Publication of the tax gap figures by HMRC has been an annual event for about ten years now. Each year in July, a short press release confidently states how much the tax gap has gone down, with a few cheerful quotes from a Treasury minister and HMRC (see here). This is accompanied by a detailed document (typically of 90 pages or so) as well as supporting information on the statistical methodology – but let’s face it, most people don’t get beyond the press release.
And each year, the Public Accounts Committee (PAC) reviews HMRC’s performance, and comments on the tax gap. Relations between the PAC and HMRC have been strained in the past (particularly when Margaret Hodge fiercely challenged HMRC’s policy for settlements with large business, around 2013), but they seemed to have become more amicable under Meg Hillier’s chairmanship of the committee. However, this year, the PAC’s report on 16 October clearly struck a raw nerve.
To coincide with the publication of the PAC’s report, Jim Harra, head of HMRC, wrote an open letter in which he said: ‘I find the Committee’s characterisation of our work in this area to be wholly unfair and unsubstantiated’.
What was it that upset Jim Harra so much? It appears to have been the PAC’s first recommendation, where they said that HMRC ‘continues to quote figures in a way which suggests a much greater degree of precision’ than actually exists in the tax gap.
Actually, I think the PAC has a point. The tax gap report is detailed and contains much useful information which is helpful both internally to HMRC and to the tax profession. It is released with a press release headed ‘Tax gap falls to lowest recorded rate’ and which praises the ‘sustained efforts by HMRC to support the overall health of the tax administration system’.
The problem is that the headline, which is all that many people will read, boils down the tax gap to a single figure with a superficial air of accuracy. When you dig deeper into the analysis, the uncertainties become more obvious: the overall figure comprises some elements calculated on a ‘top down’ basis (mainly VAT and excise taxes), while other parts are estimated from surveys, administrative and operational data. The large business data tends to be of fairly good quality, as most large businesses are regularly audited. But the small business figures are estimated from a sample of returns – and it would surprise many people to know that the small business gap is more than twice that of large businesses.
Despite popular perceptions, avoidance represents only a relatively small part of the total (£1.7bn out of £31bn), mainly because the tax gap figure is stated net of the expected yield of compliance audits – so if an avoidance scheme is implemented, but HMRC is confident it will fail, that would be booked at nil in the tax gap estimate.
So I agree that it would be better to release the tax tap figures with a more nuanced headline, making it clear that the gap is a long term performance indicator and not an accurate figure for tax which could be collected if only HMRC were perfect. Indeed, it is quite interesting to compare the style of the tax gap press release with that used by the ONS in their publications, which are generally much more technical in style (for example, see here).
I also think that the tax gap would be more useful if it were to be reconciled to other information about HMRC’s performance. For example, HMRC calculates its ‘tax under consideration’ (TUC) figure for large business, showing the total amount of tax at stake in current audits. This information is not generally widely published, but it has been obtained on a number of occasions via freedom of information requests (for example, see my comments here). While HMRC generally expects about half of the TUC to be eventually collected, there is no way to reconcile movements in TUC to the tax gap – so effectively we have an income statement measure (the tax gap for large business) and a balance sheet estimate (the total TUC) but there is no link or explanation of how the numbers relate to each other. The TUC figure may go down because audits have been concluded, or because fewer enquiries are being opened in a particular year – if instead, the tax gap and TUC were examined together, some useful conclusions about HMRC’s performance could be drawn.
Ideally, I’d like HMRC to publish its forecast for the amount of tax it expects to collect each year, and then explain why the actual receipts differ from the forecast – just as any head of tax in industry would do. Of course, some of the factors will be outside HMRC’s control – the obvious example this year being the impact of covid-19 – but the reconciliation would put the tax gap movements into much clearer context.
Where I do have huge sympathy for HMRC is with its frustration with another of the PAC’s comments that the tax gap does not include amounts for legal, but ‘sophisticated and undesirable’ tax planning. Leaving aside the practical difficulty of how you would define ‘undesirable’ tax planning (which has not been included in the avoidance figures already reported), this would undermine the whole purpose of the tax gap, which is to measure the gap between tax which could be collected in line with current legislation and the amount actually collected.
If politicians, on the PAC or more generally, want to know how much ‘undesirable’ tax planning is being undertaken, they should examine more closely how much various reliefs are costing (which they are, belatedly, beginning to do) – and then change the law if they don’t like the answer.
Publication of the tax gap figures by HMRC has been an annual event for about ten years now. Each year in July, a short press release confidently states how much the tax gap has gone down, with a few cheerful quotes from a Treasury minister and HMRC (see here). This is accompanied by a detailed document (typically of 90 pages or so) as well as supporting information on the statistical methodology – but let’s face it, most people don’t get beyond the press release.
And each year, the Public Accounts Committee (PAC) reviews HMRC’s performance, and comments on the tax gap. Relations between the PAC and HMRC have been strained in the past (particularly when Margaret Hodge fiercely challenged HMRC’s policy for settlements with large business, around 2013), but they seemed to have become more amicable under Meg Hillier’s chairmanship of the committee. However, this year, the PAC’s report on 16 October clearly struck a raw nerve.
To coincide with the publication of the PAC’s report, Jim Harra, head of HMRC, wrote an open letter in which he said: ‘I find the Committee’s characterisation of our work in this area to be wholly unfair and unsubstantiated’.
What was it that upset Jim Harra so much? It appears to have been the PAC’s first recommendation, where they said that HMRC ‘continues to quote figures in a way which suggests a much greater degree of precision’ than actually exists in the tax gap.
Actually, I think the PAC has a point. The tax gap report is detailed and contains much useful information which is helpful both internally to HMRC and to the tax profession. It is released with a press release headed ‘Tax gap falls to lowest recorded rate’ and which praises the ‘sustained efforts by HMRC to support the overall health of the tax administration system’.
The problem is that the headline, which is all that many people will read, boils down the tax gap to a single figure with a superficial air of accuracy. When you dig deeper into the analysis, the uncertainties become more obvious: the overall figure comprises some elements calculated on a ‘top down’ basis (mainly VAT and excise taxes), while other parts are estimated from surveys, administrative and operational data. The large business data tends to be of fairly good quality, as most large businesses are regularly audited. But the small business figures are estimated from a sample of returns – and it would surprise many people to know that the small business gap is more than twice that of large businesses.
Despite popular perceptions, avoidance represents only a relatively small part of the total (£1.7bn out of £31bn), mainly because the tax gap figure is stated net of the expected yield of compliance audits – so if an avoidance scheme is implemented, but HMRC is confident it will fail, that would be booked at nil in the tax gap estimate.
So I agree that it would be better to release the tax tap figures with a more nuanced headline, making it clear that the gap is a long term performance indicator and not an accurate figure for tax which could be collected if only HMRC were perfect. Indeed, it is quite interesting to compare the style of the tax gap press release with that used by the ONS in their publications, which are generally much more technical in style (for example, see here).
I also think that the tax gap would be more useful if it were to be reconciled to other information about HMRC’s performance. For example, HMRC calculates its ‘tax under consideration’ (TUC) figure for large business, showing the total amount of tax at stake in current audits. This information is not generally widely published, but it has been obtained on a number of occasions via freedom of information requests (for example, see my comments here). While HMRC generally expects about half of the TUC to be eventually collected, there is no way to reconcile movements in TUC to the tax gap – so effectively we have an income statement measure (the tax gap for large business) and a balance sheet estimate (the total TUC) but there is no link or explanation of how the numbers relate to each other. The TUC figure may go down because audits have been concluded, or because fewer enquiries are being opened in a particular year – if instead, the tax gap and TUC were examined together, some useful conclusions about HMRC’s performance could be drawn.
Ideally, I’d like HMRC to publish its forecast for the amount of tax it expects to collect each year, and then explain why the actual receipts differ from the forecast – just as any head of tax in industry would do. Of course, some of the factors will be outside HMRC’s control – the obvious example this year being the impact of covid-19 – but the reconciliation would put the tax gap movements into much clearer context.
Where I do have huge sympathy for HMRC is with its frustration with another of the PAC’s comments that the tax gap does not include amounts for legal, but ‘sophisticated and undesirable’ tax planning. Leaving aside the practical difficulty of how you would define ‘undesirable’ tax planning (which has not been included in the avoidance figures already reported), this would undermine the whole purpose of the tax gap, which is to measure the gap between tax which could be collected in line with current legislation and the amount actually collected.
If politicians, on the PAC or more generally, want to know how much ‘undesirable’ tax planning is being undertaken, they should examine more closely how much various reliefs are costing (which they are, belatedly, beginning to do) – and then change the law if they don’t like the answer.
