ON APPEAL FROM THE UPPER TRIBUNAL (TAX AND CHANCERY CHAMBER)
MR JUSTICE NEWEY
[2013] UKUT 0623 (TCC)
Royal Courts of Justice
Strand, London, WC2A 2LL
Before :
LORD JUTICE PATTEN
LORD JUSTICE BRIGGS
and
Lord justice simon
Between :
MR DAVID STEPHEN SANDERSON | Appellant |
- and - | |
THE COMMISSIONERS FOR HER MAJESTY’S REVENUE AND CUSTOMS | Respondent |
Mr Keith Gordon and Miss Ximena Montes Manzano (instructed by Bramhall Solicitors) for the Appellant
Mr David Yates (instructed by the General Counsel and Solicitor to HM Revenue and Customs) for the Respondents
Hearing date : 15 December 2015
Judgment
Lord Justice Patten :
This is an appeal by the taxpayer, Mr David Sanderson, against a decision of the Upper Tribunal (Tax and Chancery Chamber) (Newey J) released on 6 December 2013 dismissing Mr Sanderson’s appeal from the earlier decision of the First-tier Tribunal (Tax Chamber) which upheld a discovery assessment made pursuant to s.29 of the Taxes Management Act 1970 (“TMA”). The appeal turns on whether the second condition imposed by s.29(5) TMA for the exercise of the power to issue a discovery assessment has been satisfied in this case.
The relevant facts can be stated quite shortly. On 24 February 2003 Mr Sanderson submitted late his tax return for the year ended 5 April 1999. The return disclosed a chargeable gain of almost £1.8m against which were set losses of more than £2m. Details of the losses were contained in the white space section of the return in which Mr Sanderson inserted a specific form of wording settled by leading tax counsel that had been supplied to him by Hanover Veriti Limited, the promoter of the tax scheme used by Mr Sanderson to reduce his liabilities in respect of the chargeable gain.
The disclosure was in these terms:
“EUROPEAN AVERAGE RATE OPTION (TRADE NO. 82831)
I am entitled to the loss of £1,825,663 by virtue of the provisions of TCGA 1992 s.71(2). The loss is part of a loss of £1,000,000,000, which accrued to the Trustees of the Castle Trust on 8th April 1997, on the disposal of a European Average rate Option (Trade No. 82831) relating to shares in Deutsche Telecom.
BENEFICIAL INTEREST IN THE CASTLE TRUST
On 24th November 1998, I purchased for a fee (part of which is contingently payable) from the Trustees of the Charter Trust 2.273% of their beneficial interest in the Trust Fund of the Castle Trust. The interest determined on 25th November 1998, when I became absolutely entitled to receive from the Trustees of the Castle Trust the sum of £16.04.”
The Castle Trust Scheme (“the Scheme”) was set up in 1997 and has been considered by a Special Commissioner (Mr Charles Hellier) in Corbally-Stourton v HMRC [2008] STC (SCD) 907. It is convenient (as in the Upper Tribunal decision) to summarise the Scheme by reference to the findings made in that case by the Special Commissioner:
“11. I find that scheme was intended to operate in the following manner:
“(1) On 11 March 1997 Mr Tanreer Makhdumi executed a deed under Guernsey law settling £125,000 on Legis Trust of Guernsey as trustee of the Castle Trust. The principal beneficiary was the settlor’s mother who was resident in Pakistan.
(2) It was expected by those involved in the promulgation of the scheme that, through the agency of Exco Bierbaum Securities GmbH, (a derivatives broker and member of the Frankfurt Stock Exchange) … the trustees would enter into two reciprocal derivative contracts. Under the first contract the trustees were to become obliged to make a set payment to the counterparty (PDR) if the average price of Deutsche Telecom Shares over the set life of the contract exceeded a set figure, and if the average was lower than that figure then the counterparty would make payment of the same sum to the trustees. Under the second contract the obligations to pay were the reverse.
(3) The terms of the derivative contracts expressed that both would expire on 8 April 1997 when settlement would be made. The set payment was £1 billion.
(4) On 4 April 1997 the trustees, through the agency of Exco, arranged to terminate the option which was then in the money, and in consequence £999,288,500 was to be paid by PDR to an account of the trustees with UBS. Because the trustees retained the other, out of the money, contract under which they had a contingent liability of £1 billion which would mature on 8 April 1997, the UBS bank account was assigned by way of security to PDR.
(5) On 7 April UK resident trustees were appointed in place of the Guernsey resident trustees.
(6) On 8 April 1997 the out of the money derivative matured and the trustees were to pay PDR £1 billion of which the vast majority would come from the £999,288,500 which was to have been paid to them on 4 April 1997.
12. The object of these transactions was to give rise to an allowable loss of £1 billion in the hands of the trustees when they were UK resident, but for the gain of £999,288,500 to fall outside the UK capital gains net—being realised by non-UK resident trustees for the benefit of non-UK resident beneficiaries. The next steps involved the parcelling up of the allowable loss and the making of arrangements to enable it to accrue to UK taxpayers. These arrangements relied on the provisions of s 71(2) TCGA as they stood prior to their amendment in 1999. Under those provisions, where a person became absolutely entitled to trust property as against the trustees, any allowable loss which had accrued to the trustees which was represented in that property and could not be used by the trustee in the year in which the person became absolutely entitled was to be treated as accruing to the person who so became entitled. Thus if a taxpayer acquired an absolute interest in part of the trust property he would become entitled to part of the allowable loss which would otherwise have accrued to the benefit of the trustees.
13. This parcelling up and allocation was to take place by the following steps:
(i) later in 1997 three new trusts, the Charter Trust, the Magnus Trust, and the Zennith Trust were created;
(ii) the trustees of the Castle Trust made appointments of parts of the Castle Trust property to each of these new trusts.
The appointments were made contingent upon Mr Makhdumi's mother surviving until noon on 25 November 1998;
(iii) the trustees of the three new trusts sold shares of their contingent interests in the Castle Trust to UK taxpayers;
(iv) on 25 November 1998, Mrs Makhdumi being still alive, the UK taxpayers became absolutely entitled as against the Castle trustees to parts of the Castle Trust property, and thus eligible under s 71(2) to inherit the unused allowable losses of the Castle Trust.”
One can see that the efficacy of the Scheme depended on the existence of two reciprocal derivative contracts which in effect guaranteed the existence of a net liability of £1 billion as at the expiry date in April 1997 but which facilitated a change from offshore to UK resident trustees between the termination of the in the money option and the maturity of the out of the money option. The net loss occurring to the UK trustees could then be allocated between the participants in the Scheme.
The Scheme was the subject of an investigation by HMRC’s Special Compliance Office (“SCO”) and Specialist Investigation Services from 1999 to 2007. This included a review of the tax returns of its members in which more than £200,000 had been claimed by way of capital losses. As of July 1999, the SCO had obtained from the Office of Supervision of Solicitors a list of the names and addresses of the participant members which had included Mr Sanderson and his file was forwarded to Mr Peter Thackeray, one of the officers in the SCO team.
At this stage Mr Sanderson had not filed his return and, as already mentioned, this did not occur until February 2003. Further checks on Mr Sanderson’s self-assessment records were carried out between June and September 2000 but no further checks were carried out until October 2004 when Mr Thackeray was able to obtain from Mr Sanderson’s accountants a copy of the return filed in 2003. Mr Thackeray accepted in evidence that had a search of Mr Sanderson’s records been carried out in 2003 an enquiry could and would have been made within the period allowed under s.9A TMA. It would also have been possible to make an assessment under s.9C TMA before April 2004 based on the information contained in the return. But by November 2004 the time limits for either a s.9A inquiry or a s.9C assessment had expired leaving a discovery assessment under s.29 as the only means of challenging the capital loss claimed in the return. The discovery assessment was issued on 11 December 2005.
It was common ground between the parties both before the First-tier Tribunal and the Upper Tribunal that the Scheme was not effective to reduce the amount of Mr Sanderson’s chargeable gain for the tax year 1998/1999. HMRC discovered that there was no record at Exco of the transaction in the derivatives between the trustees and PDR and on 27 November 2003, following negotiations between the trustees of the Scheme and HMRC, a closure notice was issued reducing the Trustees’ loss claim from £1bn to nil. Hanover Veriti Limited wrote to Mr Sanderson on 7 January 2004 that:
“As you are aware, the Inland Revenue challenged the Castle Trust losses on the basis firstly that the transaction leading to the loss was in law, a sham and, secondly, that it lacked a commercial purpose. The Castle Trustee took advice from Leading Tax Counsel and he expressed the view that there was insufficient evidence and witnesses to show that the payments underlying the transaction were actually effected. He was, therefore, unable to advise the Trustee to continue with its challenge of the Inland Revenue. The Trustee (and the steering committee) has reluctantly accepted that advice.”
The power to issue a discovery assessment under s.29 TMA exists as part of the legislative framework governing self-assessment. This was reviewed by Moses LJ in his judgment in Tower MCashback LLP v Revenue & Customs Commissioners [2010] STC 809 where the point is made (at [24]) that the conditions for the making of a discovery assessment were tightened on the introduction of self-assessment:
“As I have already observed, apart from a closure notice, and the power to correct obvious errors or omissions, the only other method by which the Revenue can impose additional tax liabilities or recover excessive reliefs is under the new s 29. That confers a far more restricted power than that contained in the previous s 29. The power to make an assessment if an inspector discovers that tax which ought to have been assessed has not been assessed or an assessment to tax is insufficient or relief is excessive is now subject to the limitations contained in s 29(2) and (3) (s 29(1)). Section 29(2) prevents the Revenue making an assessment to remedy an error or mistake if the taxpayer has submitted a return in accordance with s 8 or s 8A and the error or mistake is in accordance with the practice generally prevailing when that return was made. Section 29(3) prevents the Revenue making a discovery assessment under s 29(1) unless at least one of two conditions is satisfied (s 29(3)). The prohibition applies unless the undercharge or excessive relief is attributable to fraudulent or negligent conduct (s 29(4)) or having regard to the information made available to him the inspector could not have been reasonably expected to be aware that the taxpayer was being undercharged or given excessive relief (s 29(5)). There are statutory limitations as to the time at which the sufficiency or otherwise of the information must be judged. These provisions underline the finality of the self-assessment, a finality which is underlined by strict statutory control of the circumstances in which the Revenue may impose additional tax liabilities by way of amendment to the taxpayer's return and assessment.”
Section 29 TMA is designed to deal with inaccuracies in the process of self-assessment. The taxpayer (in the case of an individual) is required by s.8 TMA to make and file a return containing the information which is reasonably required in order to establish the amounts of income and capital gains tax in which he is chargeable and, for that purpose, to deliver with the return such accounts and other documents relating to the information as may be reasonably required. The return must include a self-assessment of the amounts in respect of which the taxpayer is chargeable on the basis of the information provided and taking into account any reliefs claimed: s.9(1). It must also include a declaration that the return is, to the best of the taxpayer’s knowledge, correct and complete: see s.8(2).
The taxpayer’s obligation is therefore to provide a correct assessment of his tax liabilities and to support that assessment with such information as may be necessary to substantiate the figures. The Revenue has power under s.9ZB to amend a return in order to correct obvious errors of principle and calculation. There is also an unlimited power under s.9A to enquire into a s.8 return within the time limits specified in s.9A(2). In the present case, this was the quarter day next after the first anniversary of the delivery of the return. An enquiry extends to:
“anything contained in the return, or required to be contained in the return, including any claim or election included in the return”: see s.9A(4).
Section 9C TMA gives an officer power to amend the self-assessment return during an enquiry in order to prevent the loss of tax but where, as in this case, no enquiry was commenced within the s.9A(2) time limit or an enquiry was closed then the Revenue’s only power to amend the return is by way of discovery assessment under s.29.
So far as material, s.29 (at the time in question) provided:
“(1) If an officer of the Board or the Board discover, as regards any person (the taxpayer) and a year of assessment—
(a) that any income which ought to have been assessed to income tax, or chargeable gains which ought to have been assessed to capital gains tax, have not been assessed, or
(b) that an assessment to tax is or has become insufficient, or
(c) that any relief which has been given is or has become excessive,
the officer or, as the case may be, the Board may, subject to subsections (2) and (3) below, make an assessment in the amount, or the further amount, which ought in his or their opinion to be charged in order to make good to the Crown the loss of tax.
(2) Where—
(a) the taxpayer has made and delivered a return under section 8 or 8A of this Act in respect of the relevant year of assessment, and
(b) the situation mentioned in subsection (1) above is attributable to an error or mistake in the return as to the basis on which his liability ought to have been computed,
the taxpayer shall not be assessed under that subsection in respect of the year of assessment there mentioned if the return was in fact made on the basis or in accordance with the practice generally prevailing at the time when it was made.
(3) Where the taxpayer has made and delivered a return under section 8 or 8A of this Act in respect of the relevant year of assessment, he shall not be assessed under subsection (1) above—
(a) in respect of the year of assessment mentioned in that subsection; and
(b) in the same capacity as that in which he made and delivered the return,
unless one of the two conditions mentioned below is fulfilled.
(4) The first condition is that the situation mentioned in subsection (1) above is attributable to fraudulent or negligent conduct on the part of the taxpayer or a person acting on his behalf.
(5) The second condition is that at the time when an officer of the Board—
(a) ceased to be entitled to give notice of his intention to enquire into the taxpayer's return under section 8 or 8A of this Act in respect of the relevant year of assessment; or
(b) informed the taxpayer that he had completed his enquiries into that return,
the officer could not have been reasonably expected, on the basis of the information made available to him before that time, to be aware of the situation mentioned in subsection (1) above.
(6) For the purposes of subsection (5) above, information is made available to an officer of the Board if—
(a) it is contained in the taxpayer's return under section 8 or 8A of this Act in respect of the relevant year of assessment (the return), or in any accounts, statements or documents accompanying the return;
(b) it is contained in any claim made as regards the relevant year of assessment by the taxpayer acting in the same capacity as that in which he made the return, or in any accounts, statements or documents accompanying any such claim;
(c) it is contained in any documents, accounts or particulars which, for the purposes of any enquires into the return or any such claim by an officer of the Board, are produced or furnished by the taxpayer to the officer, whether in pursuance of a notice under section 19A of this Act or otherwise; or
(d) it is information the existence of which, and the relevance of which as regards the situation mentioned in subsection (1) above—
(i) could reasonably be expected to be inferred by an officer of the Board from information falling within paragraphs (a) to (c) above; or
(ii) are notified in writing by the taxpayer to an officer of the Board.”
Mr Sanderson accepts that, in the light of what is now known about the Scheme, he cannot rely upon the losses it purported to produce in reduction of his chargeable gain and that, to this extent, the self-assessment contained in his 2003 return was insufficient for the purposes of s.29(1). There was an issue as to whether the disputed assessment was made after the Revenue had made a discovery within the meaning of the sub-section. The taxpayer’s argument was that his participation in the Scheme was well known to the Revenue long before he even submitted his tax return and that by that time they were already of the view that the Scheme was not effective. It could not therefore be said that Mr Thackeray “discovered” an insufficiency in 2004 which caused him to issue the assessment.
This argument failed before the Upper Tribunal having regard to what was said about the meaning of “discover” by the Upper Tribunal in Charlton v HMRC [2013] STC 866 and the point is not pursued on this appeal. But Mr Gordon (for Mr Sanderson) contends that “discover” and “be aware” in s.29 have a parity of meaning and I shall return to that argument a little later.
Nor is there any longer an issue about the first condition (fraudulent and negligent conduct on the part of the taxpayer). HMRC’s argument that there was negligent conduct on the part of Mr Sanderson’s advisers in not seeking to amend his 1998-1999 return once it was realised that the loss claim could not be maintained failed before the First-tier Tribunal and the Upper Tribunal and is not part of this appeal. The sole issue is whether the second s.29(5) condition was satisfied.
The power of HMRC to make an assessment under s.29(1) following the discovery of what, for convenience, I shall refer to as an insufficiency in the self-assessment depends upon whether an officer “could not have been reasonably expected, on the basis of the information made available to him before that time, to be aware of the insufficiency”. It is clear as a matter of authority:
that the officer is not the actual officer who made the assessment (for example Mr Thackeray in this case) but a hypothetical officer;
that the officer has the characteristics of an officer of general competence, knowledge or skill which include a reasonable knowledge and understanding of the law: see HMRC v Lansdowne Partners LLP [2012] STC 544;
that where the law is complex even adequate disclosure by the taxpayer may not make it reasonable for the officer to have discovered the insufficiency on the basis of the information disclosed at the time: see Lansdowne at [69];
that what the hypothetical officer must have been reasonably expected to be aware of is an actual insufficiency: see Langham v Veltema [2004] STC 544 per Auld LJ at [33]-[34]:
“33. More particularly, it is plain from the wording of the statutory test in section 29(5) that it is concerned, not with what an Inspector could reasonably have been expected to do, but with what he could have been reasonably expected to be aware of. It speaks of an Inspector's objective awareness, from the information made available to him by the taxpayer, of "the situation" mentioned in section 29(1), namely an actual insufficiency in the assessment, not an objective awareness that he should do something to check whether there is such an insufficiency, as suggested by Park J. If he is uneasy about the sufficiency of the assessment, he can exercise his power of enquiry under section 9A and is given plenty of time in which to complete it before the discovery provisions of section 29 take effect.
34. In my view, that plain construction of the provision is not overcome by Mr. Sherry's argument that it is implicit in the words in section 29(5) "on the basis of the information made available to him" (my emphasis) and also in the provision in section 29(6)(d) for information, the existence and relevance of which could reasonably be inferred from information falling within section 29(6) (a) to (c), that the information itself may fall short of information as to actual insufficiency. Such provision for awareness of insufficiency "on the basis" of the specified information or from information that could reasonably be expected to be inferred therefrom does not, in my view, denote an objective awareness of something less than insufficiency. It is a mark of the way in which the subsection provides an objective test of awareness of insufficiency, expressed as a negative condition in the form that an officer "could not have been reasonably expected … to be aware of the" insufficiency. It also allows, as section 29(6) expressly does, for constructive awareness of insufficiency, that is, for something less than an awareness of an insufficiency, in the form of an inference of insufficiency.”
(5) that the assessment of whether the officer could reasonably have been expected to be aware of the insufficiency falls to be determined on the basis of the types of available information specified in s.29(6). These are the only sources of information to be taken into account for that purpose: see Langham v Veltema at [36]:
“The answer to the second issue– as to the source of the information for the purpose of section 29(5) - though distinct from, may throw some light on, the answer to the first issue. It seems to me that the key to the scheme is that the Inspector is to be shut out from making a discovery assessment under the section only when the taxpayer or his representatives, in making an honest and accurate return or in responding to a section 9A enquiry, have clearly alerted him to the insufficiency of the assessment, not where the Inspector may have some other information, not normally part of his checks, that may put the sufficiency of the assessment in question. If that other information when seen by the Inspector does cause him to question the assessment, he has the option of making a section 9A enquiry before the discovery provisions of section 29(5) come into play. That scheme is clearly supported by the express identification in section 29(6) only of categories of information emanating from the taxpayer. It does not help, it seems to me, to consider how else the draftsman might have dealt with the matter. It is true, as Mr. Sherry suggested, he might have expressed the relevant passage in section 29(5) as "on the basis only of information made available to him", and the passage in section 29(6) as "For the purposes of subsection (5) above, information is made available to an officer of the Board if, but only if," it fell within the specified categories. However, if he had intended that the categories of information specified in section 29(6) should not be an exhaustive list, he could have expressed its opening words in an inclusive form, for example, "For the purposes of subsection (5) above, information … made available to an officer of the Board … includes any of the following".”
Where there is more scope for argument is in relation to the level of awareness that the relevant information needs to create in order for the condition to bar the right to raise a s.29(1) assessment. In the present context, for example, is it necessary for the information disclosed to lead the notional officer to conclude on the balance of probabilities that there is an insufficiency or must he be satisfied beyond reasonable doubt? Alternatively is some quite different test to be applied? The balance of probabilities test had found support in Corbally-Stourton and has been adopted in the Scottish case of R (on the application of Pattullo) v Revenue and Customs Commissioners [2010] STC 107.
But in Lansdowne at first instance Lewison J (at [48]) preferred to take a different and more general approach:
“Mr Coleman said that this was the wrong test. HMRC had to know with reasonable certainty of the insufficiency in question otherwise the office could not have been 'aware' of it. There is, no doubt, an epistemological debate to be had about whether you can discover or be aware of something that does not in fact exist. In the present case, for example, the commissioners decided that there was no insufficiency. Had HMRC discovered or been aware of an insufficiency before their decision that there was in fact no insufficiency? Or had they been aware of it, but then ceased to be aware of it? And now that I have disagreed with the commissioners on one of the points, are HMRC aware of it again? Or have they been aware of it throughout? But I do not consider that I need to enter into this debate. In the present case the commissioners asked whether HMRC had sufficient information to make a decision whether to raise an additional assessment. That seems to me to be the right test.”
A not dissimilar test was applied in the Court of Appeal. The Chancellor said (at [56]):
“I do not suggest that the hypothetical inspector is required to resolve points of law. Nor need he forecast and discount what the response of the taxpayer may be. It is enough that the information made available to him justifies the amendment to the tax return he then seeks to make. Any disputes of fact or law can then be resolved by the usual processes. For these reasons I would dismiss the appeal of HMRC.”
To the same effect, Moses LJ said (at [69]-[70]):
“… As the Chancellor points out (at [56]), awareness of an insufficiency does not require resolution of any potential dispute. After all, once an amendment is made, it may turn out after complex debate in a succession of appeals as to the facts or law, that the profits stated were not insufficient. I have dwelt on this point because I wish to leave open the possibility that, even where the taxpayer has disclosed enough factual information, there may be circumstances in which an officer could not reasonably be expected to be aware of an insufficiency by reason of the complexity of the relevant law.
[70]I also wish to express polite disapproval of any judicial paraphrase of the wording of the condition at s 30B(6) or s 29(5). I think there is a danger in substituting wording appropriate to standards of proof for the statutory condition. The statutory condition turns on the situation of which the officer could reasonably have been expected to be aware. Awareness is a matter of perception and of understanding, not of conclusion. I wish, therefore, to express doubt as to the approach of the Special Commissioner in Corbally-Stourton v Revenue and Customs Comrs [2008] STC (SCD) 907 and of the Outer House in R (on the application of Pattullo) v Revenue and Customs Comrs[2009] CSOH 137, [2010] STC 107, namely that to be aware of a situation is the same as concluding that it is more probable than not. The statutory context of the condition is the grant of a power to raise an assessment. In that context, the question is whether the taxpayer has provided sufficient information to an officer, with such understanding as he might reasonably be expected to have, to justify the exercise of the power to raise the assessment to make good the insufficiency.”
It is important to emphasise that the decision in Lansdowne did not involve any qualification of what Auld LJ in Langham v Veltema identified as the question posed by the second s.29(5) condition. The hypothetical officer must, on an objective analysis, be made aware of an actual insufficiency in the assessment by the matters disclosed in the s.29(6) information. This is made clear by the Chancellor at [55] of his judgment in Lansdowne. The sole dispute in that case was whether the disclosures made by the taxpayer’s accountants were sufficient to cause the hypothetical officer to conclude that there was an insufficiency.
The passages in the judgments of the Chancellor and Moses LJ as to the level of the officer’s awareness were directed to the Revenue’s argument that the disclosures made required inferences to be drawn about the accuracy of the self-assessment based on certain legal assumptions and that the officer could not be expected to resolve issues of law in determining the impact of the information supplied. In the face of such uncertainties, the officer could not be taken to be “aware” of an insufficiency. The decision in Lansdowne confirmed that the officer was not required to resolve (or even be able to assess) every question of law (particularly in complex cases) but that where, as Moses LJ expressed it, the points were not complex or difficult he was required to apply his knowledge of the law to the facts disclosed and to form a view as to whether an insufficiency existed. That is a matter of judgment rather than the application of any particular standard of proof. And the reference to the officer needing to reach a conclusion which justified the making of a discovery assessment has to be read in that context.
Mr Sanderson’s case is that the Upper Tribunal over-stated the level of knowledge which needs to be imputed to the officer under s.29(5) in order to justify the making of a discovery assessment. The threshold is said to be a relatively low one and merely requires the officer to be able to justify his belief that further tax is due. Part of Mr Gordon’s argument rests on eliding the requirement in s.29(1) for an officer to “discover” that there is an insufficiency in the return with the condition in s.29(5) that the notional officer could not have been reasonably expected, on the information available, to be “aware” of that insufficiency. Unless, it is said, the threshold of knowledge is set relatively low it would be difficult, if not impossible, in most cases for the Revenue to be able to raise an assessment under s.29(1).
I do not accept that ss.29(1) and (5) import the same test and that the Revenue’s power to raise an assessment is therefore directly dependent on the level of awareness which the notional officer would have based on the s.29(6) information. The exercise of the s.29(1) power is made by a real officer who is required to come to a conclusion about a possible insufficiency based on all the available information at the time when the discovery assessment is made. Section 29(5) operates to place a restriction on the exercise of that power by reference to a hypothetical officer who is required to carry out an evaluation of the adequacy of the return at a fixed and different point in time on the basis of a fixed and limited class of information. The purpose of the condition is to test the adequacy of the taxpayer’s disclosure, not to prescribe the circumstances which would justify the real officer in exercising the s.29(1) power. Although there will inevitably be points of contact between the real and the hypothetical exercises which ss.29(1) and (5) involve, the tests are not the same.
I turn then to the operation of the s.29(5) condition in this case. Because the scope of the available s.29(6) information is not agreed, the taxpayer’s arguments before the Upper Tribunal were staged in three parts, each of which assumed a different level of information and knowledge on the part of the hypothetical officer, and Mr Gordon helpfully adhered to that formula in presenting the appeal.
Scenario (1): the officer has only the tax return
In essence, Mr Gordon’s submission on this scenario is that the disclosure in the tax return was sufficient to engage the notional officer’s knowledge of the Ramsay principle and to lead him to conclude that the Scheme would be open to challenge on that basis. The disclosures made by the return revealed (1) the disparity between the size of the loss claimed (£1,825,663) and the income derived from the Castle Trust (£16.04); (2) the fact that the loss claimed was comparable in amount to (and therefore cancelled out) the taxable gain; (3) that the loss could be seen to be derived from an asset which Mr Sanderson had held for only a day; (4) that the loss was part of a very large round sum (£1bn); and (5) that the £1bn loss was attributable to the disposal of a derivative.
These factors alone are said to have been sufficient to reveal to the notional officer that these were artificial losses generated by a Scheme to which the Ramsay principle would apply. It was of no consequence, Mr Gordon submitted, that the return failed to disclose the existence of the counter-option. The factors identified above were sufficient to cause the officer to conclude that the claimed losses were unlikely to survive scrutiny.
It is not, I think, necessary for the purposes of this appeal to attempt to set out a comprehensive history of the development of the Ramsay principle and its various formulations in relation to particular transactions. It is now well established that Ramsay is not the broad spectrum antibiotic designed to kill off all tax avoidance schemes that Lord Hoffmann memorably refers to in MacNiven v Westmoreland Investments Ltd [2003] 1 AC 311 at [49]. Schemes which incorporate transactions that are self-cancelling and were instituted only to achieve a particular fiscal purpose are not ipso facto disregarded but fall to be considered in relation to the proper construction of the relevant tax legislation. The question is whether a loss produced in that way was contemplated as allowable by the capital gains tax regime in force at the time.
In this case the focus would have been on what constituted an allowable loss for the purposes of s.71(2) and s.16 of the Taxation of Chargeable Gains Act 1992. The argument for the Revenue would be that the trustees of the Castle Trust had not incurred any real loss because they had entered into two reciprocal derivative contracts. In Barclays Mercantile Business Finance Ltd v Mawson [2005] STC 1 Lord Nicholls of Birkenhead, referring to the speech of Lord Wilberforce in the original Ramsay decision, said:
“31. The application of these two principles led to the conclusion, as a matter of construction, that the statutory provision with which the court was concerned, namely that imposing capital gains tax on chargeable gains less allowable losses was referring to gains and losses having a commercial reality ("The capital gains tax was created to operate in the real world, not that of make-belief") and that therefore:
"To say that a loss (or gain) which appears to arise at one stage in an indivisible process, and which is intended to be and is cancelled out by a later stage, so that at the end of what was bought as, and planned as, a single continuous operation, there is not such a loss (or gain) as the legislation is dealing with, is in my opinion well and indeed essentially within the judicial function." (p. 326)
32. The essence of the new approach was to give the statutory provision a purposive construction in order to determine the nature of the transaction to which it was intended to apply and then to decide whether the actual transaction (which might involve considering the overall effect of a number of elements intended to operate together) answered to the statutory description. Of course this does not mean that the courts have to put their reasoning into the straitjacket of first construing the statute in the abstract and then looking at the facts. It might be more convenient to analyse the facts and then ask whether they satisfy the requirements of the statute. But however one approaches the matter, the question is always whether the relevant provision of statute, upon its true construction, applies to the facts as found. As Lord Nicholls of Birkenhead said in MacNiven v Westmoreland Investments Ltd[2003] 1 AC 311, 320, para 8:
"The paramount question always is one of interpretation of the particular statutory provision and its application to the facts of the case."”
This decision, however, postdates the expiry of the enquiry window (30 April 2004) and at that time the most recent guidance from the House of Lords on the Ramsay principle was the decision in MacNiven where Lord Nicholls had said:
“7. The Ramsay principle or, as I prefer to say, the Ramsay approach to ascertaining the legal nature of transactions and to interpreting taxing statutes, has been the subject of observations in several later decisions. These observations should be read in the context of the particular statutory provisions and sets of facts under consideration. In particular, they cannot be understood as laying down factual pre-requisites which must exist before the court may apply the purposive, Ramsay approach to the interpretation of a taxing statute. That would be to misunderstand the nature of the decision in Ramsay. Failure to recognise this can all too easily lead into error. In particular, the much-quoted observation of Lord Brightman in Furniss v Dawson[1984] AC 474, 527, seems to have suffered in this way. Lord Brightman described, as the 'limitations of the Ramsay principle', that there must be a pre-ordained series of transactions, or a single composite transaction, containing steps inserted which have no business purpose apart from the avoidance of a liability to tax. Where those two ingredients exist, the inserted steps are to be disregarded for fiscal purposes.
8. My Lords, I readily accept that the factual situation described by Lord Brightman is one where, typically, the Ramsay approach will be a valuable aid. In such a situation, when ascertaining the legal nature of the transaction and then relating this to the statute, application of the Ramsay approach may well have the effect stated by Lord Brightman. But, as I am sure Lord Brightman would be the first to acknowledge, the Ramsay approach is no more than a useful aid. This is not an area for absolutes. The paramount question always is one of interpretation of the particular statutory provision and its application to the facts of the case. Further, as I have sought to explain, Ramsay did not introduce a new legal principle. It would be wrong, therefore, to set bounds to the circumstances in which the Ramsay approach may be appropriate and helpful. The need to consider a document or transaction in its proper context, and the need to adopt a purposive approach when construing taxation legislation, are principles of general application. Where this leads depends upon the particular set of facts and the particular statute. I have already mentioned where this led in Ramsay. In Furniss v Dawson[1984] AC 474 it led to the conclusion that, within the meaning of the Finance Act 1965, the disposal of shares was in favour of Wood Bastow and not, as the taxpayer contended, in favour of Greenjacket.”
Even if the notional officer can be supposed to have read these passages in deciding whether to challenge Mr Sanderson’s self-assessment, he is likely to have concluded that a Ramsay-based challenge would require a careful analysis of all the component parts of the Scheme and that no one aspect of it was likely to be determinative. The tax return failed to disclose the simultaneous entry into the counter-option; the termination of the in the money contract on 4 April 1997 which created the gain that largely funded the liabilities on the out of the money contract and the change from Guernsey to UK trustees on 7 April 1997.
The Upper Tribunal took the view that without this information it would not have been possible for the officer to form the view that the Scheme as a whole lacked commercial reality and would not be treated as creating an allowable loss. Newey J said:
“50. I accept that submission. It seems to me that the tax return might have alerted the hypothetical officer to the fact that Mr Sanderson was seeking to take advantage of a tax scheme, but it did not contain enough information to make the officer aware of an “actual insufficiency” or to justify the making of an assessment. Mr Yates said that any assessment wouldhave been based on a mere whim. It would at any rate have been speculative. The mere fact that Mr Sanderson’s loss was attributable to a tax scheme would not have meant that it was open to challenge. The hypothetical officer would have been “aware that some tax schemes work and deliver the benefits claimed” (to use words of Mr Hellier in Corbally-Stourton, at [66]). The fact that it had been felt necessary to amend section 71 of the TCGA in 1999 might have led an officer to believe that the scheme Mr Sanderson used was one of those that (prior to the passing of the Finance Act 1999) worked.”
The reference to the amendment of s.71 is to an amendment which took effect in June 1999 and was introduced to block similar schemes for subsequent tax years. Mr Gordon said that the officer would have had knowledge of this change and this would have confirmed to him the artificiality of schemes of this kind. But the fact that future schemes were blocked by legislation is as likely to suggest, as the Upper Tribunal observed, that some at least of the schemes might have survived a challenge on Ramsay principles. Mr Sanderson is not, in my view, assisted by that point.
I think the Upper Tribunal was entitled to conclude, for the reasons it gave, that the information contained in the return was not enough to have made the notional officer aware of an insufficiency in the self-assessment. Mr Yates is right in his submission that this was not a simple case as presented in the return and that the non-disclosure of the self-cancelling nature of the transaction was not compensated for by the other factors that were disclosed. The fact that the information contained in the return might have been sufficient to cause the officer to ask further questions is not enough for the reasons already explained. I would add that as this part of Mr Sanderson’s case was raised for the first time in the Upper Tribunal, it was for the Upper Tribunal to make an assessment of whether the s.29(5) condition was fulfilled and this Court can only interfere with the conclusion if it discloses an error of law. The decision which the Upper Tribunal reached was fully open to it on the facts and discloses no error of principle in relation to the application of s.29(5).
Scenario (2): the officer also has knowledge of HMRC’s publicly stated views about the Castle Trust
The additional information attributed to the notional officer in this scenario consists of HMRC’s published views about the Scheme as at 30 April 2004. Mr Gordon referred us to correspondence between the SCO and the taxpayers involved in the Scheme in which Mr Thackeray expresses the view that the arrangements may amount to a sham because of the lack of evidence to support the carrying out of the transaction on which the Scheme was based. As we know, the Scheme eventually foundered for this reason quite apart from any application of the Ramsay principle.
Mr Gordon submitted that the knowledge and understanding of the notional officer must extend to include knowledge of HMRC’s then published thinking about the effectiveness of the Scheme. To label the views expressed by Mr Thackeray in private correspondence as published is somewhat to overstate the position. As Newey J observed in the Upper Tribunal, the evidence does not establish anything more than that the SCO had taken up a particular position about the effectiveness of the Scheme which in turn depended upon a correct assessment of the relevant facts and the law. There was no published determination about the effectiveness of the Scheme until the decision of the Special Commissioner in Corbally-Stourton in 2008. In these circumstances, I can see no proper legal basis for attributing to the notional officer what may have been the thinking at the time in the SCO. The exercise postulated by s.29(5) is a consideration by the officer of the information disclosed by the taxpayer by reference to the relevant legal principles: not by reference to what some particular department or officer at HMRC may at the time have thought about the efficacy of the Scheme then under investigation.
Scenario (3): the notional officer also has attributed to him the results of HMRC’s investigations into the Castle Trust
This part of Mr Gordon’s argument is based on s.29(6)(d)(i) TMA. He submits that the notional officer should have attributed to him the benefits of HMRC’s internal research into the Scheme. The officer, he says, could easily infer from the return that Mr Sanderson was not the only taxpayer with an interest in the Scheme and that the Revenue was likely to have documentation on the Castle Trust which would be relevant to the claim.
Some reliance is placed on the decision of the Upper Tribunal in Charlton in which it was held that the use of a particular reference number in the taxpayer’s return was sufficient to link the contents of the return to a tax scheme in respect of which the promoters of the scheme had supplied HMRC with a form AAG1. The disclosure contained in that form was treated under s.29(6)(d)(i) as part of the information made available to the officer for the purposes of s.29(5).
In the present case, however, we are not dealing under this head of the argument with information supplied by the promoters of the Scheme which is readily available through a link contained in the taxpayer’s own disclosure and can therefore be treated as part of the taxpayer’s own disclosure. The existence of the SRN on the return meant that a form AAG1 must have been lodged and, as Newey J observed, was bound to contain information about the scheme in question. What the appellant seeks to have attributed to the notional officer here is information obtained by a department within HMRC as a result of its own investigations into the Scheme. There is nothing in the contents of Mr Sanderson’s return to indicate that the Scheme has been under investigation or that some relevant information about a possible insufficiency is in the possession of HMRC.
I would endorse what the Upper Tribunal said in [78]-[79] of its decision in Charlton:
“78. The correct construction of s 29(6)(d)(i) is that it is not necessary that the hypothetical officer should be able to infer the information; an inference of the existence and relevance of the information is all that is necessary. However, the apparent breadth of the provision is cut down by the need, firstly, for any inference to be reasonably drawn; secondly that the inference of relevance has to be related to the insufficiency of tax, and cannot be a general inference of something that might, or might not, shed light upon the taxpayer’s affairs; and thirdly, the inference can be drawn only from the return etc provided by the taxpayer.
79. As we have described, the balance provided by s 29 depends on protection being provided only to those taxpayers who make honest, complete and timely disclosure. That balance would be upset by construing s 29(6)(d)(i) too widely. Inference is not a substitute for disclosure, and courts and tribunals will have regard to that fundamental purpose of s 29 when applying the test of reasonableness.”
In this case it would have been entirely speculative rather than a matter of inference from the return for the notional officer to have concluded that another branch of HMRC might have relevant information on the effectiveness of the Scheme. There is no basis on which the existence of such information could reasonably be expected to be inferred from the limited disclosure in the return.
I would therefore dismiss the appeal.
Lord Justice Briggs :
I agree.
Lord Justice Simon :
I also agree.
Crown copyright©