Tax writer
Share this content

Stale discoveries lead to failed assessments

Two cases, concerning income tax and capital gains, illustrate how delays and foul-ups by HMRC can prevent it from collecting the tax that is otherwise due. 

4th Mar 2020
Tax writer
Share this content

Where there is an unreasonable delay between HMRC’s discovery that insufficient tax has been charged and issuing an assessment, the discovery becomes “stale” and the assessment becomes invalid.

As staleness is nearly always connected with HMRC’s inefficiency, it is not happy that staleness can void an assessment. Sadly for HMRC the Court of Appeal has agreed that this is the law in the case of Raymond Tooth (2019) EWCA Civ 826, and Clive Beagles [2018] UT. HMRC has sought permission to appeal Tooth to the Supreme Court, but that permission has not yet been granted.

Against that background, cases concerning stale discoveries continue to flow through the first tier tribunal (FTT), and continue to be decided in the taxpayers’ favour.

Bashir Jafari

This case (TC07465) involved an administrative foul-up by HMRC, which only came to light when the judge sought clarification of a throwaway paragraph in HMRC’s skeleton argument at the FTT.

This is what happened to Jafari’s 2009/10 return:

  • 16 July 2013: return submitted but rejected by HMRC
  • 26 July 2013: HMRC opened an enquiry into the return it rejected
  • 28 October 2013: tax return sent back by HMRC as unsatisfactory
  • 24 February 2016: enquiry closed, but closure appealed by taxpayer
  • 15 August 2018: discovery assessment issued, and treated as appealed on day of issue.

By 15 August 2018 someone within HMRC had realised that, in the absence of an actual return (which was rejected), it couldn’t open or close an enquiry into that return, so it issued an assessment to make good the error.

Judge James Austen found that HMRC was first aware of a tax shortfall no later than February 2016. There had been no new discovery about the tax in 2018: the only thing HMRC “discovered” at that stage was its own error.

The discovery was stale. At least 30 months had elapsed (probably more) before the assessment was issued. The judge cancelled the assessment and urged HMRC to cancel the associated penalties.

The judge went on to criticise HMRC’s lack of candour in preparing this case. Rather than address the issue of staleness directly, it had sought to “gloss over decisions which HMRC knows but dislikes and to proceed as if they do not exist”.

Given that there were appeals pending to the higher courts which might change the law, HMRC’s only acceptable stances should have been either:

  1. To make an application to stay proceedings before FTT pending the outcome of the (Tooth and Beagles) appeals; or
  2. Candidly to acknowledge the position, accept the inevitable adverse decision, and apply for permission to appeal.

HMRC’s stance in this case “failed to meet its obligation to ‘help the Tribunal to further the overriding objective’ of dealing with cases fairly and justly”.

Oriel Developments Ltd

The tax dispute at the centre of this case (TC07306) was a claim for chargeable gains roll-over relief on the proceeds of compulsory purchase. The FTT ruled the claim had failed as the proceeds were not reinvested in acquiring other land or any other qualifying asset.

Nonetheless Judge Tony Beare concluded that the assessment to claw back the relief was invalid due to staleness.

This is what happened:

  • 27 November 2009: company sells land in a compulsory purchase, and includes a provisional rollover relief declaration in its 2010 CT return
  • 16 April 2012: company reinvests proceeds, and includes final claim to rollover relief in its 2012 CT return
  • 15 May 2014: HMRC enquires into the 2012 return, challenging the validity of the reinvestment
  • 31 August 2014: deadline for making a valid reinvestment
  • 18 September 2015: HMRC closes enquiry into 2012 CT return, and advises that it will raise an assessment for 2010
  • 6 October 2015: HMRC amends the 2010 CT return.

HMRC’s statutory review of the appeal against this amendment was delayed pending the upper tribunal’s decision on Benham (2017) UKUT 389.

The judgment in Benham was that HMRC has no free-standing power to amend prior year returns in rollover relief cases. If a self assessment remains under enquiry, it can be amended as usual, but in all other cases HMRC must issue an assessment. The Benham judgment was issued in October 2017, and HMRC issued an assessment in May 2018.


The HMRC case file had passed through a number of officers’ hands. In 2015 the 2010 self assessment was amended by Ms Rodgers, while in 2018 the discovery assessment was issued by Dr Coffey.

It was clear that the “discovery” made by Coffey was identical to that made 30 months earlier by Rodgers. Judge Beare felt that it was the earlier discovery which ought to be considered, raising the possibility of staleness.

HMRC “reserved the right to argue, should this appeal proceed further, that it is not possible for a discovery assessment to be held to be invalid on staleness grounds”. (This, by contrast with Jafari, is exactly how to do it!)

HMRC did try to suggest that it could not have made a protective assessment any earlier, but the judge rightly rejected that suggestion.


HMRC had made a discovery in 2015 (or even earlier), but had failed to make an assessment until 2018. The discovery had gone stale so that – even though the FTT agreed with HMRC that the rollover relief was not due, the assessment to claw it back was invalid.

HMRC does not like the concept of “staleness”. However, unless it manages to get Tooth and Beagles reversed by the higher courts, it is stuck with it.

One possible solution might be to improve its systems such that HMRC officers are required to assess promptly upon making a discovery.

Replies (2)

Please login or register to join the discussion.

By Justin Bryant
04th Mar 2020 10:06

Or, like the bad losers they are, they could simply amend the legislation (retrospectively).

Thanks (0)
By mdoodney
06th Mar 2020 13:08

I had a discussion with an HMRC reviewing officer last week who confirmed that the HMRC policy remains to ignore staleness on the grounds that it doesn't exist despite the fact that the CoA says it does and this approach is beginning to attract the ire of the FTT, so I don't expect this approach to change any time soon.

Thanks (0)