Finance and Constitution Committee

Meeting date: Wednesday, October 5, 2016

Official Report 429KB pdf

---

Land and Buildings Transaction Tax

Item 2 is to continue to take evidence in our inquiry into the first year of operation of land and buildings transaction tax. We will have two separate sessions on LBTT today. In the first session, we will consider the Scottish Fiscal Commission’s 2015-16 outturn report, for which we are joined by all three members of the commission. I welcome to the meeting Lady Susan Rice, who is the chair of the commission, Professor Campbell Leith and Professor Charles Nolan.

I know that members have received copies of the correspondence from the clerks. Does Lady Susan Rice want to make an opening statement?

I will just say a few words.

I thank the committee very much for having us at the meeting. Most of you met us during your business planning day, but we thought that it was important for all three of us to be here. The previous committee saw Campbell Leith and me regularly, but this is Charles Nolan’s first appearance before the committee, so he is really in the hot seat.

That means that you get to answer all the hard questions, Charles.

Charles Nolan is our interim appointee, approved by members, until 1 April next year. He came on board at the beginning of the summer and has jumped in with two feet and added huge value right from the beginning. The transition has been quite seamless for us. He is a professor of economics at the University of Glasgow, but he worked in forecasting at the Bank of England in the early part of his career. Therefore, he brings a different perspective to our conversations, which is very good.

Campbell Leith is also a professor of economics at the University of Glasgow. Although there were lots of inputs into the outturn report, it was absolutely led and shaped by him in the late spring and the early summer.

I thank both individuals.

I will give a tiny bit of context. I remind the committee—I know that the committee knows this, but I remind people who have any interest in the matter of this all the time—that the commission is in the third year of acting as a body that scrutinises forecasts that are produced by the Government. We scrutinise forecasts and report on our findings; that is our only remit now. We do not do the forecasting, but we are getting ready to begin that as of next April.

In recent evidence sessions, there has been chat about all the different reports that seem to reflect property transactions and the LBTT take. I give a quick reminder in layman’s terms that, for any property transaction, we have the date of the transaction, which leads to the requirement that some tax be paid, but the payment of the tax is generally subsequent to the date of the transaction. If the property transaction is towards the end of the month, the tax might well be recorded by Revenue Scotland in the following month. Therefore, there is a bit of a disparity if we look at the month figures. Registers of Scotland then registers the transaction, but it waits until the tax return has been filed and the tax is in. There is a sequence.

Registers of Scotland will include all properties that do not attract the requirement for LBTT, including all transactions under £145,000. Obviously, those of us who look just at tax do not need to look at those transactions.

Finally, we believe that the Scottish Property Federation, which also publishes statistics, works on a calendar year rather than the fiscal year, so there are different timeframes.

We remind ourselves about those differences, but it is probably important to restate them here. We believe that, if we go to the raw data behind Revenue Scotland’s and Registers of Scotland’s data, we can reconcile those numbers with ours, so we are not troubled by what appear to be disparities in the published reports.

If the committee is interested in forestalling as a phenomenon, it should probably not use the Registers of Scotland data because the way in which it smooths the activity does not give as clear a picture and it does not collect data to reflect forestalling.

I think that that is all I have to say. The outturn report that I think the committee wants to consider is based, as it states, on Revenue Scotland outturn numbers, so that is the relationship that we have used. We look forward to the committee’s questions.

Thank you, Lady Rice. We have asked questions about the compatibility of different sets of figures in previous meetings, so your explanation of how different sequences from different organisations produce an apparent disparity in the figures is very helpful to me and, I am sure, the rest of the committee.

I am interested in the forecasting model. I assume from what you have said, Lady Rice, that it is a Government forecasting model in the meantime. However, I am interested in how it operates in practice, because it involves both mean and median house prices as well as volumes of transaction forecasts. We can see that all those things can be brought together, but I am struggling to understand how they are applied thereafter. It would be helpful if someone could explain to me how that methodology or modelling works.

I will take this one. As you said, three economic determinants go into the residential LBTT forecast: the average house price, the median house price and the volume of transactions. Those determinants are forecast separately by Scottish Government analysts and are then converted into revenue forecasts that are based on the application of a probability distribution. This is how that works. Imagine all the transactions taking place in the economy: some will be for relatively inexpensive houses and some will be for expensive houses. If we plot all those transactions for a given period of time, we can fit a curve to all the transactions that explains the patterns that we see in the data. The curve that fits the data is called a log-normal distribution. That means that, if we were to pick randomly a transaction from the Registers of Scotland database, the log-normal distribution would give us the probability of that transaction having a particular price. There would be a far higher probability that the transaction would be relatively cheap and a very small probability that it would be a very expensive house. However, that distribution of price transactions seems to fit the data.

The Scottish Government forecasts average house prices and median house prices, and uses those to shift the distribution forward. As average prices go up, there will be more transactions at the upper end of the market and fewer at the lower end. The Government forecasts average and median prices, which relocates the probability distribution. The Government then forecasts the total volume of transactions and then multiplies that by the probability that each transaction is going to be at a certain price point. The Government then applies the tax rate that would apply to a house of that value and calculates the tax revenue as a result.

Okay. It is complicated. The forecasts were not quite where the outturn turned out to be, so where in the forecasting process did that disparity, if that is the right word, arise?

Table 1 in our outturn report shows exactly what we tried to do. The first column in table 1 gives details of what was forecast for the relevant economic determinant. The second column gives what the outturn was and shows that the average house price was forecast to be £174,000 but turned out to be only £166,000. The third column then runs that revised piece of data through the probabilistic model that has been developed for forecasting, and computes a revised forecast.

By combining all these different ways in which the forecast economic determinants were not quite the same as the outturns, we can compute a revised forecast. The combined effect of those would be that you would expect revenues of £196 million, not the £235 million that was originally forecast. If you add in the additional dwelling supplement forestalling of between £5 million and £7 million, that means that the probabilistic model in aggregate fits relatively well once you forecast average and median house prices and transactions correctly.

Before I start, I draw members’ attention to my entry in the register of interests with respect to property investments.

There is obviously quite a lot in this, so I will try not to chew up too much of the time. Basically, you have compared the outturn to the forecast, which is fine as far as it goes, but the problem with that is that it might mean only that the forecast was wrong. I know that the forecast was changed through the year, so if the forecast is wrong, that is all the comparison tells you. I am more interested in the comparison with previous years and in whether the total tax take has gone down compared with the take in previous years as a consequence of the move to LBTT. It might be interesting to hear your comment and we can drill into it in a bit more detail.

I understand what you are saying about the modelling of the forecasting but, again, there is clearly no historical data to base it on because of the change. My statistics is a bit rusty, but I think that the log-normal distribution will depend on the value that you have for the standard deviation. If that changes, the profile will be different. All the data is telling you is that that is what has happened. The forecast aspect of the log-normal distribution has not fitted to what has happened because you have not forecast it correctly and picked the right values there.

The average and the median have taken a downward shift, and you are right that there is a tilt away from the upper end of the house spectrum because the difference between those two is bigger than it was before. They are both down and that says that there is a clearly an impact from the forecast being wrong, because house prices in general—not just the distribution—were expected to be higher than they were. You could comment on the forecasting.

As I say, the biggest thing that I want to talk about is how the impact compared with the impact in previous years. In the data that I have seen, the number of transactions in the key £375,000 to £750,000 range looks to be at least as high as it was previously. I do not think that there is any issue with the number of transactions in the top range apart from forestalling; the number has held up very well, as far as I can see. You might want to comment on that.

There are a lot of points there; I will try to remember each of them.

Table 1 does not just compare the forecast with the outturn; it tries to decompose the forecast error by rerunning the forecasting methodology with actual outturn data piece by piece. That suggests that the biggest impact on the forecast was through failing to forecast average house prices. The fact that average house prices did not rise relative to median house prices is indicative of a shift in the transactions away from the top end of the market towards the middle and bottom. That is largely the reason for the fall in revenues relative to expectations, which is what we try to analyse throughout the next section of the report.

On the standard deviation of the log-normal, once you know the mean and the median, those tell you the standard deviation. You can calculate the standard deviation from that. The fact that average house prices have not risen relative to median house prices is telling us that the standard deviation has shifted and the shape of the distribution has moved towards the lower end of the market.

Comparing with historical data is difficult because this is a completely different tax regime. When the tax regime was first implemented, it was designed to be revenue neutral. Then, when the rest of the United Kingdom adjusted its tax regime, it lowered the tax take that you would have expected and tax bands and rates were adjusted for Scotland, so you would not expect revenues from this tax regime to be comparable historically.

We have looked to decompose the forecast but, in the evidence that it provided to the committee a week or so ago, the Scottish Government looked at transactions historically using Registers of Scotland data. That is in table 1 of that document, if I remember correctly. The Scottish Government found that, over a two-year period, there was substantial growth in many of the price bands, except for the £325,000 to £750,000 band, in which there was growth of only 1 per cent. In last year’s report, we referred to the large fiscal drag effect in the Scottish Government’s forecasts. That arises because, as house prices rise, more properties are pulled into the £325,000 to £750,000 price band. Therefore, we would expect transactions in that zone to increase over time rather than growing by just 1 per cent.

09:45  

Basically, the data that I have—which I have not had time to compare to your log-normal distribution—says that the number of transactions in the £325,000 to £750,000 range is about 7.5 per cent of the total number of transactions, and that it has been between 7 and 8 per cent throughout the past few years, except during the quarter in which the forestalling kicked in prior to the change, when it was more than 10 per cent. That suggests that the number of transactions in that range as a percentage of the total has not changed. I think that you are saying that you would have expected it to have increased because of the drag on the market.

The average numbers pre and post-financial crisis were a bit higher and lower. If we looked at those two relative averages, I would expect that the share would be marginally lower, but we would expect it to rise over time, for the reasons that I have just given.

Okay. Are you saying that you cannot answer the question, “Has the change to LBTT hurt the tax take to the Scottish Government?”—which I think is the key question—because we are comparing apples and oranges?

Yes. It is a different tax regime, so we were not expecting it to generate the same amount of tax.

I will pursue some of the questions that Ivan McKee raised. Table 3 in your paper shows the outturn revenues compared with the forecast revenues. It is clear, as you mentioned, that the most sensitive price bracket is the £325,000 to £750,000 bracket, in which, relative to the other brackets, there has been a substantial reduction in the outturn revenue compared with the forecast revenue.

In the past, I have bored members of the committee with my anecdotes about the property market in the country areas, and the committee has had quite a lot of evidence from people on the slowdown in the property market in larger properties, particularly in the country areas. Do you accept that there is an issue with the breadth of the £325,000 to £750,000 band? If the number of transactions in that bracket is holding up, that might be because there are more transactions grouped at the lower end in the £300,000s and fewer transactions over £500,000. That would mean that the level of transactions was the same, but that the average price has fallen, with the result that the tax take has fallen, too. Would that be fair?

Yes, that is possible. We have Revenue Scotland data only for the entire band; we do not have data on sub-bands within the band. If we had such information, it would be useful to do that analysis.

In a previous session, the committee picked up that having the ability to interrogate the data further to understand what is happening within such a broad band would be extremely useful.

In paragraph 3.14, you comment on the forestalling effect in March 2016 when, because of the impact of the additional dwelling supplement, there was a jump in the tax revenues. Is it fair to say that, without that forestalling impact, the outturn would have been worse?

Yes. Transactions appear to have been brought forward into March, which was a particularly buoyant month. As the situation was not obviously as buoyant in the following month, it looks as though there is an element of forestalling there. If you took that away, revenues would be lower.

So, in the absence of that forestalling effect in this financial year and subsequently, we might expect that to disappear and a more normal pattern to appear.

Yes. In the report, we look at March and then we look at April to identify whether the buoyant returns in March were sustained into April. Our tentative conclusion at the moment is that they were not, but we will continue to monitor outturn data for this year as it emerges.

My final question, which goes to the heart of the matter, is about your comments in paragraph 3.24 about what has been happening in the market, particularly in the £325,000 to £750,000 band. Is it your view that the rate at which LBTT has been set has affected properties in that band and has had an impact on reducing activity and therefore reducing the tax take?

We cannot draw that definitive conclusion based on the limited data that we have at the moment. The report notes that that is the band in which revenue seemed to be lower. There could be various reasons for that. It could be a behavioural response, or there could be other shifts in the market. It could be that the log-normal distribution does not fit the top end of the distribution as well as we would like it to. There are various explanations. At present, we cannot rule out the possibility that you mention and we cannot definitively conclude that that is the case. We continue to monitor the data as it emerges.

At what point will we have sufficient data to be able to draw conclusions?

We would need to rule out all other options. We would need to rule out the possibility that damage to the property market in the Aberdeen area is causing the issue. We would need to look at whether the log-normal distribution fits the top end properly. We continue to explore the issue. In effect, by ruling out all other options, we would conclude that we cannot reject that hypothesis.

In principle, that could take quite a long time. Even with the additional dwelling supplement, it will be 18 months down the line before we have the final data, because there is a long period of reclaim. That affects what we can take from the data. At the moment, we have only one year’s data. After the ADS is all sorted, we will have two and a half years’ data and we will then be in a better position. We probably cannot definitively claim that that will be enough, but we will certainly be in a better position, I would think.

You are keeping a close eye on the issue and monitoring it.

Yes. Most of the tax take comes from the top two bands so, clearly, in forecasting the tax take, you want to get those two right.

You have just had a discussion with Murdo Fraser about the breakdown of the figures in the £325,000 to £750,000 band. We have discussed that with other witnesses. Have you had any discussions with Revenue Scotland about how a breakdown of that information might be made more readily available?

We requested data broken down as much as possible. The data that was received was broken down by price band. I do not know whether further breakdown would be available. Earlier this week, we asked the team in Governor’s house to ask for the next round of data broken down as much as possible, not only by month and price band but ideally by region as well. We want to look at the regional distribution so that we can start exploring whether there are other explanations.

We have witnesses from Revenue Scotland next, so we can follow that up with them.

I am so thankful that I studied maths and statistics for three years at university a number of years ago—ultimately, it has come in handy. I think that I followed most of the previous discussion on the log-normal distribution and so on.

At our meeting last week, there was some discussion about the impact of the higher tax take on the band from £750,000 and above. We discussed evidence and contributions from colleagues that additional measures needed to be introduced to help that side of the market. However, that does not appear to be borne out by the data that you have presented in table 3, which Professor Leith referred to earlier. In fact, the bracket over £750,000 was the best performing of all the bands in terms of tax return. I do not think that data was available to us in previous weeks, but that is clear. Will you confirm that I am reading that table correctly and that the higher value property bracket was in fact the best performing against forecast?

The graph in figure 2 shows the position more clearly.

What page is it on?

It is just above paragraph 3.20. The solid black line in the bottom right-hand corner is the tax take by month throughout the year and the darker dotted line—

You have a black and white copy of it.

As an academic, I only do black and white.

It might help if I gave you a colour version.

I have a colour copy here.

My colleagues are more sophisticated than I am.

The solid red line is the tax take by month throughout the year, and when you compare it to the blue line—which is the revenue that we expected by month throughout the year—you see that there is a significant amount of forestalling in the first part of the year. However, the revenues then recover quite sharply and the tax take is above the prediction for that tax band. There is then a big jump in March, potentially because of the forestalling effects of ADS.

There appears to be a forestalling effect in the early part of the year, but it is more than compensated for by the performance for that price band throughout the rest of the year.

I am reminded of the snooker incident, when the commentator said, “For those of you watching in black and white, the blue ball is behind the red one.”

Professor Leith, you mentioned earlier that there was a shift towards the lower end in the distribution curve, but the best-performing group is still the highest-value property range. What would explain that? If there is a shift towards the slightly cheaper end of the market, what would cause the best performing range to be the higher value one?

It is more a kind of twist in the distribution. The top end is holding steady, but somewhere within the £325,000 to £750,000 distribution things are tilting. We cannot know for sure exactly why—there could be a variety of reasons. The top end may just not be sensitive to payment of that tax.

Have I got time for a second question, convener?

Yes.

In table 8—there are no page numbers—we see the difference between the forecasts from the Scottish Government and the Office for Budget Responsibility. It is quite difficult to interpret the table but, if you were to plot the figures, you would see that there was a significant reduction in the OBR’s forecast in November 2015 compared to the Scottish Government’s forecast, which remained pretty steady. The outturn was pretty close to the Scottish Government’s forecast. Can you offer any explanation why there was such a downturn and volatility in the OBR’s forecast?

I think that the OBR’s forecast in November 2015 was based on the outturn data for the year to date, and that was being hit by the forestalling effect, so it was down. You would need to ask the OBR, but I think that it was extrapolating the forecast for the rest of the year from that data, using the data to revise its forecast.

If both the Scottish Government and the OBR had the same data, why did they produce such radically different forecasts?

The Scottish Government did not use the in-year data for any forecast revisions.

Yet the more accurate and stable forecast seems to have come from the Scottish Government.

The Scottish Government’s revised forecast is—

It is for 2016-17 budget.

Yes. As a by-product of producing the forecast for the 2016-17 budget, the Scottish Government came up with another forecast for 2015-16. It used the methodology that it used to produce the 2015-16 budget, so its forecast was bound to be pretty close.

I have a different set of questions about your reflections on the experience of devolved taxation so far, particularly how you see the key institutional relationships working and where you think the strengths and weaknesses are.

I am thinking in particular of the relationship between Revenue Scotland and the Scottish ministers. Is that the right model? What about the relationship between Revenue Scotland and Her Majesty’s Revenue and Customs? Do not feel that you have to confine your answers to those relationships if you think that others are even more important.

The committee is looking at LBTT partly because we want to understand how it is working and partly because we have an eye on the future as well as on the past and are thinking about the onward progression of fiscal devolution, particularly through the first half of this session of Parliament. Are there any key lessons that we should learn about institutional design and institutional relationships?

10:00  

I will not be able to respond to your question on key lessons without having to think for a minute or two, but you have raised a really important point. The relationships among all the different agencies matter, and it matters that those relationships are formalised and the interactions done in a disciplined way, typically through memoranda of understanding. It is important that there is good co-operation where it is needed.

It is easier for me to respond a little bit through anecdote. Because our role will be changing, our relationship with the agencies, too, will change quite deeply come 1 April. As a result, we have been very thoughtful about this issue. Up to this point—including this year—the OBR, in doing its forecasts in the areas that involve the devolved taxes here, has invited the commissioners to join its challenge meetings, which are the meetings that it has en route to coming up with its forecasts. A good—if informal—working relationship has developed between the OBR and the SFC over the two and a half years of our existence, and we anticipate that that relationship will continue to be mutually supportive when appropriate. Indeed, the legislation requires us to work together, and we think that that means working together properly.

With regard to income tax, the OBR typically gets data from HMRC. Like Revenue Scotland, the HMRC has, I believe, a legal obligation to protect identity, so with regard to what is delivered to the OBR, the data will be analysed by HMRC staff and an answer to a particular question from outwith the HMRC will be given in a way that does not reveal individual identities. Revenue Scotland has to follow those kinds of disciplines, and it is our understanding that it does. We are not critiquing Revenue Scotland—that is just the way it behaves.

As questions come up and, for us in Scotland, as the devolved taxes, each a little different from the others, increase in number, we have to work out how these relationships will need to work, and we do that through engagement and then the development of a memorandum of understanding for each agency. You ask whether these things work well, but to be very honest I think that it is very hard for us to make that judgment. Nevertheless, I really want to emphasise the importance of that question and of our working to make those relationships work well.

So you have no particular points of concern or anxiety about the institutional framework of devolved taxation that you would like to bring to our attention.

Perhaps my colleagues will speak out, but, to be honest, I find the question so broad that it would trouble me to say that I had a thought-through, evidence-based response to it at this stage. We might well think more about it.

Charles, do you have anything to add?

No. As you said, it is quite a broad question and I would need to think about it before I could say something sensible.

Perhaps I can make one point. At the moment, we are going through the transition from being a body that scrutinises the Scottish Government’s forecasts to being a body that produces its own. The OBR, for example, has relationships with various other bodies; it provides certain key macroeconomic determinants to HMRC, which then delivers a forecast of the tax revenues implied by those determinants. We are developing a kind of operating model to explore whether we will do that the same way or some other way with the various bodies—the Scottish Government, HMRC, the Department for Work and Pensions and Revenue Scotland—and the key principle underpinning that is the need to ensure that the Fiscal Commission has independence, owns the forecasts that it produces and owns the models and the data analysis, to the extent that that is logistically resource-feasible without too much reliance on Scottish Government analysts.

Thank you very much.

As no one else has indicated that they wish to ask a question, I thank the witnesses for their evidence this morning. We are very grateful for your time, and we look forward to speaking to you again over the coming years.

I suspend the meeting for a changeover of witnesses. Please feel free to get a refreshment if you want one.

10:05 Meeting suspended.  

10:11 On resuming—  

We continue our consideration of LBTT’s first year of operation by taking evidence from Revenue Scotland, which is responsible for collecting the tax. We are joined by Elaine Lorimer, who is the chief executive of Revenue Scotland. Elaine is also a member of the budget process review group, which had its first meeting last week. It is an important organisation as far as the committee is concerned. We are also joined by Chris Myerscough, who is the head of tax. I warmly welcome the witnesses to the committee’s proceedings. I believe that Elaine has a short opening statement.

I thank the convener and the committee for inviting Chris and I to join you this morning. I am conscious that the committee will have seen the written evidence that we have submitted, so I will use this opportunity to highlight key points from it.

Most of the issues that I want to talk about are to do with our organisation’s delivery over its first year of operation. In our first year, we have delivered a robust and secure online system for collection of tax, with 98.1 per cent of tax returns having been submitted online. We have established strong collaborative working relationships with the main professional bodies in Scotland, as well as with the Scottish Government, the Scottish Fiscal Commission and Her Majesty’s Revenue and Customs. We have developed legislative guidance, including on use of case studies were possible and appropriate, and we have involved in its production the Law Society of Scotland, the Chartered Institute of Taxation, the Institute of Chartered Accountants of Scotland and others. That remains an on-going dynamic process as we delve deeper into application of our legislation to technically complex transactions.

We have collected £425 million in land and buildings transaction tax, which has gone into the Scottish consolidated fund to contribute to funding public services in Scotland. We have introduced the additional dwelling supplement against a challenging timeframe and we have collected about £35 million of that supplement since April. We have commenced our approach to compliance, using the powers that have been granted by Parliament, to protect the revenue. We delivered all that with a budget of less than 1 per cent of the tax that we collected.

It is a privilege to be the chief executive of Scotland’s first tax authority since the Acts of Union in 1707. We at Revenue Scotland are very aware of the responsibility that Parliament has given us to deliver our statutory remit, and we want to deliver it in a modern, efficient and robust way, but with a strong sense of public service at its heart.

I hope that the evidence that we give this morning will be helpful to the committee. We are ready to provide you with further information, should you require it of us following our discussion today.

Thank you very much, Elaine. I am very grateful for that opening statement. Ash Denham would like to ask a question.

You might be aware that the committee has held a number of meetings on the operation of LBTT, at which Revenue Scotland has come up a number of times. Many of our witnesses have said that they are happy with the operation of Revenue Scotland, and would probably characterise it as modern and efficient, as you said. However—this is probably a small area, but I am interested in your comments on it, nonetheless—in its submission, the Law Society of Scotland said that guidance on the more technical areas might be lacking. It said that in some areas there was guidance from HMRC but

“no equivalent guidance from Revenue Scotland”,

and that there are areas in which Revenue Scotland had indicated that it did not agree with HMRC but had not issued its own guidance. I am interested in your comments on that.

10:15  

I am aware of what the Law Society said on that. Over our first year of operation, we have been working hard to produce guidance. It is fair to say that it would be impossible for us to provide definitive guidance on every aspect of our tax. We have to work with the legislation that Parliament sets us, and we have to apply it to the facts as they present themselves to us. As you can imagine, it would be very difficult for us to produce definitive guidance about some of the more technical transactions, because they are so facts-specific.

I am conscious that the Law Society and, I think, the Institute of Chartered Accountants of Scotland would like us to be more definitive in relation to where, on the face of it, our legislation is similar to the legislation in England and Wales, and where HMRC has taken a view on the legislation there. That is quite difficult for us to do so early in our operation because although, on the face of it, our legislation looks similar, it operates in an entirely different context. Our view is that we need the opportunity to consider the transactions—we are talking here about very technical and facts-specific transactions. It is only once we have had the experience of working through a number of them that we might feel that we are able to update our guidance.

We work so closely and collaboratively with the professional bodies, so we are aware of a number of technical areas on which they would like us to produce not necessarily guidance, but maybe a more technical view. We have been working over the summer to produce something like that, and it is our intention to publish soon—I hope within the next month or so—what we would call a technical update. We will not go so far as to put it into our guidance, but it will provide a bit more guidance on the technical issues that the Law Society has raised.

As Ash Denham said, the majority of submissions that we have received suggest that Revenue Scotland is performing well in administering and collecting the tax. However, a couple of concerns have been raised. The Law Society has talked about negative feedback on your opinions service and said that that may be due to “a lack of funding”. The Scottish Property Federation said:

“Revenue Scotland needs to invest in awareness of its services and to prepare for the implementation of the rent reassessment process for non-residential leases.”

Are you adequately resourced for your functions and will you need more resources, going forward?

That is a lovely question to ask a chief executive. Most of us would say that of course we would love extra resources.

Seriously, though, our opinions service, which is the root of your first question, has a 25-working-day target—which we set ourselves—to turn opinions around. We have achieved that with the vast majority of opinion requests. I think that we did not meet that deadline for only three, and that was because they were so technical that we needed legal advice and even counsel’s opinion on them.

We have demonstrated, in our achievement of our targets in our first year of operation, that we have been adequately resourced. There would always be more that we could do with extra resources, including more investment in our website or more outreach work—although that is not to say that we have not been able to invest in our website or do outreach work. As far as I am concerned, the resources that we have had in our first year of operation have been demonstrated to be sufficient for us to do what Parliament asked us to do.

However, I am conscious that expectations of our organisation will grow—we will be asked to deal with additional taxes. The air passenger tax, for example, is likely to come our way. Through the discussions that we will have with the Government on how we would wish to implement that tax, I need to ensure that we are given sufficient additional resource to invest in our systems and in a small number of additional staff so that we can take on new taxes.

You talk about needing

“a small number of additional staff”.

How many people are currently employed in Revenue Scotland and roughly how many do you think you will need for you to take on those additional responsibilities?

We have around 50 staff in the organisation at the moment. We also fund additional staff in the Scottish Environment Protection Agency. Our organisation is held up as an example because of our efficient lean approach to design of the organisation. That approach has included considering which other organisations in Scotland we can work with in partnership. In fact, we have used our powers to delegate functions. We have delegated some aspects of our work to SEPA and Registers of Scotland, rather than building up a big central machine. Apart from the 50 staff within our organisation, we fund about a dozen staff in SEPA, who work on landfill tax with us.

I want to pick up on the evidence from the Law Society that Ash Denham has already raised with you. One of the phrases that the Law Society used in oral evidence to us, which I found very striking as an academic lawyer, was the idea that we are “taxed by statute” and “untaxed by” extra-statutory “concession”. I know that that is the model that HMRC has been using for a very long time indeed. Is it the right model for Scotland? Is it desirable? Is it inevitable? Are you content with it?

It is not at all the model that we have in Scotland. Personally, I am not sure that we would want to follow that model. Chris Myerscough knows more about the matter than I do, so I may get her to comment. It is really important that the legislation is clear and that it sets out publicly the parameters of the remit of the tax, and what our role in relation to collection and management of the tax would be. We do not have the powers, as things stand.

Is the legislation on LBTT sufficiently clear to meet that test?

The proof of the pudding will be in the eating. We have had a successful first year working with the legislation. As with any tax legislation, it will be tested in the light of experience and in the light of the more complex technical questions that might come our way. Only when we have those questions to consider will we be able to see whether the legislation is clear enough for us.

That is where the role of the Tax Tribunals for Scotland may come into play; there will be aspects—as there always are with interpretation of legislation—on which judgment calls are being made and the legislation needs to be put to the test. The only place where we can get a definitive answer on interpretation of legislation is the court or—in our case—the Tax Tribunal for Scotland. As things stand, we are able to work with the legislation as we find it.

Thank you. Does Chris Myerscough want to add anything?

We, too, picked up on the comment from the Law Society’s representative. I think what she said was that the Law Society did not like the idea of being charged tax by the legislation but then being, in effect, uncharged by guidance.

We see our guidance as being to explain to people how legislation applies, not to define how legislation applies. If it is good legislation—at the moment, we have no reason to believe that it is not—it should stand on its own in that respect.

I will pan out from that and think about institutional structures a little bit more broadly. We are interested in LBTT partly for its own sake and partly with a view to there being more fiscal devolution coming in the first half of this parliamentary session. Have we got the institutional structures broadly right? I am thinking in particular of the relationship between Revenue Scotland and the Scottish ministers, and the relationship between Revenue Scotland and HMRC. Are there any aspects of the institutional landscape that you would like to draw to our attention with a view to our doing better than we currently do, or are you quite content?

In the main, the relationships work well. I had the benefit of listening to Lady Rice’s earlier evidence; she talked about the need for discipline in the conduct of those relationships because the bodies all have their own clearly defined remits. It is fair to say that we have those structures in place. For our relationship with the Scottish Government, we have a clearly defined framework document that sets out in detail—while recognising our independence—how the relationship between us and the Scottish Government should work.

Clearly, HMRC and Revenue Scotland have mutual interests because taxpayers do not stop at the border: we might want to talk to HMRC about particular taxpayers. We have information-sharing agreements that are properly regulated and we have a memorandum of understanding that sets out clearly what the relationship should be and—which is important from the taxpayers’ perspective—how taxpayer information is protected within those boundaries.

We have delegated some of our functions to SEPA and Registers of Scotland, so we also have to define properly our relationships with those bodies because Revenue Scotland is asking them to undertake some of our functions for us. We also have a relationship with the Scottish Fiscal Commission.

There is a complicated landscape, but we have in place with all the players in that landscape clearly defined relationships that are underpinned by some form of agreement or memorandum of understanding. We have structures in place, but what really matters is the relationships—we can have the underpinning framework in place, but we need the relationships to work. I have certainly not seen anything since I have arrived that would cause concern, because everybody recognises where they are coming from and wants to make the system work.

I am interested in the compliance activities that you noted in your submission. I welcome the information that you have collected £450,000 from those activities. Just to test how that works, if I purchased a property for, say, £450,000 but just completely ignored making an online return and decided that I was not going to pay any tax on it, how would you catch me? [Laughter.]

You have just admitted it on the record.

They will be checking your record today.

I am not sure how to answer that question. [Laughter.]

As you will have gathered, we formally find out about a transaction in the first instance when a tax return is submitted, so I guess your question is, “What happens if I don’t make a tax return?” Chris Myerscough will keep me right on this, but on the basis that you would wish to register your ownership of the property with Registers of Scotland, we would find out that you had done that and therefore we would be looking to undertake compliance procedures.

You would have difficulty in registering your property because the keeper of the registers of Scotland requires confirmation from us that the tax return has been received and that tax has been paid before she is able to accept an application for registration. If you were happy to take the risk of not having your title registered, so be it. However, the chances are that if you did something like that regularly, sooner or later Revenue Scotland or HMRC would catch up with you because we and HMRC obviously have a considerable amount of other data and intelligence at our disposal. Buyer, beware, as they say. [Laughter.]

That national insurance number will get you.

From our previous evidence-taking sessions, we are aware that Revenue Scotland publishes on its website outturn data for LBTT, but we have heard evidence that it would be helpful to have that information not just in terms of the tax bands but also in terms of what happens within the tax bands and regionally. Could Revenue Scotland produce that?

10:30  

Everything is possible. However, I remind the committee that production of some management information does not involve simply pressing some buttons. Over our first year of operation, we have learned that the level of data and the way in which they can be manipulated are of increasing importance to organisations including the Scottish Fiscal Commission and the Scottish Government. The data that we publish are the basic, factually correct and accurate data that we need for our purposes as a tax authority. We are being asked to produce other kinds of data that, on the face of it, would not necessarily be critically important for us as a tax authority.

We have been working with the Scottish Government, the Scottish Fiscal Commission and the Office for Budget Responsibility, which also asks us for data, to come to a consistent view about what the additional calls on our data would be, so that we can look into how achievable it is to provide the data. We already have some of the information that is required, although it would require some additional time on the part of our statistician to pull it together in a way that he would feel was robust enough for him to feel comfortable about releasing it. A system change might be required for us to provide other information that we have been asked for that is not quite so straightforward.

We are able to provide some extra levels of detail on the information that we provide at the moment. We are working with the other bodies to get a feel for whether there is anything else that they want us to do, so that we can plan into our workload the release of that further information.

Transparency is really important to us because of our independence. We would want to ensure that, where we are being asked regularly to produce data by band or by local authority, for instance, we publish that alongside the data that we currently publish, so that there is absolute visibility.

Are you considering that issue?

Yes.

For us, it is essential that the data that we publish are accurate and can be relied on. That means that, as well as the work that is required to cut the data separately, additional validation work would be required. For example, the Scottish Fiscal Commission said that it would like information broken down by area. Certainly, we collect local authorities’ codes, but we would need to ensure that those figures matched up with the postcode of the property, and that the right area code had been selected. We want to ensure that the information that we give out is accurate so that people can rely on it.

You said that the additional dwelling supplement was introduced to a challenging timeframe. Previous witnesses have told us that we might not have quite got the exemptions correct, and there have been suggestions that there should be different exemptions or different timeframes within which to claim repayments. Has that caused difficulties in collection and management of that tax? Do we need to consider making refinements?

It is difficult for me to get into questions of policy—clearly, that is not my role. I would say, however, that in terms of lessons learned from introducing that tax, we have been able to demonstrate that we could introduce something like it without a whimper. It just happened. It involved system change and a huge amount of work on our part, in collaboration with the Law Society of Scotland and others, to produce guidance.

It was not a timeframe within which I would like to have to operate too often. That is not so much to do with our organisation being ready; it is more to do with advisers externally—Scotland, essentially—being ready for a new tax. Lessons learned for us in relation to ADS were more to do with the amount of time that we need to introduce properly a new tax into the system. That is the sort of information that we are feeding in to our colleagues in the Scottish Government, for example in relation to planning for the air passenger tax.

I am mindful of some of the issues that came up during the debate on the legislation for ADS. I am conscious that commitments were made—for example to reconsider the grace period. We stand ready with our data to assist Scottish ministers in their assessment of whether the policy is working, because we are aware that they have a commitment to come back to Parliament about that.

I guess that that comes back to Adam Tomkins’s question about institutional issues. If a tax change is being introduced in Scotland and legislation is coming through the Scottish Parliament, you will be visible. However, if Westminster decides to change a tax at the drop of a hat—which seems to have happened in this circumstance—that gives organisations in Scotland less time to react and, although there might have been preparation time in the UK Government, the same might not apply here.

That is right. Such things place huge stress on an organisation such as Revenue Scotland. However, we were able to deliver—at the cost, I suppose, of spending more time developing our guidance. We had to take a view on what our organisational priorities were for that last period.

In terms of always being ready, part of the challenge of being a tax authority is that change will always happen; taxation is a very dynamic environment. Thinking about the resources that I need in my organisation, we have a small, permanent resource that is skilled in quickly putting together a programme of change—which is essentially what the new taxes are—to redesign our systems, produce our guidance and so on. It is acknowledged that because tax is so dynamic, we need, if we are going to have to do things quickly, to have the capability as an organisation to react and respond to change so that we can deliver.

Thank you very much for coming along and giving us very clear evidence this morning; we are very grateful to you.

The next meeting of the committee will take place on Wednesday 26 October, when we will have a final evidence session on LBTT with the Cabinet Secretary for Finance and the Constitution. We agreed earlier to take item 3 in private.

10:38 Meeting continued in private until 10:57.