Andrew Berry v The Commissioners for HM Revenue and Customs: [2011] UKUT 81 (TCC) [2011] UKUT 81 (TCC)

UPPER TRIBUNAL
TAX AND CHANCERY CHAMBER
[2011] UKUT 81 (TCC)Case No FTC/29/2010
Between
Andrew BerryAppellantThe Commissioners for HM Revenue and CustomsRespondent
Before
MR JUSTICE LEWISONMs Aparna Nathan and Ms Marika Lemos (instructed by RBC International Wealth Planning) for AppellantMr Malcolm Gammie QC (instructed by the General Counsel and Solicitor to HM Revenue and Customs) for Respondents - - - - - - - - - - - - - - - - - - - - -Date 25 February 2011Category: Tax
[1]Mr Berry appeals against the decision of the First Tier Tax Tribunal (the “FTT”) (Sir Stephen Oliver QC and Mr Ian Menzies-Conacher) that he was not entitled to relief against income tax for losses that he claims to have suffered as a result of “Gilt Strip Planning”. Ms Aparna Nathan and Ms Marika Lemos appear on his behalf. Mr Malcolm Gammie QC appears for HMRC.[2]As the FTT explained, a gilt strip is a UK Treasury Government stock issued by the Bank of England. “Strip” is an acronym for Separate Trading of Registered Interest and Principal flows which can be traded separately as zero-coupon gilts. Thus a three year gilt will have seven individual cash flows comprising six (semi-annual) coupon payments and a principal repayment. The strip market began in the UK on 6 December 1997; and broadly all new issues of conventional fixed coupon gilts are “strippable” (with a delay after the first issue date). The object of the Gilt Strip Planning is to create an income tax loss representing the difference between the amount paid by a person for the particular gilt strip and the amount payable when a transfer is made of that gilt strip. Mr Berry claims that the Gilt Strip Planning created a loss of £390,000. He says that the amount paid by him for the gilt strip in question was £6.5 million, while the amount payable on his onward transfer was £6.1 million (both figures have been rounded). His income tax loss was therefore £400,000. HMRC, the Respondents, say there was no such loss. The Scheme[3]The FTT described the scheme in detail (§§ 9-44). I need not repeat all that detail. I can summarise the scheme as follows.[4]The Gilt Strip Planning scheme was devised by Abacus Wealth Planning (“Abacus”) and marketed to potential customers. Mr Berry became aware of the Gilt Strip Planning idea at a presentation in the latter part of November 2003. The presentation had taken the form of a slide show supported by some explanatory material. The explanatory material stated that “the planning” had been devised by Abacus to be implemented by Abacus and a counter party bank in the Channel Islands. The material explained that:
“Under the planning, the taxpayer realises a capital gain that is exempt from capital gains tax and a loss that is allowable as a deduction for income tax purposes. The planning … involves the sale and purchase of gilt strips to and from a special purpose vehicle provided by the Bank.”
[5]The material went on to explain how the planning made use of a “principal strip” of a government gilt, being the right to receive the capital due on redemption of the gilt. It stressed that there must be uncertainty about the outcome:
“In order for the tax planning to be effective, there must be a real possibility that the market price of the gilt strip will, during the option/contract period, fall below a level at which the special purpose vehicle will wish to exercise its call option. … On the basis of legal precedents and the advice of leading Tax Counsel, we consider that the statistical probability of this happening must be at least 7%. ”
[6]It went on to say that in the event that the price did fall below “the water line” (i.e. the strike price under the call option):
“… the purchase contract will provide that the taxpayer may default from completion, and instead may pay the special purpose vehicle fixed compensation approximately equal to the option premium … . In this way the taxpayer’s economic position is broadly preserved, as is the special purpose vehicle.”
[7]The material then summarised the steps to be undertaken. These started with the customer identifying the amount of shelter he required. He was then to sign a letter engaging Abacus and pay Abacus a fee of 1% of the required shelter. He was to be provided with a “letter of suitability” as to the choice of a suitable gilt strip. (No evidence of such a letter was provided and Mr Berry could not recall ever having received one.) A meeting was to follow at which the customer would complete a call option agreement and a purchase contract and at which the bank would provide him with a facility letter. The option premium was to be deposited in an account that the customer had opened with the bank. The customer was told – “At the end of the option/contract period, your purchase of the gilt strip, and its sale under the option agreement, will occur in quick succession (it is anticipated that the interval will be mere minutes). Because you will buy the gilt strip before selling it, you will need to borrow funds from the Bank for a short period using [a facility letter provided to you by the Bank]. The loan will be repaid from the proceeds of sale of the gilt strip under the option agreement. The transaction costs will also be taken by the Bank at this time, from the funds already deposited by you for this purpose [i.e. an amount deposited by the customer equivalent to the anticipated costs of the planning]. After the transactions are completed the Bank will supply you with the relevant contract notes. Abacus will provide you with the information required to complete your tax return and claim loss relief in due course.”[8]In the explanation of “costs” are these words:
“The precise amount of your costs will depend upon the size of the loss required and the length of the option/contract period. In general, costs will be a lower percentage the larger the transaction, but will increase for shorter option/contract periods. Abacus will advise you of the approximate cost of the transactions when you confirm the amount of the required loss. If the SPV does not exercise the call option then only the bank’s 0.5% fee for the borrowing facility, together with the Abacus fee of 0.85% plus VAT, is non-refundable. If the SPV does not exercise its option as a result of movements in market values, you will be invited – as far as it is practicable and possible to do so – to participate in the planning afresh, in which case Abacus will make no further direct charge to you, and the bank’s borrowing facility will roll over at no further cost to you.”
[9]It will be noted that the precise amount of the customer’s costs would depend on “the size of the loss required”. So also would the nominal amount of gilt strips purchased, as later documentation clearly revealed:
“We understand that the size of the gilt strip you intend to acquire will vary, depending on which gilt we advice you to be suitable. To determine this, we need to know the amount of income you wish to shelter.”
[10]Stripped to its bare essentials the scheme worked as follows. Taxpayers subscribing to the Scheme would agree to purchase gilt strips for a price (‘£X’) under a forward purchase contract and at the same time would grant in consideration of a premium (‘£Y’) an option to buy them from the taxpayer for £(X minus Y). As the FTT said (§ 11):
“Underpinning the project was the legal proposition that the grant of an option is separate from and not part of the transaction that occurs on the exercise of the option; the option price will not, therefore, come into the reckoning as an amount payable on the transfer of the gilt strip.”
[11]If both the contract for purchase were completed and the option were exercised the taxpayer would claim a loss of £Y. However, the contract contemplated that the taxpayer might default, in which case he would pay the seller £Y. Since the seller of the gilt strips and the option holder (who had paid £Y for the option) were both under the control of Abacus, the economic effect was that neither the taxpayer nor Abacus would have suffered any loss. This was all subsequently explained in the contemplated application to the Bank for a loan facility. The FTT quoted from the application:
“In the unexpected event that the current market value of the gilt strips is below the exercise price, SPV will not wish to exercise its option. The purchase contract will be drafted in such a way that there is no specific performance requirement enforceable upon Mr X but rather Mr X can pay damages to SPV for the difference in the exercise price and the fixed price rather than completing the contract. This will equal the amount of the Option Premium paid to him originally by the SPV. In either event the position for SPV and Mr X is certain and neither party, nor the bank, can suffer an economic loss as a result of the fluctuations in the price of the gilt strip.”
[12]It follows from all this that the nominal amounts of gilt strips to be bought and sold were driven by the tax loss that the taxpayer wanted to create; and that whether or not the option was exercised, the taxpayer’s overall economic position would not change (apart from the costs of the scheme).[13]Mr Berry signed up to the scheme in late November 2003. On 2 December 2003 he appointed four attorneys to act on his behalf. Once he had done that his active part in the scheme was over.[14]On 26 November 2003 S G Hambros wrote to the Jersey Financial Services Commission for approval of a £62 million overnight facility to be granted to a special purpose vehicle to enable a series of purchases and sales of gilt strips to be effected. To avoid any market risk the purchases and sales were to take place contemporaneously: the market maker was to take a pre-agreed turn on the price. The letter stated that the special purpose vehicles were owned by a charitable trust but that control of the gilts remained with S G Hambros throughout. Approval was given on 2 December “for an exposure to Amaryllis Pride Ltd to the sterling value of £62 million for the purchase and sale of UK gilt strips until 10 December 2003”.[15]SG Hambros operated a custodian account with a third party bank which in turn operated a nominee client account arrangement with CREST. SG Hambros maintained a separate sub-account for each client to record that client’s holdings of securities. In effect that meant that transactions were recorded as electronic entries in these sub-accounts; and that legal title to the securities remained with the bank throughout.[16]3 December 2003 was a day of activity. It was the First Business Day. On that day: i) SG Hambros opened a bank account in Mr Berry’s name; ii) Mr Berry, acting through his attorneys, entered into the forward purchase contract to purchase, by 5.00pm on the Fourth Business Day after the date of the forward purchase contract gilt strips with SEDOL Code 0913904 with a nominal value of £24,953,210 at a price made up of four elements totalling £6,496,308 plus a “Spread Component”. Amaryllis Pride was the other party to that contract. Clause 2.1 of the forward purchase contract required Mr Berry to deliver a purchase notice in the prescribed form to Amaryllis Pride by 5.00pm on the Fourth Business Day after the date of the forward purchase contract together with evidence of the receipt of the purchase price said to be payable to Amaryllis Pride. In the event of the failure to deliver a purchase notice and/or the evidence of receipt of the purchase price, Mr Berry was liable to pay an amount, referred to as “the Liquidated Sum”. This was made up of £390,000 plus the Spread Component. iii) Mr Berry, through his attorneys, sold a call option to Amaryllis Pride and they entered into an option agreement, under which £390,000 was to be paid by Amaryllis Pride. The call option was exercisable in a 24 hour window beginning at 5.00 p.m. on the Fourth Business Day. iv) S G Hambros wrote to Mr Berry offering a credit facility (“the Facility”) to borrow up to £6,102,308 or such other sum as might be agreed by S G Hambros from time to time for the purpose of assisting with the purchase of the gilt strips. No interest was chargeable but an arrangement fee was payable to S G Hambros. In accordance with the terms of the Facility, at Mr Berry’s request to draw on the Facility an advance of sums would be made to Amaryllis Pride in discharge of the purchase price of the gilt strips.[17]8 December 2003 was the Third Business Day. On that day Amaryllis Pride acquired gilt strips from an independent party for delivery on 9 December. At the same time it entered into a contract to sell the same nominal amount of gilt strips to an independent third party for delivery on 10 December. There was a small difference between the acquisition price and the sale price.[18]9 December 2003 was the Fourth Business Day. On that day: i) At 4.36 p.m. (i.e. 24 minutes before the end of the “sale date” in the forward purchase contract) Mr Berry’s attorneys faxed a signed copy of the purchase notice required by the forward purchase contract; ii) An “accounting set inquiry”, timed at 4.41 p.m., recorded the “purchase of gilts” by Andrew Berry from Amaryllis Pride: the “amount” is recorded as £6,502,308. A transaction advice showed £24,953,213 nominal of 4¼% Treasury Strip 7 June 2032 as being “delivered” from the Amaryllis Pride account to an account held in the name of Andrew Berry; iii) An accounting set inquiry of 5.15pm on 9 December recorded Amaryllis Pride’s “exercise of option over gilt strip”. The amount was shown as £6,102,308. A Transaction Advice recorded that £24,953,213 of 4¼% Treasury Strip 7 June 2032 was transferred from Amaryllis Pride to Mr Berry’s account.[19]The scheme was now complete.[20]The difference between the price shown on the purchase of gilt strips by Mr Berry and the price shown on the purchase of the gilt strips by Amaryllis Pride from Mr Berry just over half an hour later is £400,000. That is the same amount as Amaryllis Pride paid on the grant of the option. Mr Berry argues that the difference in the two prices is a loss on which he is entitled to claim income tax relief. The legislation[21]The legislation, as it stood at the relevant time, was as follows. Section 18 of the Income and Corporation Taxes Act 1988 charged “discounts” to tax under Schedule D Case III. Section 64 provided that:
“Income tax under Case III of Schedule D shall be computed on the full amount of the income arising within the year of assessment, and shall be paid on the actual amount of that income, without any deduction.”
[22]The particular case of relevant discounted securities was dealt with by Schedule 13 of the Finance Act 1996. Paragraph 14 of that Schedule says in terms that a gilt strip is a relevant discounted security. Paragraph 1(1) of the Schedule provides that:
“Where a person realises the profit from the discount on a relevant discounted security, he shall be charged to income tax on that profit under Case III of Schedule D or, where the profit arises from a security out of the United Kingdom, under Case IV of that Schedule.”
[23]Paragraph 2 used to contain a corresponding provision relating to losses, but it had been repealed by the time of the events in question on this appeal. Instead, paragraph 14A dealt specifically with strips. It provided, so far as relevant:
“(1) A person who sustains a loss in any year of assessment from the discount on a strip shall be entitled to relief from income tax on an amount of his income for that year equal to the amount of the loss. … (3) For the purposes of this paragraph a person sustains a loss from the discount on a strip where— (a) he transfers the strip or becomes entitled, as the person holding it, to any payment on its redemption; and (b) the amount paid by him for the strip exceeds the amount payable on the transfer or redemption (no account being taken of any costs incurred in connection with the transfer or redemption of the strip or its acquisition). The loss shall be taken to be equal to the amount of the excess, and to be sustained in the year of assessment in which the transfer or redemption takes place.”
[24]The expression “transfer” is defined by paragraph 4 which says:
“in this Schedule references to a transfer, in relation to a security, are references to any transfer of the security by way of sale, exchange, gift or otherwise”
The FTT’s approach to construction[25]The FTT summarised its approach to construction as follows (§ 47):
“Adapting the words of the House of Lords in Barclays Mercantile Business Finance Ltd v Mawson (2003) 76 TC 446 and [2005] STC 1, the question for us is whether paragraph 14A of Schedule 13, construed according to its purpose, intended to apply to the transactions comprised in the Gilt Strip Planning viewed realistically.”
[26]They added: “49. The decision in BMBF establishes that it is irrelevant to the construction of the particular provision that the arrangement is designed to produce a tax advantage or, absent any specific anti-avoidance wording, that one or more component parts of the transaction have no commercial or investment purpose but have been inserted for tax reasons. Nonetheless, in determining the reality of the transaction, the fact that the arrangements were designed and implemented as a scheme or package, whether to achieve a tax or some other advantage, will be a relevant consideration in determining as a factual matter the reality of the transaction.[50]The House of Lords in IRC v Scottish Provident Institution supra 76 TC 538 was concerned with the question whether “entitlement” to gilts relied upon to found a loss claim could be established in circumstances where the right upon which the entitlement depended could be cancelled by the exercise of an option. The option in question had been matched by a separate but distinct cross-option. The special commissioners had found as a fact that there had been an outside but nonetheless real possibility that circumstances might occur in which the two options would not be exercised so as to cancel each other out. The House of Lords nevertheless went on to ignore the contingency that had been built into the scheme to produce that finding of fact and construed the relevant legislation by reference to the reality of the arrangement and whether the relevant statutory provisions, construed purposively, were intended to apply to that reality. In this connection we mention the decision of Arden LJ in Astall v HMRC 2009 EWCA 1010 Civ at para 34 where she said of the fact that “a real commercial possibility has been injected into a transaction does not mean it can never be ignored”:
“It can be disregarded if the parties have proceeded on the basis that it should be disregarded.”
The appeal[27]Under section 11(1) and(2) of the Tribunals Court and Enforcement Act 2007 a party to a case has a right to appeal to the Upper Tribunal “on any point of law” arising from a decision of the First Tier Tribunal. Section 12 gives the Upper Tribunal certain powers; but they only arise “if the Upper Tribunal … finds that the making of the decision concerned involved the making of an error of law.” This formulation, as it seems to me, invites the application of the familiar test in Edwards v Bairstow [1956] AC 14 to the question whether an error of law has been made. However, I think that there is one difference between the approach of the High Court on appeal by way of case stated from the Commissioners (either General or Special) and the Upper Tribunal’s powers under section 12. Section 12 (2) says that:
“(2) The Upper Tribunal— (a) may (but need not) set aside the decision of the First-tier Tribunal, and (b) if it does, must either— (i) remit the case to the First-tier Tribunal with directions for its reconsideration, or (ii) re-make the decision.”
[28]Importantly, section 12 (4) says that:
“(4) In acting under subsection (2)(b)(ii), the Upper Tribunal— (a) may make any decision which the First-tier Tribunal could make if the First-tier Tribunal were re-making the decision, and (b) may make such findings of fact as it considers appropriate.”
[29]Thus under the new appellate system, there is a specific fact-finding power conferred on the Upper Tribunal which did not exist in the High Court on an appeal by way of case stated. However, that fact-finding power only arises if the First Tier Tribunal has made an error of law. The Ramsay principle[30]I was expertly guided by both Ms Nathan and Mr Gammie QC through the origins and development of the Ramsay principle in the House of Lords from its birth in WT Ramsay Ltd v Commissioners of Inland Revenue (1981) 54 TC 101 to its maturity in Commissioners of Inland Revenue v Scottish Provident Institution (2004) 76 TC 538. I will state my own conclusions on this array of learning as shortly as I can.[31]In my judgment: i) The Ramsay principle is a general principle of statutory construction (Collector of Stamp Revenue v Arrowtown Assets Ltd (2004) ITLR 454 (§ 35); Barclays Mercantile Business Finance Ltd v Mawson [2005] STC 1 (§ 36)). ii) The principle is twofold; and it applies to the interpretation of any statutory provision: a) To decide on a purposive construction exactly what transaction will answer to the statutory description; and b) To decide whether the transaction in question does so (Barclays Mercantile Business Finance Ltd v Mawson (§ 36)). iii) It does not matter in which order these two steps are taken; and it may be that the whole process is an iterative process (Barclays Mercantile Business Finance Ltd v Mawson (§ 32); Astall v HMRC [2010] STC 137 (§ 44)). iv) Although the interpreter should assume that a statutory provision has some purpose, the purpose must be found in the words of the statute itself. The court must not infer a purpose without a proper foundation for doing so (Astall v HMRC (§ 44)). v) In seeking the purpose of a statutory provision, the interpreter is not confined to a literal interpretation of the words, but must have regard to the context and scheme of the relevant Act as a whole (WT Ramsay Ltd v Commissioners of Inland Revenue (1981) 54 TC 101, 184; Barclays Mercantile Business Finance Ltd v Mawson (§ 29)). vi) However, the more comprehensively Parliament sets out the scope of a statutory provision or description, the less room there will be for an appeal to a purpose which is not the literal meaning of the words. (This, I think, is what Arden LJ meant in Astall v HMRC (§ 34). As Lord Hoffmann put it in an article on Tax Avoidance: “It is one thing to give a statute a purposive construction. It is another to rectify the terms of highly prescriptive legislation in order to include provisions which might have been included but are not actually there”: See Mayes v HMRC [2010] STC 1 (§ 30)). vii) In looking at particular words that Parliament uses what the interpreter is looking for is the relevant fiscal concept: (MacNiven v Westmoreland Investments Ltd [2001] STC 237 (§§ 48, 49)). viii) Although one cannot classify all concepts a priori as “commercial” or “legal”, it is not an unreasonable generalisation to say that if Parliament refers to some commercial concept such as a gain or loss it is likely to mean a real gain or a real loss rather than one that is illusory in the sense of not changing the overall economic position of the parties to a transaction: WT Ramsay Ltd v Commissioners of Inland Revenue (1981) 54 TC 101, 187; Inland Revenue Commissioners v Burmah Oil Co Ltd (1981) 54 TC 200, 221; Ensign Tankers Ltd v Stokes [1992] 1 AC 655, 673, 676, 683; MacNiven v Westmoreland Investments Ltd (§§ 5, 32); Barclays Mercantile Business Finance Ltd v Mawson (§ 38). ix) A provision granting relief from tax is generally (though not universally) to be taken to refer to transactions undertaken for a commercial purpose and not solely for the purpose of complying with the statutory requirements of tax relief: (Collector of Stamp Revenue v Arrowtown Assets Ltd (§ 149)). However, even if a transaction is carried out in order to avoid tax it may still be one that answers the statutory description: (Barclays Mercantile Business Finance Ltd v Mawson (§ 37). In other words, tax avoidance schemes sometimes work. x) In approaching the factual question whether the transaction in question answers the statutory description the facts must be viewed realistically. (Barclays Mercantile Business Finance Ltd v Mawson (§ 36). xi) A realistic view of the facts includes looking at the overall effect of a composite transaction, rather than considering each step individually: (WT Ramsay Ltd v Commissioners of Inland Revenue (1981) 54 TC 101, 185; Carreras Group Ltd v Stamp Commissioner [2004] STC 1377 (§ 8); Barclays Mercantile Business Finance Ltd v Mawson (§ 35). xii) A series of transactions may be viewed as a composite transaction where the series of transactions is expected to be carried through as a whole, either because there is an obligation to do so, or because there is an expectation that they will be carried through as a whole and no likelihood in practice that they will not: (WT Ramsay Ltd v Commissioners of Inland Revenue (1981) 54 TC 101, 185). xiii) In considering the facts the fact finding tribunal should not be distracted by any peripheral steps inserted by the actors that are in fact irrelevant to the way in which the scheme was intended to operate: (Astall v HMRC (§ 34)). xiv) In considering whether there is no practical likelihood that the whole series of transactions will be carried out, it is legitimate to ignore commercially irrelevant contingencies and to consider it without regard to the possibility that, contrary to the intention and expectation of the parties it might not work as planned: (Commissioners of Inland Revenue v Scottish Provident Institution (2004) 76 TC 538, 558 § 23). Even if the contingency is a real commercial possibility it may be disregarded if the parties proceeded on the basis that it should be disregarded: (Astall v HMRC (§ 34)).[32]Although I have referred to rather more authority than the FTT (in deference to the very detailed submissions that were made to me) I do not consider that there is any difference in the general approach between my summary and the test that the FTT applied. So far, therefore, I do not consider that the decision under appeal involved any error of law. Application of the general approach to the legislation Identification of the purpose[33]The FTT said (§ 48):
“The purpose of the statutory provision is identified in the opening words of paragraph 14A(1), namely – “A person who sustains a loss in the year of assessment from the discount on a strip shall be entitled to relief from income tax on the amount of his income for that year according to the amount of loss.””
[34]It was this identification of the purpose that was at the heart of Mr Berry’s appeal. Ms Nathan said that either the Ramsay principle was “closed out” in interpreting this particular statutory provision; or that the purpose was to be found in the prescriptive way in which the legislation was drafted. The legislation told you both when a loss was sustained and also how much the loss was. There was simply no room for an appeal to some wider statutory purpose.[35]She relied particularly on the decisions of the Inner House of the Court of Session in Commissioners of Inland Revenue v Scottish Provident Institution (2004) 76 TC 538; the Special Commissioners in Campbell v Inland Revenue Commissioners [2004] STC (SCD) 396; Proudman J in Mayes v HMRC [2010] STC 1 and Henderson J in Gripple Ltd v HMRC [2010] STC 2283.[36]If, as I think, the Ramsay principle is a general principle of statutory construction I do not see how it can be disapplied in attempting to understand a statutory provision. Moreover, in Astall v HMRC Arden LJ rejected the submission that a purposive construction should not be applied to paragraphs 1, 2 and 3 of Schedule 13. Paragraph 14A is the replacement of paragraph 2, and is expressed in much the same terms, so the same principle must apply. If and in so far as the Special Commissioners held otherwise in Campbell v Inland Revenue Commissioners (which was referred to in Astall v HMRC), I should follow the Court of Appeal by whose decision I am bound. I therefore reject the submission that the Ramsay principle is to be disapplied in interpreting paragraph 14A.[37]Ms Nathan’s alternative submission is more promising. In Commissioners of Inland Revenue v Scottish Provident Institution the legislation in issue was section 155 of the Finance Act 1994. That provided:
“(1) Where, as regards a qualifying contract held by a qualifying company and an accounting period, amount A exceeds amount B, a profit on the contract of an amount equal to the excess accrues to the company for the period. (2) Where, as regards a qualifying contract held by a qualifying company and an accounting period, amount B exceeds amount A, a loss on the contract of an amount equal to the excess accrues to the company for the period.”
[38]It then went on to define amount A and amount B. The argument for Scottish Provident (§ 34) was that:
“The word ‘loss’ in the present context was a statutory construct, being the difference between amount A and amount B, if A was less than B. It followed that ‘loss’ in the present context was a legal concept, being one which had a specific statutory meaning. Accordingly there was no justification for treating the amount which resulted from the difference between amount A and amount B as anything other than a loss for taxation purposes, on the ground that it was not a loss in commercial terms. As Lord Hoffmann observed in MacNiven at para 58: ‘If a transaction falls within the legal description, it makes no difference that it has no business purpose. Having a business purpose is not part of the relevant concept’.”

The Lord President, giving the judgment of the court, said:

“42 It is clear from section 155 of the 1994 Act that the ascertaining of profit or loss is to be carried out by reference to the particular qualifying contract and particular accounting period. The section does not address the setting off of a profit on one such contract against a loss on the other, or visa versa. Next, it is clear that whether there is a profit or loss on a particular qualifying contract for a particular accounting period does not depend on the application of a concept of profit which is independent of what is provided in the section. Subsections (1) and (2) stipulate what is to be regarded as profit or loss, and require that amount A and amount B are to be ascertained on one or other of two bases; and in the case of the mark to market basis, each of these amounts, and hence the difference between them, depend on changes in value, if any, over the accounting period and payments, if any, due and payable to or by the company in that period. 43 In these circumstances we consider that senior counsel for the respondents [i.e. Scottish Provident] was well-founded in submitting that section 155(2) employs a legal concept, being a construct which has a specific statutory meaning. In our opinion the artificial framework for which the section provides does not indicate that a commercial meaning falls to be given to ‘loss’, let alone that the relationship between one qualifying contract and another has to be considered from a commercial viewpoint, in order to determine whether there was any true ‘loss’.”
[40]This was a case in which the legislation clearly defined amounts A and B; and explicitly said that if A exceeded B “a profit … accrues” and if B exceeded A “a loss … accrues”. The Inner House were doing no more than focussing closely on the particular statutory provision with which they were concerned. Amounts A and B were clearly defined, and the consequences of one exceeding the other were clearly spelled out. In those circumstances there was no escape from applying the literal words of the section. It was a case in which any general expectation that when Parliament uses a word like “profit” or “loss” it means a real profit or loss was displaced by the express terms of the statute. In fact the decision of the Inner House was reversed by the House of Lords. But since the House of Lords reversed the Inner House on a different point, this particular point did not fall to be considered.[41]Campbell v Inland Revenue Commissioners concerned the same Schedule as is in issue on this appeal. On 23 December 1999 Mr Campbell borrowed £3.9 million from the bank. On the following day he subscribed £3.75 million for loan notes in a newly formed company of which he was sole shareholder and director. That company made investments in other companies. Some three months later, on 15 March 2000, he executed a deed of gift of the loan notes in favour of his wife. The market value of the loan notes at the date of the gift was £1.5 million. By virtue of paragraph 8 of Schedule 13 Mr Campbell was deemed to have received the market value of the loan notes on the occasion of the gift. He claimed that the difference between the amount of his subscription and the market value of the notes at the date of the gift was a loss on which he was entitled to tax relief. The argument for HMRC (§ 51) was that transactions which have the sole purpose of complying with statutory provisions are to be disregarded; and that Mr Campbell’s subscription for loan notes and the subsequent gift to his wife had that sole purpose. Consequently both should be disregarded. If they were disregarded, then Mr Campbell had suffered no loss. The Special Commissioners began their discussion by finding as a fact that Mr Campbell had a commercial non-tax purpose in making investments through the medium of the company, although his main purpose was to obtain tax relief (§ 65). But his gift to his wife was wholly tax motivated (§ 67). HMRC had accepted that the entire subscription monies had been paid for the loan notes and had also accepted that the gift was a transfer. Thus the only issue for the Commissioners (§ 68) was:
“whether the difference between the subscription price for the Loan Notes (£3.75m) and their market value at the time of that transfer by gift (£1.5m) is a 'loss' within the meaning of para 2(2) and para 2(3), or whether the Appellant's tax motivation in subscribing for the Loan Notes in the form in which they were issued to him denies relief on the application of the Ramsay doctrine.” (Emphasis added)
[42]The Commissioners then discussed the Ramsay principle in detail, for which they were commended by the House of Lords in Barclays Mercantile Business Finance Ltd v Mawson (§ 38). They expressed their conclusion as follows: “86. In this case, we are concerned with the terms of Sch 13, para 2 in circumstances in which the Inland Revenue accepts that the subscription price was entirely paid in respect of the acquisition of the Loan Notes and that there was a transfer by the Appellant to a connected person. Paragraph 2(3) is an entirely mechanistic provision which calculates the 'loss' by deducting the subscription price 'paid in respect of [the] acquisition of [the Loan Notes]', within para 2(2)(b), from the market value deemed by para 8 to be obtained on the 'transfer', within para 2(2)(a), and deducting any relevant costs.[87]Once an amount paid in respect of a relevant discounted security is ascertained and the amount received (or deemed to be received) on transfer or redemption is determined, there is a 'loss' where the former exceeds the latter. There is no room for the purpose of the holder of the relevant discounted security to inform the construction of the term 'loss'. In other words, once the terms 'amount paid … in respect of [an] acquisition of [a relevant discounted security]' and 'amount payable on … transfer or redemption [of the relevant discounted security]' have been construed in the context of para 2(2), the 'loss' is also automatically ascertained. This is confirmed by the terms of para 2(3) which provides that 'For the purposes of [Sch 13] the loss shall be taken … to be equal to the amount of the excess increased by the amount of any relevant costs …'. Paragraph 2(3) confirms that the term 'loss' is, to use the terminology of The Lord President (Lord Cullen of Whitekirk) (at para 43) in Scottish Provident a 'construct which has a specific statutory meaning', so that, like s 155 of the Finance Act 1994, in Scottish Provident, para 2(2), of Sch 13 is 'an artificial framework … [which] does not indicate that a commercial meaning falls to be given to “loss”'.[88]The artificial (legal) meaning of the term 'loss' in Sch 13, para 2(1), (2) is further reinforced by the statutory mechanism which quantifies a 'loss' for these purposes. Firstly, the 'loss' is increased by the 'relevant costs' incurred by a taxpayer (being the costs incurred 'in connection with the acquisition of the [relevant discounted security]' and costs incurred 'in connection with [any] transfer or redemption of the [relevant discounted security]': see Sch 13, para 1(4) and para 2(3)(a). Secondly, as we have already observed, the transferor of a relevant discounted security is deemed to receive an amount equal to its market value, when the transfer is to a connected person, even though he may receive no such sum. These factors, while not at all conclusive in themselves, confirm that the term 'loss', in the context of para 2(1) and (2), is far removed from any 'commercial' sense of the term,[89]Here, the amount paid by the Appellant in respect of the Loan Notes exceeded the amount which he was treated as obtaining on the transfer to his wife. It follows that by the express words of para 2(2) he sustained a loss for the purposes of Sch 13.[90]Put another way, para 2 is a provision which is sufficiently 'closely articulated' (on Lord Millett's terminology in Arrowtown) or 'legal' (using Lord Hoffmann's terminology in Westmoreland) to be unaffected by the purpose of the Appellant in subscribing for the Loan Notes.”[43]The ground of the decision, as I read it, is that “There is no room for the purpose of the holder of the relevant discounted security to inform the construction of the term “loss””. In other words Mr Campbell’s motivation did not automatically deny him his tax relief. They were not saying that the fact finding tribunal should ignore the reality of the transactions that in fact took place. Moreover, as they went on to point out, the Commissioners’ finding of fact that Mr Campbell had a commercial purpose in subscribing for the loan notes meant that HMRC’s argument failed on the facts (§ 92). At the start Mr Campbell was the holder of the loan notes, for which he had paid in real money borrowed from the bank. At the end, Mr Campbell no longer owned the loan notes; but he still owed the money to the bank. His economic position had changed for the worse. In ordinary terms, he had suffered a loss. The loan notes had not disappeared. They still existed, but they were owned by his wife. There were, therefore, two real events separated by several months which existed in the real world. In his written submissions Mr Gammie QC submitted:
“Thus, once the Commissioners had decided (or the Revenue had conceded) that no part of the £3.75m was in reality a gift to the company, there was only one possible answer to the statutory question posed by paragraph 2(2)(b) of Schedule 13 — what amount did Mr Campbell pay to subscribe the securities? Similarly, once it had been determined (or conceded) that the reality of the arrangement was that Mr Campbell subscribed the securities and then, as a separate matter, gave them to his wife, paragraph 8 of Schedule 13 supplied the answer to the question – what amount did Mr Campbell receive on transferring the securities? Paragraph 8 directed that this was their market value. The relevant point about Campbell is that the provisions of Schedule 13 were too closely articulated in relation to the reality of the taxpayer’s transactions in that case. It is not that the provisions of Schedule 13 are too closely articulated to exclude the application of the Ramsay principle and to prevent one deciding in any other case what is the tax reality of the taxpayer’s transactions.”
[44]In effect this is how the FTT dealt with Campbell. They said (§ 63):
“The present transactions are different and distinguishable from the two separate transactions found in Campbell, supra. There, the reality of the taxpayer’s transaction comprised first, a subscription for “relevant discounted securities” at an overvalue and, second, a gift of those securities to his wife. The reality of the transaction was not a gift to the company issuing the securities nor was it a gift to his wife.”
[45]I agree with Mr Gammie’s submission and I consider that the FTT were right to distinguish Campbell in the way that they did.[46]Mayes v HMRC concerned tax relief on the surrender of a bond. It was agreed that certain events comprised a tax avoidance scheme. The question was whether it worked. Proudman J held that it did. She held that the court must apply a purposive construction to the statute (§ 7) and recorded that the dispute was how to apply that purposive construction (§ 8). The case is certainly no authority for the proposition that the principle of purposive construction should be disapplied. Proudman J held that the legislation sated its purpose explicitly (§ 14); and then went on to examine closely how the legislation worked on the basis of a number of hypothetical examples. She concluded that the legislation could produce unfair or arbitrary results (§ 31), which led her to the conclusion that Parliament had legislated by reference to form rather than substance (§ 32). She concluded (§ 33):
“To pose the question implicit in the speech of Lord Wilberforce in WT Ramsay Ltd v IRC…, “Is this a tax on gains, or gains less losses, or is it a tax on arithmetical differences?”
I agree with Mr Furness that the legislation is not designed to allow a line to be drawn between legitimate and illegitimate avoidance on the basis of pre-ordained transactions. It exacts tax on a formulaic approach to the transaction without any overriding purpose that some types of transaction may not count. It is common ground that the tax avoidance motive is insufficient by itself to raise the Ramsay principle. It does not seem to me that the legislation includes terms capable of being construed as business or commercial concepts rather than (as Lord Hoffmann put it in MacNiven at [33]) held within the confines of purely juristic analysis.”[47]She repeated later in her judgment (§ 44) that the legislation did not seek to tax real or commercial gains.[48]To my mind this is a case in which a close examination of the statutory language coupled with an examination of how it worked in practice led to the conclusion that there was no option but to construe it literally. I do not consider that it is of direct assistance in construing the provisions in issue on this appeal. (I understand that the Court of Appeal have heard an appeal from the judgment of Proudman J and that the outcome of the appeal is still awaited).[49]Gripple Ltd v HMRC concerned tax relief for expenditure on research and development. Henderson J said (§ 12):
“I would, however, make the general point that the provisions form a detailed and meticulously drafted code, with a series of defined terms and composite expressions, and a large number of carefully delineated conditions, all of which have to be satisfied if the relief is to be available. The schedule runs to 26 paragraphs, and occupies ten pages in Tolley's Yellow Tax Handbook for 2005–06. I emphasise this point because one of Mr Gordon’s submissions for Gripple is that the schedule evinces a general intention to provide enhanced relief for expenditure on R & D, and that a generous construction should where possible be adopted in order to further that general aim. I am unable to accept this submission. It seems to me, on the contrary, that a detailed and prescriptive code of this nature leaves little room for a purposive construction, and there is no substitute for going through the detailed conditions, one by one, to see if, on a fair reading, they are satisfied.”
[50]In my judgment what Henderson J was being asked to do was (in Lord Hoffmann’s words) to rectify the terms of highly prescriptive legislation in order to include provisions which might have been included but were not actually there. That exercise plainly goes much further than any principle of interpretation (whether purposive or not) will allow.[51]As I have said, the FTT held that the purpose of paragraph 14A was the general proposition stated in sub-paragraph (1) viz.:
“A person who sustains a loss in the year of assessment from the discount on a strip shall be entitled to relief from income tax on the amount of his income for that year according to the amount of loss.”
[52]In my judgment the FTT were right to identify the purpose of the paragraph in that way. This is not a case in which Parliament has used algebra (amount A and B) to create a notional profit or loss. It has used words which have a recognised commercial meaning; and it is to be expected that Parliament intended to tax (or relieve) real commercial outcomes. The FTT were right not to adopt a slavishly literal “tick-box” interpretation of the legislation. This is precisely how the Ramsay principle is meant to operate. I thus conclude that the FTT made no error of law in identifying the purpose of the legislation. Looking at the facts realistically[53]The only possible error of law under this head would be if the FTT had not looked at the facts realistically. Otherwise an evaluation of the facts is (obviously) a question of fact involving no question of law. Since the FTT expressly directed themselves to look at the facts realistically, it would be surprising if they had not done so.[54]Although the FTT expressed some doubt about whether Mr Berry acquired the gilt strips at all, they proceeded on the assumption that he had (§ 52). Thus if and in so far as Ms Nathan’s criticisms of the FTT’s discussion of “legal” ownership of the strips were well-founded, they do not matter. The FTT then considered (§ 53) whether the forward purchase contract, the option and the exercise of the option should all be regarded as a single transaction. This in turn required them to consider whether Mr Berry had taken a position in the gilts market with a real financial risk.[55]They noted that Abacus had been advised that the scheme should include a real possibility that the option might not be exercised; and that the risk had been statistically quantified at 7 per cent (§ 54). They noted also that the amount of gilt strips to be acquired and then sold was driven by the amount of tax loss that the taxpayer wanted to generate (§ 55). Their evaluation of the real possibility of risk was as follows: “56. The risk was said for Mr Berry to be based on the probability that by the end of the Fourth Business Day the option would not be exercised because the gross value of the gilt strips had fallen below the “water line” and Mr Berry would be left holding the strips. But, until the purchase notice has been served, Mr Berry has no gilt strips and so is exposed to no risk. If the value of the gilt strips held up he would serve the purchase notice and buy the gilt strips from Amaryllis Pride and Amaryllis Pride could then exercise its option to purchase them. Mr Berry would then have lost nothing, and only incurred the cost of buying the package. If the value of the gilt strip had fallen below the water line by the Fourth Business Day, Mr Berry would have chosen not to serve the purchase notice and Amaryllis Pride would have decided not to exercise the option to purchase them. This was contemplated by the parties who did not envisage that Mr Berry would have to complete the purchase: see, for example, the credit application referred to in para 28 above. Mr Berry (who had already been credited with £390,000 as the option price) would be debited with that amount plus the spread component as the so-called liquidated damages under the Forward Purchase Contract; overall he would have lost nothing. S G Hambros were completely protected by the matching sale of the gilts purchased on the Fourth Business Day and completely protected by the charges in their favour.[57]In theory the Four Attorneys might have served a purchase notice and bought gilt strips from Amaryllis Pride and Amaryllis Pride might have decided not to exercise its option to purchase them. In theory the Four Attorneys might have chosen not to serve the Purchase Notice while Amaryllis Pride might have exercised its option to purchase them. Bearing in mind that S G Hambros was completely protected and that it had complete control over all the funds put in to make the Gilt Strips Planning work and that Amaryllis Pride stood to lose nothing, it is inconceivable that those eventualities would have happened. It would have required Amaryllis Pride and the Four Attorneys to have acted so irrationally as to defy credibility. The reality is that the parties proceeded with the Planning on the basis that the possibility of Mr Berry being exposed to any real risk of being left holding the gilt strips in a falling market should be disregarded.”[56]Their ultimate conclusion on the question of risk was (§§ 59, 60):
“This has indicated to us that any risk to the Bank, to Amaryllis Pride and to any of the participators in the Gilt Strip Planning was a façade. It had no reality. It was built into the Gilt Strip Planning scheme, but the parties proceeded on the basis that it should be disregarded. …This is confirmed by the oral evidence.”
[57]Having reached that conclusion the FTT were then in a position to reach a conclusion on the question whether the forward purchase contract and the option contract (and its exercise) should be regarded as a single transaction. On that question they said (§ 62):
“The implementation of the Gilt Strip Planning was exactly as the parties had intended. The self-cancelling result was entirely in line with their expectations which were that Amaryllis Pride and the Bank would be protected by the matching purchase and sale of gilt strips in the market and that Mr Berry would pay his fee. This leaves the option contract and the forward purchase contract and the associated debits and credits entries as ingredients in a single scheme. The scheme was designed to ensure that no “real” money ever reached Mr Berry. The £390,000 “premium” may have been credited to him; but it was retained under the Bank’s control for use in financing the exercise of the forward purchase contract. The proceeds of sale when the option was exercised were to be applied in discharging Mr Berry’s intra-day liability to the Bank. The gilt strips were kept securely under the control of the Bank and away from the hands of Mr Berry; they were needed to meet Amaryllis Pride’s agreement to sell them to the outside third party.”
[58]Ms Nathan said that the FTT had recognised that Mr Berry had a liability to the Bank to discharge, and that there was therefore a real transaction. She said that accounting entries were the normal way in which changes in credit and debit balances in bank accounts were recorded; and that in the case of securities held in CREST accounting entries in sub-accounts were the normal way in which changes of ownership were recorded. Accordingly she submitted that the FTT were wrong to conclude (§ 63) “that no loss was sustained by Mr Berry; nor was any amount paid by him for the gilt strips, nor was there any transfer of the strips.” In my judgment it is really only the first of these conclusions that matters, having regard to the purpose of paragraph 14A that the FTT (rightly in my judgment) identified. Once the FTT had reached the conclusion that there was no element of real risk and that the anti-Ramsay device (consisting of the possibility that the option would not be exercised) could be disregarded because that is how the parties had proceeded, the FTT was left with a self-cancelling scheme. Looked at realistically, if and in so far as Mr Berry bought and sold gilt strips the purchase price and the sale price were the same. The option fee was no more than a refundable deposit. His overall economic position before and after had not changed (apart from the fact that he had paid the fees required to participate in the scheme).[59]This is exactly the situation that Lord Wilberforce described in Ramsay itself:
“In each case two assets appear, like “particles” in a gas chamber with opposite charges, one of which is used to create the loss, the other of which gives rise to an equivalent gain which prevents the taxpayer from supporting any real loss, and which gain is intended not to be taxable. Like the particles, these assets have a very short life. Having served their purpose they cancel each other out and disappear. At the end of the series of operations, the taxpayer’s financial position is precisely as it was at the beginning, except that he has paid a fee, and certain expenses, to the promoter of the scheme.”
[60]In my judgment the view of the facts that the FTT took was a realistic one. If they did, they made no error of law. The scheme in issue in this appeal fell squarely within the Ramsay principle as described in Ramsay itself. The subsequent elucidation of the principle has done no more than to explain what that principle entails. So, in the words of Little Gidding:
“… the end of all our exploring Will be to arrive where we started And know the place for the first time.”

Result

[61]In my judgment the FTT made no error of law. The appeal must be dismissed. MR JUSTICE LEWISON Release date: 25 February 2011

Cited in 19 later judgments

and 10 more of the most senior