UK ministers in this portfolio have changed a few times, so maybe the Chief Secretary to the Treasury did not know because so many ministers have been dealing with the matter. In a letter to me on 19 July, which I will read it out because it is pertinent to the question, Mel Stride said:

“In our conversation you expressed your wish to notify the European Commission formally for a Highlands and Islands exemption for your new ADT. I want to reiterate the serious concern I expressed in our call about this approach.”

I have, I am afraid, a series of letters in which, as I have described to the committee, I highlighted the issue that is of concern. We wanted a resolution from the UK Government. I have a list of correspondence that backs up what I have told the committee. The UK Government, to give a whole answer, said that it had reservations about notification, for the reasons that we understand: it did not think that the proposal was compliant with EU rules, so it did not want to proceed with notification.

The UK Government then wrote to me saying that if the Scottish Government were to take on the liability—bearing in mind that there is historical non-compliance—it might approach Europe. Would you seriously expect me to sign up to taking on the historical liability for potential non-compliance of an EU Commission matter? Of course you would not. We focused on what other resolution we could get to the air departure tax issue. As I said, I have a correspondence trail that backs that up.

I am trying to be fair to the Chief Secretary to the Treasury. The Financial Secretary to the Treasury has dealt with the issue, the Exchequer Secretary to the Treasury has dealt with the issue and the Economic Secretary to the Treasury has dealt with this issue. I have consistently dealt with the issue. If the committee wishes to see further correspondence, so be it. I have been totally accurate in what I have said to Parliament.

Are you saying that the UK has not made an approach to the European Commission on the matter?

To be fair, I point out that only the UK Government can do that because the UK is the member state. The Scottish Government cannot do the notification, in any event. I approached the UK Government with a way forward, but it was reluctant to do it.

To understand the impact that that might ultimately have on the airlines in Scotland, I note that if they had to pay back any support, it would have a horrendous impact on the Highlands and Islands. Therefore, together in partnership we tried to work on other ways to resolve the issue. I have made suggestions including using the rates and bands, for example. We have continued to work on the issue. That is true and I can prove it. It is not true to say that I did not approach the UK Government about notification. It has written back to me in those terms.

There are other ADT questions, before we get on to VAT.

In your letter to the committee yesterday, you said:

“I would like to put on record that the suggestion that there has been no engagement with the UK Government on the issue of notification is simply untrue.”

There is, of course, an important distinction between the Scottish Government engaging with the UK Government on ADT and the Scottish Government making a formal request that the UK Government notify the European Commission. You just mentioned that you wrote to the Financial Secretary to the Treasury. I formally request that you provide the committee with that letter and any other relevant letters and emails that you have sent.

Okay.

Thank you for that. You have also said that the Scottish Government does not want to take on the liability or risk from the UK Government, in notifying the European Commission. You said the UK Government had made an assessment. What assessment has the Scottish Government made of the chances of the European Commission approving an exemption for Highlands and Islands?

First, I say that, again, there is an issue about legal advice. It is about the public interest. Probably a number of people are now aware that our assessment is that the scheme may well not be compliant. Therefore, you are asking for notification of something that you think might not be compliant. That is our assessment.

There is either a case to be made or there is not. You seem to be saying that the scheme may not be compliant. Is that your assessment?

That is my assessment. Incidentally, I think that it is the UK Government’s assessment too, which is why it has not notified the EC.

As I said before, there is a distinction between engaging and formally requesting. I do not think that that distinction has been clear.

Let me be clear. I formally requested notification in a letter and the UK Government formally responded to me in writing.

You have been happy to give the impression that you want to find a solution as soon as possible, but you are not necessarily supporting actions to find that solution as soon as possible because, as Patrick Harvie said last week, that would make your budget process harder.

Last week the Treasury said that it is working with the Scottish Government on developing options for devolving APD and delivering an exemption for Highlands and Islands passengers. What alternatives to notifying the EU Commission are you working on, how serious are you about those alternatives and when is the air departure tax likely to be devolved?

On 17 July 2017, I wrote to Mel Stride, who was then Financial Secretary to the Treasury. I said:

“As you know, under APD there is an exemption for all passengers flying from the Highlands and Islands airports. However, as we discussed, following careful and detailed consideration of legal advice, it has been clear for some time that an exemption needs to be notified and approved by the European Commission under state-aid rules before it can be implemented for ADT.”

That is further evidence that I raised the matter with the UK Government. That is clear. There was then the letter that I have referred to, the reply to my point and a subsequent conversation.

I wonder whether Neil Bibby wants to reflect on the inaccurate information that he has put in the public domain. In terms of my attempts to resolve the issue, we have been working very hard. The current arrangements for air passenger duty are, of course, delivered by the UK Government. If it wants to reduce the rates, it can do so now.

I rehearse, convener, that we cannot implement laws in Scotland that are contrary to EU law. We do not know how the world will change post-Brexit, but as things stand right now we cannot continue the Highlands and Islands exemption like for like—although it is our aspiration to do that—because to impose the tax, for the first time, on the Highlands and Islands, would have, in the terminology that was used by Highlands and Islands Airports Ltd, a “catastrophic” impact. That is why it is so important that we resolve the issue.

Of course, none of this, as far as I know, was raised through the course of the Smith commission and the fiscal agreement. It was all new to us at the point at which we tried to implement the air departure tax in Scotland. We have, with the UK Government, generally found ways through on identification of issues with state-aid notification. The UK is the member state, so the UK Government must pursue notification. Short of the issue being resolved, there has been no other suggestion from the UK Government that would work and be compliant.

One suggestion that I made to the UK Government that would substantially resolve the issue, would be for it to allow us to use the powers under the rates and bands. That would work particularly for short-haul flights. That would, however, require a zero rating for all of Scotland, which would come at a cost and it would breach the no-detriment principle in the Smith agreement, so we are at an impasse with the UK Government.

Beyond that, what have we tried to do? I have set up a Highlands and Islands working group so that all interests can come together to see whether there are other ways forward, recognising that I had thought that there was consensus on addressing the Highlands and Islands exemption issue, so that there was a like-for-like exemption, going forward. I am happy to share the membership of that working group with the committee, if that would be helpful. That group is trying to exhaust the issue to see whether there is another way forward such that we can take on devolution of the tax in a satisfactory manner.

I was asked for the detail and I have tried to provide it. We are working very hard to try to resolve the issue, but I am not going to sacrifice the Highlands and Islands of Scotland by imposing the tax upon them when we are genuinely trying to work on a solution to the issue, which is a problem that I have inherited from the UK Government. I think that there is a responsibility on the UK Government to find a solution.

Neil Bibby was not in the last discussion, so I will let him in here again very quickly.

I also asked when the tax is likely to be devolved.

I cannot answer that because a solution has not been found yet.

I will ask about VAT. Last week Liz Truss said that it is a good time to be talking about the fiscal framework. I asked her whether it is a good time to talk about possible future assignment of VAT to Scotland. She said that we have a lot on our plate at the moment and to let existing powers “bed down”, which is fair enough. The reason why we could not be assigned VAT, as I understand the situation, is that the European Union would not allow different rates within member states. Post-Brexit there will be no real reason why VAT could not be fully assigned to Scotland, will there?

That is a good point. We would be assigned VAT, not the power on rate setting. I think that that power would be more useful, but it is not in the agreement. Willie Coffey has made the fair point that the world has changed. Brexit circumstances will mean that the constraints on the UK on varying rates will—assuming that we come to some arrangement—become more flexible.

The assignment issue that we are facing is about the methodology that is published and how we think we arrive at a number. I know the committee has expressed concern about going, essentially, from estimate to estimate on VAT, for which we never get an outturn figure, unlike income tax or other devolved taxes. That is quite challenging and could be quite volatile.

I continue to have concerns about that. We are still trying to stay true to the agreement. I have raised my concerns about volatility with the Chief Secretary to the Treasury. Maybe there will be an opportunity to revisit the issue. We will publish the methodology shortly, but there are still outstanding concerns about the fact that it is estimate based.

11:00  

I think that the Chief Secretary to the Treasury, in evidence to the committee, ruled out any other administrative alternative. We want to make sure that the estimates are absolutely robust. I say to the committee that even though there are transitional arrangements, one year’s data is probably insufficient, especially when that year will be subject to Brexit turbulence. We will have to look at a longer timeframe to understand the impacts of Brexit on the economy, VAT receipts and so on. We have to take the time to get that right; I have flagged that up to the chief secretary.

I will ask about the land and buildings transaction tax, but first I will ask a follow-up question on APD. The cabinet secretary said—quite fairly, I think—that the issue hangs on whether there should be notification to the European Commission of the Highlands and Islands exemption. The Treasury had said that it would do that, but only if the Scottish Government took full liability for all the associated risks—historical and future. That was not acceptable to the Scottish Government. If the Treasury were to say it would agree to notify if you were to accept responsibility for just future risks, would that be acceptable?

First of all, we are running close to the clock on Brexit anyway, so the UK Government would be notifying an organisation that we are, apparently, leaving.

I think that the chief secretary, when she came last week, said that she expects that any future arrangement with the EU will include an element of the UK adhering to state-aid rules.

It will, of course, depend what the state-aid rules are. All this is the subject of negotiations. I am being asked a hypothetical situation without knowing any of the legal parameters, so it would be ill judged if I were to try to answer. I am genuinely trying to be as helpful as possible. If there were future liability issues, the problem that we would still have, as the legal advice is that we are not compliant, is that the measure cannot be passed by the Scottish Parliament. We would still not pass go; the law officers would not allow it. That hypothetical situation would not even get past the consideration of the law officers. What we do will have to be compliant in order for us to take it forward in legislation in the Scottish Parliament.

You are saying that the whole question of who has the risk is irrelevant because even if the Treasury were to underwrite the entire risk, because the law officers tell you there is a compliance issue, you cannot proceed.

We could not proceed, if it was illegal.

The rest, therefore, is irrelevant.

We are in the realm of total speculation about a hypothetical situation. However, some way to resolve the problem that would address legal compliance, the fiscal problems, the historical legacy and any future risk with a like power, would be more attractive than where we are right now. I hope that the Highlands and Islands working group will come up with a more practical solution than the one that has been presented to me by Mr Fraser.

Thank you. That was not the question that I was going to ask.

I am relieved.

I want to ask you about LBTT. There was an announcement on Saturday by the Prime Minister that the UK Government is publishing a consultation on changes to stamp duty land tax and is considering introducing a surcharge on SDLT for non-UK-resident purchasers of property. The suggestion is that it would be between 1 per cent and 3 per cent. I appreciate that it is only a consultation that is at a very early stage, and I do not expect you to reveal at this stage what might be in your budget, but if the UK Government proceeds down that route, is that something on which you think Scotland should follow suit?

I genuinely cannot answer that question without seeing the detail, but I advise the committee that I have asked for the information from the UK Government to see how it thinks that would work. I will look at the contents of the consultation. Due to the nature of the fiscal framework and the relationship that we have, if the UK Government does something on LBTT and we do not match it, there will be a financial outcome, so I will have to look very closely at what it is proposing. Once I have the detail, I will certainly look at it. I had no early awareness that the announcement was coming. I will certainly react when I have all the information before me.

I realise that there are potentially supplementary questions from an earlier debate from Tom Arthur.

The cabinet secretary spoke earlier about how the Scottish Government supports the work of the independent Scottish Fiscal Commission—for example, in pressing the HMRC to provide sufficient data when required. Are you satisfied that the amount and robustness of data that is provided by HMRC are sufficient? Is there enough data on, for example, non-devolved taxes such as income tax from dividends, which can act as an indirect indicator of behavioural change—of people shifting their income to dividends or setting up limited companies?

There are many elements to that question. Fundamentally, I ask officials whether we are assured that the SFC has the data that it needs. It has been clear about the data that it requires. We are working through that and are trying to support it to get what it needs from HMRC. I heard the earlier evidence from HMRC, which described the level of accuracy on income tax as being about 98 to 99 per cent. We have a service-level agreement with it. HMRC described comprehensively how it is identifying Scottish-rated taxpayers.

The matter beyond compliance, which Mr Fraser raised, is the wider issue of whether some people go for incorporation and pay corporation tax rather than having non-savings and non-dividend income. We will have to watch that very closely to see the behavioural effects.

That said, the Scottish Fiscal Commission analyses and quantifies what it thinks the tax behaviours are. It will have to look very closely at all the relevant data to make sure that it feels that it is robust in the assessments that it gives us. Notwithstanding everything that you heard earlier, the income tax figures that we have take into account the behavioural impacts of our policy, as well.

That concludes questions. I thank the cabinet secretary’s officials for their evidence. At the start of the meeting we agreed to take the next items in private. Thank you, cabinet secretary.

11:06 Meeting continued in private until 12:25.