Joye v HMRC: top slicing relief rules

In the March issue of Tax Adviser, I discussed the case of Sarah Yaxley v HMRC [2025] UKFTT 51 (TC) (‘The power of a single word’). The background to that case was a challenge to HMRC’s calculations of top slicing relief and how Mrs Yaxley fell foul of procedural hurdles preventing her from obtaining the full amount of relief to which she was entitled under the top slicing relief rules (when correctly applied).

The fact that HMRC had been misapplying those rules became clear in the First-tier Tribunal’s decision in Silver v HMRC [2019] UKFTT 263 (TC), which was published on 18 April 2019, although concerns had been raised in the professional press previously. Following the Silver case, the law was amended (or clarified) and HMRC’s official position subsequently changed so as to be aligned with the Silver decision.

This article concerns another case in the top slicing relief story: Joye v HMRC [2025] UKFTT 664 (TC).

The facts of the case

Mr Joye submitted his 2017-18 tax return in July 2018. His return included a computation for top slicing relief of about £10,800. I infer that his return was submitted electronically and, as a result, the calculation of top slicing relief was required to conform with HMRC’s interpretation of the law (whether or not that interpretation was in fact correct).

I also infer that the return was not the subject of any enquiry by HMRC, the enquiry window ending in July 2019. Furthermore, it seems that Mr Joye did not seek to amend his tax return, which is something he could have done up to 31 January 2020. Therefore, the return can be considered to have become final on 31 January 2020.

On 5 May 2020, however, Mr Joye made a claim for overpayment relief in accordance with the rules in the Taxes Management Act (TMA) 1970 Sch 1AB. Mr Joye’s claim was based on his belief that the top slicing relief should have been just over £66,000 and therefore he sought a repayment of over £55,200.

By the time that the case reached the tribunal, HMRC agreed that the original top slicing relief calculation was wrong and that Mr Joye’s increased claim for top slicing relief was based on a correct interpretation of the legislation. It appears, however, that initially HMRC resisted this – although the tribunal’s decision is not totally clear on this point. As a result of HMRC’s acceptance (whenever it took place) that Mr Joye’s revised calculation was correct, the focus of the dispute turned on the availability of overpayment relief and the Sch 1AB rules themselves.

Schedule 1AB provides a time-limited opportunity (four years) for taxpayers to recover overpaid tax over and above a taxpayer’s right to amend a Self Assessment return. However, presumably to provide some sense of finality within the Self Assessment system, a taxpayer’s right to overpayment relief under Sch 1AB is subject to a number of restrictions. They are set out in para 2 of the Schedule as eight exceptions (or ‘Cases’) which, if engaged, prevent a claim from being validly made.

Of particular importance in this case is Case G, which prevents an overpayment relief claim being made when the original (but now accepted to be incorrect) calculation of the taxpayer’s tax liability ‘was calculated in accordance with the practice generally prevailing at the time’.

The First-tier Tribunal’s decision

The case came before Tribunal Judge Geraint Williams and Jane Shillaker.

The first matter that the tribunal had to decide was the date on which the tribunal has to determine the prevailing practice. Without much hesitation, the tribunal stated that in a case such as this, it is the date on which the original tax return was submitted; therefore, July 2018.

The tribunal then turned to the case law on prevailing practice. In recent years, this has focused on cases where taxpayers have sought to avoid a discovery assessment. This is because TMA 1970 s 29(2) provides a defence against discovery assessments, available to taxpayers if their return was made in accordance with the then prevailing practice.

The logic is that taxpayers should not be subject to any additional tax simply because, subsequent to their return being submitted, HMRC’s views have changed. (It should be noted that this defence does not apply in cases where the return was subject to an enquiry. In such cases, HMRC is entitled to depart from previous views of the law, although any earlier prevailing practice will, of course, be relevant should penalties be imposed.)

The tribunal focused on the First‑tier Tribunal’s decision in Boyer Allen Investment Services v HMRC [2012] UKFTT 558 (TC), which set out propositions that will allow a court or tribunal to determine whether a practice can be said to be generally prevailing. In summary, those propositions are:

1. The alleged practice should be capable of being readily ascertained, have substance and be sufficiently precise.
2. It will be readily ascertainable, either through published statements or by way of settled practice.
3. An internal HMRC practice will not be generally prevailing until it can be identified with sufficient precision by taxpayers and their advisers.
4. The same quality of clarity and precision must be shared by HMRC and taxpayers (and their advisers).
5. To be generally prevailing, it must have been adopted by HMRC and generally (if not universally) by the taxpayer community.
6. The practice must be settled and applied in a consistent manner.

The tribunal considered a number of different sources to ascertain what (if any) was the prevailing practice. In particular, the tribunal accepted that HMRC’s published manuals (as at August 2016) reflected evidence of an ascertained substantial and sufficiently precise practice. However, the picture was not so clear in relation to whether the practice was adopted generally outside HMRC. Industry articles published after March 2020 did acknowledge HMRC’s change of practice post-Silver but these were of little assistance to Mr Joye, as he needed to show the absence of a generally prevailing practice when he submitted his return in July 2018.

What the tribunal found particularly relevant in this regard was a number of articles published in Taxation. There were a couple of articles by Richard Curtis in the noughties which looked at top slicing relief. In particular, in September 2009 Richard wondered whether ‘it is time for HMRC to adopt a new interpretation’. He asked himself: ‘Am I on to something here or would that still just be a novel interpretation of the legislation?’

Whilst it can be said that Richard had (correctly, so it transpires) identified flaws in HMRC’s interpretation of the legislation, the tone of the article suggested that his was a lone voice (very much like the child in the story of the emperor’s new clothes) and therefore insufficient to displace the practice generally prevailing at the time.

The tribunal next considered another article from Taxation, this time written by Tim Good (who represented Mr Joye in the present case). Tim’s article was published in September 2017 and he set out a number of ways in which, he believed, HMRC’s interpretation was wrong. From my recollection, most of these difficulties arose because of the various changes in the taxation of savings income a few years earlier. The article’s standfirst (quite possibly written by Richard Curtis, who was then the editor of Taxation) said: ‘Tim Good … suggests that it is time to follow the legislation rather than existing practice.’ The tribunal considered this to be further evidence that there was indeed a generally prevailing practice at the time of publication.

However, the tribunal continued and concluded (correctly in my view) that Tim’s article ‘was the catalyst for the taxpayer community challenging HMRC’s practice’. This conclusion was further evidenced by articles published in 2019 both before and after the Silver decision.

The tribunal therefore decided that the previous generally prevailing practice came to an end on the date of publication. As that predated Mr Joye’s tax return, the tribunal concluded that the calculation in the return was not ‘calculated in accordance with the practice generally prevailing at the time’. As a result, Case G did not apply and Mr Joye was entitled to his overpayment relief.

Commentary

This is the first case that I can remember where the taxpayer has won a prevailing practice argument. The case is different from the others of the past two decades, though: the other cases involved taxpayers seeking to establish the existence of prevailing practice, whereas this case involved a taxpayer trying to show the absence of a prevailing practice. However, what is particularly helpful about this case is that it demonstrates that there was a prevailing practice but that this later ceased to be the case.

What I found of particular interest was the date that the tribunal chose to represent the end of the previously prevailing practice – being the date of the publication of Tim Good’s article in September 2017. That approach certainly has the advantage of precision in that it is possible to point to the date of publication with some certainty (if one overlooks the fact that Taxation articles are generally published online a day before the formal date on the cover of the printed magazine).

However, it takes time for a practice to change. I am sure that of all the people who read Tim’s article, only a small minority read it on the date of publication, even if they read it within a week or so. Of course, this assumes (and I am sure the publishers of Taxation would agree) that the readership is so significant a proportion of the population that it can be sufficient to negate the prevalence of a particular practice. It is at least a mainstream publication within the tax profession.

It seems to have helped Mr Joye that Tim’s article was widely disseminated amongst tax advisers across the country and that Tim had managed to collate evidence of challenges to HMRC’s interpretation from as early as late 2017.

In short, as the tribunal stated, Tim’s article was the catalyst for change. However, possibly for pragmatic reasons, the tribunal also decided that the date that the previous practice ceased to be generally prevailing was the date of publication.

Even if those two statements are considered to be inconsistent and that the change must inevitably have taken place after the date of publication, the tribunal referred to sufficient evidence to support the view that the change predated the submission of Mr Joye’s return. Of course, if it were decided that the change took place on a later date, there would inevitably be some arbitrariness and subjectivity as to the date chosen. For example, suppose a tribunal had chosen 1 May 2018 to be the date when the effect of Tim’s article had reached a tipping point; that would be unhelpful to a taxpayer who had submitted a tax return on 30 April 2018. The tribunal’s approach, as I have said, removes some of this capriciousness.

What to do next

The case shows it can be possible to challenge interpretations of the law, even if they are deeply embedded within the tax community, and to do so successfully. However, what perhaps proved to be crucial in the present case was the volume of evidence that demonstrated how Tim’s article had set off a chain reaction amongst tax advisers.

As a result, if you are proposing to demonstrate the absence of a generally prevailing practice, it will be very helpful to show that the view you are advocating is more widely held and is being acted upon by others. As might be said, there are two types of taxpayer who appear before the tribunal: those who are loaded with evidence; and those who lose.

© Getty images