Notification of uncertain tax treatments: round two

Last year the government consulted on plans to require large businesses to notify HMRC of uncertain tax positions. The proposals were met with criticism from respondents, largely owing to the relatively low threshold for reporting and the subjective proposed test for ‘uncertain’ treatments, which required taxpayers to predict whether HMRC would challenge a position knowing all of the relevant facts (even if that challenge was unlikely to be successful).

The second round of the consultation seeks to address these concerns through raising the reporting threshold and setting out more prescriptive tests for notification. The consultation closes 1 June 2021.

As part of the wider publication of consultation documents on "tax day", HMRC has released the second round of its consultation on the notification of uncertain tax treatments. Interested parties have until 1 June to provide a response.

The government intends to introduce legislation implementing the regime in Finance No.2 Bill 2021/22 with effect for tax positions taken in any returns due to be filed after April 2022. Subject to any further delay, this means that corporation tax positions being taken today fall within the new regime. In addition, it is implied by the consultation that some pre-existing ongoing tax treatments will also need to be notified even when the treatment in question has been applied for many years.

What is the purpose of the proposal?

The stated intention of the proposal is to reduce what HMRC call the "legal interpretation tax gap", the difference between the tax that would have been collected under HMRC’s interpretation of the law (whether or not that interpretation is correct) and the tax that was actually collected. The cost of the legal interpretation tax gap was estimated at £4.9bn in HMRC’s 2019/20 report.

This £4.9bn annual figure is out of all proportion to the annual increase in revenue projected from the measure. The extra tax take peaks at £45m for the tax year 2023/24 and settles at £40m in subsequent years. Given those numbers, it is not surprising that some in the industry estimate that the overall compliance cost for HMRC and taxpayers may exceed the extra revenue raised.

What was proposed in the first consultation, and what has changed?

The first round of the consultation proposed a notification requirement for large business undertaking transactions with an uncertain tax treatment where the tax at stake exceeded £1m per annum. The proposals were met with fierce criticism from respondents. Criticism focused on the proposed definition of "uncertain" treatments, which required taxpayers to consider not the view of a hypothetical court (i.e. whether a particular tax treatment is more likely than not) but whether a hypothetical HMRC officer might challenge the treatment (even if that challenge was not ultimately successful or even likely to succeed), a difficult task with no objective answer. HMRC aims to provide further clarity through the more prescriptive "triggers" set out in the new consultation.

The latest proposals reflect the comments made in the first consultation and envisage that the new regime will:

  • apply only to corporation tax, employment taxes and VAT;
  • have a notification threshold of £5m; and
  • apply to "large businesses" with the same criteria as the existing SAO regime, i.e. business with either a turnover exceeding £200m or balance sheet total exceeding £2bn. This will not be restricted to entities falling within the Companies Act 2006, so the measures will include general partnerships and LLPs. The proposed penalty for non-compliance with the new regime remains £5,000, but this will now only be levied on the business rather than on the individual person(s) responsible.

However, it is the definition of "uncertain" which has seen the most significant change since the first consultation, and which will be of most interest to taxpayers and practitioners.

How will an uncertain tax treatment be defined in the proposed regime?

The consultation provides for seven possible "triggers", which will require a business to report transaction(s) to HMRC if the tax treatment applied:

  1. is different to HMRC’s known position;
  2. is different to published industry practice (published by either HMRC or a trade body);
  3. is a change from the treatment applied to an "equivalent" transaction in a previous return, and that change is not due to a change in legislation, case law or HMRC policy;
  4. is "in some way novel such that it cannot reasonably be regarded as certain";
  5. has resulted in an accounting provision to recognise the risk of a different tax treatment being applied;
  6. results in a deduction exceeding the amount incurred by the business or in income received which is not reflected for tax purposes (unless HMRC is known to accept this treatment); or
  7. has been the subject of professional advice (not subject to legal privilege) which contradicts previous professional advice or which the business does not follow.

Of these, we think triggers (4), (6) and (7) present the greatest risk of overreach or unintended consequences.

In relation to "novelty", the consultation suggests that notification would be required where "there is a new or novel product, transaction or business structure where there are various ways that it can be treated and HMRC’s position is not known". This is a broad test which will require notification for any new product launched or transaction entered into by a business where the tax treatment is not obvious, or which does not explicitly fall within HMRC’s guidance. VAT will be fertile ground for these types of issues, since new products often do not fit neatly into any of the established categories for exemption or reduced rates. Indeed, whether a new food item is subject to VAT or not is given as an example of this in the consultation.

A second example in the consultation introduces some confusion about what counts as "novel". This is the scenario where a UK bank has a branch of an overseas services entity which it has included in its VAT group (with the benefit that otherwise irrecoverable VAT is not incurred in relation to services between members of the VAT group). This type of structure is widely understood in the financial services industry, and banks had open discussions with HMRC about it for several years. In fact, VATA 1994 s 43(2A)–(2E) was introduced in FA 1997 to combat similar arrangements. It does not really make sense to think of this as a "new" structure, but rather a longstanding arrangement where HMRC have more recently started to challenge the VAT treatment.

Some in the industry estimate that the overall compliance cost for HMRC and taxpayers may exceed the extra revenue raised

Trigger (f) – for deductions which exceed the amount incurred etc. – reminds us of an equivalent concept used in the GAAR (FA 2013 s 207(4)(a)-(b)) and the loan relationships regime TAAR (CTA 2009 s 455D(1)(a)–(b)) where in both cases it is used as an indicator of potential abuse or avoidance. But crucially, under the GAAR or the TAAR, there is no abuse or avoidance where the result is consistent with the principles on which the relevant legislation is based or its underlying policy objectives. In trigger (f), this test is subtly changed so that it is not the legislative principles or policy (enacted by parliament) that must be considered but instead whether HMRC is known to accept the treatment. This does not sit easily with the opening statement in the consultation that the notification regime "is not intended to promote any assumption that HMRC’s interpretation is correct, nor that HMRC is a final arbiter of tax law."

By asking taxpayers to compare the size of a deduction with the amount incurred, this test also leaves open the question of what it means to incur an amount. Amounts recognised in company accounts would seem to be a sensible starting point, but the example given in the consultation of something that meets trigger (f) involves an amount which is recognised in the accounts (a foreign exchange loss) but which HMRC might say is not a real expense.

Finally, trigger (g) has the unattractive feature that it may put two different taxpayers who undertake exactly the same arrangement in different positions as regards the notification requirement, depending on whether advice was taken, or was taken from more than one adviser, or was taken from an adviser whose work is covered by legal professional privilege.

To what extent will different taxpayers be affected by the proposed changes?

The impact of the proposals on taxpayers will depend in part on whether they already have a dedicated customer compliance manager (CCM) and the nature of their relationship with that CCM. For some large businesses, particularly those with a relatively open relationship with their CCM, the proposals are unlikely to make a significant difference in their interaction with HMRC, not least because the consultation recognises that there should be no requirement to disclose arrangements which have already been raised with the CCM in sufficient detail.

For many businesses without a CCM, the main barrier to engagement with HMRC is not a desire to avoid scrutiny, but simply an inability to make contact with an appropriate officer or technical specialist with the authority to make a decision. If HMRC really do open a new route for discussions around tax uncertainty for those businesses without a dedicated CCM (as suggested in the consultation), that in itself would be an improvement for many taxpayers.

What should taxpayers do now?

Taxpayers within the regime should consider whether any ongoing or anticipated tax positions will fall within the new legislation following its publication. In particular:

  • Any tax position which exceeds the £5m threshold and will be included in a return filed after April 2022 should be identified and reviewed for a possible disclosure requirement. Advice provided on any prospective significant transactions should take the new disclosure rules into account.
  • Ongoing tax positions which exceed the £5m annual materiality threshold should be identified and reviewed in light of the proposed triggers. For example, the VAT treatment of any material recurring revenue items, if that treatment is not consistent with HMRC’s current guidance, should be scrutinised as it is likely that disclosure will be required.

When working with businesses to review potential sources of tax uncertainty, we have found that HMRC’s view can itself be uncertain or subject to change, which can compound the difficulty in applying the suggested new triggers. For example, HMRC’s published practice in relation to the VAT treatment of compensation payments underwent a sudden and significant change in September 2020, before changing again in January 2021, leaving some of the substantive issues open for resolution at some point in the future. How should a business that now receives a large compensation payment assess whether the VAT treatment is uncertain, as tested by the triggers described above? We think the answer cannot be known for sure until the notification requirement arises (e.g. in April 2022), by which point HMRC may or may not have a sufficiently clear and settled position in their guidance.

A further difficulty is that identifying ongoing tax positions which exceed the materiality threshold is not always straightforward, particularly for VAT where an ongoing contract might be a minor part of a VAT return (or not included altogether other than as a net statistical figure). Obtaining certainty that all ongoing tax positions have been identified therefore risks becoming an exercise in proving a negative for some businesses (albeit that we would expect the higher proposed materiality threshold will reduce this burden for most).

One way of reducing the legal interpretation gap might simply be for HMRC to offer a faster and more responsive non-statutory clearance procedure on a wider range of tax issues

What else could HMRC do to address the legal interpretation gap?

Many responses to the first consultation focused on the difficulties taxpayers without a CCM experience when engaging with HMRC. One way of reducing the legal interpretation gap might simply be for HMRC to offer a faster and more responsive non-statutory clearance procedure which is willing to engage on a wider range of tax issues. Taxpayers without a CCM would appreciate the ability to discuss a tax treatment directly with an HMRC specialist to shape information requests and to avoid protracted correspondence.

One sidenote on this: where HMRC is unable or unwilling to give a non-statutory clearance, we have seen a number of responses from HMRC declining to provide an answer on the basis that HMRC does not think that there are genuine points of uncertainty (this being one of the grounds on which the clearance team can legitimately refuse to provide an answer). It will be interesting to see what happens if HMRC subsequently decides that the tax treatment in question is uncertain enough to require notification under the new regime, even though it was not uncertain enough to require an answer from HMRC to the clearance request.

It would also be helpful if the proposals were introduced in conjunction with HMRC providing more (and clearer) guidance on longstanding issues of uncertainty (in fairness, this was recognised by the government in its response to the first consultation).

Finally, it would be helpful if the government considered simply reducing areas of uncertainty in the tax system. For VAT in particular, there is an opportunity to tackle the uncertainty that surrounds the borderlines of the VAT exemptions, much of which is due to outdated exemptions and narrowly drawn CJEU decisions which the UK is no longer obliged to follow. To take an obvious example, an update to the financial services exemptions based on modern regulatory definitions and outsourcing practice would sweep away current areas of uncertainty.

We will be providing input to the consultation, both in an independent capacity and through industry groups, and we encourage taxpayers to do the same. 

This article was first published by Tax Journal.