Hancock & Anor v Revenue & Customs [2017] EWCA Civ 198

[2017] EWCA Civ 198Case No A3/2016/1763
IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE UPPER TRIBUNAL
TAX AND CHANCERY CHAMBER
MRS JUSTICE ROSE DBE (CHAMBER PRESIDENT) AND
JUDGE HOWARD NOWLAN
[2016] UKUT 0081 (TCC)
Venue Royal Courts of Justice, Strand, London, WC2A 2LLDate 25/05/2017
Before
LORD JUSTICE LEWISONLORD JUSTICE KITCHINLORD JUSTICE FLOYD
Between
ANTHONY HANCOCKAppellantsTRACY LEE HANCOCKAppellantHM REVENUE & CUSTOMSRespondents
Michael Sherry instructed by direct access for AppellantsMichael Gibbon QC and Elizabeth Wilson (instructed by The General Counsel and Solicitor to HM Revenue & Customs) for RespondentsHearing Hearing date: 13 March 2017
Judgment Approved
[1]This appeal raises a question of statutory interpretation of the Taxation of Chargeable Gains Act 1992 (“TCGA”). It concerns the interaction of the rules which relate to corporate reorganisations and those which relate to qualifying corporate bonds (“QCBs”).[2]The appellants, Anthony Hancock and his wife Tracy Hancock (“the Hancocks”) appeal from the decision of the Upper Tribunal (Tax and Chancery Chamber) (Mrs Justice Rose DBE, Chamber President and Judge Howard Nowlan) dated 18 February 2016 ([2016] UKUT 0081 (TCC)) which allowed the appeal of the respondents (“HMRC” or “the Revenue”) from the decision of the First-tier Tribunal (Tax Chamber) (Judge Berner and Mr Nigel Collard) dated 21 July 2014 ([2014] UKFTT 695 (TC)). The First-tier Tribunal (“FTT”) allowed the Hancocks’ appeal against closure notices to their self-assessment tax returns for the year to 5 April 2004. The appeal is with permission granted by the Upper Tribunal (“UT”) itself.[3]In broad terms, under the TCGA, a reorganisation of share capital is not treated as a disposal of the shares for the purposes of the capital gains tax (“CGT”) legislation. Instead, any gain on the original shares is “rolled over” into the new holding, and will normally fall to be taxed on a subsequent disposal of that holding outside the reorganisation rules. Those rules, which operate to defer the payment of CGT, also apply to the conversion of securities, for example loan notes and corporate bonds, as they apply to a reorganisation of share capital, but with any necessary adaptations.[4]Gains on QCBs are, by contrast, not chargeable to CGT at all. The exemption was one which was introduced to stimulate the market in British corporate bonds: see the explanation given by Buxton LJ in Weston v Garnett (Inspector of Taxes) [2005] STC 1134 at [40] to [41]. In accordance with that objective, to qualify for the exemption, the bonds must, amongst other things, not be redeemable other than in sterling.[5]The combined operation of the reorganisation provisions with those applying to QCBs could mean that when shares are converted into QCBs, so that the gain on the shares is rolled over into the QCBs, the taxation of the gain would not simply be deferred, but would escape CGT altogether. For that reason, section 116 of the TCGA provides that, where it applies, conversion into a QCB creates a deemed capital gain, frozen at that date, which falls to be taxed on the subsequent disposal of the QCB. The central issue in this appeal is whether the Hancocks have structured their arrangements so as to avoid that consequence. If they are correct, the redemption for cash of loan notes, which had an aggregate nominal value of £9,724,651, did not bring into charge any accrued capital gain, save in respect of a small part of the gain which represented Revised 03/01 Loan Notes – see further below.

The facts

[6]I can take the agreed facts from the judgment of the UT. The Hancocks held, between them, 100 per cent of the share capital in a limited company, called Blubeckers Ltd. On 24 August 2000 Blubeckers Ltd was sold to Lionheart Holdings Limited (“Lionheart”). At the date of disposal the issued share capital of Blubeckers Ltd consisted of 5,219 £1 ordinary shares. Mr Hancock held 2,611 shares and Mrs Hancock held 2,608. The initial consideration payable by Lionheart took the form of loan notes issued by Lionheart to the value of £9,270,000, with provision for payment of additional consideration depending on the subsequent performance of the business.[7]The loan notes issued on 24 August 2000 were: i) £500,000 A Loan Notes 2007, which were issued to Mr Hancock; ii) £4,137,664 B Loan Notes 2004, which were issued to Mr Hancock; and iii) £4,632,336 B Loan Notes 2004, which were issued to Mrs Hancock.[8]The appeal is concerned only with the tax consequences arising from subsequent actions taken with the B Loan Notes 2004. The UT referred to the tranche of B Loan Notes 2004 issued in August 2000 as the “08/00 Loan Notes”, and I will continue to do so.[9]The 08/00 Loan Notes provided that the note holder could require repayment in US dollars, with the exchange rate to be the spot rate obtained or obtainable by Lionheart twenty days before repayment. It is common ground that the provision for payment in a currency other than sterling and at an exchange rate other than that prevailing at redemption prevented the 08/00 Loan Notes from being a QCB (see section 117 of the TCGA).[10]Additional purchase consideration became payable and was paid on 22 March 2001, as follows: i) £477,516 B Loan Notes 2004 were issued to Mr Hancock; and ii) £477,135 B Loan Notes 2004 were issued to Mrs Hancock.[11]Following the UT, I will refer to these additional B Loan Notes 2004 as the 03/01 Loan Notes.[12]On 9 October 2002 deeds of variation removing the rights to redemption in US dollars from the 03/01 Loan Notes were executed. The UT referred to these amended notes, as I will, as the “Revised 03/01 Loan Notes”. Corresponding action was not taken in respect of the 08/00 Loan Notes. The overall effect was that the Revised 03/01 Loan Notes were now QCBs whereas the 08/00 Loan Notes were not QCBs (“non-QCBs”). It is common ground that the effect of the conversion of the 03/01 Loan Notes into the Revised 03/01 Loan Notes was to freeze the gain on those notes and give rise to a charge to CGT on their subsequent disposal.[13]On 7 May 2003 the 08/00 Loan Notes and the Revised 03/01 Loan Notes beneficially owned by the Hancocks were exchanged for two Secured Discounted Loan Notes 2004 (“SDLs”). One of these SDLs had a nominal value of £4,615,180 (in the case of Mr Hancock) and one had a nominal value of £5,109,471 (in the case of Mrs Hancock). The SDLs were QCBs. The terms of the exchange (or exchanges) effected on 7 May 2003 were set out in a single document. The SDLs issued to each of the Hancocks were issued in their aggregate amounts, without those attributable to the 08/00 Loan Notes and the Revised 03/01 Loan Notes being separately identified.[14]The SDLs provided for redemption on 30 April 2004 or for early redemption on either 30 June 2003 or 31 December 2003 on 30 days’ notice. An early redemption notice was given and the SDLs were redeemed on 30 June 2003 for cash, together with the payment of the associated redemption premium.[15]In summary, therefore, the chronology was as follows: i) 24 August 2000 – the 08/00 Loan Notes were issued (non-QCBs). ii) 22 March 2001 – the 03/01 Loan Notes were issued (non-QCBs). iii) 9 October 2002 – the 03/01 Loan Notes (non-QCBs) were converted into the Revised 03/01 Loan Notes (QCBs). iv) 7 May 2003 – the 08/00 Loan Notes (non-QCBs) and the Revised 03/01 Loan Notes (QCBs) were converted into SDLs (QCBs). v) 30 June 2003 – the SDLs (QCBs) were redeemed for cash.[16]The UT recorded that they had been told that the tax at stake here was some £830,000, and that as a result of the overall scheme of which this was an example, about £3.5 million in tax was known to be at stake.

The statutory framework

[17]Debts are assets for the purposes of the TCGA: see section 21(1)(a). By section 251(1) no chargeable gain accrues on a debt incurred by a person “except in the case of a debt on a security as defined in section 132”. Section 132(3)(b) defines “security” as including “… any loan stock or similar security whether of the Government of the United Kingdom or any other government, or any public or local authority in the United Kingdom or elsewhere, or of any company, and whether secured or unsecured.”[18]I consider first the group of provisions which apply when companies undergo corporate restructuring which involves swapping existing (original) shares or securities for new shares or securities. The overall effect of these provisions is that there is no disposal of the original shares at the time of the swap. This is referred to as the “no disposal fiction”. Rather, the original shares and the new shares are treated as the same asset, and the new shares are deemed to have the same acquisition date and cost as the original shares (“the same asset fiction”). I will call the dual fiction involved in this treatment “the reorganisation treatment”.[19]The key provision is section 127. It provides so far as material:
“127 Equation of original shares and new holding Subject to sections 128 to 130, a reorganisation shall not be treated as involving any disposal of the original shares or any acquisition of the new holding or any part of it, but the original shares (taken as a single asset) and the new holding (taken as a single asset) shall be treated as the same asset acquired as the original shares were acquired.”
[20]Section 126 provides some definitions, for use, amongst other places, in section 127:
“126 Application of sections 127 to 131 “(1) For the purposes of this section and sections 127 to 131 “reorganisation” means a reorganisation or reduction of a company's share capital, and in relation to the reorganisation— (a) “original shares” means shares held before and concerned in the reorganisation, (b) “new holding” means, in relation to any original shares, the shares in and debentures of the company which as a result of the reorganisation represent the original shares (including such, if any, of the original shares as remain). (2) The reference in sub-section (1) above to the reorganisation of a company's share capital includes- (a) any case where persons are, whether for payment or not, allotted shares in or debentures of the company in respect of and in proportion to (or as nearly as may be in proportion to) their holdings of shares in the company or of any class of shares in the company, and (b) any case where there are more than one class of share and the rights attached to shares of any class are altered.” (a) “original shares” means shares held before and concerned in the reorganisation, (b) “new holding” means, in relation to any original shares, the shares in and debentures of the company which as a result of the reorganisation represent the original shares (including such, if any, of the original shares as remain). (a) any case where persons are, whether for payment or not, allotted shares in or debentures of the company in respect of and in proportion to (or as nearly as may be in proportion to) their holdings of shares in the company or of any class of shares in the company, and (b) any case where there are more than one class of share and the rights attached to shares of any class are altered.”
[21]These provisions refer to the original holding and the new holding as being shares (or, in the case of the new holding, debentures) in the company. However section 127 also applies where there is a “conversion of securities”. Thus section 132 (as amended by the Finance Act 1997) provides:
“132. Equation of converted securities and new holding (1) Sections 127 to 131 shall apply with any necessary adaptations in relation to the conversion of securities as they apply in relation to a reorganisation (that is to say, a reorganisation or reduction of a company's share capital). … (3) For the purposes of this section and section 133— (a) “conversion of securities” includes any of the following, whether effected by a transaction or occurring in consequence of the operation of the terms of any security or of any debenture which is not a security, that is to say— (i) a conversion of securities of a company into shares in the company, and (ia) a conversion of a security which is not a qualifying corporate bond into a security of the same company which is such a bond, and (ib) a conversion of a qualifying corporate bond into a security which is a security of the same company but is not such a bond, and (ii) a conversion at the option of the holder of the securities converted as an alternative to the redemption of those securities for cash, and (iii) any exchange of securities effected in pursuance of any enactment (including an enactment passed after this Act) which provides for the compulsory acquisition of any shares or securities and the issue of securities or other securities instead.” (a) “conversion of securities” includes any of the following, whether effected by a transaction or occurring in consequence of the operation of the terms of any security or of any debenture which is not a security, that is to say— (i) a conversion of securities of a company into shares in the company, and (ia) a conversion of a security which is not a qualifying corporate bond into a security of the same company which is such a bond, and (ib) a conversion of a qualifying corporate bond into a security which is a security of the same company but is not such a bond, and (ii) a conversion at the option of the holder of the securities converted as an alternative to the redemption of those securities for cash, and (iii) any exchange of securities effected in pursuance of any enactment (including an enactment passed after this Act) which provides for the compulsory acquisition of any shares or securities and the issue of securities or other securities instead.”
[22]Section 115 provides for the exemption from CGT for QCBs. It provides:
“115 Exemptions for gilt-edged securities and qualifying corporate bonds etc (1) A gain which accrues on the disposal by any person of- (a) gilt-edged securities or qualifying corporate bonds, or (b) any option or contract to acquire or dispose of gilt-edged securities or qualifying corporate bonds, shall not be a chargeable gain.” shall not be a chargeable gain.”
[23]Section 117 contains the definition of a QCB:
“117 Meaning of “qualifying corporate bond” “(A1). For the purposes of corporation tax “qualifying corporate bond” means any asset representing a loan relationship of a company; and for purposes other than those of corporation tax references to a qualifying corporate bond shall be construed in accordance with the following provisions of this section. (1) For the purposes of this section, a “corporate bond” is a security, as defined in section 132(3)(b) – (a) the debt on which represents and has at all times represented a normal commercial loan; and (b) which is expressed in sterling and in respect of which no provision is made for conversion into, or redemption in, a currency other than sterling, … (7) Subject to subsections (9) and (10) below, for the purposes of this Act, a corporate bond – (a) is a “qualifying” corporate bond if it is issued after 13th March 1984; and (b) becomes a “qualifying” corporate bond if, having been issued on or before that date, it is acquired by any person after that date and that acquisition is not as a result of a disposal which is excluded for the purposes of this section, or which was excluded for the purposes of section 64 (4) of the Finance Act 1984.” (a) the debt on which represents and has at all times represented a normal commercial loan; and (b) which is expressed in sterling and in respect of which no provision is made for conversion into, or redemption in, a currency other than sterling, (a) is a “qualifying” corporate bond if it is issued after 13th March 1984; and (b) becomes a “qualifying” corporate bond if, having been issued on or before that date, it is acquired by any person after that date and that acquisition is not as a result of a disposal which is excluded for the purposes of this section, or which was excluded for the purposes of section 64 (4) of the Finance Act 1984.”
[24]Thus, on the basis of the provisions set out so far, a gain on shares which is rolled over into a QCB under section 127 would escape CGT altogether on disposal of the QCB by the application of section 115. Section 116, where it applies, seeks to avoid this result. It has the effect of freezing the gain which could have been achieved at the date of conversion of the original shares into the QCB, and deem that gain to accrue as a chargeable gain on a subsequent disposal of the QCB. Section 116 provides, so far as material:
“116 Reorganisations, conversions and reconstructions (1) This section shall have effect in any case where a transaction occurs of such a description that, apart from the provisions of this section— (a) sections 127 to 130 would apply by virtue of any provision of Chapter II of this Part; and (b) either the original shares would consist of or include a qualifying corporate bond and the new holding would not, or the original shares would not and the new holding would consist of or include such a bond; and in paragraph (b) above “the original shares” and “the new holding” have the same meaning as they have for the purposes of sections 127 to 130. (2) In this section references to a transaction include references to any conversion of securities (whether or not effected by a transaction) within the meaning of section 132 and “relevant transaction” means a reorganisation, conversion of securities or other transaction such as is mentioned in subsection (1) above, …. (3) Where the qualifying corporate bond referred to in subsection (1)(b) above would constitute the original shares for the purposes of sections 127 to 130, it is in this section referred to as “the old asset” and the shares or securities which would constitute the new holding for those purposes are referred to as “the new asset”. (4) Where the qualifying corporate bond referred to in subsection (1)(b) above would constitute the new holding for the purposes of sections 127 to 130, it is in this section referred to as “the new asset” and the shares or securities which would constitute the original shares for those purposes are referred to as “the old asset”. (4A) … (5) So far as the relevant transaction relates to the old asset and the new asset, sections 127 to 130 shall not apply in relation to it. (6) In accordance with subsection (5) above, the new asset shall not be treated as having been acquired on any date other than the date of the relevant transaction or, subject to subsections (7) and (8) below, for any consideration other than the market value of the old asset as determined immediately before that transaction. … (9) In any case where the old asset consists of a qualifying corporate bond, then, so far as it relates to the old asset and the new asset, the relevant transaction shall be treated for the purposes of this Act as a disposal of the old asset and an acquisition of the new asset. (10) Except in a case falling within subsection (9) above, so far as it relates to the old asset and the new asset, the relevant transaction shall be treated for the purposes of this Act as not involving any disposal of the old asset but— (a) there shall be calculated the chargeable gain or allowable loss that would have accrued if, at the time of the relevant transaction, the old asset had been disposed of for a consideration equal to its market value immediately before that transaction; and (b) subject to subsections (12) to (14) below, the whole or a corresponding part of the chargeable gain or allowable loss mentioned in paragraph (a) above shall be deemed to accrue on a subsequent disposal of the whole or part of the new asset (in addition to any gain or loss that actually accrues on that disposal); and (c) on that subsequent disposal, section 115 shall have effect only in relation to any gain or loss that actually accrues and not in relation to any gain or loss which is deemed to accrue by virtue of paragraph (b) above.” (a) sections 127 to 130 would apply by virtue of any provision of Chapter II of this Part; and (b) either the original shares would consist of or include a qualifying corporate bond and the new holding would not, or the original shares would not and the new holding would consist of or include such a bond; (a) there shall be calculated the chargeable gain or allowable loss that would have accrued if, at the time of the relevant transaction, the old asset had been disposed of for a consideration equal to its market value immediately before that transaction; and (b) subject to subsections (12) to (14) below, the whole or a corresponding part of the chargeable gain or allowable loss mentioned in paragraph (a) above shall be deemed to accrue on a subsequent disposal of the whole or part of the new asset (in addition to any gain or loss that actually accrues on that disposal); and (c) on that subsequent disposal, section 115 shall have effect only in relation to any gain or loss that actually accrues and not in relation to any gain or loss which is deemed to accrue by virtue of paragraph (b) above.”
[25]In Harding v Revenue and Customs Commissioners [2008] EWHC 99 (Ch); [2008] STC 1965 Briggs J (as he then was) explained, at paragraphs 17 to 22, how the statutory scheme dealt with the interaction of the reorganisation treatment and QCBs. He said:
“20. Since the central feature of a QCB is that a disposal of it for cash is not chargeable to tax, it was necessary to make special provision for an exchange of an old chargeable asset (such as a share) for a QCB, in the context of a takeover. The relevant special provision is contained in s 116. By sub-s (5), s 127 is disapplied, so that there is no rollover. By sub-s (10) such an exchange gives rise to no immediately chargeable disposal of the share but the chargeable gain or allowable loss which would then have arisen is calculated, and then charged to tax or (if a loss) allowed at the time of the disposal of the QCB, and at the rate then in force. This is generally known as a ‘frozen gain’ mechanism. 21. In relation to the latent gain inherent in the old asset immediately prior to exchange, the frozen gain regime achieves in relation to an exchange for QCBs substantially the same result as the rollover arrangement achieves in relation to an exchange for non-QCBs. In both cases the latent gain in the old asset is taxed upon the disposal of the new asset. As might be expected, gains or losses attributable purely to the new asset are, if it is a QCB, left out of account. The key to the successful operation of the frozen gain regime is that it displaces the ordinary rule that no chargeable gain or allowable loss occurs on the disposal of the QCB.”
[26]The issue on this appeal arises out of the drafting of section 116(1)(b), and whether or, perhaps more accurately, how it applies to the transactions carried out by the Hancocks. The UT expanded the words of section 116(1)(b) so as to include the words that were there by implication, and to expose the two limbs, which I will denote A and B:
“either (A) the original shares would consist of or include a QCB, and the new holding would not consist of or include a QCB or (B) the original shares would not consist of or include a QCB, and the new holding would consist of or include a QCB.”
[27]The Hancocks’ case is that, given that all the SDLs were QCBs, it is only limb B which could apply here. The “new holding” consists of QCBs. However, viewing the exchange of 7 May 2003 as a single transaction, it cannot be said that the original shares/securities did not “consist of or include a QCB”. The Revised 03/01 Loan Notes were still QCBs on that date. Section 116 therefore simply does not apply, with the following consequences. The first consequence is that section 116 cannot bring into charge any frozen gain on the 08/00 Loan Notes on disposal of the QCB. A further consequence is that sections 127 to 130 are not disapplied. Thus, applying the literal words of section 127 to the transaction entered into on 7 May 2003, what one would have thought to be the chargeable gain or loss latent in the holding of 08/00 Loan Notes is rolled over into the tax exempt SDLs, and, on disposal, any gain is exempt under section 115.[28]HMRC’s case is that this is not the correct way to apply the statutory provisions. On their proper application there are two conversions. The first is the conversion of the 08/00 Loan Notes into SDLs. This first conversion is one to which limb B of section 116(1)(b) applies, because it is a conversion of non-QCBs into QCBs. Sections 127 to 130 are disapplied by section 116(5). The conversion does not fall within section 116(9), and so section 116(10) applies. This has the effect that the chargeable gain is frozen, and there is a charge to tax on that gain upon ultimate redemption.[29]The second conversion is a conversion of the Revised 03/01 Loan Notes into SDLs. That conversion is not within section 116(1)(b) as QCBs are being exchanged for QCBs. However, it is common ground that section 116 applied on the earlier removal (on 9 October 2002) of the dollar redemption provision from the 03/01 Loan Notes, converting those notes from non-QCB to QCB status. Accordingly, on the redemption of the SDLs, the frozen gain on the 03/01 notes would be brought into charge as well.

The decision of the FTT

[30]The FTT recognised that the starting point for the application of section 116 was that the transaction in question must be one to which, apart from the application of section 116, sections 127 to 130 would apply. The FTT also recognised that sections 127 to 130 “set the scene” for the construction of section 116. Considering the reorganisation rules, the FTT concluded that there was no bar in principle to a conversion of securities being a single conversion encompassing a conversion of more than one class of security into a different security. Turning to section 116 itself, the FTT considered that there was “an unfortunate mismatch” between section 116(1)(b), which used the words “consist of or include” and which would appear to permit the new holding or original shares not to consist entirely of a QCB and subsections (3) and (4) which used the words “constitute”, which would appear to require QCBs alone.[31]In the end however the FTT favoured the Hancocks’ construction and that section 116(3) and (4) had to give way:
“47. In our judgment s 116(3) and (4) should be construed so as to apply both where the original shares or the new holding comprised only the QCB, and where the original shares or the new asset merely included a QCB. Only in this way could effect be given to circumstances that s 116(1) makes clear are intended to be governed by s 116. Given the meaning of “original shares” and “new holding” within s 126, as modified for s 132 purposes, the true construction of s 116(3) and (4) is, in our view, to encompass any QCB that, respectively, forms part of the description “original shares” or “new holding”, whether or not there is another asset included within the same description in respect of the same reorganisation or conversion.”
[32]The FTT was, however, very clear that this construction did not give effect to Parliament’s intention, and that the construction which it favoured represented an unintended loophole. In paragraph 52 and 53 of its decision the FTT said:
“52. Although, in the light of [counsel for HMRC’s] argument, we have taken the view that Parliament cannot have intended to allow the non-QCB element of a conversion of securities into QCBs to escape taxation, that in our judgment is the effect of the clear words of s 116(1)(b). Whilst, as we shall describe in more detail later, the approach to be taken is one of purposive construction, that does not mean that we can ignore the clear words, and seek to re-write legislation based on what may be discerned as the true result intended by Parliament. We do not consider that s 116(1)(b) can be construed otherwise than on its own terms; it is the legislative expression of what Parliament enacted as the scope of s 116, and no purposive construction can fill the gap created by the fact that certain circumstances that might be thought to have been intended to be within s 116 fall outside it according to the clear words of s 116(1)(b). 53. That this leaves a loophole in s 116 TCGA is something we have considered in reaching our conclusion in this respect. But we do not consider that the language of s 116(1)(b) admits of an interpretation that can avoid what may be perceived as an injustice or absurdity (see Harding v Revenue and Customs Commissioners [2008] STC 3499, per Lawrence Collins LJ at [51] and the cases there cited). So far as s 116 is concerned, this is, in our judgment, the case like Revenue and Customs Commissioners v Bank of Ireland Britain Holdings Limited [2008] STC 398, at [44] where an anomaly cannot be avoided by any legitimate process of interpretation.”
[33]Accordingly, the FTT concluded that in the present case there was a single agreement for the conversion of the 08/00 Loan Notes and the Revised 03/01 Loan Notes into the SDLs and neither of the conditions in section 116(1)(b) was met. The transaction therefore avoided CGT on the gain which was latent in the 08/00 Loan Notes.[34]The FTT also dismissed an alternative argument advanced by HMRC that the exchange on 7 May 2003 and redemption on 30 June 2003 were to be viewed as a single transaction, applying the principle in W T Ramsay Ltd v IRC [1982] AC 300.

The decision of the UT

[35]The UT considered that the starting point for the analysis was not section 116(1)(b) but rather section 132. This meant that one did not start by asking whether section 116 applied to the transaction, but by asking what the transaction was. Although the word “transaction” in section 116 can encompass more than one conversion of securities, that was not so where the route to section 116 was through section 132. Such a construction would be inconsistent with section 116(2) which provided that references to a transaction include “references to any conversion of securities (whether or not effected by a transaction) within section 132”. The UT also noted that in section 116(2) it is provided that references to “relevant transaction” meant a conversion of securities or other transaction such as is mentioned in subsection (1). These words were a clear pointer that the transaction referred to in section 116(1) was the conversion of securities as defined in section 132 and not some broader transaction that could include one or more such conversion. The remainder of the UT’s reasoning is as follows:
“39. …Any other construction creates the question that has generated the difficulty in this case namely: how is one to decide whether a series of closely connected conversions of securities is to be treated as one transaction for the purposes of section 116 or as more than one transaction and if more, then how many? The absence of any statutory mechanism for answering that question enables the Hancocks to assert that it is a question to which they can choose the answer by drafting either one or more contracts. We do not accept that that can have been Parliament’s intention and it is a not a construction compelled by the wording of the provisions. 40. We therefore hold that the ‘transaction’ referred to in those opening words of section 116(1) (as expanded in section 116(2)) is intended to be the conversion for the purposes of section 132, where the relevant route in to section 116(1)(a) is section 132. Each ‘conversion’ for the purposes of section 132 is a different ‘transaction’ for the purposes of section 116(1). 41. The next question is therefore whether there was one or more conversion of securities in this case. For the answer to that, one must go to section 132. Section 132 covers ‘the conversion of securities’. That term is defined, non-exhaustively, as including a conversion of a security which is a non-QCB into a security which is a QCB (s 132(3)(a)(ia)) and a conversion of a security which is a QCB into a security which is a non-QCB (s 132(3)(a)(ib)). 42. We agree with HMRC that it is significant that the examples given in section 132(3)(a) only encompass what they call unmixed conversions. We consider that section 116(3) and (4) are also pointers to conversions only being unmixed conversions. 43. In our judgment, on the proper constructions of sections 132 and 116, each original single asset or single security should be treated as the subject of a conversion whenever this is provided for by section 132. This accords not only with the natural meaning of the wording in section 132 but with the overall structure of the provisions in sections 126, 132, 135 and 136 for the rollover of gains in the case of reorganisations, conversions, takeovers and schemes of arrangement. The common feature of all these provisions is that they are addressing the situation in which there would be a chargeable disposal of some asset for capital gains purposes, and their purpose is to nullify that disposal but then to attach the latent gain or loss in respect of the "original shares or securities" to the new asset or assets that then represent the original shares or securities. 44. The provisions can only operate sensibly if there is separate treatment of each asset that might otherwise have been the subject of a disposal or part disposal. We fully accept that on any form of reorganisation, conversion etc, the new holding might well be composed of two or more shares or securities. The legislation has always contemplated this and dealt with the two different ways in which the gain or loss should be calculated if there is a later disposal of only one of the two or more new assets that replaced the original asset. But since the various provisions are designed to nullify a disposal of the original asset, they must, in our judgment, be applied separately by reference to that asset, and not by reference to some composite aggregation of two or more assets which might have quite different acquisition dates and costs.”
[36]In relation to the drafting of section 116(1)(b), the UT said this at paragraph 55:
“We acknowledge the conundrum that is posed by this provision. We do not need to assert that there is no situation in which the notion of a mixed conversion advocated by the Hancocks could apply without being inimical to the structure of the legislation. It may be that we have failed to identify that situation. All that we say is that on the facts of this case our application of the reorganisation rules leads to the conclusion that there were two separate conversions at the relevant stage of the transactions in this case.”
[37]The UT therefore allowed HMRC’s appeal. The UT also rejected HMRC’s secondary, Ramsay argument and there is no appeal from that part of the UT’s decision.

Approach to construction

[38]In Luke v Inland Revenue Commissioners [1963] AC 557, the House of Lords was faced with a section of the Income Tax Act 1952 which, if read literally, “would in many cases defeat the obvious intention of the section”. Lord Reid said at 577:
“How, then, are we to resolve the difficulty? To apply the words literally is to defeat the obvious intention of the legislation and to produce a wholly unreasonable result. To achieve the obvious intention and produce a reasonable result we must do some violence to the words. This is not a new problem, though our standard of drafting is such that it rarely emerges. The general principle is well settled. It is only where the words are absolutely incapable of a construction which will accord with the apparent intention of the provision and will avoid a wholly unreasonable result, that the words of the enactment will prevail.”
[39]In Mangin v Inland Revenue Commissioners [1971] AC 739 at 746, Lord Donovan made three important points about the approach to construction of tax legislation. Firstly, he said that the words were to be given their ordinary meaning, not some other meaning simply because their object or effect was to frustrate tax avoidance. Secondly, citing Rowlatt J in Cape Brandy Syndicate v Inland Revenue Commissioners [1921] 1 KB 64 at 71 he said: “… one has to look merely at what is clearly said. There is no room for intendment. There is no equity about a tax. There is no presumption as to tax. Nothing is to be read in, nothing is to be implied. One can only look fairly at the language used”.[40]Thirdly, he said:
“… the object of the construction of a statute being to ascertain the will of the legislature it may be presumed that neither injustice nor absurdity was intended. If therefore a literal interpretation would produce such a result and the language admits of an interpretation which would avoid it, then such an interpretation may be adopted.”
[41]In Ramsay v Inland Revenue Commissioners [1982] AC 300, Lord Wilberforce explained at 323 C to D:
“What are “clear words” is to be ascertained upon normal principles: these do not confine the courts to literal interpretation. There may, indeed should, be considered the context and scheme of the relevant Act as a whole, and its purpose may, indeed should, be regarded.”
[42]More recently in Billingham (Inspector of Taxes) v Cooper, Edwards (Inspector of Taxes) v Fisher [2001] EWCA Civ 1041; [2001] STC 1177 at paragraph 35, Robert Walker LJ said:
“Whatever the difficulties the court has to do its best to make sense of the statute, and that means not only making grammatical sense of the text but also finding a rational scheme in the legislation. That is not to say that the court should start off with preconceptions about what it expects to find, or that it should shrink from saying so in the rare case where a tax statute has “plainly missed fire” (the expression used by Lord Macmillan in IRC v Ayrshire Employers Mutual Assurance Association (1946) 27 Tax Cas 331, 347). But as Viscount Simon LC said in Nokes v Doncaster Amalgamated Collieries [1940] AC 1014, 1022 (which was not a tax case, but has often been cited in tax cases): “ … if the choice is between two interpretations, the narrower of which would fail to achieve the manifest purpose of the legislation, we should avoid a construction which would reduce the legislation to futility and should rather accept the bolder construction based on the view that Parliament would legislate only for the purpose of bringing about an effective result.”” “ … if the choice is between two interpretations, the narrower of which would fail to achieve the manifest purpose of the legislation, we should avoid a construction which would reduce the legislation to futility and should rather accept the bolder construction based on the view that Parliament would legislate only for the purpose of bringing about an effective result.””
[43]In cases of plain drafting mistakes, in order to give effect to the obvious intention of Parliament, the court can go beyond resolving ambiguities in language, and can omit or insert or substitute words: Inco Europe Limited and others v First Choice Distribution (a firm) and others [2000] UKHL 15; [2000] 1 WLR 586. The court is not thereby exceeding its interpretive role but merely making as much sense of the statutory provision in context as it can. The power to proceed in that way is limited to cases where the court is, as Lord Nicholls explained in that case:
“abundantly sure of three matters (1) the intended purpose if the statute or provision in question, (2) that by inadvertence the draftsman and Parliament failed to give effect to that purpose in the provision in question; and (3) the substance of the provision Parliament would have made, although not necessarily the precise words Parliament would have used, had the error in the Bill been noticed. The third of these conditions is of crucial importance.”
[44]In Jenks v Dickinson (Inspector of Taxes) [1997] STC 853, Neuberger J (as he then was) was addressing the predecessor CGT legislation to that with which we are concerned. Having set out the then current provisions for dealing with exchanges of QCBs for shares in the context of a reorganisation, he said at page 867:
“It will be seen that the intention of the legislature in enacting Sch 13 was to ensure that in a case where qualifying corporate bonds were exchanged for shares (and vice versa) in the context of a reorganisation, any capital gain enjoyed in relation to the bonds was not to be charged to tax, but any gain enjoyed on the shares was to be charged to tax. Where the old asset was qualifying corporate bonds, the exchange of those bonds for shares was deemed to be a disposal (which did not give rise to tax because any gain on the qualifying corporate bond was to be exempt); when the shares were sold, the gain on the shares would be assessed by deducting the net proceeds of sale on the disposal of the shares from the cost of acquiring the shares, which was the value of the bonds at the time they were exchanged for the shares. However, where the old asset was shares, which were exchanged on the reorganisation for qualifying corporate bonds, that exchange was not treated as a deemed disposal of the shares. On the disposal of the bonds, capital gains tax was chargeable only on the gain made on the shares; in order to calculate that gain, one had to deduct the cost of the acquisition of the shares (in the normal way) from the consideration received for the shares, which would be the value of the bonds at the date of the exchange. Thus, there was a deferred chargeable gain on the shares which accrued on the exchange and which crystallised on the disposal of the bonds.”
[45]Neuberger J considered that both the taxpayer’s and the Revenue’s cases gave rise to an anomaly. The major anomaly on the taxpayer’s construction was that it was:
“… contradictory to the evident purpose of the relevant statutory provisions, viewed as a whole, viz that capital gains made on qualifying corporate bonds should be exempt from tax, whereas capital gains made on shares should be subject to tax. In the circumstances, principle, common sense, and authority show that the court is ‘entitled, and indeed bound, to… adopt some other possible meaning’ if it exists (to quote Lord Reid (see [1963] AC 557 at 579, 40 TC 630 at 648)).”
[46]Neuberger J also made observations in Jenks about the correct approach to deeming provisions, noting the following passage from the decision of Peter Gibson J (giving what amounted to the judgment of the Court of Appeal) in Marshall (Inspector of Taxes) v Kerr [1993] STC 360, which had the express approval of Lord Browne-Wilkinson in the House of Lords:
“… I take the correct approach in construing a deeming provision to be to give the words used their ordinary and natural meaning, consistent so far as possible with the policy of the Act and the purposes of the provisions so far as such policy and purposes can be ascertained; but if such construction would lead to injustice or absurdity, the application of the statutory fiction should be limited to the extent needed to avoid such injustice or absurdity, unless such application would clearly be within the purposes of the fiction. I further bear in mind that because one must treat as real that which is only deemed to be so, one must treat as real the consequences and incidents naturally flowing from or accompany the deemed state of affairs, unless prohibited from doing so.”
Arguments on this appeal The appellants’ submissions

Arguments on this appeal

[47]Mr Sherry for the taxpayer, whilst realistically accepting that the arrangements made by the taxpayers were structured in significant part in order to avoid the incidence of CGT, nevertheless emphasised that the agreement for conversion of the 08/00 Loan Notes and the Revised 03/01 Loan Notes was part of a wider commercial settlement between the Hancocks and Lionheart. It was therefore realistic to regard it as a single transaction and not several conversions.[48]Section 127 was there to deal with reorganisations which could involve the holder of shares starting with one or more type of share and finishing with a holding of other shares and debentures. It was not right to regard reorganisation as a series of conversions. It was a process by which one set of holdings was converted into another.[49]The list of conversions in section 132 was not exhaustive, see for example Klincke v Revenue & Customs Commissioners [2009] SFTD 466 at paragraph 48. The references in section 132 to “securities” contemplate aggregation of securities through the use of the plural case. Section 132(1) was referring to the process of conversion rather than applying to each individual security.[50]Mr Sherry stressed that section 127 dealt not only with conversions within section 132, but also with exchanges within section 135 and reconstructions within section 136. All these could involve a set of holdings on the one side and a different set on the other side. Nothing within these provisions required one to apply the provisions separately to each class of share held before the reorganisation, conversion, exchange or reconstruction.[51]Section 116(1) was clear, and its meaning could not be overridden or displaced by reference to other provisions such as 116(3) and (4). The conundrum identified by the UT at paragraph 55 was not a conundrum: the drafting of 116(1) fitted with the scheme of the reorganisation provisions. The Revenue’s construction of section 116 was not clear, and as the upshot of the section was a charging provision, section 116(10), it offended against the principle that the taxpayer was only to be taxed on clear words, and not the intendment or equity.[52]It was accordingly inapt to describe the transaction entered into by the Hancocks as two transactions for the purposes of section 116. Section 116 was the gateway into the operative provisions, and it was closed on the taxpayers’ construction to a transaction of the type entered into on 7 May 2003.

The respondents’ submissions

[53]Mr Michael Gibbon QC, who appeared for the Revenue with Ms Elizabeth Wilson, stressed the purpose behind the legislation at various levels. At the highest level, the purpose of the legislation was to tax gains. The purpose of the reorganisation provisions was to ensure that tax was deferred but ultimately payable when there was a disposal for cash. The QCB regime was grafted onto this basic architecture, but was only intended to exempt gains on QCBs. He relied heavily on the judgment of Neuberger J in Jenks from which I have quoted at paragraphs 44 to 46 above. The result, if the taxpayers were right, flatly contradicted the scheme of the legislation which was to tax gains on non-QCBs.[54]When considering the interaction between section 116 and 132 one started with section 132, as the UT had stressed at paragraph 37. The direction in section 132 is to apply section 127-130 “with any necessary adaptations”. If one starts with a conversion within section 132 of a single class of securities, one would read sections 127-130 accordingly. It was significant that section 132 referred only to unmixed conversions and not to mixed conversions, i.e. those where the old or new holding includes both QCBs and non-QCBs. The taxpayers’ argument ignored the fact that in general the singular includes the plural and vice versa. Consistently with that, section 116 was framed so as to operate on unmixed conversions as well.[55]In broad terms, therefore, Mr Gibbon supported the essential reasoning of the UT.

Discussion

[56]It is convenient when describing the various types of transaction and conversion involving shares and QCBs to use the terms “input side” and “output side” to describe the two sides of the transaction and the terms “input” and “output” to describe what goes into the transaction and what comes out of it, even though the statute does not use these terms.[57]Section 116(1) begins by saying that the section:
“… shall have effect in any case where a transaction occurs of such a description that, apart from the provisions of this section– (a) sections 127 to 130 would apply by virtue of any provision of Chapter 2 of this Part;”
[58]The first enquiry is therefore to determine whether the transaction is of such a description that sections 127 to 130 apply to it. Sections 127 to 130 do not apply directly to a transaction such as that entered into by the Hancocks, as those sections require shares on the input side and shares (or debentures) on the output side. The Hancocks’ transaction involved QCBs and non-QCBs on the input side and not shares. It is worth observing at this stage that, in the reorganisation context, in contrast to the situations envisaged by section 116, there is no question of the shares on the input side being subject to different tax treatment i.e. of having mixed taxable and non-taxable assets on that side. The shares are all taxable assets, and the effect of section 127 is simply to defer the accrual of the gains or losses on them all. It is therefore understandable that, in the context of a reorganisation of share capital, the term “original shares” defined in section 126 permits aggregation of holdings of different classes of share. To put it another way, the input side of a reorganisation of share capital is fiscally homogeneous.[59]When this judgment was circulated in draft I considered that there was a further difference between a section 127 reorganisation and a transaction such as that entered into by the Hancocks, namely that a reorganisation did not allow for the possibility that the fiscal character of the shares on the input side could be different from the fiscal character of the shares or debentures on the output side. It was, however, pointed out to us by both parties on the receipt of the draft judgment that, because of the mention of debentures as a possible component of the output side of the reorganisation, there existed the possibility of the output side consisting of or including QCBs. In other words, it is possible that the output side of a reorganisation can be fiscally mixed. Both parties agree, however, that the input side of a reorganisation can only include shares and is therefore fiscally homogeneous.[60]It is the case therefore that sections 127 to 130 do not apply directly to the transaction under consideration. However, Section 116(1)(a) does not only refer to sections 127 to 130 applying to the transaction in question, but refers to those sections applying by virtue of any provision of Chapter 2 of Part IV of the TCGA. There are several sections in Chapter 2 of Part IV which apply sections 127-130 to transactions other than a reorganisation of share capital. These include section 132, as well as section 135(3) (exchange of securities for those in another company), section 136(2) (scheme of reconstruction involving issue of securities).[61]Thus section 132(1) provides for the application of sections 127 to 131 “with any necessary adaptations” in relation to the conversion of securities “as they apply in relation to a reorganisation…”. Section 135(3) applies those sections “with the necessary adaptations” to exchanges of securities in company A and company B “as if company A and company B were the same company and the exchange were a reorganisation of its share capital”. Section 136(2)(b) applies the sections “with the necessary adaptations” in relation to schemes of reconstruction involving the shares or debentures of company A and company B “as if company A and company B were the same company and the exchange were a reorganisation of its share capital”.[62]It is clear from section 132(3)(a) that a “conversion” of securities may be of taxable securities (in the sense that they are securities capable of giving rise to a chargeable gain or loss) and of non-taxable securities, namely QCBs, which are exempt from CGT under section 115. This is a respect in which conversions differ from reorganisations, which, because they only involve shares on the input side will always be concerned with taxable securities. This difference immediately raises the question of whether, in a conversion of securities as distinct from a reorganisation, it is permissible to aggregate securities together, so that the input side of a conversion may include both taxable and exempt securities. I will refer to this question as that of whether the term “conversion” allows for mixed inputs.[63]As I have said, when applying section 127 to reorganisations to which that section applies directly, there is no difficulty with allowing aggregation of different classes of holding on the input side because, for tax purposes, all the shares will be treated in the same way. In applying section 127 to conversions of securities, however, it is not clear that it is justified to aggregate securities together, when different classes of security will need different tax treatment. It is important to bear in mind that we are to apply sections 127 to 130 (and therefore the definitions of section 126) with “necessary adaptations”. It does not follow therefore that one must simply substitute the word “conversion” for “reorganisation” wherever the latter appears.[64]Thus, in section 126, the “original shares” to which section 127 refers means, at least in the context of a reorganisation, “shares held before and concerned in the reorganisation”. This clearly permits aggregation. However this definition is not apt when applying section 127 to a conversion of securities within section 132. A conversion is not, as Mr Sherry submits, an overall process (as a reorganisation plainly is). In a reorganisation, the overall process of reorganisation necessarily involves all the shares on the input side. There is just one reorganisation. By contrast, if one starts with two assets which meet the definition of securities in section 132, it does not follow at all that there is just one conversion. Each of the assets on the input side is converted to something else. It makes perfect sense to speak of each asset being involved in its own conversion rather than both assets being involved in an overall conversion.[65]This takes one back to section 132. Although section 132 is not an exhaustive list of possible conversions, it does not list any conversions with mixed inputs, and particularly no conversions with inputs with mixed fiscal status. That fact fortifies the conclusion that term “conversion” does not allow for mixed inputs.[66]Even before one comes to section 116, therefore, it would be apparent that the two types of conversion, which are specifically picked out in section 132(3)(a)(ia) and (ib), may need to be treated separately and not aggregated. That approach, in my judgment, can be justified as a “necessary adaptation” as contrasted with the application of section 127 to shares in a reorganisation, where everything put into the reorganisation has the same taxable status, and where aggregation is permissible.[67]When considering whether this is the right approach, it is in my judgment legitimate to examine the impact of permitting the aggregation of fiscally different inputs on the operation of the scheme as a whole. If the Hancocks’ argument is correct, transactions which include QCBs and non-QCBs on the input side and QCBs on the output side are excluded from the operation of section 116 through the operation of section 116(1)(b). The effect would be that the chargeable gain which has accrued on the non-QCB would escape CGT altogether. This would be, as Neuberger J put it in Jenks, “… contradictory to the evident purpose of the relevant statutory provisions, viewed as a whole, viz that capital gains made on qualifying corporate bonds should be exempt from tax, whereas capital gains made on shares should be subject to tax…”[68]The result so achieved would not be possible if one understands the combined effect of sections 132 and 127 as requiring one to treat conversions of QCBs and non-QCBs separately because of their different tax status. Section 116(1)(b) does not then block the application of the section to those separate conversions.[69]I recognise that limb A of section 116(1)(b) expressly countenances a “transaction” in which the input “includes” a QCB and therefore allows for a fiscally mixed input. However the use of the different term “transaction” is significant. Section 116(1)(b) is not saying that a conversion may include mixed inputs. That a conversion may include mixed inputs is an essential link in the taxpayer’s argument.[70]A further essential link in the taxpayers’ argument is that the deeming of section 127 applies to a conversion with mixed inputs. It is this deeming which allows the Hancocks to say that the gain inherent in their 08/00 Loan Notes is rolled over into the tax exempt SDLs. The effect is, again, to cause one to step back and ask whether it can have been intended to allow conversions with mixed inputs. To allow the deeming provision to be applied in that way would defeat the “policy and purposes” of the TCGA in the way identified in the passage from Marshall v Kerr which I have referred to above. The statutory fictions introduced by section 127 require restriction in order to avoid this obviously unintended result.[71]I accept that the drafting of section 116(1) is anomalous, and Mr Gibbon did not advance any explanation of what situations the broader notion of “transaction” was intended to cover. However, once the overall scheme is understood, the point on section 116(1)(b) is exposed for what it is, namely an isolated drafting anomaly inconsistent with the rest of the scheme. To allow section 116(1)(b) to upset the rest of the scheme in the way propounded by the taxpayer is to treat a simple drafting anomaly as determinative of the proper interpretation of the statutory scheme, in a way which is completely contrary to its overall purpose. The fact that one is left wondering as to the purpose of the additional words in the sub-section is not fatal to the Revenue’s construction. As Mr Gibbon submitted, no “counter-mischief” is identified. To accept the taxpayer’s construction exposes a far greater anomaly, namely that by the simple expedient of structuring a transaction which mixes QCBs and non-QCBs in any proportion, chargeable gains on non-QCBs escape CGT altogether.[72]It is not therefore necessary to scrutinise the later sub-sections of section 116 to see whether it operates on the basis of unmixed conversions only. The UT thought that there was some support for their interpretation in section 116(2). That sub-section points out, firstly, that a reference to a transaction includes a reference to any conversion of securities whether or not effected by a transaction. It is true that those words were introduced by amendment in order to make it clear that conversions which took effect by effluxion of time were covered: see the explanation in Harding v Revenue and Customs Commissioners [2008] EWCA Civ 1164; [2008] STC 3499 at [27] to [29]. It does not follow that this was not the meaning of the section before the amendment, however. The sub-section points out, secondly, that “relevant transaction” means “a reorganisation, conversion of securities or other transaction”. That does not take the argument much further, but there is certainly nothing in section 116(2) which is inconsistent with the transaction in question being, in each case, the conversion and with that term excluding mixed inputs.[73]The language of sections 116(3) and (4), which refers to the QCB “constituting” the original shares or the new holding respectively, with the consequence that the QCB is either the “old asset” or the “new asset” is also consistent with the idea that transactions involving QCBs are not mixed. The language does give rise to a further puzzle, however. As I have mentioned, the output of a reorganisation, where section 127 applies directly, may include QCBs. This gives rise to at least a suspicion that the draughtsman considered the word “constitute” to be capable of including mixed assets. Alternatively he may have overlooked the possibility. It is not likely to be fruitful, therefore, to pursue this anomaly further.[74]I do not overlook the fact that section 127 is also called into play by sections 135 and 136 (exchanges and reconstructions), which may involve more than one asset on the input side. I note however that section 137 has an anti-avoidance provision which applies to exchanges and reconstructions within sections 135 and 136. Avoidance schemes based on those sections, as opposed to section 132, may have further obstacles in their way.[75]Drawing this together, in my judgment the correct approach to conversions of securities within section 132 is one in which conversions are not permitted to have mixed inputs. The UT was right to treat the gateway provision as section 132 and to apply it in this way, and therefore prevent the deeming effect of section 127 leading to a result not intended by Parliament. Accordingly section 116(1)(b) does not operate as a barrier to the application of the remainder of the section. The Hancocks’ transaction falls to be treated as two separate conversions in the manner contended for by the Revenue. Indeed I did not detect any real dispute that if that was the correct approach to sections 132, 127 and 116, then that would be the consequence.[76]Mr Sherry submits, however, that the transaction which the taxpayers entered into on 7 May 2003 was a single transaction, which, in addition to effecting a conversion of the loan notes, also settled certain commercial matters. It resulted in a single tranche of SDLs. It therefore cannot, as a matter of fact, be split into its individual components. The transaction therefore has to be treated as one in which QCBs and non-QCBs are converted into QCBs.[77]I cannot accept that argument. Once it is accepted, as I do, that the statutory scheme is intended to operate differently on the two types of conversion, it cannot be open to the taxpayer to avoid the operation of the scheme by the way he structures the transaction. To do so would be a triumph of form over substance.[78]I would therefore dismiss the appeal.[79]I agree that the appeal should be dismissed for the reasons given by Lord Justice Floyd and Lord Justice Lewison.[80]I agree that the appeal should be dismissed for the reasons given by Floyd LJ. Since our reasoning differs to some extent from that of the UT I add some short observations of my own.[81]Although it appeared from the decision of the UT and the skeleton arguments before this court that the dividing line between the parties was whether the process of interpretation began with section 116 or section 132 it became clear during the hearing that it was common ground that the correct starting point was section 132.[82]Section 132 applies sections 127 to 131 to conversions of securities “with necessary adaptations”. Necessary for what? The answer to that question is, in my judgment, in two parts. One part is to make adaptations necessary to give effect to the policy underlying the Act as a whole and, in particular, to that part of the Act which deals with corporate reconstructions. The other part is to make adaptations necessary to make sections 127 to 131 work in the same way as they work for corporate reconstructions.[83]For these purposes the key feature of sections 127 to 131, as they apply to corporate reconstructions, is that the fiscal character of the assets on what Floyd LJ has called the input side is that they are chargeable or fiscally homogeneous. The effect of the “rollover” provisions is that a latent charge to CGT is merely deferred: it does not disappear. If sections 127 to 131 have a different effect on a latent charge in relation to a “mixed conversion” of securities then those sections will operate in relation to conversions of securities in a quite different way. That would subvert the evident intention of Parliament.[84]A number of judges have described the policy underlying the fiscal treatment of QCBs. In Jenks v Dickinson (Inspector of Taxes) [1997] STC 853 Neuberger J referred to:
“… the evident purpose of the relevant statutory provisions, viewed as a whole, viz that capital gains made on qualifying corporate bonds should be exempt from tax, whereas capital gains made on shares should be subject to tax.”
[85]In Harding v Revenue and Customs Commissioners [2008] EWHC 99 (Ch); [2008] STC 1965 Briggs J put it thus:
“The key to the successful operation of the frozen gain regime is that it displaces the ordinary rule that no chargeable gain or allowable loss occurs on the disposal of the QCB.”
[86]Is it possible to arrive at an interpretation that gives effect to the twin objectives of: i) Making sections 127 to 131 work for conversions of securities in the same way that they work for corporate reconstructions and ii) Give effect to the evident intention of Parliament?[87]I believe that it is. The steer is given by section 132(3) which sets out categories of “unmixed conversions”. If there is an unmixed conversion then the fiscal character of the assets on the input side of the conversion is fiscally homogeneous. If the old asset is a QCB there is a deemed disposal under section 116 (9), but no charge to tax. In any other case the latent gain is preserved by section 116 (10). That accords with how the scheme is meant to work.[88]Section 116 (3) and 116 (4) apply the definitions of “old asset” and “new asset” only to conversions which “constitute” unmixed conversions. The application of those definitions does not apply to mixed conversions. Section 116 (1)(b) cannot, in my judgment, be regarded as clear; not least because as the FTT itself pointed out there is a “mismatch” between that sub-section and sections 116 (3) and (4). In Luke v IRC [1963] AC 557 Lord Reid said:
“To apply the words literally is to defeat the obvious intention of the legislation and to produce a wholly unreasonable result. To achieve the obvious intention and produce a reasonable result we must do some violence to the words. This is not a new problem, though our standard of drafting is such that it rarely emerges. The general principle is well settled. It is only where the words are absolutely incapable of a construction which will accord with the apparent intention of the provision and will avoid a wholly unreasonable result, that the words of the enactment will prevail.”
[89]Given the choice between, in effect, ignoring words in section 116 (1) (b) and writing words into sections 116 (3) and (4), the FTT chose at [47] to write words into the latter two sub-sections. Since, as the FTT itself concluded, that reading did not give effect to the evident intention of Parliament I consider that the FTT made the wrong choice. I consider that the approach described in Luke enables us to disregard the phrase “or includes” in section 116 (1) (b) in the circumstances of this case. In my judgment therefore the potential gain held within the non-QCBs was frozen on the conversion and did not disappear in a puff of smoke. I, too, would dismiss the appeal.

Cited in 2 later judgments