Sanderson v HMRC – Court of Appeal rules insufficient disclosure to prevent discovery assessment

Mon 29 Feb 2016

The case of Sanderson v HMRC considers whether HMRC were able to raise a discovery assessment under section 29 Taxes Management Act 1970 (all further legislative references are to this Act).

A discovery assessment can only be raised in limited circumstances, broadly, where the ‘normal’ enquiry window has passed, there is a loss of tax and one of the following conditions is met:

• a loss of tax was brought about by fraudulent or negligent conduct (now reworded as ‘careless or deliberate action’, broadly with effect from 1 April 2010) of the taxpayer – section 29(4); or

• by the time the ‘normal’ enquiry window closed, a ‘hypothetical officer’ could not have reasonably been expected to be aware of the loss of tax, based upon the information available to them – section 29(5).

Before the earlier Courts, it had been found that neither Mr Sanderson nor his advisors were negligent (and therefore within section 29(4)), and this point had not been appealed. As such, the case before the Court of Appeal focussed on the second of these two conditions, effectively, whether sufficient disclosure had been made in Mr Sanderson’s tax return.

The invalidity of the tax planning was not in question – Mr Sanderson accepted that this had failed and so his only possible defence was that he had made sufficient disclosure for an HMRC officer to be aware that the self-assessment was insufficient.

Background to the case
Mr Sanderson entered into a tax scheme in order to shelter a £1.8 million capital gain – the gain, reduced by the loss generated under the scheme, was reported within his 1998/99 tax return. The return also included two short paragraphs (drafted by counsel engaged by the scheme promoter) within the white space disclosing the nature of the loss and how it had arisen, although the information provided was fairly limited. The due date for filing this return was 31 January 2000, although it was not submitted until February 2003.

In January 2004, the scheme provider advised Mr Sanderson that agreement had been reached with HMRC that the scheme did not create the capital losses anticipated. However, it was not until 2008 that the Special Commissioners found in favour of HMRC in a similar case (Corbally- Stourton).

HMRC’s Special Compliance Office (SCO) were investigating the scheme from 1999, and by July of that year, had received details of all taxpayers who had used the scheme, including Mr Sanderson. Although the SCO made enquiries to Mr Sanderson’s tax office, as he had not submitted his return at that time, no action could be taken.

Because Mr Sanderson filed his return late the statutory enquiry window was extended to 30 April 2004. However, it was not until October 2004 that the SCO managed to obtain a copy of the return, and in December 2005 a discovery assessment was raised.

Mr Sanderson claimed that HMRC were precluded them from making a discovery assessment after 30 April 2004 because by then they had sufficient information to prompt an enquiry based on both the content of the return and the other information available to HMRC through the SCO enquiry.

Mr Sanderson’s appeals to the First-tier Tribunal and the Upper Tribunal on this point both failed.

Findings of the Court
The case turned on whether a hypothetical officer of HMRC, of general competence, knowledge and skill, could have been expected to have been aware of an insufficiency of tax, based on the information made available to him, at the end of the enquiry window (section 29(5)).

This analysis considered three different levels of information and knowledge available to the hypothetical officer:

• Contents of the tax return only
  The taxpayer’s counsel argued that the disclosure included within the white space was sufficient for a hypothetical officer to consider that the scheme was likely to be subject to a Ramsay-based challenge.  This argument was based on the disclosed disparity between the claim for a loss of £1.8 million compared to the level of income from the trust involved of only £16, the fact that the capital loss claimed cancelled out a gain of similar amount and that the loss arose from the disposal of a derivative held for only one day.  Therefore, all this information should have alerted the notional officer to the fact that these were artificial losses generated by a scheme.

The Court found that, even if the notional officer had read the disclosures listed above, to arrive at the decision that a Ramsay based challenge was appropriate would have required careful analysis of all of the component parts of the scheme, and certain elements of the scheme which would have been required to conclude on this were not in fact disclosed.

More importantly, although the information in the return may have been sufficient to cause a hypothetical officer to ask further questions, it was not sufficient for him to identify that there was an insufficiency in the self-assessment.

• Based on tax return and HMRC’s publicly stated view of the tax scheme
  The fact that the Special Compliance Office (SCO) of HMRC set out its view that the scheme might amount to a sham in the course of correspondence with the promoters did not amount to HMRC’s publicly stated view at that time. Indeed, the Special Commissioners’ decision that the scheme was ineffective was not published until 2008 with the decision in Corbally-Stourton.

Section 29(5) required consideration of the information disclosed by the taxpayer and did not extend to the view of another officer or department at that time as to the efficacy of the scheme.  Therefore, court found that the notional officer could not be attributed with the knowledge that the SCO had, so again the taxpayer’s argument failed.

• Based on tax return and HMRC’s knowledge from their investigation into the tax scheme
  The Court upheld the decision of the Upper Tribunal that knowledge of HMRC’s internal research should not be attributed to the hypothetical officer. There was nothing within Mr Sanderson’s return that made any reference to indicate that the scheme was under investigation by HMRC or other specific information, such as a DoTAS scheme reference number disclosed on a tax return (as relied on in the Charlton case).

The hypothetical officer could only speculate that there might have been a possible insufficiency in the self-assessment, rather than being able to infer it from the return. So again, the taxpayer’s argument failed.

What constitutes sufficient disclosure?
It is clear from this case, as with earlier cases, that the threshold for meeting the condition of full disclosure is very high – it is not sufficient to make brief comment about the planning, it requires sufficient disclosure for the hypothetical ‘ordinarily competent’ officer to be aware, not merely suspect, that there was a potential loss of tax.

What is sufficient in one case may not be sufficient in another, and therefore care is always needed when considering such matters.