Public Accounts Committee

Oral evidence: The effective management of tax reliefs, HC 892

Wednesday 14 January 2015

Ordered by the House of Commons to be published on 14 January 2015

Watch the meeting: http://www.parliamentlive.tv/Main/Player.aspx?meetingId=16927

Members present: Margaret Hodge (Chair); Mr Richard Bacon, Guto Bebb, Stephen Hammond, Chris Heaton-Harris, Meg Hillier, Mr Stewart Jackson, Dame Anne McGuire, Stephen Phillips, John Pugh, Nick Smith

Sir Amyas Morse, Comptroller and Auditor General, Rob Prideaux, Director, Gabrielle Cohen, Executive Leader, National Audit Office, and Richard Brown, Treasury Officer of Accounts, were in attendance.

Witnesses: Jim Harra, Director General, Business Tax, Lin Homer, Chief Executive and Permanent Secretary, HMRC, and Indra Morris, Director General, Tax and Welfare, HM Treasury, gave evidence.

              Q1 Chair: Welcome and happy new year. I think this is a really interesting and very good Report. I hope we can have a constructive dialogue on the issues that have been raised in it and, hopefully, make some progress at the end.

              To start with, this is the second time we have looked at tax reliefs. In all the work on tax, it has emerged as a key issue around many of the things that have concerned the Committee. Last time you appeared before the Committee on the issue, we asked you who was accountable for value for money in the cost of expenditure tax reliefs. Can you say, Lin, are you accountable or is Treasury accountable? Who is the accounting officer on this?

              Lin Homer: Thank you for inviting us and happy new year to all the Committee.

              I’ll have a go at answering; Indra may want to add something. I think this is an important area and there is much in the Report, as with many of the Reports we get from the NAO, that we can work on and where we can benefit from the recommendations. The Report asks how well we collect and report data, how well we watch whether things are working out as they are intended and how we identify early abuse. On all of those things, it is entirely right to hold me and HMRC accountable for taking what comes out of the policy work that the Government and HMT do and hand over to us.

              I am sure we will spend some time today talking about how we do that. I think there are some differences of opinion between us and Amyas on the best way to do it, but we would accept that responsibility for implementing what we would say are tax reliefs in the round. Some of them look and act a bit more like expenditure than others, but all of them need to be assessed, evaluated and monitored by us, and we should be held to account for that. But the question about the narrower issue of the cost of a relief—

              Chair: The big issue, actually.

              Lin Homer:—sometimes goes to policy, rather than to implementation. When it is in that space, our answer would be that it is the Government of the day, accountable to Parliament.

              Q2 Chair: Let me challenge you on that.

              Lin Homer: I thought you might.

              Chair: I want to challenge that, because this goes to the heart of our concerns. Government sets a policy. It then goes to Parliament and, in getting the agreement of Parliament, it sets down an estimate of how much that policy will cost. We are focusing this afternoon on what we call the expenditure tax reliefs—the ones where Government introduces something expecting a change of behaviour. In some cases, as in the Report, the cost goes down, but we are focusing on the ones where it goes massively above the estimate. That is a cost to the taxpayer. It is not a question of whether the policy is all right—somebody has to be accountable to Parliament for its value for money. That is not challenging the policy, but making it accountable to Parliament for value for money and for the estimate’s being so different from the final outcome in costs.

              I completely understand that you are accountable for the administration, and we will come to that, but we are talking about really big sums here and somebody has to be accountable to Parliament for them. I do not mind whether the Treasury say that they are or you say that you are, but we need somebody to be accountable because that will enable Parliament, through our Select Committee mechanisms and other ways, to hold you to account for them.

              Indra Morris: Might I come in on this? Accountability is different between tax and spending. You touched on this in the last discussion with Sir Nicholas: spending and tax are not the same. We operate within long-standing arrangements on taxation—the double mechanism, if you will—to the Floor of the House via the Finance Bill and then on policy to the Treasury Committee. That accountability is different from the spending arrangements and accountability on spending. That is because tax and spending are different.

              Q3 Chair: I have to challenge you on that. I am talking about expenditure tax reliefs. This goes to the heart of the issue. We are talking about in excess of £100 billion—that is common ground between us—just on expenditure tax reliefs. We agreed that last time.

              Let me take one example of where you are trying to say that something is a tax relief, but it is an expenditure tax relief—so it is there to change behaviour and to change something in the economy. Let us take DCMS, whose budget I know a little about. At the moment, there is a political debate about the cuts in the Arts Council budget—£83 million, and everybody is being ping-ponged across the politics. On the other hand, if you look at this Report, there are a whole set of reliefs—the film tax relief, the high-end TV tax relief, the orchestra tax relief, the video games tax relief, the children’s television tax relief—and they all have a cost. They are all expenditures that are supposed to impact on what happens in the economy.

              In a sense, what Parliament wants is an open debate about expenditure and choices. You are saying to us that you can have a debate about the £83 million on the Arts Council, because that is in DEL, but you cannot have an open debate about these other expenditures—choices made by Government that we cannot challenge on value for money. That is not transparent government, and it is not holding you to account for the expenditure that you incur.

              All I am asking is that there should be transparency. It is not like personal tax reliefs, say, which are a totally political choice. This is an expenditure to impact on the economy in the same way as the Arts Council budget is an expenditure to impact on the economy. The two have to be considered side by side; they cannot be considered separately.

              Indra Morris: I am puzzled by this view that there is no accountability because the creative tax reliefs, for want of a shorthand—

              Q4Chair: How much did you spend on those? How much did you spend on film tax relief last year?

              Indra Morris: I don’t know off the top of my head.

              Q5 Chair: How much have you spent?

              Lin Homer: It is in our December publication. We can give you a note.

              Q6 Chair: Let us take any of them. How much have you spent on the video games tax relief?

              Lin Homer: I was going to try to take this discussion forward with the use of an example. It is not a creative industry one, which I know are dear to your heart, but if you take agricultural property relief, whose costs the Report highlights have gone up significantly, what is happening—

              Q7 Chair: I know you are trying to take me off this. The reason why I have chosen my example is that it is a really clear example of where the Treasury chooses. I know this because you negotiate it with the Treasury and it is a real expenditure. We are talking about over £100 billion. Yet one is accountable; the other is hidden; and you cannot even tell us this afternoon.

              Lin Homer: Film tax relief is £225 million in 2013-14, and expected to be £230 million this year.

              Q8 Chair: So £225 million on the film tax relief, and we are argy-bargying about—

              Indra Morris: Animation tax relief is £12.3 million, and high-end TV £14.8 million (in 2013-14).

              Lin Homer: These are clear on the face of propositions when they go into Parliament.

              Q9 Chair: No, they are clear as estimates.

              Lin Homer: The point I wanted to make is that there are various influences on costs that mean that simply thinking that an increase in cost is a either a sign of abuse of a sign of lack of accountability is problematic. I think what we should be held accountable for is making sure that we are watching those things and giving advice, but I do not think you can see tax reliefs as exactly the same as spending, even in the tax expenditure—

              Chair: I am not saying it is a sign of abuse. We will come to abuse. I am not arguing that; my argument is entirely about expenditure.

              Q10 Mr Bacon: I would like to check something Indra Morris said earlier. You drew a distinction between tax and spending and said they were not the same thing. Do you think tax and tax reliefs are the same thing?

              Indra Morris: Tax reliefs are part of the tax system—

              Q11 Mr Bacon: No, I am asking a very specific question. You said tax and spending are not the same thing. My question is: do you think that tax and tax reliefs are the same thing, or do you think they are different from one another?

              Indra Morris: I think tax reliefs are part of the tax system.

              Q12 Mr Bacon: Are tax reliefs and tax the same, or are they different from one another?

              Indra Morris: They are part of the same thing.

              Q13 Mr Bacon: They are part of the same thing. It is a very simple question, and I would like a simple answer. You said tax and spending are not the same thing—that is how we got into this. I want to know if you think tax and tax reliefs are the same, or if you think tax and tax reliefs are different. Which is it? I know what I think. I think they are different. You think they are the same—yes or no?

              Indra Morris: I think they are one way in which Government make decisions—

              Q14 Mr Bacon: They are not one thing; they are two things. One is tax; the other is tax reliefs. I am asking you if you think tax and tax reliefs are the same thing—yes or no?

              Indra Morris: Tax reliefs are a key mechanism for designing the tax system—

              Q15 Mr Bacon: I am asking you to contrast tax reliefs with tax. Tax is where you get money in, by taxing things. Tax relief is where you make a conscious decision to forgo that income. It is different, isn’t it?

              Indra Morris: No, I don’t agree.

              Q16 Mr Bacon: It’s not different? I know Treasury people are very bright and can talk about angels on a head of a pin until the cows come home, to mix my metaphors, but are you seriously telling me now, here, that getting the money in and not getting the money in are the same thing?

              Indra Morris: I think they are part of—

              Q17 Mr Bacon: Are you seriously telling me that they are the same thing? They are different. At the end of one, I have some money; at the end of the other, I don’t have some money. They are plainly different.

              Lin Homer: I would welcome being able to ask Jim to use air passenger duty as an example, to show Mr Bacon how they can be the same thing. We are in danger of making something that is necessarily complex more simple than it is.

              Q18 Mr Bacon: You already have 17,000 pages of tax code, so you have done a very good job of making it complicated. Sometimes, people need to come along and try to make things a bit simpler.

              Lin Homer: I think it would be worth thinking about it—

              Chair: Sorry to interrupt, but I have two Members who want to declare an interest.

              Chris Heaton-Harris: Sorry, I didn’t realise I had to. I am unpaid chairman of a regional theatre that produces and therefore possibly—hopefully—will be

in receipt of a very generous tax relief at some point in the near future.

              Lin Homer: I thought you were all going to declare an interest and say you had to pay tax.

              Stephen Hammond: I also should declare, as per my entry in the Register of Members’ Financial Interests, I have a holding in Harwood Film Partnership.

              Jim Harra: I think tax reliefs can perform a variety of different functions within the tax system—

              Q19 Chair: We understand that. It is why we are talking about expenditure reliefs. Don’t talk about the others. That is why I want to focus on the expenditure reliefs. We understand they have different functions.

              Jim Harra: I don’t think it is possible to definitively categorise tax reliefs in that kind of way.

              Q20 Mr Bacon: In what kind of way?

              Jim Harra: There is this international concept of tax expenditure as a form of tax relief, which is not terribly well defined but has been around for well over 20 years, and on that basis we do, in our own publications, try to identify those reliefs that we think have primarily characteristics of a tax expenditure, those that have mixed characteristics and those that do not have characteristics of a tax expenditure.

              Q21 Chair: Jim, I am sorry to interrupt you, but some tax reliefs such as the theatre ones are brought in for a specific purpose—to have an impact and encourage more original productions in regional theatres. That is why that one was brought in. Film tax relief was brought in to encourage more British production of films. They are brought in for a specific purpose.

              In Germany, which I have looked at a little, they have reliefs of that nature—what I call expenditure tax relief. Under the German system, they are then assessed alongside other forms of expenditure in the equivalent of the DCMS budget. The principal underlying thing—this is why I do not want to get on to a new tax relief; the DCMS examples bring us to the core of the argument—is that what we are doing here as a country is giving money to DCMS’s area of concern, the arts sector, in part through the Arts Council grant and in part through tax relief.

              The two should be considered together and there should be total transparency around that; we will come later to whether it is properly administered and whether there is abuse. As a principle, the two should be taken together, seen together and accountable to Parliament and the public together. You need to kill that argument in my brain to convince me that I am wrong—and you haven’t.

              Indra Morris: These reliefs are brought to Parliament through the Finance Bill to be debated and voted on.

              Chair: Nobody is arguing with that, but the reliefs then disappear into the system. If they had a theatre tax relief in Germany, they would consider it alongside their budget. To be blunt, German public expenditure is in a slightly better condition than ours, so it might not be a bad thing to do that.

              Q22 Austin Mitchell: Surely it is reasonable that both should be accountable to Parliament. If you cut a grant to the Arts Council, as we are doing, you can assess the effects of that, restore it next year if it is disastrous or cut it further if it leads to a flourishing of art production. That is the accountability that Parliament has. But we cannot be accountable on a tax relief because it is just there—it goes on year after year, rising or falling or whatever, and nobody tells us whether it is effective or doing what it says it is doing. Nobody tells us how we can account for it or how far we should count that as part of the budget for creative media or whatever. They need to be accountable and known in the same way.

              Indra Morris: What I am saying is that the accountability is there. It is different from spending; as I said, there is the double mechanism both to the Floor of the House through the Finance Bill and to the Treasury Committee on policy. I think that in part what you are saying is that the missing bit is looking at them alongside each other. I am not convinced by that, but—

              Q23 Chair: Hang on a minute. The accountability is there when you consider the policy. What we then do as a Committee is look back at the expenditure and see whether it has given value for money. You look at this Report being produced today by the NAO and you find—I don’t think it is that difficult to define, but even accepting some difficulties at the margin—that for half the tax expenditure the information is not there and that the information in many of the areas is inaccurate. All you do is provide the information when the policy is introduced: you find estimates. Draw my attention, Rob, to the half on which information is not provided and the ones where the information is inaccurate. We are not getting the full picture here.

              Rob Prideaux: In terms of the costs that are published, figure 2 on page 18 shows that of the 200 or so that we think have characteristics meaning that they are trying to achieve a particular objective, there are published data from tax returns on 60 on a cost basis; there are 55 with the numbers estimated from other sources; and there are 53 for which data are not collected. There are also 28 where data come in through tax returns, but they are not actually used or published in any way.

              Q24 Chair: That covers both points. You have very kindly provided the ones that I asked for, but it is not true that you are giving Parliament a proper position on all of them, and that is why I want to know who is accountable for that.

              Lin Homer: As I have said before, we would accept the accountability for publishing information. I think we want to look very carefully at some of the recommendations in the Report about whether we can improve the quality of what we publish, and we think we could do better, particularly in the annual publication we make. However, we do think that that forms a basis, picking up Mr Mitchell’s point, which does allow for accountability. An element of that accountability is to Parliament itself, to the Treasury Committee as well as to the other Select Committees, but my personal view is that the basis of information and data that we provide is as rich as, if not richer than, any tax administration you are comparing us to. We publish over 600 pages of national statistics as the tax administration, and the quality of that is recognised by NAO as very good in lots of places. What I would challenge is that there is not a simple dynamic that says that an increase in cost means a value-for-money issue, and I think many of those go to a policy perspective—

              Q25 Mr Bacon: Hang on. You are missing a step here. I think most of us would probably agree with what you have just said, namely that an increase in cost does not necessarily point to a value-for-money issue. It may merely point to the fact that the policy is working. Let us say—to take a simple example; I know that there are more complex ones, but let us stick with a simple one—that there is a policy, which this Committee would not challenge, to have a tax relief to encourage more regional theatre, as a result of which the costs incurred on tax relief for regional theatre are higher. That is not a surprise, and it would have been expected. It does not necessarily, in and of itself, create a value-for-money issue. You immediately went on to say that it was about the policy objective, but there is an intermediate stage, where you ask the legitimate question: “How much extra regional theatre has it caused there to be, and is it value for money?” Is the policy, which this Committee would not question, of having a tax relief to encourage regional theatre, effective, efficient and economic? That is a legitimate question, is it not?

              Lin Homer: It depends on the policy objective.

              Mr Bacon: No—

              Lin Homer: I think it does, Mr Bacon.

              Q26 Mr Bacon: Hang on. Let me just rephrase my question. It should not depend on the policy objective. The question of whether the pursuit of a policy is effective, efficient and economic should not depend on what the policy is. We do not question the policy. We are questioning whether the policy, the “it”—whatever the “it” is—is implemented effectively, efficiently and economically. I am merely asking: is it not, surely, a legitimate question to ask whether this particular policy, pursued through the implementation of this tax relief, was done economically, efficiently and effectively? You replied “It depends on the policy objective.”

              Lin Homer: No, I didn’t.

              Mr Bacon: Yes, you did.

              Lin Homer: If you want me to try to answer the question—

              Chair: Let her say what she wants to say.

              Dame Anne McGuire: Could we allow the witness just a bit of space to answer the question? You might not like the answer, but I think she deserves a wee bit of space to answer it in her way.

              Q27 Mr Bacon: I would love to hear the answer, but you definitely said, “It depends on the policy objective.” My whole point is that no, it doesn’t.

              Lin Homer: You changed your language between the start of what you said and the end, and it was a very important change. I am sure it was not intentional effectively to ask me a different question, but that is what you did. It is entirely acceptable for you to ask me whether we are implementing a policy in a way that is efficient, effective and value for money. It is me who is accountable for that, as the accounting officer of HMRC, and it is predominantly HMRC that carries that weight. HMT is useful to us because it is part of our internal challenge. The first question you asked was whether the costs had increased—not the implementation of them, but the costs—and that is something different. That may be a policy question.

              Chair: May be.

              Q28 Mr Bacon: That speaks to the first part, or the third part. It speaks to whether it is economic or not. What I was trying to say—

              Lin Homer: I am very sorry if this sounds ruder than I mean it to be, but that in my view is not for you as a Committee. That is for Parliament.

              Q29 Mr Bacon: Let us pursue this for a second. The point that I was trying to make was, of course if you have a tax relief, as a result, money is forgone by the Treasury in income, because the tax relief is working. There is a spike, which of course could simply mean that the policy is working, that people are seeing the tax relief and the behaviour is changing accordingly—

              Lin Homer: All the underlying values have changed, as would happen in agricultural property relief.

              Mr Bacon: That behaviour change in itself is absolutely fine. I was not disputing that for one moment. I was merely adding that if there is a cost and then an increased cost, just like there would be for the expenditure of a certain amount of money on a policy and then the increased expenditure of a certain amount of money on a policy because the policy had been expanded, there is also around that a legitimate question of whether it is effective, efficient and economic. In other words, how well is the money being spent to achieve the policy? That is all.

              Lin Homer: Not value for money—

              Mr Bacon: Sorry, can you say that again?

              Lin Homer: I really think examples are useful, because in your implementing space I do not think that we think that there is any example in this Report where our approach to implementing a relief has given a significant rise to a problem of value for money—

              Chair: Let me give you an example in the Report.

              Stephen Phillips: May we listen to the answer first and then go on to the next question?

              Lin Homer: There are several examples where a decision by Parliament to expand the relief has increased the cost. First, an entrepreneurs’ relief would be an extremely good example of that, where the lifetime allowance has been increased three times by Parliament in a way that will inevitably increase the cost. Secondly, inheritance tax has been held down to the cap—again, a decision of Parliament which inevitably increases the cost. That is included in the Report as one of the prime examples of a shifting cost that I think is entirely related to policy.

              Chair: The problem then is that you do not know.

              Lin Homer: We do know actually.

              Q30 Chair: May I go to Amyas and then we will move on, because I think we are making a bit of progress?

              Sir Amyas Morse: I hope you don’t think that we imagined it, but we do not think that simply an increase in the cost of relief is indicative of something having gone wrong. We do not think that at all and we are not trying to suggest that in any way in this Report. But unless you are 100% comfortable that you already know why that increase has happened, we do think you should inquire—

              Lin Homer: We would agree with that.

              Sir Amyas Morse: I am trying to establish grounds of agreement, because I think there are quite a lot of them.

              Lin Homer: So do I.

              Sir Amyas Morse: We simply think that it is a pretty good indicator that you need to understand slightly better what is happening if you thought costs were going to be one level and it turns out that they are at another, or if costs have gone up and you have a strong suspicion on the ground that there might be a significant element of tax avoidance. All we are saying is, we think that in these circumstances you need to have reasonably prompt means of finding out what is going on. That is really what we are trying to say in the Report, to a large extent.

              Lin Homer: We absolutely accept that, if you like, having an inquiring mind is probably one of the most significant asks of my department. We would say that our approach to tax reliefs requires us to think about the customer, their behaviour and the product in question. Within that range of things, on occasion what is happening to the overall cost will be an element. We do not think that it is either the leading element or even a necessary element in all reliefs. I would like to illustrate that with some examples. If I may, I will give you two facts about agricultural property relief, which kind of show that there is a lot of complexity in this. This is not me doing this; this is some of my cleverer tax people, and I think that they are really good at making these inquiries.

              In agricultural property relief, if you just look at the cost—it is on your graph, figure 5, which shows the 11 reliefs that NAO has looked at most closely. It shows the 11 that have increased in cost by 25% over six years. Agricultural property relief has increased in cost from £256 million in 2007-08 to £334 million in 2011-12, which is quite a significant increase in cost. It is about 30%. If you look at the number of claims, it almost hasn’t moved at all. It has moved from just over 2,000 to probably about 2,200. That doesn’t suggest that anything significant has changed in the objective of the relief, which is to keep farms at the point of death—

              Q31 Stephen Phillips: Is that right? Those 200 people might not—

              Lin Homer: It goes up and down year on year. There are some mergers and some increases, but it is a 10% shift in claims at most. It is probably even less than that, whereas it is a 30%—

              Q32 Stephen Phillips: But might not those 200 be very significant claims?

              Lin Homer: But we know they aren’t because we look at every claim. We know what has happened in that time, and I think at one point it was suggested that we sometimes go for the obvious answer, but that is because sometimes the obvious answer is right. In that case, two things have happened: agricultural values have gone up significantly; and inheritance tax levels have remained the same. As a consequence of that, more estates are subject to relief. What you are seeing is that the normal number of agricultural estates is broadly the same. The relief is working as intended. I could spend quite a lot of money doing a lot more detailed work on costs, but looking at customer behaviour, and the costs within that, tells me everything that we need to know.

              Share loss relief is much bigger, and it is one that you might want to talk about. The Report wrongly criticises us for not noticing something until 2013. Actually, almost immediately after the relief went in, my tax experts spotted that some people, between 60% and 80% of people, were using it abusively. Just like with some of the charitable relief, those claims were refused. The claims have not been paid, and the tax at risk has not been lost. Yes, tribunal cases take an enormously long time to work through the tribunal, but that is one of the reasons why we have now brought in accelerated payments. We think that our risk approach is ensuring that we inquire and that we inquire successfully. Should we do a bit more, and a bit more systematically, on costs? Probably we should. Should we turn our whole system into being obsessed with costs? We strongly don’t think we should. That is the difference between them.

              Q33 Chair: We are moving forward. Cost is used as the basis for understanding what is happening with a particular tax relief.

              Lin Homer: No, it is one of a—

              Q34 Chair: I accept that it is one of the key bits of information that you need. I hope that what we have established is that you don’t always have it and, where you do have it, it is not always accurate. If what you are saying to us is that you accept that you have to have consistent data on costs and that those data should be accurate, I think that is one essential.

              Lin Homer: We accept that we could improve. I want to be clear that we do not always think it would be proportionate to collect detailed costs on all taxes.

              Q35 Chair: We will come back to that, because the Report says on page 19, paragraph 1.8 that you asked for it inconsistently, which is one of the things that we would say.

              Lin Homer: No.

              Q36 Chair: The Report says that you asked for information on seven tax reliefs where the cost is less than £2.5 million.

              Lin Homer: We will sometimes ask for costs to help us understand behaviour, and we will sometimes have costs because it is easy to collect.

              Q37 Chair: If you sometimes do it, why not—The purpose of this whole argument is that these are tax reliefs that are supposed to influence behaviour, and you have to understand whether they are effective. You are saying that you sometimes do it on a whim.

              Q38 Chris Heaton-Harris: Are you talking about the same costs? I think you are talking about costs as in the total amount of money spent on the relief, and I think Lin is talking about administrative costs.

              Lin Homer: No, I am talking about the cost of the relief. We do not think it is always necessary or proportionate to know it in the level of detail that I think you would like. Jim could probably help.

              Jim Harra: The guiding principle that we follow when we ask taxpayers for data about reliefs is that we need it in order to administer the relief. We would not normally ask people to return data to us that are only required for statistical reporting. We would do that only in exceptional circumstances because of the burden that it places on people. You will find instances where we ask people for data on their tax returns in order to claim a relatively small relief because we need that data to give them their relief. You will see other instances where we do not need the data to give people the relief, so we do not ask for it. In those cases, we try to get at the cost of that relief from other data sources. An example of that, which is mentioned in the Report, would be principal residence relief in CGT, where we do not ask people to return nil returns when they sell their home, but we do get data about the sale of homes that enable us to estimate what the cost of that relief is.

              Q39 Stephen Phillips: If one looks at figure 2, there are a number of reliefs where data are not collected at all. Is that right, Mr Harra?

              Jim Harra: Yes. An example of that would be principal residence relief. If I sell my home tomorrow, it will qualify for that relief, and I will be exempt from capital gains tax.

              Q40 Stephen Phillips: So, in relation to those, the cost of the tax forgone is simply not known. Is that right, Ms Homer?

              Lin Homer: It is not known accurately. The judgment that we are making there, to go back to Mr Bacon’s point, is that when Parliament decided to make the principal residence exempt from capital gains tax, it was thinking philosophically, if you like, rather than thinking, “We’ll invest X, but not Y.” If any Parliament asked us, we could go back and gather more, and we would tend to do that, because this is when we do detailed costing, at a point when a Government—

              Q41 Meg Hillier: It seems to me, with things like real-time data for employers and the digitisation of the process, that the data will surely be much easier in time. Can you outline what the time scale is for that? Once you have that, you can run a report very easily, so politicians can make a policy decision based on actual numbers, see how the trends work and make adjustments to policy if there is a problem with it. At the moment, it seems like you are flying blind, frankly.

              Lin Homer: We are not flying blind, but you are absolutely right that we are making judgments all the time about the admin burden on individuals on businesses on one side—

              Q42 Meg Hillier: The admin burden will be reduced by digital, won’t it?

              Lin Homer: It could very well be. You know that we are testing digital accounts at the moment; we have pilots going with both individuals and businesses. Our initial experience is that small businesses are very happy to give us information if it helps us serve them well. So we are seeing, for instance, that in the case of  VAT, where we would traditionally have kept to a minimum the data collection that we might need because there is lots of form-filling, in an environment where we are interacting more regularly, getting a customer to agree to more richly populate what we know about them so we can serve them is a kind of contract people are prepared to enter into.

              I think that over the next two to five years, as we expand our opportunity to have more personalised accounts, we are going to be able to play with this, but I still want to make the point of principle that sometimes Parliament is doing things without a strong reference to cost. It is a principled matter, and it is not always tax forgone. Sometimes the state is saying, “We don’t tax in that space.” They are not saying—

              Q43 Mr Bacon: You are selling the principal private residence as a really good example of that.

              Lin Homer: It is.

              Q44 Mr Bacon: But what we are really interested in is those tax reliefs that are specifically designed to change behaviour and produce new kinds of behaviour. Let me give you a different example. Years ago, this Committee looked at the Government scheme to reduce the student drop-out rate. Over a period of four years, the Government spent £800 million trying to have higher retention of students in universities, so that more people completed their degrees. The drop-out rate beforehand was 22%, and after the four years and the £800 million expenditure, the drop-out rate was still 22%. So whatever else it was, it was not a very effective, efficient or economic use of the £800 million.

              That was just money doled out in one way and another, but one can imagine that a policy might have been designed to achieve the same result—namely, to reduce the drop-out rate—or a different result in a different policy space by the use of a tax relief, rather than by doling out £800 million in expenditure. Then the question becomes whether in this case, just as with the £800 million spent on trying to reduce the drop-out rate, the expenditure—in this case, the money forgone—was forgone efficiently, effectively and economically. Did it produce the desired result? That is a perfectly sensible question, is it not?

              Lin Homer: I think the business premises renovation allowance is quite similar. It was brought in with the specific objective of encouraging the renovation of unused commercial buildings in disadvantaged areas. Very soon after it was brought in, our people saw abusive features being used.

They interacted quickly through our policy partnership with the Treasury. We made changes very soon after we had originally legislated and tightened up the scheme. Your core challenge that we should be looking for things that work is completely right, but that change did not come about because we were tracking cost; it came about because we were tracking behaviour—

              Q45 Chair: Right, let me get Amyas in, because I think you are doing this deliberately.

              Lin Homer: It is just more rounded, that is the difference.

              Sir Amyas Morse: What you are saying is perfectly fair. Please do not think that we are arguing something that no one has actually suggested. We are not saying that cost is the sole indicator, but there are times when costs against what you expected the costs to be are a significant indicator. That must be true and logical. If you take things such as share loss relief, you suddenly find that you have a huge spike. The question we are interested in is this: in some instances, your people who are working on the cases will pick things up, but in essence there is a time lag—it may be more; it may be less—and the need to change legislation probably depends on the rate of growth that you see. I suggest that you are going to find that out only by looking at the whole population, not just at individual case files.

              We are saying that actually knowing how fast things are going by aggregating numbers in a significant number of important cases probably gives you a faster warning that things are really moving fast and you might need to legislate, rather than simply relying on picking things up and dealing with them retrospectively, which, after all, costs a lot more in terms of administration and time. You would think that that was something with which you would want to be furnished. Is that not reasonable?

              Jim Harra: I agree; I think it can be useful. I would like to pick up two contrasting examples. In the case of share loss relief, we did not have to rely on or wait for data about the spike in costs of the relief to know that abuse was going on because we were seeing large numbers of avoidance cases, which we had already picked up before we had measured the spike in costs. A contrast with that is seafarers’ earnings deduction, where we saw an increase in the costs but had no operational intelligence to tell us what was going on. In that case, the increase in the costs provoked us to investigate in order to find out what was going on. There was no abuse, but it is an example of where costs did drive us to look at what was happening.

              Sir Amyas Morse: That is really all we are asking about. We are not trying to take you further than that.

              Q46 Chair: I want to ask you some questions about share loss relief because I think it is really interesting. Let’s pursue it for a minute. The Report tells us that you do not publish data on its cost because it is not robust. Is that right?

              Jim Harra: We know the value of the share losses on which relief is claimed, but we cannot estimate the amount of tax relief that flows from that because there are so many variables in there that we would have to look case-by-case and do a calculation.

              Q47 Chair: I don’t know how you would deal with that issue technically, but I would put it to you that we are talking about a lot of money here—more than £2 billion.

              Lin Homer: I think it increased from £320 million to £1 billion[1].

              Q48 Chair: It is not good enough; if you are going to be accountable to Parliament, you need to know the cost. You cannot just say that it is really difficult to collect the costs and get it accurate. If we are going to hold you to account on the effectiveness, efficiency and economy of this particular expenditure tax relief, we need to know the cost, and you do not give it to us. That is one thing. Perhaps you can go away and think about that.

              The second thing is that you say that you knew in 2006-07 when you got that big spike in the expenditure. Tell me if I am wrong, but the Report tells us that you did not do anything until 2013. There was a DOTAS registration around one or two schemes, but there was plenty of abuse going on elsewhere.

              Jim Harra: I am afraid that you are wrong, Chair. From 2005-06, we knew from DOTAS disclosures and our own operational intelligence that avoidance schemes were being used that were seeking to exploit share loss relief. We opened investigations into all those cases and did not agree the claims. Where repayments were involved, we withheld repayments. In 2013, we measured the cost of the claims in 2006-07, but by that time, we had long had them all under investigation. We have already had one case go to tribunal on that, which we won and which is now under appeal, and we are making progress on it. Accelerated payments will also help us to accelerate the resolution of that. That is an example of a relief where we did not have to wait or rely on measuring costs to know that there was abuse going on and to tackle it.

              Rob Prideaux: We do not contest that at all. We are not challenging whether the action you have taken to tackle those claims was right or wrong. It certainly started soon after the 2006-07 peak. What we would say is that you did not know until 2013 that, in that year, 80% of the claims for share loss relief you were investigating for suspected avoidance—you knew you were investigating a lot of claims, but you did not know it was 80% of the total. That added bit of information should be useful to you when thinking about whether the reliefs are working as intended and whether some sort of legislative fix may be appropriate.

              Jim Harra: I acknowledge that, and in previous hearings, we have acknowledged that our management information on marketed avoidance schemes is not what we would want it to be. We have described the work we are doing to improve.

              Lin Homer: It is getting better.

              Jim Harra: Yes.

              Sir Amyas Morse: And now you are monitoring the quantum of that relief on an ongoing basis, I think I am right in saying.

              Jim Harra: Yes, and a key reason why we have put additional monitoring in on that relief is that it is now subject to the cap on reliefs that was introduced in 2011, and we expect that to have a major effect on whether it is being used for avoidance.

              Q49 Chair: I simply make the point that I do not know why it took you until 2013, and what you were doing between 2006-2007 and 2013—maybe you would like to tell us, because it is a long, long time. Even in the best of all worlds, even with staff cuts, you could have done it earlier. One of the reasons why I think it is really important that you monitor this stuff earlier and tell us the costs, and not just the estimates, is that early intervention is much better than ending up in a tribunal. You want to stop it happening before you end up in litigation and tribunal.

              Lin Homer: Sometimes—and in the other example I gave you—we have absolutely done that. We have gone for early change to the legislation. In the case of share loss relief, we have a targeted anti-abuse rule, which is an intervention and a change, and that was brought in, but we think the underlying legislative provision is sound. What we think has happened here—you have seen it and you have had the promoters in front of you—is an attempt at very aggressive, very targeted abuse, which we think has failed. The question is: do you throw out your good idea because of that attempt or do you do it down? As for what we were doing between 2006-07 and now, we were trundling through the process, which can be very delayed, of seeking information disclosures. One case had tens of thousands of pages of information to be disclosed.

              As Jim has already mentioned, one of the reasons why we are pleased to have been given the power on accelerated payments is that in those very aggressive, very abusive marketed schemes, we can increasingly say, “We do not think this works. We are not going to give you the benefit of holding the money until seven years later, when we have proved it in court. You are going to pay up front.” I have said to you before that we genuinely believe that is also a game-changer.

              If you move on and the abusive marketed avoidance disappears, we think the basic objectives of share loss relief remain intact, and it would be sad for Parliament if that initial objective were not allowed to progress because of that interlude. That is the judgment we made. In others, we go, “No, this is too risky. We are not sure enough about our ground. Let’s legislate early”, but those are judgments that we are making in each case.

              Of course I accept Amyas’s point. Where we can get early information on cost, we should ask ourselves why we would not. We have already said that our view is that, in terms of our annual publication, we should go away and look at whether our “not known” category can be challenged. Let me be clear: sometimes I think we are going to come back and say such things as, “not relevant”, or, “not proportionate to acquire”, but it is a fair challenge to say to us that just having a category that looks as though we do not care is not the place that a good tax—

              Q50 Chair: Lettings relief is a case in point in the Report, where the cost of the tax is £200 million, which is a considerable amount of public money, and you do not think it is worth collecting the information on it.

              Jim Harra: Lettings relief is part of principal private residence relief, so its cost is reflected in our published cost of that relief. We do not specifically ask taxpayers to make returns when they are claiming lettings relief if it means they have no chargeable gain. Lettings relief is capped at £40,000, so if it does not completely extinguish their gains, they do have to make a return and show a computation of how they have arrived at the relief.

              Q51 Chair: How do you know it is being used for the purpose that was intended by Parliament? How do you know that the £200 million we are spending is achieving the purpose intended by Parliament? Do you know? You don’t know.

              Lin Homer: This goes to Meg’s question earlier. When information has to be gained by making a specific inquiry of an individual and we have 48 million individual customers, it can be a very costly business for the individual and for us. We make proportionate judgments about when information is so valuable and necessary to us that we must have it—when it is useful and when it is not needed. We just do.

              Q52 Meg Hillier: May I follow up my previous point about digitisation? This relates to real-time data. I should declare that I am a small employer of a part-time babysitter and one other babysitter who occasionally does stuff on the payroll, so it is not very big but because it is a small amount the burden for me could be enormous. But I have found that there is a burgeoning of interesting software out there that is available for free to people like me, so I simply plug in the hours and the hourly rate and up, magically, pops all the other information and it enables me to do all my returns online and automatically to HMRC. That has been a major breakthrough in the last few years. If it is possible to do that, surely it is possible to do it for all the others, so my question is, how fast is digitisation going and when will that sort of facility be available to home owners who are selling, entrepreneurs, farmers and all the rest?

              Lin Homer: I would say that during the next five years, we are moving into that field, but I—

              Q53 Meg Hillier: So in five years’ time, all of this will be pretty automated is what you’re saying?

              Lin Homer: Maybe not all, but a lot more. I would not want you to think we are sitting there thinking, “Don’t need to know that.” We have one of the richest data warehouses you can imagine. An enormous amount of that data is shared by 600 publications. The ONS takes a great deal of its information from us. We are increasingly sharing it through two research approaches—not only with people who do research for us, but through what we call the Datalab with other academic institutions, so that this quite big, rich database is just being put out there. So we can see the benefit, not just for us but for others, of making what we know broadly available.

              Q54 Chair: Lin, all I want is us and the Treasury Select Committee to hold you to account on value for money for these reliefs. You do have rich data. The issue is getting it in a form so that we can test whether this is meeting the objectives set by Parliament; whether you are administering it efficiently, which is the other aspect of it; whether it can be deemed value for money; and whether it is open to abuse. Those are the sorts of thing we would be looking at. At the moment, we can’t. You are coming back and saying, “We’re going to get better.” You have left yourself a little bit—

              Lin Homer: I am saying we are also, actually, very good at it now. When you say, “At the moment, we can’t,” that is not fair.

              Q55 Chair: At the moment, “we” can’t—“we” can’t.

              Lin Homer: I think we share a great deal of information with you.

              Chair: No, “we” can’t. If you go back to figure 2, you’ll see we have not got sufficient information on all those expenditure tax reliefs to enable us to do it. “We” can’t. You may be doing it internally. You have to be accountable, through us and through the Treasury Select Committee, to the public. I’m going to bring in Austin, who has been really patient.

              Q56 Austin Mitchell: I have been patient while the Chair has exploded my bombshell questions, so all I can do is pick around in the debris. This is a question about the share loss relief. Claims rose from £385 million to £1,206 million in one year; these are ones that were not dealt with until later. But leaving that aside, the problem it produces, I feel, is that you do not often know where there is a problem and there is abuse, because you do not collect the totals. We do not have the totals—the costs of these reliefs—and you do not calculate them all. That is borne out, I think, by what Jim said about the discovery of this share loss relief claim in 2005-06. You said it was partly from your own investigations, but it was also from a DOTAS report. The NAO Report says at paragraph 3.18 that they have looked at the speed of response to other claims where your response was not much quicker—three cases—and in most of these the trigger for action was a DOTAS disclosure. In other words, they are saying to you this is being used as part of a tax fiddle, a tax avoidance scheme, and you are then taking action, but you don’t know that it is being used in that way because you don’t keep the totals.

              Jim Harra: In the case of share loss relief, we did know from the outset when it was being abused, because the promoters of those schemes were having to disclose the fact that they were doing that through DOTAS or because our own operational intelligence was telling us the schemes—

              Q57 Austin Mitchell: But the other schemes were business premises renovation, share loss relief, and entrepreneurs relief.

              Jim Harra: I think the BPRA one and the share loss relief are good contrasts. In the case of share loss relief, the abuse does not involve the interpretation of the legislation. It involves the creation of contrived or artificial losses. Therefore, we concluded that we did not need to amend the legislation in order to tackle that abuse. There was a targeted anti-avoidance rule brought in in 2006, and we feel that we can use that and other arguments to tackle that abuse.

              In the case of BPRA, the abuse does involve the interpretation of the legislation, and although we do not think it works, and we are tackling it, we did decide that we should clarify the legislation as well, to put the matter beyond doubt. That is a judgment—it is part of a standardised process that our product owners have to follow. Where there is attempted abuse going on, they have to conclude, does this or does this not require legislative change as a way of tackling it? Those decisions are governed by an anti-avoidance board, which, in both of these cases, agreed the action that we were taking. So they are contrasts, but they are quite deliberate contrasting choices that we made.

              Q58 Austin Mitchell: Yes, you have taken action, thank heavens, but what the Report is saying, it seems to me, is that the trigger for that action came from DOTAS, not from your collection of the figures and your knowledge of the abuse.

              Jim Harra: No, in the case of share loss relief, I think there are about 25 schemes altogether, four of which we found through DOTAS, and the other 21 we found through our own operational intelligence, but we did not identify that through tracking the costs of the relief. It was after the event we knew the costs.

              Q59 Chair: Are you going to get all £800 million back, then?

              Lin Homer: We haven’t paid it out.

              Jim Harra: We haven’t agreed any of those claims. To the extent that repayments were required we withhold the repayments. To the extent that people reduce their liability, until we had accelerated payments that was a cash-flow advantage that people were able to get, but now we have a tool that says you cannot get that cash-flow advantage either, so we have not lost any of the tax in share loss relief. We are fighting it. We are either holding on to it, if it was a repayment case, or we are fighting to get it from the tax take.

              Q60 Guto Bebb: I want to pick up some things that have been thrown out during the hearing. In particular, Lin said that the figures in relation to entrepreneur relief were entirely, in your view, covered by changes in the legislation that has been undertaken by Parliament. In view of the fact that this is a bone of contention between the NAO and yourselves, I just wanted to allow you to finish that comment, because I don’t think you were able to finish it.

              Lin Homer: I think it would probably be rash of me to say “entirely” and if I did use that word you would be right to challenge me, but what I think we are clear about is that a significant element of the cost variation comes about because of elements that we can see very clearly. The first of those is the changes in the scope of the relief, which is a judgment made on a number of occasions. Of course all this follows on from the removal of taper relief from capital gains tax, so in a sense it is not an entirely new space. This is something that followed on the removal of taper relief across capital gains tax.

              So the initial decision was that entrepreneurs or people investing in entrepreneurial businesses could have a lifetime allowance of up to £1 million of relief. That was then extended to £2 million, and then £5 million and then £10 million. Now £10 million compared to £1 million is a 10-times increase in the value of that relief. That inevitably alters the scale of it, and the point I was trying to make earlier is that that was not an unknown thing.

              Jim may want to say something about the behavioural consequences, but the second element of that is that capital gains tax levels also changed during that time, so the relief was to give you a 10% tax rate on sale of an entrepreneurial business, rather than be subject to capital gains tax. When it was introduced, that was the difference between 10 and 18, and that was the difference between 10 and 28. Those two factors account for an enormous amount of the change.

              Q61 Guto Bebb: This is my question. We are looking in the Report at a £2.4 billion difference. Would you be willing to hazard a figure as to how much of that difference is explained by policy changes?

              Lin Homer: I would prefer for me not to make that up. We could ask KAI, our knowledge analysis team, to have a go at that, but you can see that that must be the majority of that.

              Q62 Guto Bebb: But in terms of allowing the Committee to have an understanding that is accurate, it would be valuable if you could challenge the figure we currently have in front of us, because we have no reason to doubt the figure of £2.4 billion, unless you have figures that would counteract that.

              Lin Homer: But the other element, which is a bit harder to explain—I may need to make way for my tax expert sitting to my left—is that one of the things they were trying to do was to change the behaviour of entrepreneurs and get them, in a sense, to let go of their business and let it grow. If you take the relief away, it is not that they would then have paid 28% on that. Some of them would not have sold their businesses, so you have to take into account the behavioural shift that also would not have occurred. If you look at the capital gains tax growth during that period, the relief and the growth have a relationship. So you could not just say that if we had kept the lifetime allowance lower or if we had made it 15% instead of 10%, it would have cost us less, because the consequence might have been that you lost some capital gains entirely, and that is the complexity that we are trying to get over.

              Jim Harra: I would just add that entrepreneurs’ relief is one of those reliefs where we have got very good operational intelligence, so we do not rely on monitoring costs to understand whether it is being abused. Every single claim is risk profiled, so we know from our own work whether we have concerns about what is going on. In addition, for every single taxpayer who claims it, we police their lifetime limit to make sure that they cannot exceed it. We can see no evidence that it is being systematically abused.

              Q63 Chair: Is there any abuse?

              Jim Harra: There have been two schemes disclosed under DOTAS. In one of those schemes, no user has ever taken it up. In the other scheme, it is an employer who is trying to get an extra tax benefit for their employees through inserting a company on which they can then claim entrepreneurs’ relief. We know about that scheme. From our point of view, one of the advantages of entrepreneurs’ relief in terms of tracking what goes on is that it involves you becoming a shareholder in a vehicle in which you then sell the shares. We can get our hands on the shareholder register, so, as soon as we see a company being used as a vehicle for avoidance, we can identify every shareholder.

              Q64 Chair: That is the one instance that you know about. Can you say to the Committee there is no other abuse?

              Jim Harra: We are confident that there is no widespread or systematic abuse.

              Q65 Chair: “Widespread” is different. In that £2.4 billion, is there abuse?

              Jim Harra: I cannot put my hand on my heart and say there is not a single instance of it, but it is low risk compared with, for example, share loss relief.

              Sir Amyas Morse: It is not as if we are putting gross numbers out there. We have done our best to reflect HMRC’s own estimates of what they thought the relief would cost, including being revised for the changes in the legislation that you have described so fully. It is not as if this is something different. When we look at those estimates and compare them with what actually happened, there is a big delta. I am listening and I agree. If it was all explained by structure, you would expect the numbers to look similar, but they do not, so there is a slight problem in that explanation.

              Jim Harra: I do agree. I just want to make a comment about figure 8 in the Report in relation to that, but I do agree. When we were making the policy changes in this relief, it was very difficult to forecast, without a wide margin of error, what the impact of that was going to be, because we were seeing policy change being accreted on policy change each year, and we were basing our forecasts on costs from previous years when those were not there, so it was very difficult to get that right. Obviously, the actual costs are significantly different from what we forecast. When we did publish our forecast, both we and the OBR said that it was a tentative forecast, subject to a wide margin of error.

              We have two graphs in figure 8, but they are not measuring exactly the same thing, so you cannot compare them directly. The red line—the upper line—is measuring the gross cost of this relief; the yellow line is measuring the net Exchequer effect of the policy changes. In some cases, although you will be giving more relief to entrepreneurs as they sell their shares, you are getting more capital gains tax, albeit it 10% rather than 28%, because more people are selling shares. That is part of the explanation of why there is a delta—they are not comparing the same thing.

              Lin Homer: Although I know that the Chairman hates us doing this—I don’t plan to do it—we felt that there was a better graph, which we would have liked to have been included in the Report, that we think shows the correlation much more clearly. I am very happy to send it to you. I do not propose to introduce it today, but I think it makes the point that Jim is making.

              Q66 Chair: To be honest, we would prefer you to sort that out and agree a graph beforehand.

              Rob Prideaux: May I just add something? We did look at that—the Department shared those data with us, but we did not demonstrate the correlation sufficiently strongly for us to feel that we could put it in the Report. The other point that I wanted to make, which is in the Report, is that the Treasury told us that with so many policy changes in a relatively short time, it was difficult to investigate fully whether the cost increase might have been influenced by misuse of the relief, but HMRC has carried out checks of the claims, as Jim described. The problem is that it is very hard to know all the reasons and to explain them—

              Q67 Chair: Whether they have been abused a lot or whether it has been explained?

              Sir Amyas Morse: But that should mean that they are constantly updating it, not just letting it stand.

              Lin Homer: Absolutely.

              Sir Amyas Morse: If you will forgive me, in one of the years, the amount was double what you were expecting. Quite honestly, we are not trying to go further than to say that you need to have more constant and better information in these areas. That is where we are going, not further.

              Indra Morris: Costings for this have been published every year since 2011, so the information is there.

              Sir Amyas Morse: I know, but the expected numbers are not in line with it. I am not arguing—

              Indra Morris: That is partly because—

              Sir Amyas Morse: It is a normal management approach that if you estimate something and you are consistently finding very big differences, you need to try to work to understand the difference. That is what we are saying to you.

              Indra Morris: And we said at the very outset—we were very transparent—that this was highly uncertain. We made no claims about how much investment or how many people. There was no quantified measure, if you like. We have kept an eye on it, and, as my HMRC colleagues have said, there is a risk-based approach: checking and keeping an ear to the ground. It is a key part of the ongoing discussion—

              Sir Amyas Morse: But all the estimates have been over.

              Q68 Chair: I think all we are saying is that £2.4 billion is a lot of money and we want to understand whether it is being spent in a value-for-money way.

              Indra Morris: What we are saying from the checks is that we think it is. Parliament has made a number of decisions this Parliament to increase the value of it, and you have another opportunity in the next Finance Bill to debate entrepreneurs even further if you want to.

              Q69 Chair: You don’t know—that is what you hope.

              Lin Homer: No, we think we do know.

              Q70 Chair: You obviously didn’t know well enough to convince the NAO.

              Lin Homer: Chair, there are points of disagreements between ourselves and the NAO about what level of inquiry is appropriate and what information is valid. We have a clear view on this. There is a difference of opinion between us about whether we could or should have known more and whether the fact that this very wide range of costs has been exceeded is good, bad or indifferent. We think that that goes to the policy debate about whether you keep it in. We have no evidence to suggest that there is systematic abuse and we have no evidence to suggest that Parliament’s intention is not being achieved by this.

              Q71 Chair: Sorry, but although it’s good to hear that, my problem is that it is a negative. All we would say is that in monitoring and assuring us of the value for public money you have to say, “We have looked at this. There is no abuse, apart from one case, and this is meeting the intention of Parliament.” It is very simple. At the moment, you are saying, “We don’t know; we haven’t had any evidence,” which is a negative.

              Lin Homer: No, I am very happy to turn it round into that positive.

              Indra Morris: And some qualitative research on this was conducted last year—I think it is due to be published this year. That will be useful.

              Q72 Guto Bebb: I want to clarify something. At the start of the hearing, the Chair highlighted the figure of £100 billion of tax relief. I want to ensure that I am correct in my own mind about this: would that include the £12.5 billion or whatever on principal private residence?

              Lin Homer: Yes, I think it would. I think we had that debate last time that our own view is that there are a number of quite non-contentious things in that figure.

              Q73 Guto Bebb: That’s fine. The final point is parochial, so I apologise for that, but a lot of my constituents are complaining that you are not doing enough on the agricultural side of things. They complain that small, Welsh family farms are being bought by people who keep a few horses then claim the reliefs. I want to know that you are investigating this with a hard nose.

              Lin Homer: So, Jim, are you out in the Welsh hills enough?

              Jim Harra: I should pass to Indra as a horse rider. I take it that you are talking about agricultural property relief in terms of inheritance tax.

              Q74 Guto Bebb: I accept your point about the value going up, but there is some concern.

              Jim Harra: Yes. What we do is monitor that that relief is being claimed correctly. There are rules in there about how long property must be held before you can qualify for the relief, which is two years for an owner-occupier who is doing the farming themselves, and seven years for someone who lets to a tenant farmer.

              Q75 Guto Bebb: It is the farming that they are concerned about.

              Lin Homer: It has to be worked.

              Jim Harra: It is for a much richer set of data than just the cost to say what is actually happening out in the Welsh hills or wherever as a result of this relief—what kinds of behaviour are being produced and, for example, are two years and seven years the right figures, or whatever? I would say that this relief, or something similar, has been there for 90 years and survived successive radical reforms of this tax and its predecessors, so successive Parliaments over 90 years have kept it.

              Q76 Chair: I have a few more questions, but I will rush through them because I know that I am the one who is obsessed with this topic. First, I will make a plea for consistency between your definitions and those of the Office of Tax Simplification. If you do not have an agreed definition of what tax reliefs are, that makes things really difficult. When I meet the Office of Tax Simplification, they tell me that it has gone up to 1,140, but when I come down here you are at 398 or something. Can I make a plea for that, or am I asking for the impossible?

              Lin Homer: It is quite difficult to define what a tax relief is. I know, for instance, in the work that John Whiting and OTS have done recently, they count every element, so I think there are something like 18 reliefs in and around banking levies, whereas it might be that the NAO and ourselves will go away and count that as one.

              Q77 Chair: I am just asking for agreed definitions. It might be that you give some to them or the other way around, but at least then we will all be looking at the same data.

              Lin Homer: I think we all think that there are about 400 major reliefs—

              Q78 Chair: I met with them yesterday and that is not what they said to me.

              Lin Homer: I know, but they are very deep tax experts and I think they capture literally everything that has moved since the dawn of time, don’t they?

              Jim Harra: They certainly count some very small things and, ironically, they say that entrepreneurs’ relief is not a relief but a rate of tax. So it is most definitely a term of art rather than a term of science. Having said that, we do have a regular dialogue with the OTS and I am quite happy to look at whether there are things that should be added to our list or taken away.

              Indra Morris: I am slightly surprised, because I looked at this quite recently and it looked to me like we had broadly the same numbers. There was a bit of catching up to do from the last fiscal event, but even in their 2011 report, I think they focused on only 155 and decided to leave or ignore about 800.

              Q79 Chair: They quantify more—

              Indra Morris: Yes, we have probably got a number, about 140, since.

              Chair: The other interesting thing is that, if you can get that agreement, then we can see what we are getting. All of us want simplification, but the trend is towards more rather than less. It makes it easier to see that.

              Q80 Austin Mitchell: We heard from, I think, Shire pharmaceuticals, which set up an office—two people—in Luxembourg and gave them an enormous amount of money. That office then loaned money to Shire in Britain and America and so on because it could claim tax relief on the interest. That must be going on on a considerable scale to reduce tax obligations in any one specific country. Do we know the total of tax forgone owing to that relief, not just from Shire, but everyone?

              Lin Homer: As Jim said, we have tax under investigation in relation to share loss relief. We do not think we have lost tax. There are two big aggressive schemes where action is—sorry, we have had 21 disclosed schemes, haven’t we, and we are working through those, so—

              Q81 Austin Mitchell: But you must have lost tax.

              Lin Homer: No.

              Q82 Austin Mitchell: Well, they aren’t paying tax on the—

              Lin Homer: I think—sorry, Jim.

              Jim Harra: I think we may not be talking about share loss really, I suspect.

              Austin Mitchell: No, I’m not.

              Jim Harra: I think you are probably talking about controlled foreign—

              Austin Mitchell: Interest tax relief on debt.

              Lin Homer: Oh, right. Sorry.

              Jim Harra: I think this goes back to something we discussed before the Committee, which is the controlled foreign companies regime, which applies to UK-parented groups that have overseas subsidiaries, but it does not apply to groups that are not parented in the UK. So there are some groups who may be benefiting in the way you described, but if they are not parented in the UK they are not subject to the UK’s CFC rules. However, if they are parented in the UK they are subject to those rules, which protect us from that kind of avoidance.

              Q83 Austin Mitchell: But we do not know the total of money not received by the Revenue because people are claiming tax relief on this—

              Jim Harra: I think what you are asking for is, do we know how much we might get if we had different rules from the ones that we have, which is—

              Q84 Chair: Do you know?

              Jim Harra: Sorry.

              Chair: We don’t know?

              Jim Harra: No, that’s not something that we have.

              Q85 Chair: Let me just go on to various other things.

              One issue raised with me is, why don’t you have more sunset clauses with some of these tax reliefs, so that you can then test whether or not they are there for the purpose that was intended and are working?

              Lin Homer: Well, some do.

              Q86 Chair: How many?

              Indra Morris: Not many.

              Lin Homer: Not many. Business premises renovation allowance does.

              Indra Morris: Social investment tax relief. BPRA has already been extended.

              Q87 Chair: Would that be a trigger to force an assessment as to whether or not it was meeting the purpose intended?

              Indra Morris: The BPRA one was extended in this Parliament, actually, for another five years. That is running until 2017.

              Q88 Chair: Why not more often? It is a way of controlling it.

              Indra Morris: It’s a fair debate to have, but let us not give ourselves false reassurance that somehow some of these reliefs are not partly—it comes back to this point about objectives and the extent to which tax is also about sending a signal about what is valued as much as it might be, in some cases, about saying, “We would like to get a certain amount of response, whether that’s employment or investment.” So you could do, but there is a debate to be had about, where would it work.

              Q89 Chair: I am just talking about this group: the ones where you do want to have an impact somewhere, either on a particular industry or a particular group of people, or whatever it is.

              Jim Harra: I think it is a valid debate. The small number of reliefs where we do have sunset clauses, like the corporate venturing and enterprise investment schemes, they have common characteristics: they are intended to achieve very specific aims and they are usually approved state aids, where you do want to make sure that they still meet the conditions for that approval or, if they have achieved objectives, they are stopped. But it is quite valid for Parliament to decide whether to have them or not.

              Q90 Chair: Okay, maybe we can reflect on that.

              The second one was R and D tax relief. When I looked at the detail of that in one of the appendices, looking 2001 to 2011, the costs of the tax relief went up from £100 million—I hope I am reading this right. It is indexed at 100 in 2001 and goes up to 1,060 in 2011-12, but the actual amount—total business expenditure on R and D; the purpose of this is new increased investment in R and D—stays more or less static. If you start at 100 in 2001, it goes up—the cost to the taxpayer of this relief has increased tenfold. The actual amount of R and D activity in the private sector has scarcely moved. That is back to the purpose that Parliament intended and the effectiveness of this relief, which is a very expensive relief: over £1 billion, I think. I haven’t a clue. How much is it, R and D? A lot of money?

              Rob Prideaux: Yes, in one of the reports—

              Chair: And again, that is part of the accountability that I am seeking—that we are seeking.

              Lin Homer: So, research and development, there are two schemes: one for SMEs and—

              Q91 Chair: I know the schemes. I am just asking about the effectiveness, Lin. If you can’t answer it, you can’t. They are stark figures.

              Lin Homer: This is an area we have looked at very carefully. Again, we don’t believe there is any systematic abuse in this area. We think it’s achieving its objectives.

              Q92 Austin Mitchell: How is it achieving its objectives—the amount spent on research and development?

              Indra Morris: That’s not the only counter-factual. If it was not there—

              Q93 Chair: There has been no increase—

              Lin Homer: But it may have gone significantly down in the absence of relief. We do not know the counter-factual.

              Q94 Chair: Let me put it to you again. Of course, it “may”. The whole purpose behind this sitting this afternoon is that we don’t know. You may be 100% right, but if you’d just done a little bit of research and investigation then you could have come to us and said to us, “Actually, the taxpayer is still getting value, and if we hadn’t had this incredibly generous tax relief, R and D by private sector companies would have gone to nought.” I would have been happy with that. You can put it as a thesis to me; you can’t prove it.

              Indra Morris: Lin mentioned the data lab earlier and two of the evaluation projects that are using that data are looking at R and D tax credit and one of them is specifically looking at the decision making within small businesses, and how that is affected by R and D tax credits.

              Q95 Chair: And?

              Indra Morris: We don’t know yet—that is the evaluation. That is research; it takes time.

              Q96 Austin Mitchell: The R and D in pharmaceuticals has gone down.

              Jim Harra: First of all, one of the reasons why it is key that that research is done is that some of the recent policy changes that have been introduced in R and D are intended to make it have a greater behavioural effect than it has had. For example, creating an above-the-line relief for businesses is precisely intended to make sure that it is something they take into account when making decisions about how much to spend on R and D, because some of our evidence was that it was a relief claimed by their tax department but it was not featuring in their internal business cases to decide whether to invest in R and D. So we need to evaluate whether those policy changes are having that impact.

              Rob Prideaux: In answer to your question, Chair, in the last estimate, for 2014-15, it is £1.7 billion a year.

              Q97 Chair: A lot of money, and all we’re asking is that the taxpayer should know whether we are getting value.

              Indra Morris: We are supporting about £13.2 billion of investment.

              Q98 Chair: All I was reading was the graph in the appendix and all I am doing is lifting from the graph. I know there is some investment, but I am asking the question, as an MP, and it doesn’t look value for money to me.

              I will ask about three other issues that are slightly on the edge, but I gave you notice, Lin, of one or two. First, it wouldn’t surprise you if I asked you what on earth you think you are doing such that people get their information about their tax affairs in tweets. Would you like to have the opportunity to explain that to us? As for phone answering, I will put it on the record that it is dire, Lin. The figures I have are that 34.5% of calls to the HMRC helpline in September were going unanswered, which was an increase. The figure was 2.4 million in September, but it was 1.2 million a year before that. And the average wait to get a call answered is around 11 minutes in the September figures, and that is more than double the average wait of more than four minutes 12 months before. It is dire, and then to expect people to just tweet you if they want information—I wouldn’t tweet for my personal tax information. And I do now tweet.

              Lin Homer: Yes, we’re well aware. [Laughter.]

              Chair: Don’t tell me you are following me?

              Mr Jackson: I think they deleted you.

              Lin Homer: Let me deal with the more serious part of that first. I am back in front of you next month and I am sure that the customer will be quite a big part of that conversation. We are not doing well enough. We are, roughly, maintaining the level that we got last year. That is significantly up on two or three years ago, but it is not good enough. And we are really striving—

              Q99 Chair: That’s not what the figures say. Are my figures wrong?

              Lin Homer: Month on month, it goes up and down. You remember that when we talked to you at the end of last year, we had achieved around about 80% overall in the year. We think we will be there or thereabouts. We could well be 2% or 3% lower, but it won’t be very far off.

              The issue for us is that we had been falling into a pattern of the first six months of the year being much poorer than the second six months, and this year we’ve been trying—with only partial success—to get a more consistent performance. Now, the good elements of that, which I think you have given us credit for, are that we did much better on tax credit peak this year than we did last year. About 80% of calls were handled in the last week of tax credit. You will remember that, on the last day the year before, the figure was 16%, so we have had some significant improvements. Overall, our waiting times are evening out, and they are not 11 minutes all of the time. The consequence of trying to be more consistent is that we have done less well in the second half of the year than we normally do. The first half has been better, and the second half has not been so good.

              We have been making some overall investments in customer service. We now have all the 0300 numbers, and we have changed our telephony. We think that, overall, that is a good thing, but there have inevitably been some teething issues. You might have read the Which? report, which says that our voice recognition is not instantly picking up new things such as VAT MOSS, which we thought you might ask about at the next hearing. Overall that is a good thing. We have also been upgrading the technology on call handlers’ desks, which is a good thing, but it means that they have all had to train and get used to what we are doing. We are doing what we now call “flexible resourcing” so there will be 1,500 people helping out through SA peak, which would not normally be in the call centres. July, August and September 2014 were not quite as good as in 2013.

              Q100 Chair: It was pretty lousy.

              Lin Homer: It was a couple of per cent. lower. It wasn’t massively different. What I would say, because SA peak is now two weeks away, is that there are still quite a lot of people who need to fill in their tax return. It is always that way at this stage and, as of yesterday, 5.2 million people had done so, which means that there is still half to go. That is always the way. It is 140,000 more than at this point last year, but you won’t be surprised when I say that, if lots of those people leave it to the last day or so, they put themselves and us under tremendous pressure.

              Q101 Chair: Do you expect them to tweet?

              Lin Homer: I was going to go on to tweeting. I sat in front of your Committee a few years ago and suggested that we should use texting, and I think you all laughed at me when I said that. We now regularly use texting as a day-to-day part of our business, and it really works. In the tax credit peak, when people were queuing to renew their tax credits, we texted them to say that they can go online, and half a million people did. In the same way, we think that tweets, on which I am not as expert as you are, give a chance for us to give an answer to a frequently asked question that thousands of people can see. That is what it should be used for; it absolutely should not be used for, “This is my unique reference and this is my bank account. Can you take my tax payment?” Of course you are right that it is not a space to do personal stuff and it is not a space to do detailed stuff. We find that tweets and webinars, which Jim’s team do brilliantly, give a chance for a wide answer to a question. Because people who are followed get re-tweeted, a good answer to a tweet actually reaches thousands of people, so we think it is one small part of an armoury. On our core need to answer the telephones well, we are still not doing well enough. I would expect us to spend some time with you on that in the next year.

              Q102 Chair: How many tweets do you do a week?

              Lin Homer: Goodness knows—probably not as many as you.

              Q103 Chair: There are two others. I have been asked about Scotland—I have now lost the bits of paper.

              Lin Homer: I am hopeful that you will get a letter by the end of the week. I am sorry that we weren’t able to get one to you today. Broadly speaking, we believe that we are on target for the introduction of landfill and land duty this year, and on target for the introduction of the income tax changes next year. We are publishing an annual report, and we are due to publish another one in May. We are undertaking a significant number of campaigns to make sure that people know what we need to do. Edward Troup is leading for us on that one. You will get a letter within the next 48 hours, and if that generates more questions, we would be very happy to hear from you.

              Q104 Dame Anne McGuire: What is the relationship between HMRC and the Scottish Government, who are obviously going to have responsibility for the tax? It is quite a complicated deconstruction of a unified system—I see nods of agreement—and I wonder how many staff you have working on that. How are you progressing the deconstruction? Has there been any agreement on where people are taxed from? Most MPs, for example, are taxed from Cardiff even though, for tax purposes, I am actually resident in Scotland. There may be policy issues that you might not wish to answer, but it is coming quite quickly and we need to get a feel for how many people are actually working on this, given that it is a major change in how we collect taxes in the UK.

              Lin Homer: I do not have all those detailed questions here, but I am sure we will have made a note. We will make sure we add them to the letter. We have been working on this for a long time. You know that this pre-dates the referendum and the wider devolution conversation. Edward has been accounting directly to their Select Committees. We think the working relationship is strong, positive and going very well. We are having discussions about both the level and the pace of deconstruction. Our intent is to provide support in whatever way is appropriate. There will be choices to be made about when we are doing something under direction for the Scottish Government versus when they are choosing to take it over themselves. I believe we are working very productively on that.

              We will try to pick up on both the questions that the Chair gave me notice of and the things you have spoken to, but if there were a desire to have a more thorough discussion with the Committee on that, in a public or a private sitting, as we have done before, we could perhaps offer an information-giving session on that. There is some very good-quality work. I do not believe it is a big team. I will not try to guess off the top of my head, but I think we have put good-quality people in, and I think the relationships with our colleagues in Scotland are good and strong.

              Q105 Guto Bebb: In relation to the answer you have just given, obviously some of the tax is currently being devolved in a Scottish context and, in some cases, will be devolved in a Welsh context. In terms of future-proofing, are you having discussions with the Administration in Cardiff as well?

              Lin Homer: Edward is Mr Devolution. We are talking with Northern Ireland, Wales and Scotland. To pick up your point more directly, we will give the best advice to each of those regional Governments about the best way to do things, which we sometimes think will be by continuing to keep experts in small groups, but there will be choices to be made and we will do our best to give good informative advice about the choices in good time for decisions to be made.

              Q106 Meg Hillier: Just for the record, can you tell us who Edward is?

              Lin Homer: Edward Troup, second permanent secretary. We felt this was a sufficiently important matter for the devolved Governments to have someone they could hold accountable.

              Q107 Meg Hillier: So he’s a civil servant. He is your number two.

              Lin Homer: Yes, he’s my second permanent secretary. He is himself a tax expert and he is personally being held to account in the Scottish environment in their Select Committee. He and Amyas have appeared together.

              Sir Amyas Morse: Yes, I have had that pleasure. Thank you; it is nostalgic for me.

              Q108 Chair: And my very final point: I have read that you have given a contract to the TDX Group to support debt collection. Is that right?

              Lin Homer: We are working with the Cabinet Office on the introduction of a civil service-wide debt management integrator. We are working on that contract. It is coming into effect—it is not in operation yet—and we are completing the arrangements for that.

              Q109 Chair: A completely reliable source that always gives me the best of information, Private Eye, asserts that the TDX Group has very much the structure of the ones in our PwC session—that Shire has—so, in essence, it is an American-based company that has a structure in Luxembourg to avoid paying UK tax. I have wonderful maps, which I could have brought and showed you. It seems to me that before the tax authority of Government gives a contract to a firm to do a very important job, should you not be assuring yourself that that firm does not indulge in a complicated maze of company structures to avoid paying UK tax on the business that you will be giving them, which will be funded by the taxpayer?

              Lin Homer: Chair, you know we won’t talk about individual taxpayers with you.

              Q110 Chair: No, I’m talking about a contract.

              Lin Homer: I can assure you that following, I think, Danny Alexander’s request, both ourselves and the Cabinet Office are applying the assurance standards that Government asked us to. Actually, the Cabinet Office has shown itself to be very strong with its commercial arrangements where any existing or future provider is not prepared to sign up to the arrangements. We will apply what we are asked to, and if people don’t match up either at the start or during a contract, action is taken. It is not always possible to go as far in applying commercial law as some commentators would want.

              Austin Mitchell: You can pick and choose—

              Lin Homer: We might have to beg to differ about whether Private Eye always gets everything correct—

              Chair: They usually do.

              Lin Homer: But I am well aware that it’s a good source of information.

              Q111 Chair: Can you write to us and let us know—give us assurances about this particular contractor? I am seeking assurances. It is a contractor that will get money from the taxpayer.

              Lin Homer: I am happy to take away and discuss that with the Cabinet Office, and we will share with you whatever we think we properly can. Some of this is commercially sensitive information, but I am very happy to take that one away and discuss it with the Cabinet Office.

              Chair: Okay. Thank you very much indeed.

              Lin Homer: See you again next month.

              Oral evidence: The effective management of tax reliefs, HC 892                            28