HMRC fails to establish "practice generally prevailing" in top-slicing relief dispute

In Roger Joye and David Sumners v HMRC [2025] UKFTT 664 (TC), the First-tier Tribunal (FTT) allowed the taxpayers' appeals against HMRC's decisions to deny overpayment relief based on incorrectly calculated top-slicing relief (TSR).

Background

Roger Joye and David Sumners filed self assessment tax returns (SATRs) for tax years 2016/17 and 2017/18, respectively, each claiming TSR, which applies when an individual realises the benefit of a life assurance policy. It recognises that the accumulated growth of the policy has taken place over several years and it would be unfair to tax the whole gain in one year because it may push the individual into the higher rate or additional rate tax bands for that year. The amount of TSR is broadly the difference between the tax payable on the full gain and the grossed-up tax payable on annualised 'slices' of the gain.

Mr Joye and Mr Sumners had used HMRC's tax calculator program to calculate the available TSR, but subsequently realised that the calculations were incorrect. They made overpayment relief claims in the amounts of £55,201.63, for Mr Joyce, and £40,758, for Mr Sumners.

HMRC opened enquiries into the overpayment relief claims and subsequently issued closure notices to Mr Joyce and Mr Sumners, denying entitlement to the increased relief. The closure notices were appealed to the FTT.

FTT's decision

The appeals were allowed.

HMRC accepted that its tax calculator program was incorrect. In fact, it had been determined by the FTT in Marina Silver v HMRC [2019] UKFTT 0263 (TC), that HMRC's calculations were not in accordance with the legislation due to the use of a shortcut which, in some instances, did not take into account the taxpayer's personal allowance. Rather than apply the personal allowance to each annualised slice of the gain, HMRC's calculator assumed that, if the taxpayer's income in the year the gain was realised meant the personal allowance was reduced or removed, that should follow for every year the TSR applied.

HMRC sought to argue that, even if its calculation methodology was incorrect, it was the "practice generally prevailing" (PGP) at the time the taxpayers' SATRs were filed and therefore HMRC was not liable to give effect to the overpayment relief claims. The FTT confirmed that, as HMRC was seeking to rely on a PGP argument, the burden of proof was on it to establish that a PGP existed.

In HMRC v Household Estate Agents Ltd [2007] EWHC 1684 (CH), Henderson J summarised PGP as a practice that is: "relatively long-established, readily ascertainable by interested parties, and accepted by HMRC and taxpayers' advisers alike".

Mr Joyce and Mr Sumners were able to provide a significant amount of evidence which showed that HMRC's methodology was not accepted by taxpayers' advisers and the FTT therefore rejected HMRC's PGP argument.

Comment

This decision provides helpful guidance as to how the FTT is likely to approach the concept of PGP, a concept which is referred to in several places in the tax legislation. 

The taxpayers were successful on this occasion because they were able to provide the FTT with both publicly available articles criticising HMRC's practice, and contemporary correspondence, to demonstrate that this criticism was widespread among professional advisers.

The decision can be viewed here.