The discovery assessment provisions in s. 29 of TMA 1970 permits HMRC to make an assessment for underpaid tax in certain situations. The time limit for raising the assessment depends on the behaviour that led to the underpayment. The standard time limit is four years from the end of the relevant tax year, but this increases to six years in cases of carelessness. Where there has been deliberate behaviour leading to an underassessment, the time limit is 20 years.
A discovery assessment is a valuable tool in HMRC’s arsenal, as these time windows are considerably longer than those for raising enquiries. However, it is important to understand that s. 29 only gives HMRC the power to raise an assessment, not to open an enquiry outside the window.
In the case of Raymond Tooth (R), a number of aspects of the legislation were considered that are important to understand in practice. A discovery assessment can only be made if there is a genuine "discovery" of an inaccuracy leading to an underpayment of tax (this can include an omission of income). R had entered into a tax planning arrangement that involved employment losses. A disclosure was made in the additional information pages, and the losses were recorded in the partnership pages of the return due to limitations with the software used.
HMRC opened an enquiry, but were later informed that they had done so under the wrong statutory provision. Being out of time to open a valid enquiry, a discovery assessment was raised. Due to the length of time that had elapsed since the end of the relevant tax year, HMRC needed to show there was a deliberate inaccuracy in order for the s. 29 assessment to be valid.
The case progressed all the way to the Supreme Court, via both tax tribunals and the Court of Appeal. All four courts found in favour of R, but for slightly different reasons on each occasion.
On the matter of whether there was a deliberate inaccuracy, the Supreme Court stated that a "deliberate inaccuracy" means one that is deliberately inaccurate, rather than a deliberate statement which transpires to be inaccurate. The difference is subtle, but important. The Court held that for a statement to be deliberately inaccurate "there will have to be demonstrated an intention to mislead the Revenue on the part of the taxpayer as to the truth of the relevant statement or, perhaps, (although it need not be decided on this appeal) recklessness as to whether it would do so".
The Court decided that there was no inaccuracy, due to the quantum of the figures being correct when looking at the return as a whole – albeit some were in the wrong place. Had R simply included the figures on the partnership pages with no explanation, the decision may have gone the other way – which underlines the importance of making a disclosure in the white space whenever there is any potential for confusion. In any case, the judges also said that even if the use of the "fudge" did constitute an inaccuracy, they were not convinced it would have been deliberate, meaning the shorter time limit applied and HMRC would have been out of time with the assessment anyway.
The Upper Tribunal had found that there was no valid discovery because HMRC had taken the view that there was an inaccuracy five years prior to raising the assessment (for the reasons explained above). The discovery was therefore "stale" due to the time that elapsed. This view was upheld by the Court of Appeal. The Supreme Court dismissed the concept of staleness, pointing out that there is no mention of it in the statutory provisions or in previous case law. The judges’ view was that there is no need for such a concept, as the taxpayer is already protected by the statutory deadlines.
It is important to understand that this represents the opinion of the judges, but as it was not essential to the crucial facts of the case in hand, it is merely an observation made "in passing", so it does not set a legal precedent. However, it does indicate the likely way that a case that did rely on an argument of staleness would go if heard.
Finance Bill changes
The Finance Bill 2021/22 contains clauses to bring the High-Income Child Benefit Charge (and certain other charges that are collected via self-assessment but are not directly attributed to income or gains) firmly within the remit of s. 29. This follows the Upper Tribunal decision of Jason Wilkes, which held that there was no legal right under s. 29 to collect these charges.
Controversially, the provisions will be retroactive – a move which is very rarely popular, though HMRC’s argument is that the change is merely a clarification of the law. The forthcoming Court of Appeal decision will consider the matter in detail based on the existing law.
The legislation will not apply retrospectively to those individuals who previously received a discovery assessment and:
– who submitted an appeal to HMRC, on the basis of the arguments considered by the upper tribunal, on or before 30 June 2021(the date at which the Upper Tribunal handed down its decision in the relevant case.); or
– whose appeal, made on or before 30 June 2021, has been stood over by the Tribunal pending the final outcome of the relevant litigation.