Tooth – discovery assessment invalid as no inaccuracy in return

In HMRC v Tooth [2018] UKUT 38, the Upper Tribunal (UT) has confirmed that a discovery assessment issued by HMRC was invalid as the taxpayer's self-assessment did not contain an inaccuracy and in any event there was no deliberate intent by the taxpayer to bring about an insufficiency of tax.

Background 

Mr Raymond Tooth (the taxpayer) sought professional advice on how he might reduce his income tax liability for 2007/08 and was advised that a tax planning arrangement known as Romangate might help him achieve this aim. The arrangement was designed to produce an income tax loss in 2008/09 which the taxpayer was advised could be set against his 2007/08 liability. 
Unfortunately, the taxpayer's advisers were unable to complete his self-assessment tax return for 2008/09 using HMRC's approved software. Due to a technical issue with the software they were unable to enter the income loss. The taxpayer therefore entered the loss on the partnership pages of his return and made a 'white space' disclosure informing HMRC of what he had done and explaining that it was an employment loss and not a partnership loss that was being claimed. 

In 2009, HMRC enquired into the claim but did not open an enquiry into the 2007/08 return because of the uncertainty at the time between enquiries into claims and returns which was the subject of ongoing litigation in HMRC v Cotter [2013] STC 2480.

In 2013, following the Supreme Court's judgment in Cotter in favour of HMRC, HMRC wrote to the taxpayer stating that as a result of the Cotter decision, income tax was overdue from the taxpayer because it had rejected his claim to offset the employment losses from the Romangate arrangements against his other income. The taxpayer argued that Cotter required HMRC to enquire into the return rather than the claim, which HMRC accepted. In 2014, HMRC then issued a discovery assessment pursuant to section 29, Taxes Management Act 1970 (TMA), for 200708 claiming the taxpayer's return was inaccurate and that the mistake was deliberate. My claiming that the loss was brought about deliberately, HMRC was able to rely on the 20 year extended time limit for raising a discovery assessment under section 36(1A), TMA.  

The taxpayer appealed against the discovery assessment, relying on the following two grounds:

(1) HMRC had not made a 'discovery'; and

(2) the assessment was out of time because there was no deliberate inaccuracy.

In order to successfully challenge the validity of the assessment, it was only necessary for the taxpayer to succeed on one of these grounds.

The First-tier Tribunal (FTT) found in favour of the taxpayer and allowed his appeal. Whilst acknowledging that HMRC had made a 'discovery', the FTT held that the situation had not been brought about deliberately. Section 29(4), TMA, was not satisfied and the discovery assessment was therefore invalid.

HMRC appealed to the UT.

UT decision

The appeal was dismissed.

The UT considered whether there was in fact an inaccuracy. It was accepted that the deduction the taxpayer sought to make in his return was not permitted. So the question was whether an entry in a document that is explicitly based on a bona fide, albeit controversial, interpretation of the law, which is later found to be incorrect, amounts to an inaccuracy. The UT decided that it does not. The taxpayer had clearly stated the position he was taking in his return and there had been full white space disclosure. Although  the entry on the partnership pages of the return was clearly inaccurate, in the overall context, the UT concluded that the approach taken by the taxpayer, to force his interpretation into the return in a way that was precluded by HMRC's approved software,  did not constitute an inaccuracy. 

The UT's decision on whether there was an inaccuracy was enough to conclude the matter, as the pre-conditions to the operation of section 29 had not been satisfied. However, as the other points had been argued by the parties, the UT addressed those as well. 

With regard to whether, if there were inaccuracies, they were deliberate, the UT concluded that any such inaccuracies were not deliberate. Again, it looked at the overall context and came to the view that because the taxpayer had taken steps to draw the purported inaccuracies to the attention of HMRC, he had not acted deliberately.

Finally, the UT considered whether, if the return and computation contained deliberate inaccuracies, there had been a 'discovery' by HMRC. The UT formed the view that any discovery was made in 2009, when all the facts were known to HMRC and it first raised a challenge (when an enquiry was opened under Schedule 1A, TMA). If a discovery had been made then, it had become 'stale' by the time of the issue of the discovery assessment in 2014. 

Comment  

This is an important decision as the UT has confirmed that an entry on a return is not inaccurate if it is based on a bona fide interpretation of the law, notwithstanding that that interpretation is controversial and is later found to be wrong.

In addition, the UT's comments on what constitutes a discovery for the purposes of section 29 are equally important. HMRC's conduct in this case was heavily criticised by the UT. The UT did not approve of HMRC's attempt to use the discovery legislation as a 'replacement' for a Schedule 1A enquiry. HMRC had not opened the right enquiry at the right time and could not seek to use its discovery powers in order to circumvent the difficulties it faced.

The UT commented that on making a discovery, HMRC should act expeditiously in issuing an assessment. A discovery can only be made once and the taxpayer should be protected from HMRC relying upon a 'stale' discovery. In this case, it was clear to the UT that the first officer made the discovery in 2009; the second officer simply found out something that was new to him. If the first officer determined not to issue an assessment, that outcome was binding on HMRC. The concept of 'staleness' is an important and developing area of the law and is something which was discussed in Pattullo v HMRC [2016] UKUT 270 (TCC) and  Hicks v HMRC [2018] UKFTT 22. A link to our recent blog on Hicks is available to view here.

A copy of the decision can be viewed here.