Item 2 is evidence on the Comptroller and Auditor General’s report, “The administration of the Scottish Rate of Income Tax 2014-15” and the Auditor General for Scotland’s report, “The administration of the Scottish Rate of Income Tax 2014/15—Report to the Scottish Parliament’s Public Audit Committee”.
I welcome our panel of witnesses: Caroline Gardner, the Auditor General for Scotland; Sir Amyas Morse, the Comptroller and Auditor General; Steven Corbishley, director of the National Audit Office; and Sarah Walker, deputy director and head of the devolution unit at Her Majesty’s Revenue and Customs. I understand that none of the witnesses wishes to make an opening statement, so we will move straight to questions.
First, I ask Sir Amyas Morse to give us an overview of the document that he has provided, particularly in relation to some of the challenges that will be faced in connection with the implementation of the information technology arrangements.
We were given very full access to the documentation, and we have necessarily looked at what is available at this point in time. We published our report just before the communications round that HMRC undertook to alert Scottish rate of income tax payers that they were considered to be in that category. However, there were a number of things that we did not know, such as what the announced rate would be. Some of our remarks in the report are, therefore, slightly anticipatory of what might happen.
You will see from the report that we have a positive view of the IT arrangements and that we give a warning about the level of effort involved in the most difficult aspect of the whole set of arrangements, which is establishing clearly who should pay the Scottish rate of income tax. I will make just one remark on that, if I may. It is important to push on with identifying those people, and we remark in the report—although we are not critical—that not an awful lot has happened and not a huge amount of resource has been put into that area so far.
It is important to push on now, while the rates are neutral, so to speak. If there were to be a differentially higher rate in the future, people might be a little more backward in coming forward to identify themselves and to co-operate. However, we are in slack water, and that is the time to establish as clearly as possible what the population is. I do not disagree with HMRC on that, but I think that now is a valuable time to get a good grip on whom we are talking about. Establishing that population, certainly among those people who are working outside Scotland, will be a considerable challenge.
Can you confirm whether there is evidence that tax advisers are currently giving advice to clients on how they should be domiciled and how they should approach the matter?
I am very much a former tax adviser, so I am not reliable in this area. However, I believe that the only advice that they could give would depend on there being a positive or negative variance in the Scottish rate. I do not see what other advice people could reasonably give. If there was an advantage, a tax adviser might indicate that, but no reputable tax adviser would be involved in concealment or dishonesty. It is a question of seeing how things develop.
I have nothing to add to that, convener. As the Comptroller and Auditor General said, it will depend very much on whether there are different tax rates in the future. Like the NAO, we do not look directly at what tax advisers are doing, but it is something that we will keep an eye on, and I imagine that HMRC will do the same.
Just to play this out, if different tax rates existed in Scotland, England and Wales, is it feasible that we could end up as the Cayman Islands of the United Kingdom? Is that a possibility?
The resemblance between Scotland and the Cayman Islands may be a little remote. We have seen other tax jurisdictions make use of their tax regime to attract economic activity. You do not need to go to the Cayman Islands to see that—you could just go to Ireland. I do not mean to suggest that that will happen in Scotland, but such possibilities exist. It is clear that how you decide to divide the fiscal burden or benefit between the state and taxpayers will probably have an effect on a fairly small proportion of the taxpayer population. Although they might have a high economic value, only a very small proportion of people have a residence in Scotland and live outside Scotland or are tax residents in Scotland but are high earners and temporary visitors. It is fair to say that those are the two populations in which volatility is likely to exist.
I will stop now and let Sarah Walker talk about the matter.
There are two important aspects to establishing Scottish taxpayer status, the first of which is the mass market. We have done a lot of work in that area over the past few months to ensure that we are contacting everybody who is on our books to confirm their Scottish taxpayer status. I can give you more details about that if you like.
Once the rates start to diverge, as Amyas Morse and Caroline Gardner have said, the second important aspect is our focusing on the people to whose income that might make a significant difference and trying to anticipate their behaviour. The fact that the rate will apply only to earned income and not to investment income will make it easier to get a grip on that, in a sense. We will be looking at those people whose earnings from their activity are in play.
A piece of preparatory work that we have done in the current year has focused on our high-earning population—the people whom our high-net-worth units deal with. That work looks specifically at high earners who are resident in Scotland in order to make sure that we understand their residency status when there is a risk that they may have two addresses and might, therefore, want to shift their residence from one to the other.
It is true that having no difference between the rates in the first year of the regime will make it easier for us, because we will be able to do the work to establish where people are resident against a background of there being no incentives for them to try to convince us that they are in one place rather than the other. Once we have that baseline, and once a different rate comes in, we will observe people’s behaviour to see whether there is a shift in response to that different rate. We are doing everything that we can to be prepared for that scenario.
What is the expected cost of the implementation of the IT systems, and what is the cost so far? Will the system be UK based, or is there an option to devolve it?
The figures that we have published for the cost of IT change are between £10 million and £15 million for the entire implementation of the Scottish rate of income tax. We are reviewing those figures, and we will publish an updated figure in the annual report that is due to come out later this year.
The changes are all to our UK-based system; there is no self-contained Scottish-rate system that could be based elsewhere. The important thing about the Scottish rate is that it will be totally integrated into the income tax system, which is run as a UK system. That will make the changes low risk. The changes to our basic pay-as-you-earn and self-assessment systems are just to take account of a different rate; there are no self-contained SRIT IT systems.
Figure 4 on page 14 of the NAO’s report is the “SRIT Project Board risk register”. It is quite reassuring to see that, out of the 13 separate risks, eight are moderate and some are low. I draw your attention to the three high-risk areas. The first is the area that the convener has just mentioned:
“Communications are unclear for those who are not Scottish taxpayers (eg non-UK residents living in Scotland, or the rest of the UK population) resulting in possible confusion and erroneous customer contact.”
I thought that a person had to be domiciled in Scotland to be a Scottish taxpayer. The example that was given when Mr Morse—Sir Amyas, I should probably say—was here previously—
I am happy to be on first-name terms if it suits you.
10:15
The example was given of oil workers from Sunderland who worked in the North Sea, who would not be domiciled in Scotland. However, there is something that I am not clear about. We are all looking at our manifestos for the Scottish Parliament elections, and there are already commitments from my party, the Labour Party and others on varying tax rates. If someone who lived in Scotland had an address in England, for example, and the tax rate on earned income rose significantly in Scotland, could they just use their English address or would they have to be domiciled in England for so many months of the year? I am not familiar with tax exiles or the Cayman Islands, but if someone in Scotland was a bit annoyed about paying more tax than a person who earned the same income in England, how easy or difficult would it be for them to get an address in England to avoid paying that tax? What are the criteria?
As you have said, the rule is that a person must have their main residence in Scotland for the majority of the year.
What does “main residence” mean? If I had a house in Edinburgh and a house in London and I dropped between the two, which would be the main one? What is the definition of the “main residence”?
In most cases, it would be a matter of facts. One place would be the person’s home and the other place would be a temporary residence. We would look at where the person’s children went to school, where the person was registered with a general practitioner and where the person considered their main residence to be for their general living arrangements. For the vast majority of people, those things would be very easy to determine.
A person may not have children at school, and they would be entitled to be registered with a GP wherever they lived. For 20 years, I have been registered with two GPs, one in Inverness and one in Edinburgh.
Yes. We have not set out a specific list of criteria, but we have published guidance on our website that explains the type of things that we will take into account in determining where a main place of residence is.
If you have not established definite criteria and the higher tax rate goes up to 50p or 60p in the pound, or whatever, it will potentially be easy for a Scottish taxpayer to say that they have an address in Carlisle when it could be a single room. That will be quite easy to do to become a tax exile in England.
Allow me to reach into what I know about the matter. There have been a lot of tax cases to do with residencies, and there are quite well-established criteria for judging the facts. Obviously, each situation would be looked at on its merits, but it is not as if inspectors of taxes would find the matter incredibly difficult to understand. We have had all this with people who have tried to show that they are resident or non-resident in the UK tax jurisdiction, so it is not new territory. The approach that will be applied in Scotland is well understood.
But the approach is new between Scotland and England, and that is what my question is about.
I do not disagree with that at all. I am simply trying to reassure you that, from the point of view of how we go about establishing the main residence, we are not on terribly new ground. It has been necessary to establish domicile and main residence for tax purposes for ever, so there is quite a lot of expertise in how that is gone about. People are not making it up as they go along.
It is true that there is no open-and-shut answer. The circumstances of each person and all the various things that have built up over the years that are used to determine their main base, if I can call it that, are looked at. That is how things work.
That is right. There is a fallback. If it is impossible to establish in a particular case that a particular place is the main residence, we can count the number of days that are spent in England rather than in Scotland over the year. We think that there will be very few cases in which we will need to go into that level of detail, but that is a fallback if we need it.
That will be quite difficult to prove if a person drives up and down the road. The eligibility criteria are not clear, and the example of a person being registered with a GP—we have a GP with a bad cold sitting here—is not the best one. Also, not everyone has children who go to school or nursery. Therefore, as things stand, the issue does not seem clear to me.
As Amyas Morse has said, this is not a new issue for HMRC, but it is something that we will need to keep under review and develop. We have developed guidance in consultation with tax professionals in Scotland, and we have issued examples involving students, people in the military and other people who move around a lot. We have tried to address the obvious questions relating to people who might be in doubt, but we will keep the matter under review.
I appreciate that the issue is not new, but it is new within the United Kingdom.
Yes.
We are looking at new territory here.
Page 14 of the report says:
“Communications are unclear for those who are not Scottish taxpayers (eg non-UK residents living in Scotland, or the rest of the UK population) resulting in possible confusion and erroneous customer contact.”
Which people are in the group that has been determined to be high risk?
We are not talking about high-risk people; the rating in the column concerns the degree of effectiveness of our controls.
The effectiveness of your controls is high in that regard.
Yes.
That is good news. In that case, I should be looking at lines that read “Very low”.
The reference to non-UK residents is there simply because someone has to be a UK resident before they are in the frame to choose whether they are a Scottish or a UK resident. If they are a non-UK resident, they should not have to worry about the Scottish rate at all. The risk that we are talking about concerns a situation in which people are confused and contact us even though they ought not to be concerned about the Scottish rate at all.
Given that there is not a high risk, are we saying that everything is good and working well? Everything that I am reading seems positive, and there seem to be good working relationships between HMRC and the Scottish Government. That is what we were hoping for and had been promised, so that seems good. Have you identified any risks to that?
We are monitoring all the risks carefully. The most important thing is to ensure that we have a high level of confidence that we have accurately identified Scottish taxpayers. We are in the middle of the process of initial identification. We have written to 2.6 million people in Scotland and we are now dealing with address changes that have been notified to us. As a result of that, we have produced some more advertising that is due out in February, which is designed to reach people who might not have had a letter from us although they should have. We are continuing to pay a lot of attention to that area. As we move into the new financial year, we will think about what continuing monitoring we have to do and we will prepare for compliance work where we think that is necessary.
I want to clarify a couple of things. The Scottish rate of income tax is not actually a devolved tax, is it? The Department for Work and Pensions paper makes that clear. It is still part of the UK tax system. I see that our witnesses are nodding.
Paragraph 1.3 of the NAO report says:
“There are a number of exceptional cases whereby a taxpayer would always be deemed Scottish, for example if they are a member of Parliament for a Scottish constituency.”
I assume that that includes MEPs, list MSPs and so on, and that it is not restricted to someone who represents a constituency. Again, I take our witnesses’ nods as affirmative.
On the issue of relief at source, which affects the pension companies, there is a two-year transition period in place as of this year, as far as I understand. Presumably, there have been talks with the pension industry on the issue. It does not seem to me to be a terribly complex issue, but perhaps you can give me a bit more information on it.
The thing about relief at source is that, for the first time, the pension providers will need to be able to distinguish between their contributors on the basis of whether they are potentially eligible for relief at a different rate. For example, if the Scottish basic rate is different, Scottish taxpayers will be entitled to a different amount of tax relief.
In relation to the generality of the Scottish rate, with PAYE, employers have always been able to distinguish between their employees and apply different tax rates because of the tax code system that we have. In relation to pension relief at source, we have never had a mechanism that would enable us to distinguish between contributors. Therefore, we do not have an automatic system in place that allows the pension scheme to calculate the claim for tax relief based on the different status of different taxpayers. Those systems have to be developed. A wide variety of companies—insurance companies and various pension companies—provide pensions, and they all have different IT systems and different challenges in relation to how they keep their records of their customers. We are allowing them extra time to ensure that they are able to distinguish and identify their taxpayers on the basis of the rate that they are liable for, so that they can make claims at the right rate.
But surely pension companies already deal with a multiplicity of jurisdictions, so it cannot be that complicated for them.
In order to do what they have to do, they have to be able to identify each contributor by their national insurance number and to use that to confirm with us whether the person is a Scottish taxpayer. That involves two-way communication. We have to ensure that the companies are giving us the right information with regard to the identity of their contributors, and we help them to understand the rate that they should charge. It is a major change for pension schemes, against a background of an awful lot of other types of change for the industry. That is why we agreed that we would give those companies extra time to get ready for it.
Is the two-year period reasonable?
I think that it was agreed with the industry at the time, following extensive consultation.
Clearly, there will be complications if, at the end of that period, we do not have an arrangement in place.
If the rate is different, we will have an interim arrangement in place. This issue arises only if the basic rate in Scotland is different from that in the UK. If the basic rate is different in 2016-17 and 2017-18, we have an interim solution that involves HMRC making changes to PAYE codes to adjust the amount of relief. That is a bit messy, but it will get the right amount of money to the right people.
It would be accurate, and there would be no loss of money to the Scottish Government.
That is correct. It would get the right amount of money to the claimants.
We hope that, once we get the system in place, along with the pension industry, the system will be much smoother and more automatic. The plan is that that will be in place in the event that the basic rate is different in 2018-19.
In the section of the report entitled, “Key findings”, paragraph 4 says that HMRC
“has put only limited resources into specific planning for potential tax loss as a result of SRIT.”
Could you elaborate on that?
That relates to the resources that we were putting into planning for compliance in 2014-15. More recently, we have increased the amount of work that we are doing on compliance. The work that I mentioned in relation to high-net-worth individuals is an example of that. We are also doing quite a lot of work with major employers to ensure that we have an assurance that their PAYE systems will operate the Scottish rate correctly. We are ramping up the work that we are doing on compliance.
The risk is clearly less acute when the rates are the same, as opposed to there being a differential rate between England and Scotland. However, it is still important for us to make sure that the allocation between Scottish taxpayers and rest-of-UK taxpayers is as accurate as we can make it. We are making sure that there are plans in place to support that.
10:30
Do you have any idea what percentage of income tax is collected? I presume that a small amount is not collected, for one reason or another.
We believe that around 99 per cent of the tax on earned income—which is what we are talking about—comes in automatically, so the tax gap is relatively small.
We have already talked about tax evasion and avoidance and so forth. If the tax rates move, it is clear that a lot of inventive reasons for such activity will be given on both sides of the border. We have been talking about Scottish tax rates going up and tax rates south of the border remaining the same, but it could be the other way round. We need processes in place to catch tax evasion on both sides. Is that being looked at in such a way that there is not just an assumption that the Scottish rate will go up and that there will be a need for measures to address Scottish taxpayers trying to relocate or to become tax residents south of the border? It could be vice versa.
Absolutely. The important thing for us is to get it right, rather than to maximise tax on one side or the other. The sorts of compliance strategies that we are looking at are equally applicable whether the rate in Scotland is higher or lower.
I want to ask about the DWP side of things, if I may. I do not know whether our witnesses are comfortable looking at that.
The costs on the HMRC side are becoming clearer as things develop, but the memorandum of understanding with the DWP simply says that the Scottish Government will be charged quarterly in arrears. Do we have any idea of what those costs are likely to be?
I am afraid that that is not within the scope of the work of the National Audit Office. At some point, we can look at that issue separately, but we have not looked at it for the purposes of today’s meeting. It is not in the report.
I wonder whether Caroline Gardner has any comments.
Thank you, Dr Simpson. You will have noted the exchange of letters relating to that. It is clearly another adjustment that will be required around the Scottish rate of income tax. As you would expect, we will be doing a proportionate amount of audit work on that from the point of view of the Scottish Government, as we are doing around the HMRC interaction. The exchange of letters is very recent, so it will come into our planning for the audit year that is about to start. We will provide the committee with more information on that at an appropriate point. We expect that very small amounts of money will be involved.
In October or November last year, the Finance Committee received a list of the risks. Now we have the new report. Have there been any shifts in the risks between the report to the Finance Committee and the report to this committee?
Not as far as we can see; there is nothing that I need to alert you to. So far, the conversation has related to the main risks as I see them. The need to successfully identify the right group of taxpayers is way above everything else.
Good morning. I am going to refer back to the risk register. This has partly been answered in response to the question that my colleague Richard Simpson has just asked.
There are a couple of things in the assessment of effectiveness in the risk register that are rated as “Moderate”. I am looking for a bit of clarification, and I apologise if I am going over old ground. The issue is to do with the IT systems.
One of the risks listed—it is the seventh item—is that
“The project does not manage the IT complexity and planned technological change”.
Another is the fact that
“The Department for Work & Pensions and HMRC are not joined up in their delivery of SRIT.”
Among the difficulties that are faced, both those risks are seen as being “Moderate”.
Perhaps it is to do with the fact that we have seen a number of IT projects in which it has been claimed that they are going to plan but they are not. Given that you have rated the risk as “Moderate”, how confident are you that that assessment is absolutely correct, going by the systems that have been used before by various Government departments, in terms of the effectiveness of delivery?
The way that the risk register has been laid out has the risk of being misleading. What we are describing as “Moderate” is HMRC’s assessment of the effectiveness of the controls of the risk, not the level of risk. That is quite important.
Right. I beg your pardon.
We are saying that this is probably the least risky route to getting an IT capacity in this area. It will involve using a system that is working well and simply modifying that—not to a very great extent—in order to deliver SRIT. That is far less risky than what would have happened if you had said, “We’re going to go off and write a new system.” Frankly, that would have been a waste of money, and much more risky.
As IT projects go, we do not regard this as a risky project. What is stated in the risk register is simply that the controls over the project are moderate rather than very high. That probably reflects a reasonably confident attitude on the part of HMRC about carrying out the project.
Yes, indeed. All the changes to our IT systems that are necessary in order to launch the tax in April—the changes to PAYE to cope with the special tax codes that are being given out to people who are liable for the Scottish rate and to enable us to register and take account of changes of address and register the date of changes of address—have been put into the system. They are there and working already.
The next tranche of IT changes will relate to the self-assessment system. People will not put in their self-assessment returns for 2016-17 until after the end of that year, so we have longer to make sure that those changes are in place. However, those are relatively routine changes for us. They go with the grain, if you like, of the PAYE and self-assessment systems and we are confident that the process will go smoothly.
I have one more question, which relates to the eleventh risk that is listed in your risk register. You say:
“Other organisations publish information about SRIT that is incorrect, misleading or wrongly timed, leading to customer confusion and contact.”
I do not quite understand which organisations you are referring to. Could you expand on that point?
If, for instance, there is an article in the press about how SRIT works and it is wrong or misleading, that can cause a big rush of inquiries and phone calls to HMRC and we need to deal with that confusion and help people to understand. Some of the risks are about levels of contact and levels of phone calls that we might need to be prepared for. That is one reason why we might get a rush of phone calls that we would have to be able to cope with.
It is just that it was not clear whether you were talking about the media or other Government departments.
It could be anybody, but the media is one example.
Good morning. Sir Amyas, I would like to ask you about your memorandum of understanding with the Auditor General. It tells me a story, which I think is correct. In paragraph 5, you say that Audit Scotland does not
“have any statutory right of access”
to HMRC data. Equally, at paragraph 10, you say that you are
“independent of government”,
which I very much welcome. At paragraph 31, you say that you and the Auditor General are both
“committed to working together in an environment of professional respect and promoting a culture of openness”,
which of course is absolutely what we want to hear.
At paragraph 6, you say that you are
“happy to work together to ensure that the Parliament receives appropriate reports”.
My difficulty lies in understanding what appropriateness might be. Who is going to draw the line? Frankly, what am I and my successors on the committee going to be able to do if you and your successors draw the line somewhere where we are not comfortable with it? You hold the pen that decides where the line goes. Could you explain how you see that?
I will answer that on two levels, if I may. Once we understand what the rules are and exactly what the implementation of the Scotland Act 2012 calls for, it will be possible for us to have a conversation about the nature of the assurance that we can or should give, which will flow from the nature of the information to a large extent. You would know that the approach was not working correctly if we did not agree on it—that would give rise to something, and we will strive hard to avoid that.
If there was an eventuality on which we did not agree, the most likely reason for that would be that there was some clash between my preserving my position in regard to the UK Parliament and what I might agree to for the purpose of giving assurance to the Scottish Parliament. I am sorry if that sounds a bit theoretical. What I am trying to say is that my primary job is in providing assurance to the UK Parliament; I am sure that you do not have to be troubled with my work up here, but I try and do that in a pretty forthright way, so my independence is extremely important to me. Therefore, the most likely reason for our having to work our way through something would be that it was likely to have an unplanned effect on my independence in relation to the UK Parliament. At that point, we would just have to work our way through it.
I am with you there, and I respect your independence just as I respect the Auditor General’s independence. You can be complete pains to the Government because that is what we pay you to be. I have no trouble with that.
Thank you.
Nor do I have any trouble with the fact that you might occasionally disagree. What worries me is what would happen if that disagreement came down to somebody having to pay for making an effort to do some work. I suspect that you and your successors will find a way of working with the Auditor General, but what would happen if there were a disagreement about who was to do the work to check something—to dig into something—and you said, “Hang on—I don’t want to pay,” but the Auditor General could not pay because you held the purse strings? I am trying to explore that kind of operational issue. I am not for a moment suggesting that that will occur between you two good folk, but it might occur some way down the track between your successors.
That is a valid point. Who is resourcing whom? We need to find a way of discussing that with the commission that funds me. At the moment, it funds me primarily to do UK Government work, and it would be quite a big change for it to say that it was prepared to fund me to do work in another way. It would also depend on the agreed interpretation of the rules and legislation to do with accountability at a Government level and, to be honest, I am rather keen not to get out in front of that discussion, as it is not my job to determine such things. Nevertheless, I understand that there might be issues that it would need to be possible to determine.
What you suggest is unlikely to be a big problem. If this committee, for example, wanted a lot of work to be done within my remit that meant that I had to do a lot of additional work, I would have to discuss with the Public Accounts Commission what it regarded as reasonable and unreasonable. However, I am quite sure that its attitude would be that, unless the work was going to make a big difference to my budget, it would not rush to be difficult about it. If the work was substantial, some agreement would have to be reached between Westminster and the Scottish Parliament, and that agreement would have to be brokered at a level above me.
If we identified that as a block, we would pull it up and say that it needed to be resolved at parliamentary level, because it is not my job to be arbitrary about such matters. I do not want to be arbitrary; I want to ensure that the matter is discussed properly at governmental level.
10:45
I am grateful for that. It is useful to put that on the record, because we might get there, although I hope that we do not, of course.
I will add to that briefly, because it is important that you hear from me, too. As Amyas Morse said, his primary responsibility is to provide assurance to the UK Parliament, and my primary responsibility is to provide assurance to the Scottish Parliament. In the past, those two roles have had very little to do with each other in many ways.
Because of the way in which the Scottish rate of income tax has been set up, we now have a shared interest in relation to HMRC’s collection of the tax on the Scottish Government’s behalf. I am clear that, as the memorandum of understanding sets out, the responsibility for auditing HMRC and the rights that go with that sit with the C and AG and not with me. That is why I was happy to agree to the committee’s request to provide an annual report that would provide additional assurance on the work that has been planned and carried out and the conclusions from it.
It is possible that, in the future, I will feel that the Scottish Parliament wants or would benefit from more assurance than the C and AG feels it is appropriate to provide. That is exactly the sort of issue that I would draw to the committee’s attention in my report that sits alongside the C and AG’s report. Resolving that would be a matter that is above our pay grades, but you would have the assurance that I thought that work needed to be done or, speaking hypothetically, that I disagreed with the conclusions that the C and AG had reached on the basis of the work that he had done. Therefore, that is an important check and balance in the system, which reflects the new arrangement of shared dependency on HMRC to collect the tax.
I have a couple of questions about the risk register. The first issue has been touched on. Risk 13 is that
“Communication products designed to inform the UK population in relation to SRIT are not delivered to time and specification.”
Colin Keir asked about that, and Sarah Walker gave the example of the media. I am interested in the panel’s opinion as to whether additional work should be undertaken with the media so that they are fully informed about SRIT and how it will be delivered and they can then be part of the solution by getting the correct information out to the wider population. I am sure that inaccurate information from the media would have an adverse effect on HMRC, as it would receive a huge number of additional phone calls.
Yes. The media have been a big part of our communication strategy. When we did the big mailshot in December and sent out letters to the Scottish tax-paying population, we did quite a lot of proactive work with the media in Scotland. We issued background material and briefed journalists to ensure that they had material to put in their publications to explain to people what the Scottish rate of income tax is all about, why they were about to get a letter and what they needed to do. That led to a number of articles in the Scottish press, and quite a lot of activity online on things such as Twitter.
As I said, we will follow that up with direct advertising in the next couple of months. There will be advertising in the press and online and coverage on the radio. The message that we are getting out is simple: people should make sure that HMRC has their up-to-date address so that we can correctly apply the Scottish rate.
The announcement of the rate in the draft budget in December was another opportunity for us to draw attention to the Scottish rate. The fact that the rate will not be different attracted a certain amount of publicity, and we used that opportunity to give background information to journalists so that they understood the situation. We were pretty happy with the coverage that we got in December and we were certainly happy with the level of accuracy.
That is helpful. My second question concerns the first point on the risk register, which is that
“The Scottish Parliament does not agree the Scottish Government’s Scottish Rate Resolution, resulting in a late change of rate.”
That risk is marked as “Low” under your
“assessment of the effectiveness of controls in place”.
Will you provide a bit more information on that, please? I pose the question because who will form the Scottish Government after the May election is probably not an issue. Obviously, apart from that, the electorate will determine what the make-up of the Parliament is.
The legislation requires the Scottish Parliament to set a rate by the end of March—by the end of the financial year—so that a rate is in place for the new financial year. If that decision were to be taken right at the very end of the financial year, we could still apply the Scottish rate. However, that would give rise to additional costs, because we would have to change people’s tax codes after the start of the financial year. That would involve extra costs for us and more disruption for employers.
The memorandum of understanding that we have with the Scottish Government asks it to give us an indication of the rate by the end of November, before the start of the financial year. That is the assumption that we will use in order to set all the PAYE codes and tax collection mechanisms that we need, and that minimises the cost and disruption.
That is the way in which we would prefer to proceed. However, the legislation allows the Scottish Parliament to not set the rate until the end of March. We can cope with that and there is no problem, but it creates additional cost and disruption, so the risk assessment reflects the fact that there would be additional work to do and additional cost if the decision on the rate was taken at the very last minute.
I assume that, when the legislation was going through the parliamentary process, HMRC raised that point with the Scottish Government.
Yes. The Scottish Government is very aware of what we need to be able to operate PAYE. That is why we asked it to give us the assumption at the beginning of December. The Scottish Government knows that if, for whatever reason, the rate was set differently at the end of March, that would cost the Scottish Government additional money and be more disruptive for employers.
The point that I raised was whether, when the bill for this to take place was going through the parliamentary process, HMRC actually raised this issue.
Do you mean the Scotland bill back in—
Yes—the 2012 act.
I am pretty sure that the issue would have been discussed at that time. The time limits are in the 2012 act, so the process for having a Scottish Parliament resolution to set the rate and the implications of the timing of that certainly would have been discussed at the time.
It would have been quite embarrassing if we had had two auditors general today and you had worked together to prepare for the introduction of the tax and gone way over budget. However, that is not the case.
Caroline Gardner’s report states that you have worked well together and that the
“estimated cost ... of setting up the Scottish rate of income tax has decreased by £10 million”.
We are very used to hearing of increases in this committee. Is that decrease a result of incredible efficiency on HMRC’s part or did you overestimate in the first place?
I think that it is down to us. The original estimate of £40 million to £45 million was made back in 2010, so it is a very old estimate. It was done at a very early stage when we had not done the work on what would be involved. We are now down to an estimate of between £30 million and £35 million, which we published in early 2015. We have information about the amount of contact and the amount of activity that we have had to do for the mailshot. That has been less than we expected, so I think that we will reassess the overall estimate.
The best explanation is probably that the original estimate was quite generous, because we did not have a good idea of what was going to be involved. We have put a lot of work into refining the plans for the IT change; we also had to make an assumption about how much activity, phone calls and contact we would get. We erred on the side of caution, and the cost has turned out to be less than estimated. Therefore, the estimate has come down over time.
I appreciate that, and because of that you are 20 per cent under budget.
I have a question for Caroline Gardner. As we have heard, the cost of implementation has been 20 per cent under budget, but Revenue Scotland has gone a good 20 per cent over budget. I appreciate that the issue is more a matter for the next agenda item, but in the context of looking at savings in one budget, will you explain the situation?
I think that the convener would probably prefer me to answer that in detail under the next item. In broad terms, the answer is similar to the one provided by Sarah Walker: the estimate was made at an early stage, and as the scale of the task became clearer, the budget also became clearer. I am happy to answer in more detail under the next item.
I thank the panel for its contribution. I briefly suspend the meeting to allow a changeover of witnesses.
10:56 Meeting suspended.Next
Section 23 Report