‘Staleness’ cannot be used against Discovery Assessments – UK Supreme Court Decides in HMRC v Tooth
The UK Supreme Court commented during the recent case of HMRC v Tooth [2021 UKSC 17] that ‘staleness’ is not a concept that can be used against discovery assessments, meaning that a discovery will not lose its quality over time.
‘Staleness’ has previously been argued by taxpayers to state that a discovery needs to be new and the discovery assessment needs to be issued within a certain period of time otherwise the discovery and the discovery assessment will not be valid. The concept of ‘staleness’ has been subject to numerous cases of litigation between HMRC and taxpayers since 2008 and this recent ruling has provided clarity on this matter.
What is a Discovery Assessment?
The rules for discovery assessments are largely contained within the Taxes Management Act 1970 (“TMA 1970”) and HMRC must ‘discover’ in order to carry out a discovery assessment.
HMRC can conduct an assessment on a taxpayer if they ‘discover’:
- That an income / chargeable gain which ought to have been previously assessed has not been assessed;
- That an assessment to tax is or has become insufficient; or
- Any relief provided previously is or has become excessive.
Where a taxpayer has filed a tax return, there are further restrictions to where a discovery assessment can be made. HMRC will need to show either:
- The loss of tax was brought about by the taxpayer acting carelessly or deliberately (or on behalf of the taxpayer); or
- That the HMRC officer could not reasonably have been expected, on the basis of information available to him before that time, to be aware of the facts giving rise to the loss of tax.
There are time limits for a discovery assessment to be made and this ultimately depends on the reason for the incomplete disclosure. These time limits range from:
- 4 years from the end of the relevant tax year to which the assessment relates. This applies where the conduct was an incomplete disclosure without careless of deliberate behaviour;
- 6 years from the end of the relevant tax year to which the assessment relates. This applies where the loss of tax was brought about due to the careless behaviour of the taxpayer; and
- 20 years from the end of the relevant tax year to which the assessment relates where the loss of tax was:
- Brought about by the deliberate behaviour of the taxpayer (or a person acting on their behalf);
- Attributable to a failure by a taxpayer to notify HMRC of a chargeable income / gain; or
- Attributable to arrangements where the taxpayer failed to comply with DOTAS / POTAS notification obligations.
Facts of the case
Mr Tooth filed a self-assessment tax return for the 2007/08 tax year and had taken part in the Romangate Scheme (considered to be a tax avoidance arrangement) and this scheme was utilised to generate an employment-related loss which he could use against his income for both the 2007/08 and 2008/09 tax years.
Mr Tooth’s tax advisors prepared the return on his behalf and used software approved by HMRC. However, the software used did not allow for the tax advisors to enter the employment-related loss from the scheme in the additional information page of the return and instead, they included the loss in another part of the return and explained the reasons for this in the ‘white space’ of the return.
After legislation was passed to retrospectively make the Romangate Scheme ineffective, HMRC opened a discovery assessment in October 2014 claiming that they had discovered an insufficiency with Mr Tooth’s tax return and Mr Tooth deliberately filed an inaccuracy which caused a loss of tax. Mr Tooth disagreed with this and appealed against this discovery assessment to the First-Tier Tribunal (“FTT”) putting HMRC to proof that there had been the discovery and denied that there were any inaccuracies in the return, deliberate or otherwise.
The FTT agreed with HMRC that the discovery had been made by one of its officers in October 2014 and rejected Mr Tooth’s submission that it was appropriate to have regard to the “collective knowledge” of HMRC which included knowledge other HMRC officers had when Mr Tooth’s return was received in 2009 and first considered. However, the FTT agreed with Mr Tooth that there had been no deliberate inaccuracy and allowed his appeal.
HMRC appealed to the Upper Tribunal (“UT”) who ruled that there had been no discovery in 2014 due to HMRC forming its own view about the insufficiency of the return in 2009 and:
“…that a discovery in 2009, even if it had been pleaded, would have become “stale” by 2014…”. 
The UT were therefore highlighting, at the time, that the discovery assessment would need to issued whilst the discovery was new or the discovery would become ‘stale’. The UT also reversed the original decision of the FTT on the point of the timing of when a discovery can be made and said that if two HMRC officers independently make the same discovery at different times, only the first discovery would qualify as a discovery.
The UT also held that read as a whole document, there were no inaccuracies in Mr Tooth’s return and that if they were wrong, there would have been no deliberate inaccuracy. HMRC then appealed to the Court of Appeal, who ruled by majority, that there was a deliberate inaccuracy in the return but agreed with the UT that there was no qualifying discovery.
Supreme Court Decision
HMRC decided to appeal further to The Supreme Court to restore the FTT decision that there was a discovery and the Court of Appeal ruling that there were deliberate inaccuracies in the return.
The Supreme Court unanimously dismissed HMRC’s appeal, ruling that there was no deliberate inaccuracy in the return, nor had they intended to mislead HMRC.
The Supreme Court then dealt with the issue of the ‘discovery’. The Supreme Court stated that there is no principle that HMRC have ‘collective knowledge’ and the focus for S.29(1) TMA 1970 is the state of mind of the individual HMRC officer. Therefore, an HMRC officer can discover something that another HMRC officer has already discovered.
The Supreme Court also stated that the discovery condition under S.29(5) TMA 1970 operates by reference to the state of mind of the particular hypothetical officer of HMRC dealing with the taxpayer’s case at a particular time and does not involve collective knowledge of HMRC.
The Supreme Court importantly stated that there is no concept of ‘staleness’ – whereby a taxpayer would argue that a discovery might lose its quality over time and become ‘stale’. The Supreme Court stated that this is contrary to the ordinary use of language and would conflict with the statutory scheme.
Although the Supreme Court ruled against HMRC on the deliberate inaccuracy, the Supreme Court comments on the lack of a ‘staleness defence’ for a taxpayer will assist HMRC in future litigation against those in similar discovery assessments who try and argue that the passage of time will cause the discovery to lose its quality.