Land Securities Plc v Revenue & Customs [2011] UKFTT 599 (TC)

FTT-Tax

Land Securities Plc v Revenue & Customs

[2011] UKFTT 599 (TC) · 2011-09-14

[1]This was an ingenious scheme designed to create an allowable capital loss of £200,415,181 for the Appellant’s accounting period ended 31 March 2003. The Appeals were against the decision of HMRC to deny the entire loss. The transactions required to effect the scheme all occurred between 27 March 2003 and 25 September 2003, though since the loss could not be entirely utilised in the relevant period, this case involved formal appeals, all governed by the same points, for various later periods as well.

[2]In late 2002, the Land Securities group was conscious that its recent repayment of capital had left the group both short of finance to fund various planned property acquisitions, and also at risk of having its credit rating with the various rating agencies downgraded. It therefore entered into discussions with several banks with a view to trying to address these two concerns.

[3]One of the banks approached was Morgan Stanley. Morgan Stanley suggested a scheme to Land Securities that claimed to provide finance (albeit only very short-term finance) for Land Securities’ proposed property acquisitions, in the form of a joint venture company between the two groups, into which Morgan Stanley would inject the vast proportion of the required finance in the form of a capital contribution. The other, and in reality the far more material and realistic benefit of this scheme, was that it was claimed that it would generate a capital loss for Corporation Tax purposes for the Appellant in an amount geared to the capital provided, namely £200 million.

[4]Very briefly the proposal was as follows. The Appellant would first re-name a company that had been some form of group “name protection” company, the only 9 shares of which the Appellant had owned for years, and would call that company LM Property Investments Limited (“LMPI”). The rights attaching to 41 of the then authorised but unissued 91 shares were then changed. The 41 shares were re-classified as B Ordinary Shares (“the 41 B shares”), and they were given rights to dividends and distributions in relation to, their capital, any premium at which they were issued, and also to any capital contribution specifically made in respect of them. The 9 shares remained unclassified, and their rights were not changed, but it was suggested and expected that the effect of the change to the rights of the 41 B shares would be that capital contributions made in respect of the 9 shares would likewise confer value only on the 9 shares, and not on the 41 B shares. Another Land Securities subsidiary, Ravenseft Properties Limited (“RPL”) then subscribed the 41 B shares at a premium of £3.75 million; the Appellant made a capital contribution of £1.25 million in respect of the 9 shares, and the Appellant then sold the 9 shares to a Morgan Stanley Cayman Isles company, Morgan Stanley Canmore Limited (“Canmore”) on 31 March 2003, simultaneously granting Canmore a put option for £1 to put the shares back on to the Appellant at any time within the next 12 months at market value. The expectation was that Canmore, having acquired the 9 shares, would make a capital contribution of £200 million to LMPI. LMPI would then ostensibly rank as a joint venture property investment company between RPL and Canmore, with each shareholder having rights to appoint equal numbers of directors.

[5]Whilst the Appellant claimed that this method of funding the joint venture vehicle was attractive to the Land Securities group from a rating agencies standpoint, since it did not technically rank as debt, the Appellant conceded that the principal objective of the transactions was to create a capital loss. This was to be achieved by the re-acquisition of the 9 shares within a six-month period of their disposal, pursuant either to the exercise of Canmore’s put option, or the exercise of a call option that was granted to the Appellant after the initial steps. The re-acquisition price was to be market value, obviously reflecting the £200 million capital contribution that it was assumed Canmore would have made to LMPI prior to the re-acquisition of the 9 shares.

[6]The scheme was designed, from a tax point of view, to exploit the point, under section 106 TCG Act 1994, that if (as in fact occurred) the Appellant re-acquired the 9 shares within the six-month period after their disposal to Canmore, the Appellant would be required to match its 31 March disposal of the 9 shares for just £1.25 million, not with its historic acquisition cost of £9, but with the price paid on the re-acquisition. Since by that time the value of the 9 shares had been enhanced by the capital contribution made by Canmore, such that the re-purchase price was £202,265,179.50 the Appellant claimed a capital loss of the excess of that amount over the figure of £1.25 million.

[7]Not surprisingly, HMRC challenged the transaction. The ways in which the challenges were mounted are too complex to explain in this Introduction, but very shortly the first argument was that the disposal of the 9 shares should simply be ignored on Ramsay principles. The second argument, which became rather more involved during the course of the hearing, was that the value shifting section, namely section 30 TCG Act, operated to deny the loss. The possible application of section 30 revolved around the provision in sub-section 30(9). This sub-section modified the usual feature that section 30 applied in relation to the receipt of tax-free benefits,coupled with reductions in the value of assets, to apply as well to “increases in the value of assets” where disposals preceded acquisitions. The Appellant contended that the disposal of the 9 shares on 31 March was a disposal of the shares that it had owned for years, and not a disposal of shares that it would later acquire, so that sub-section 30(9) was not in point. In due course we considered two different contrary arguments, pursuant to which the value shifting section might be engaged. We concluded that the section was in point. That led to the final issue of what reduction, if any, would be “just and reasonable”. In this context the Appellant contended that if its loss was reduced to nil, this would not be reasonable, since on any view it would have a latent gain on the re-acquired shares in LMPI since shares then worth approximately the price paid on their re-acquisition would be left with the hitherto unallocated base cost of £9. Our conclusion was that the whole loss should be disallowed. The facts in more detail

[8]In summarising the transactions, it will make it easier for a reader to understand the steps if we add some comments as to why particular features of the transactions doubtless existed. We are adding these comments not because they are of much relevance to the decisions that we have reached, but simply because they will facilitate an immediate understanding of the scheme.

[9]The Appellant had owned the 9 issued shares (the only issued shares) out of the 100 authorised shares of LMPI for many years. The company appeared initially to have been used simply for “name protection” purposes, which resulted in the fact that it had had no apparent activity. The rights attaching to the 9 shares were not technically altered, and they were periodically referred to as “the unclassified shares”.

[10]The rights attached to 41 of the unissued shares were changed in the manner that we have described shortly in the Introduction, and that we will have to revert to below. The 41 B shares were then subscribed by RPL on 27 March 2003 as we have described. The Appellant also made the modest capital contribution on the same date in respect of the 9 shares. One of the obvious objectives of RPL in subscribing the 41 B shares, was that this subscription would explain how it was that the company remained theoretically a “joint venture” company, once the 9 shares had been sold by the Appellant to Canmore, and it would also explain why it was that both the Land Securities group and the Morgan Stanley group would have the right to appoint equal numbers of directors to the Board of LMPI.

[11]In passing, we comment that the reason why: · the Appellant chose to use its original 9 shares in LMPI for the purposes of the following steps; · the Appellant endeavoured to change only the rights attaching to the 41, rather than the 9, shares when changing the share rights to ensure that capital contributions made by the holder of either the 9 or the 41 shares would enhance the value of only the shares in respect of which the capital contribution was made; and · the Appellant boosted the value of the 9 shares by making a capital contribution which of course did not enhance its base cost for those 9 shares for capital gains purposes, rather than subscribe additional shares, was all presumably geared to enhancing the defences against a possible Ramsay contention by HMRC in relation to the steps to which we will now turn, and of course to avoid a further acquisition of shares for section 106 purposes. In other words it was obviously considered that the risks of a Ramsay attack succeeding were diminished if the Appellant at the next step disposed of shares, and theoretically “unchanged shares” that it had held for a long time, and not shares deliberately acquired for the purposes of the scheme.

[12]On 31 March 2003, the Appellant sold its 9 shares to Canmore for £1.25 million; the Appellant granted Canmore a put option for £1 to put those shares back to the Appellant at market value at any time before 29 February 2004, and pursuant to their share rights both RPL and Canmore appointed 2 directors to the Board of LMPI.

[13]Again it is worth noting that at this stage the Appellant had no call option to re-acquire the 9 shares. This doubtless reflected the desire, again from a Ramsay perspective, to avoid the feature of matched puts and calls, and was obviously feasible at this point. This was because Canmore, principally interested in being given a banking style exit for its purchase, needed the put, but until there were properties in LMPI, there was no need for the Appellant to have a call.

[14]Once RPL, as property adviser to LMPI, had identified properties that it was suggested that LMPI should purchase, and a funding requirement of a further £200 million, several key transactions occurred on or very shortly after 1 August 2003. They were that: · the Appellant acquired a call option over the 9 shares for a premium of £1.4 million, that premium to be paid as soon as Canmore had provided the envisaged further funding to LMPI, regardless of the form of that funding; · the price payable under both Canmore’s put and the Appellant’s call was to remain the market value of the 9 shares, subject to a further payment to Canmore of 3.5% of the amount of any capital loss eventually established to result from the relevant transactions; · the Appellant and a different Morgan Stanley company, Morgan Stanley & Co International Limited (“MS International”) entered into a cash settled forward agreement, the effect of which was that the Appellant would pay MS International an amount equal to the difference should the initial payment under the put or call fall short of the figure of £202,741,491, whereas in the reverse direction, MS International would pay the difference to the Appellant if the price under the options exceeded that same figure; · Canmore contributed £200 million to LMPI, this amount being attributed to the 9 shares; · a guarantee of various obligations was granted to Canmore by the Appellant and a security was also granted to Canmore over the 41 B shares held by RPL; and · LMPI loaned the £200 million to the Appellant, at interest.

[15]A few days later LMPI used the £4.5 million (which had not been lent up to the Appellant) resulting from the Appellant’s initial capital contribution and from RPL’s share subscription to purchase a property in Sunderland.

[16]On 20 August, the Appellant repaid £35 million of the loan-up referred to at the last bullet point of paragraph 14, and LMPI acquired (via an inserted company) a fully-let property referred to as 120 Cheapside and 4-9 Wood Street.

[17]On 9 September, the Appellant exercised its call option, and the price payable, ostensibly geared to the value of the 9 shares, was fixed by the directors of LMPI at £202,265,179.50.

[18]On 22 September, the Appellant repaid the balance of the up-stream loan, including interest, to LMPI in the amount of £166,544,235 such that LMPI’s assets consisted of cash, the Sunderland property and the company that directly owned the Cheapside/Wood Street property.

[19]On 25 September the 9 shares were transferred back to the Appellant, and the Appellant paid the top-up figure of £476,317.50 to MS International under the agreement referred to at the third bullet point of paragraph 14 above. We were told that the Appellant used a separate loan facility, and not the money contributed into LMPI, to fund this re-purchase of the shares, in part for “financial assistance” reasons. The points in contention and the brief contentions of the parties

[20]As indicated in the Introduction, the Appellant claimed, having identified its disposal of the 9 shares on 31 March 2003 for £1.25 million with its reacquisition of them in September, to have realised a capital loss. It accordingly treated its base cost as the price paid on re-acquisition of the shares plus the £1.4 million paid for the call option.

[21]In passing we might again note the rather odd, and fairly incidental, point that even if the put option had been exercised, a provision stated that if the call was then subsequently exercised, the call would trump the put. This was doubtless designed to ensure that the £1.4 million price given for the call was added to the base cost of the 9 shares on their re-acquisition.

[22]As also indicated in the Introduction, HMRC challenged the scheme on two different grounds.

[23]The first was on the Ramsay type basis that it was inevitable that the shares disposed of would be re-acquired, and the various terms, notably the forward agreement mentioned in the third bullet point of paragraph 14 above, indicated that in reality Morgan Stanley did not have any real interest in, or risk in relation to, the property investments that might be put into the claimed “joint venture” vehicle. Canmore and MS International would inevitably “exit” at a banking-type figure geared to the funds provided, an implicit and rather costly interest rate, and (having regard to the premium paid for the call option and the contingent further amount payable should capital losses be established) a fee for delivering the capital losses.

[24]The Appellant defended the Ramsay attack on two different bases. It conceded that the scheme was a tax avoidance scheme, but it claimed that there were commercial features to the scheme, and that the fact that Canmore provided finance of £200 million, albeit in the event only for a six-week period, and that that finance was provided in an attractive manner, meant that the scheme was not just a pure tax scheme. Evidence was given by Mr. Martin Wood (“Mr. Wood”), the Group Head of Tax & Treasury and Insurance for Land Securities Group PLC, in relation to these commercial points and we will consider these points in the course of summarising his evidence.

[25]Secondly, however, the Appellant contended that even if it had chosen not to advance these points, it was still impossible, when seeking to identify the realistic nature of the transaction, to ignore all the very real steps that had been effected. In the Agreed Statement of Facts, for instance, the Respondents had accepted that Canmore became the beneficial owner of the 9 shares when they were transferred to it on 31 March. Canmore had plainly contributed £200 million into LMPI; that amount was indisputably an asset of LMPI; LMPI had lent that amount to the Appellant, with the former receiving interest, and the latter paying interest, and when the shares were re-acquired they were worth the price paid for them. And that price was a very different price from that at which the shares had earlier been disposed of because certain real transactions had occurred and those transactions could not possibly be ignored.

[26]In addition to contending that on a Ramsay challenge HMRC could not disregard all these points, the Appellant also pointed out that the identification rule of section 106 TCGAct was designed to counter “bed and breakfast” transactions, and so was designed to accept (and then modify) the form of transactions that were designed to achieve tax advantages. Accordingly, when construing the statute in a purposive manner it was difficult to discern a purpose to the effect that the section was meant to apply only where the parties had no tax advantage motive. The section was designed to operate in just that very scenario, and it was impossible to draw a distinction between those tax avoidance purposes where HMRC accepted that the provision was meant to apply, and different tax advantages that HMRC might find less acceptable, such that they contended that the provision was not meant to apply.

[27]In the event that we dismissed the Ramsay challenges, the next issue was whether section 30 TCGAct applied, so as to diminish or eliminate the loss.

[28]It was common ground between the parties that a scheme had been effected whereunder, either before or after the disposal of an asset (the 9 shares disposed of by the Appellant on 31 March 2003), a tax-free benefit (the £200 million capital contribution) had been conferred on a party with whom the Appellant was connected (LMPI). Section 30 also commonly requires there to have been a material reduction in the value of the asset disposed of (or in the value of various other identified assets), and naturally in this case there had been no such material reduction in value. Sub-section 30(9) provides, however, that:

“(9) In relation to a case in which the disposal of an asset precedes its acquisition the references in subsections (1)(a) and (2) above to a reduction shall be read as including a reference to an increase”

. If this provision applied in relation to the Appellant’s disposal of the 9 shares on 31 March 2003, it was then common ground that all of the pre-conditions to the application of section 30 would have been satisfied because there would have been a very material increase in the value of the 9 shares, coupled with the required tax-free benefit. The question would then be whether, and by how much, it would be just and reasonable to increase the consideration for the Appellant’s disposal of the 9 shares, having regard to the scheme and the tax-free benefit in question, and the Appellant’s latent gain of roughly £201.25 million in respect of the re-acquired shares.

[29]Prior to turning to that latter issue, the critical first question is whether sub-section 30(9) applies at all in this case. The Appellant contended that it did not because, addressing the issue of whether the disposal of the 9 shares “preceded their acquisition”, that condition was not satisfied. The disposal of the 9 shares was a disposal of shares that the Appellant had held for many years; it was definitely not a disposal in a “bear” transaction where the taxpayer disposed of an asset, and that asset was only acquired after the disposal. The sub-section did not refer to a disposal preceding a “re-acquisition” of the asset disposed of. The very word “its” in the phrase that “the disposal of the asset had to precede “its” acquisition” meant that if the disposal was of an asset that the taxpayer had acquired and held prior to the disposal, section 30(9) did not apply, and therefore section 30 could not be invoked by HMRC.

[30]At the commencement of the hearing, both parties agreed that we could not reach the conclusion that the disposal preceded the acquisition, for the purposes of sub-section 30(9), by relying on the identification rule in section 106. This apparent consensus was based on the assumption that both the logic and the precedent of Park J’s judgment in the case of Davies v. Hicks

[2005] STC 850

compelled and bound us to pay regard just to the actual facts, in considering the application of section 30(9), and not to the fact that section 106 had treated the 31 March disposal as being a disposal of the shares which were acquired in September.

[31]The argument proceeded initially therefore along the lines that the Appellant contended that since the Appellant had in fact disposed of shares that it had acquired many years before the disposal, and that the disposal of the shares did not, as a matter of plain ordinary language, “precede their acquisition”, the critical sub-section did not bring section 30 into play. By contrast HMRC contended that on a proper interpretation of section 30(9), the sub-section did still apply in the present circumstances. HMRC’s argument was that when the sub-section commenced by stating that “ In a case in which the disposal of an asset precedes its acquisition”, the opening words focused attention on the transactions involved in the particular set of transactions. Thus, because the Appellant’s scheme was all geared to the disposal on 31 March and the re-acquisition in September, and because when focusing attention on those two transactions, the disposal preceded the acquisition (“ the acquisition, rather than “ its” acquisition, perhaps being the more appropriate word to use when explaining this contention), the sub-section did apply. Paraphrasing HMRC’s contention in a rather more extreme fashion, just to make the point contended clear, the contention was effectively that the words “In a case” effectively enabled or required one to read the sub-section as if it had read as follows: “Where a situation or a scheme involves particular transactions, and amongst those transactions there is a disposal and an acquisition of the same asset and the disposal precedes the acquisition ……….” 32 As indicated, the basis on which the parties had agreed that we were precluded from treating sub-section 30(9) as applying by virtue of the identification provision of section 106, was that this was suggested to result from the decision of the Special Commissioners and Park J. in the Davies v. Hicks case.

[33]The facts of that case were that an individual held a substantial holding of shares in a listed company, all standing at a gain, and he wished to dispose of at least part of the holding. He accordingly transferred the shares to a Discretionary Trust on a hold-over basis, and the resident trustees then disposed of the shares in the market. Prior to the re-acquisition of the shares, the resident trustees resigned and were replaced by Mauritius resident trustees. These trustees then re-purchased the shares, and in various later transactions sold parcels of shares. The effect of the equivalent capital gains provision to section 106 with which we are presently concerned identified the disposal by the resident trustees with the later re-acquisition by the Mauritius trustees, such that there was no material gain. In the ordinary way, the gains later realised by the Mauritius trustees, as they disposed of parcels of shares, would have been calculated by reference to the original historic base cost (so occasioning a large gain), and would generally have been traced through to the resident settlor and beneficiary. The then terms of the Mauritius Double Tax Treaty, however, operated to exempt the resident settlor and beneficiary from the charge to tax.

[34]HMRC accepted all these results but then contended that because there was an exit charge and a deemed disposal of the assets of the settlement when the settlement became non-resident, and because the resident trustees had not been treated as having disposed of the shares contributed into the settlement by the settlor, implicitly those shares were still the settlement property, so that the exit charge occasioned a chargeable disposal of those shares, and the realisation of a large gain. In fact of course, the settlement held no shares at the point when the UK trustees resigned and the Mauritius trustees were appointed because, at that point, the settlement either held the cash proceeds of the initial market disposal, or an indebtedness owed by the broker. In neither event would there be a chargeable gain.

[35]The decisions of the Special Commissioners and of Park J. were both to the effect that section 106 was just a computation provision and it did not have the wider effect of deeming shares to be held by the settlement for the purposes of the exit charge (anyway of course a quite different charging provision), when in fact at that relevant point the settlement held no shares. The Special Commissioners also remarked that the effect of section 106A was not to turn the disposal by the resident trustees into a “bear” transaction, in other words a transaction where the trustees were effecting a disposal of shares that they did not own. They did in fact dispose of the shares that had been settled, and it was only for the limited purposes of computation that the disposal was artificially treated as being of the shares later acquired by the settlement (i.e. by the Mauritius trustees).

[36]We indicated at an early point in the hearing that although HMRC was content to advance its contention under section 30(9) by reference solely to its “In a case ….” argument, and conceded that Park J’s decision bound us to ignore the provisions of section 106 in applying section 30(9), we were far from convinced that this was correct, and far from convinced that the Davies v. Hicks case was relevant to the present situation. At the end of the hearing, therefore, we asked both parties to send us their written submissions in relation to our suggestion that section 30(9) might be brought into operation precisely because section 106 treated the Appellant’s disposal on 31 March as being matched with its re-acquisition of the shares in September. It seemed to us that the very calculation of the loss that was in contention resulted entirely from that matching, and that, contrary to the situation in Davies v. Hicks, we were not giving section 106 any wider or different effect than the one (that of identifying the September re-acquisition with the 31 March disposal) that it was manifestly meant to have.

[37]In the context of requesting contentions in relation to this point, we did undertake to the Respondents that we would still address what they said remained their preferred contention, namely their “In a case …” argument.

[38]We therefore have to attend to five different points. The first two relate to the Respondents’ Ramsay contentions. Firstly, we have to consider whether there is any justification to the Appellant’s claim that the tax scheme did achieve some commercial advantages. Having clarified the facts we then have to consider the Respondents’ claim that the disposal should simply be ignored. The third and fourth points are the two contentions for bringing section 30(9) into play. The fifth point is of course the measure of adjustment, should section 30 as a whole be in point, and bearing in mind in particular that the Appellant, if denied its claimed loss in full, would still have a latent chargeable gain of approximately £201.25 million in respect of its re-acquired 9 shares in LMPI, subject to later value movements.

[39]There was, however, a sixth point that we also raised in the course of the hearing. HMRC’s counsel asked us to ignore this point because it had not been advanced at all by HMRC, and we agreed to ignore the point. We will, however, mention it now, not to raise it in any way, but to illustrate two features that seem to be relevant to other aspects.

[40]We have already mentioned that when RPL subscribed the 41 B shares on 27 March 2003, the rights attaching to those shares had been changed to confer the following rights on the relevant shares:

“(a) The right to participate in dividends, returns of capital or other distributions of LMPI, including without limitation on a winding up, by reference in each such case, to the aggregate of: (i) the nominal capital paid up on each B Ordinary Share, (ii) any premium paid up on issue or payable as a term of the issue of a B Ordinary Share, and (iii) any further capital contributions designated by the directors as referable to each Ordinary B Share, proportionately to the like aggregate contributions made with respect to each share in LMPI. (b) Save as set out in (a) above, the B Ordinary Shares ranked pari passu in all respects with the existing shares. (c) No dividend, return of capital or other distribution might be made with respect to the B Ordinary Shares unless a dividend, return of capital or other distribution was also made with respect to the other shares in the capital of LMPI proportionately in accordance with (a) above.”

[41]The point that had seemed to us to be slightly unclear was whether, when Canmore made the capital contribution of £200 million to LMPI, this contribution did, as was obviously intended, increase the value of only the 9 shares, rather than all the shares in the ratio of 41 to 9 (82% to 18%). We raised this point because the rights attaching to the 9 shares had not been changed in any way and so they did not provide for capital contributions made by the holder of the 9 shares to confer the contributed value just on the 9 shares. We assume that the argument would have been that the change to the share rights of the B shares alone did achieve the desired effect, even though the rights attaching to the 9 unclassified shares were left unchanged, because of the last line of paragraph (a) above, and more particularly the clear restriction on the B share rights contained in paragraph (c) above. For present purposes, we are not concerned to decide whether the 9 shares were indeed worth just in excess of £200 million when reacquired, or alternatively only 18% of that figure.

[42]The two reasons why we have even referred to this drafting point are as follows. Firstly, when two capital contributions were in fact made, and intended to confer value on the 9 shares (that is the contribution of £1.25 million on 27 March, and the contribution of £200 million shortly after 1 August), the distinctly odd way in which the draftsman went out of his way to change the rights attaching to the 41 B shares, and to leave the 9 unclassified shares ostensibly in their absolutely unchanged state does reflect the concern that a Ramsay contention might be rendered more problematic if a technical change had been made to the rights of the 9 unclassified shares.

[43]Secondly, we do suggest that had anything (other than the success of the tax scheme) depended on being absolutely certain that the £200 million capital contribution would have conferred value only on the 9 shares, we are far from clear that the purchaser of those shares would have been content to rely on the proposition that the restrictions on the 41 B share rights to which we have referred would definitely ensure that the 9 shares were worth the amount obviously intended following the making of capital contributions in respect of them. Quite apart from the point that we have already referred to, it certainly appeared that the very brief drafting terms would have dealt inadequately with the situation of capital contributions being made after the existing net worth of the company had been increased by retained income and unrealised gain in respect of properties. None of this, however, mattered because what was absolutely obvious was that so far as Canmore and MS International were concerned, the price at which the Morgan Stanley group would exit from LMPI had absolutely nothing to do with properties or the value of shares. So the arguments in relation to share valuations and share rights were relevant only to the efficacy of the tax scheme, and the fixation about the Ramsay risk, rather than commercial reality. The commercial reality was manifestly that the Morgan Stanley group as a whole was entirely unaffected by whether properties were acquired by LMPI or not, and by whether if they were, they might fall or rise in value. The effect of the forward agreement mentioned at the third bullet point of paragraph 14 above was that the Morgan Stanley group would exit at a price that would reflect the capital provided to LMPI, an interest equivalent return, plus a fee (some of it outright, and some of it contingent) for the provision of the tax scheme. Mr. Wood’s evidence

[44]Mr. Wood was the only witness to give evidence, and his evidence was given essentially in relation to the background to the scheme, and to the suggested commercial benefits occasioned by the scheme. Mr. Wood was subjected to extensive and skilful cross-examination by Mr. Ghosh, all designed to establish that the Morgan Stanley scheme was a pure tax scheme, and that it achieved no commercial, balance sheet, or rating agencies benefits.

[45]We should immediately record that Mr. Wood was an impressive and plainly honest witness. Indeed he was candid to the point that when Mr. Ghosh asked him to concede that particular points had been advanced in support of the commercial benefits suggested to be achieved by the scheme, and that on examination those benefits had emerged to be largely illusory, he repeatedly conceded that he accepted Mr. Ghosh’s points. Many others in his position would have persisted in suggesting that there was some, if only minor, underlying validity to the particular points in issue.

[46]Mr. Wood’s evidence in relation to the origin of the scheme, and the group’s problems at the time the scheme was planned was clearly genuine. Mr. Wood described that when he first joined the group, shortly before the events in question, he soon realised that the management was seriously concerned that its recent repayment of capital to shareholders had left the group under-capitalised, that its various credit ratings for secured and unsecured debt with the various rating agencies were under stress, and that it was far from clear that the group could actually raise the finance to fund the property acquisitions planned in its business model. We were told that a particular proposed acquisition had fallen through on account of an inability to fund the acquisition, and Mr. Wood described the way in which this had sent reverberations around the group.

[47]We also accept that Mr. Wood then approached several bankers with a view to discussing financing possibilities, and that it was in that context that Morgan Stanley suggested their scheme.

[48]Neither Mr. Wood, nor Mr. Gardiner in making his submissions, denied that once the Morgan Stanley scheme had been suggested, and once it was appreciated that the scheme might generate a capital loss of £200 million, that became a very central feature of the scheme, and of the group’s interest in the scheme.

[49]In challenging any commercial benefits of the scheme, it is reasonable to summarise that Mr. Ghosh for the Respondents demolished many of the claimed commercial benefits of the scheme. Mr. Wood accepted, for instance, that although the capital contribution method of providing finance into a company that theoretically ranked as a joint venture company involved no strict borrowing, the existence from the outset of Canmore’s put meant that the Appellant might immediately have to borrow to fund the exercise of the put, thus having an immediate need to borrow £200 million once Canmore’s capital contribution had been injected. Even ignoring the specific payments for the tax benefit (the £1.4 million premium that the Appellant paid for its call option, and the contingent further 3.5% that would be added to the put prices if the capital losses were established), the implicit borrowing costs at 200 basis points over LIBOR, contrasted with the group’s more usual 75 basis points margin was very costly, and thus disadvantageous. Although the group was troubled about a possible downgrade in its credit rating, and indeed it suffered such a downgrade during 2003, Mr. Ghosh suggested to Mr. Wood that the rating agencies were not concerned about liquidity, or the group’s inability to fund its acquisition programme. So far as we could tell, the reason for the downgrade was indeed more related to sensitivities about demand in the London commercial buildings letting market, and at the diminished prospect for healthy rent reviews. Mr. Ghosh suggested, and Mr. Wood largely accepted, that when properties were acquired by LMPI in August 2003, there was an element of wanting to get the properties into the company in order to support the tax planning. When asked whether the group would have effected the scheme if no property acquisitions had been in point, and the £200 million had been left on deposit, Mr. Wood said that he did not know, because that issue had never been considered. Mr. Ghosh also established that when the Board of the Appellant had documented their decision to exercise the call option, that decision was really dictated by the feature that the re-acquisition had to be made before the end of the 6-month period from 31 March, and that other ostensible reasons advanced for exercising the call were largely make-weight or fictitious.

[50]Whilst Mr. Ghosh did thus demolish many of the claimed commercial benefits of the scheme, we still consider that the following factors put a slightly different complexion on the transactions than any suggestion that the property acquisitions were just window dressing for a tax scheme.

[51]There was no doubt that the group was planning the property acquisitions prior to its discussions with bankers. It also very much sounded as if the Cheapside property would have been acquired at an earlier date than mid-August but for some delays in that transaction, and as if other property acquisitions that had been planned, for which the £200 million contribution would have provided temporary funding, either fell through or were delayed. Once the Appellant had embarked on the scheme, it certainly wanted to get the properties into LMPI before the put was exercised, but it is equally certainly the case that the property acquisitions were genuine, and that the business plan to acquire the properties pre-dated the tax scheme. The tax scheme was thus an expensive way (or a cheap way, if the tax scheme succeeded) of funding the property acquisitions; those acquisitions were not just “window-dressing” for a tax scheme. We also accept the claim that the Morgan Stanley appointed directors of LMPI contributed valuable property expertise from a somewhat different standpoint than the traditional Land Securities approach, when attending Board Meetings of LMPI, though we imagine that this assistance could have been secured independently of the scheme.

[52]We also accept, as Mr. Gardiner submitted, that the scheme did produce funding for a six-week period of £200 million. Once the capital contribution had been injected, there was no prospect that Canmore would immediately exercise the put, so that the Land Securities group could assume that the funding was in place until late September. Indeed had there not been the unavoidable delays, and problems, in acquiring other properties in accordance with the business plan, it rather seemed that the Cheapside property would have been acquired before mid-August, and other planned property acquisitions would have been made as well. Had that been so, the funding would presumably have been in place for much more of the six-month window period.

[53]Having given that very general summary of the evidence, however, it may be appropriate to mention two points. One is that it was the case, and the Appellant did not remotely deny this, that the Morgan Stanley scheme was, first and foremost, a tax avoidance scheme. Secondly, as we will indicate shortly, we attach relatively little importance, in considering the Ramsay challenge to the scheme, to these suggested and, in the case of many of them, largely demolished, commercial justifications for the scheme. The Ramsay issue

[54]We will turn now to the Ramsay issue, and first to the contentions of the parties. The contentions of the Appellant

[55]The Appellant contended that the approach that we should take was to apply the law in a purposive manner, and apply that application of the law to the facts, realistically ascertained.

[56]The first proposition, thus, was that section 106 had been inserted into the legislation to counteract transactions where there were disposals but nevertheless the taxpayer fundamentally wished to retain the asset rather than dispose of it. The section was thus designed to acknowledge, and then reverse, disposals designed to generate some tax benefit, and it could not be disapplied merely on the ground that HMRC objected to some tax advantages more than others.

[57]Although LMPI had been a dormant company, prior to being activated to undertake the transactions in this case, the 9 shares that were the subject of the disposal had not been acquired for the purposes of the scheme. They had been owned by the Appellant for many years.

[58]In analysing the facts realistically, it was impossible to arrive at any other analysis than that: · the 9 shares had been transferred beneficially on 31 March to Canmore, as the Agreed Statement of Facts accepted; · until 1 August the Appellant had no call option to re-acquire those shares; · the Board of LMPI operated genuinely as a Board of the joint company; · Canmore did make a capital contribution of £200 million into LMPI, this contribution being entirely financed by Morgan Stanley companies and in no way by the Land Securities group; · some properties were acquired by LMPI, and LMPI plainly owned the assets (properties plus loan to the Appellant or cash at bank) that reflected the entire net worth of LMPI; · until the Appellant re-acquired the 9 shares, the Appellant had in no way funded LMPI; and · when the 9 shares were re-acquired, that purchase was made with funds derived from a separate borrowing, and had nothing to do with the asset base of LMPI.

[59]It was notable that HMRC had agreed, in the Agreed Statement of Facts, that “pursuant to the terms of the Share Sale Agreement, on 31 March 2003 [the Appellant] sold the [9] shares to Canmore for a consideration of £1,250,000 paid in full on that date, and Canmore became the legal and beneficial owner of those shares on that date.” Having regard to the recent clarification that the Ramsay and equivalent doctrines have always only involved matters of interpretation, and have rescued tax law from an isolated application of strict literal interpretation, so conforming tax interpretation to all other principles of statutory interpretation, it was impossible for HMRC to accept that the 9 shares were transferred legally and beneficially to Canmore, and then to contend that they were not.

[60]In considering, thus, whether the facts, analysed realistically, supported the contention that the 9 shares had been disposed of to Canmore, in particular for the purposes of a section that contemplates somewhat artificial disposals where in reality the taxpayer intended to acquire identical shares immediately, or re-acquire the shares disposed of, it was impossible to ignore all the reality of the steps summarised in paragraph 48 above, and the conclusion had to be that the shares had been disposed of. The contentions on behalf of the Respondents

[61]It was contended on behalf of the Respondents that: · Section 106 was a provision “intended to apply to prevent taxpayers from crystallising losses that reflected a genuine fall in the value of an investment where a “bed and breakfast” transaction meant that there was no real change in the economic ownership of that investment; it does not apply, and was not intended to apply where the very creation of an asset, its sale, repurchase and the shift in value in between are all part of the same composite whole effected in order to create a loss. · HMRC was right not to have sought to disapply section 106 in the case of Davies v. Hicks. “Why? Because there were shares which Mr. Hicks always owned and frankly it is not Mr. Hick’s problem that 106 operated to shelter a gain rather than get rid of a loss”. · Where shares have been created for the very purpose of a tax avoidance exercise, disposed of only to be reacquired (having been artificially enhanced in value) the whole series of circular (self-cancelling) transactions should be ignored, since a realistic view of the facts means that, quite simply, there has been no disposal, no reacquisition (and no capital contribution by Canmore). · There was considerable significance in applying the Ramsay doctrine to the issue of whether the transactions related to an existing asset or an asset created or only acquired for the purposes of the scheme. This was emphasised in the case of MacNiven v. Westmoreland Investments

[2003] 1 AC 311

by the significance attached in that case to the feature that the interest that was technically paid under the scheme was genuine interest that had accrued due prior to the scheme even being considered. In this present Appeal, the 9 shares had effectively been created for the purpose of the scheme. · “The monies borrowed by [the Appellant] (from a pre-existing loan facility) to allegedly re-acquire the shares were, on any realistic view of the facts, the monies effectively used by [the Appellant] to finance LMPI’s acquisition of the properties that LMPI acquired”. · “Canmore made a capital contribution of £200 million, which Canmore got back on the exercise of the call option by [the Appellant]. [The Appellant] funded LMPI, not Canmore”. · Finally, we should treat ourselves as bound by the decision in Furniss v. Dawson to disregard the disposal. Our decision in relation to the Ramsay issue

[62]Both parties were at least agreed that the right approach for us to take was to interpret the legislation in question in a purposive manner, and to apply it to the facts, realistically analysed. We will accordingly start with the legislation, and then move to the facts. The purposive interpretation of the legislation

[63]Both parties acknowledged that section 106 and equivalent “matching” or “identification” rules were essentially introduced to undermine certain “bed and breakfast” transactions. Whatever the schemes at which they were principally targeted, there is nevertheless no denying that they created an artificial basis of calculating gains and losses (or, very commonly, neutral results where the matched re-acquisitions would lead to no material gain or loss), and the results occasioned were thus entirely the product of the statutory fictions, bearing no relationship to the actual facts. Even if the most obvious application of the rules might be to undermine artificial realisations of losses, it is perfectly obvious that if taxpayers undertook “bed and breakfast” transactions in order to realise gains that fell below the annual exemption level, so creating a “step-up” in base cost for future realistic disposals of the re-acquired shares, the provisions would manifestly operate to nullify those realisations. That, therefore, is one simple example of the situation where the relevant sections would operate in relation to the realisation of gains, rather than the realisation of losses.

[64]Whilst Mr. Ghosh sought to narrow the application of section 106, to that of the artificial realisation of genuine losses, he nevertheless considered that HMRC had been right not to have challenged the application of the section in the Davies v. Hicks case. We will defer dealing with the whole issue of whether, in applying Ramsay, there is much significance to the issue of whether the asset involved in the transactions had been a newly created asset for the purposes of the scheme, or was an existing asset, and this was of course the essential ground on which Mr. Ghosh suggested that HMRC was right to have accepted the application of section 106 in the Davies v. Hicks case , and also right to dispute it in this case. As we indicated, we will revert to the “created asset/existing asset” point below, but for present purposes we reject Mr. Ghosh’s argument that there is a coherent explanation as to why section 106A was obviously engaged in the Davies v. Hicks case, yet said not to be intended to apply to the facts of this case. Davies v. Hicks was certainly nothing to do with “bed and breakfast” transactions of the conventional variety, but was a case where the settlor of the trust set up the entire arrangements to avoid tax on large capital gains, by reliance on the then terms of the Mauritius double tax treaty. To do this, however, the scheme had to avoid the exit charge that would have applied had the trust emigrated when holding the shares, and this was only achieved by the sale and re-acquisition of the shares, such that the shares were not held at the point of emigration. It was then by unquestioned reliance of section 106A that the market disposal of the shares did not occasion the realisation of the large gain, since section 106A matched the market disposal with the later re-acquisition by the Mauritius trustees, so eliminating the gain. We fail to see that these steps and these intentions are more consistent with the purposes to which the identification rules are targeted than the facts in the present case. We agree that the technical and fictitious matching rules of section 106A were rightly assumed by all (not just held by the Special Commissioners and Park J, following any contention on the point) to be in point in the Davies v. Hicks case, but we consider that section 106 is equally in point in this case.

[65]We consider that Mr. Gardiner was right in his suggestion that in trying to interpret section 106 in a purposive manner, we should only discern that the section was designed to operate in circumstances where the taxpayer was likely to be doing something artificial, quite possibly disposing of an asset that it meant to re-acquire, and the purpose of the section was then to create an entirely fictitious basis of calculating gains and losses. In a sense the point at stake in this context is not that dissimilar to the point that influenced Proudman J. and the Court of Appeal in HMRC v. Mayes

[2011] EWCA Civ 407

where she found it impossible to discern the statutory purpose and intent of another formalistic and formulaic set of provisions, namely the “chargeable event” code for taxing partial surrenders and eventual disposals of insurance policies.

[66]We have yet to look at the facts of this case, but our conclusion on the purposive interpretation of section 106 is that it was a provision that was intended to create a statutory fiction, and it was intended to apply where the taxpayer was likely to be doing something artificial and, very probably therefore, to be disposing of an asset that it meant to re-acquire. If thus we reach the conclusion on the facts, that the Appellant did dispose of the 9 shares on 31 March, we consider that section 106 will manifestly have applied to that disposal as Mr. Gardiner has suggested, and that we cannot disapply the section on the ground that “the section was obviously not meant to apply in that situation”. The issues of whether the 9 shares were pre-existing shares or an asset created for the purposes of the scheme, and whether this distinction is particularly significant

[67]We now commence our Ramsay consideration of the factual analysis of this case by first considering the degree of significance that may attach to the issue of whether the shares allegedly disposed of were shares created for the purpose of the transactions, or existing assets.

[68]Both parties obviously attached some significance to this question because HMRC advanced it as a very material contention in support of their Ramsay case, and the attention that the Appellant gave, in ensuring that there was no technical change to the rights actually attaching to the 9 unclassified shares, confirmed that the Appellant was also conscious that its case might have been weaker, had a “shelf” company, for instance, been purchased and used as LMPI, rather than an existing company. We note the point that the “shelf” company would have been unsuitable in the one obvious respect that its shares would need to have been acquired more than 6 months prior to the disposal of the shares to Canmore in any event, but the Appellant’s sensitivity to using an existing company obviously went beyond that point.

[69]We consider that if there is much significance to the issue of whether the 9 shares were “an existing asset” or “a newly acquired asset”, then on the reality of the situation, the 9 shares looked more like a new and deliberately, albeit carefully, modified asset rather than an existing asset. It is plain that LMPI had previously been a dormant “name protection” company. This obviously achieved the section 106 point that the 9 shares, the only shares in issue until 27 March, had been acquired years before 2003, and not in the preceding 6 months. But the reality, nevertheless, was that until 27 March 2003 the company re-named LMPI had had nothing to do with property investment, or the assets that were to be contributed into it, or indeed the rights that it was claimed came to attach to the shares.

[70]Having said that and having concluded, rather as Mr. Ghosh suggested on behalf of HMRC, that the 9 shares appeared only to be an “existing” asset in a very technical sense, we also conclude that in this case there is little logic as to why the realistic analysis of whether those shares should be treated as having been disposed of to Canmore (or not, on HMRC’s contention) is that influenced by this issue. The origin of the significance attached to this point appears to be the point that influenced the courts in McNiven v. Westmoreland Investments , where it was suggested that the result (of validating the deduction for interest when the interest was fictitiously paid out of a new advance by the lender, with the lender appreciating that although it would receive the interest, it would fail to recover anything in respect of its new loan) might have been different, had not Westmoreland genuinely incurred the liability for the interest prior to the implementation of the scheme. The court thus suggested that if the scheme had involved artificial steps that had extended to creating a wholly fictitious claim for interest relief, rather than merely the technical payment of genuinely accrued interest liability, the result might have been different. Since that would have had a material bearing on the reality of the whole transaction, it was very easy to see why the court suggested that this feature might have been significant.

[71]In the present case, however, we find it difficult to see why there should be much difference in analysing the realistic nature of the steps, all of which occurred on and after 27 March 2003, according to the distinctions that the same transactions might have been effected with: · 9 shares of a “shelf” company; · the 9 shares that had been owned by the Appellant for many years; and · the 9 shares of a company that might have held substantial property investments, prior to the disposal of the shares to Canmore. In all three of those cases, it is perfectly credible that every other step in the Morgan Stanley scheme actually effected, could have been implemented in precisely the way that the steps were actually implemented in this case, whichever of the three situations might have prevailed at the outset. In view of this we attach very little significance to the issue of whether the shares here were sustained as being existing shares, or were more obviously close to “new shares”.

[72]Reverting to the facts of Davies v. Hicks , we repeat that we consider that, for the reasons given in paragraphs 64 and 65 above, it was right that section 106 applied in that case to the disposal by the resident trustees (in whose hands the shares were fairly “newly acquired” shares in any event), and we entirely understand that it was of the essence of that case that the disposal was of assets in respect of which there was a large latent gain. We fail, however, to see that when we come to consider the “realistic appraisal of the facts”, in this case, and we compare it with the “realistic appraisal” of the various transactions in Davies v. Hicks, that the realistic analysis of transactions that all occur after the “starting-point” of the cycle of transactions is that influenced by the near historic accident of whether the shares were an existing asset or a new asset.

[73]Our conclusion in this case, thus, is that the 9 shares disposed of to Canmore were technically an existing asset that had been owned for years, and whose acquisition was years prior to the period, in which the acquisition might have been relevant for section 106 purposes. The 9 shares were also in reality somewhat different in terms of their effective rights and their value, when disposed of to Canmore. But we fail to see that these features of whether they were old shares or new shares (however reconciled) should have much influence on our appraisal of the realistic nature of the transactions effected. The transactions, realistically analysed

[74]We turn now to that vital question. We first draw a distinction between the issue of whether there were unrealistic features to the Appellant’s transactions, and the different and the relevant issue of whether, realistically appraised, the 9 shares were disposed of to Canmore or not.

[75]In a general sense, we of course accept that there were features of the transactions that were unrealistic. No-one considered that a long term property joint venture was being created between the Land Securities group and the Morgan Stanley group. It plainly was not. The terms of the forward agreement made it clear that, on the Morgan Stanley side of the scheme, the objective was to recover a banking-style return for the finance that had been injected into LMPI, plus a return, some of it outright, and most of it contingent, in return for the tax scheme. On the Land Securities side, it was equally clear that the properties that were put into the joint company were properties that had been on Land Securities’ acquisition list, and once properties were to be acquired, the grant of the call option to the Appellant by Canmore revealed that from that point onwards, the Appellant wanted to have, and needed to have, the ability to resume total ownership and total control over LMPI, and the properties owned by LMPI.

[76]These features of unreality are, however, not the issue. When we have to consider whether, realistically appraised, the 9 shares were disposed of to Canmore, we need to weigh up the feature that HMRC appear to have accepted that legal and beneficial ownership of the 9 shares passed to Canmore; we must bear in mind all of the transactions that appeared to have been effected that we summarised, in recording Mr. Gardiner’s contentions at paragraph 48 above, and then we need to consider whether on a realistic appraisal some other version of the facts, advanced by HMRC, is in fact more realistic.

[77]This, it seems to us, is where HMRC’s case founders. HMRC’s counsel never identified precisely what he considered the more realistic analysis of the facts to be. The penultimate two bullet points recorded at paragraph 61 above demonstrate the type of arguments that were advanced. It was contended that it was the Appellant that funded the property acquisitions made by LMPI. It was contended that the transactions were circular, when in most senses they were not, since the shares re-acquired were manifestly very different in reality from the shares initially disposed of. It was contended, at the penultimate bullet point in paragraph 61 above, when it was mentioned that Canmore had made a contribution to LMPI (inconsistently with the proposition that it was the Appellant that had all along funded the property acquisitions) that Canmore “got back” the contribution when the Appellant bought the shares back. And it was then stated that “It was the Appellant (not Canmore) that had funded LMPI”.

[78]There are certain facts in this case that it appears to us that HMRC simply cannot ignore or dismiss. We accept that the Land Securities group would have always appreciated that it would have to use funds, or resort to an existing facility in order to fund the re-purchase the 9 shares. Until the transfer back of the 9 shares, however, we consider that HMRC cannot dispute the reality that LMPI was largely funded by Canmore. It may have been appreciated that matters would have to change fairly shortly, but certainly when the properties were acquired, LMPI was essentially funded by Canmore, and not by the Appellant.

[79]The other absolutely unchangeable fact is that from the moment the capital contribution was paid to LMPI by Canmore, Canmore had assets of approximately £205 million, and it had no debt. It certainly had that asset base when the Appellant re-acquired the shares, and whilst it was superfluous for us to know anything about the later fate of LMPI, we consider it inconceivable that there could have been any re-analysis for any purpose other than that LMPI owned the relevant assets, and that in accounting and company law terms, the bulk of those funds resulted from the making of a capital contribution.

[80]Mr. Gardiner fairly criticised HMRC’s skeleton argument, and the contentions advanced on the Ramsay analysis, for the feature that HMRC never identified what they actually said was the alternative, and the more realistic, appraisal of the facts. The following point was never, as such, advanced on behalf of HMRC, but it seems to us that the only re-analysis that actually fits the unchangeable ultimate facts in October 2003 would be to say that the Appellant owned the 9 shares at all stages, Canmore had not made a capital contribution to LMPI at all; what it had done was to make a short-term loan to the Appellant, and it was the Appellant that had made the capital contribution into LMPI. This is at least consistent with the two unalterable facts that from 4 August 2003 onwards, the net worth reflecting the capital contribution was definitely in LMPI, and it also recognises that until 25 September it was indisputably Canmore that was funding some entity on a short-term basis.

[81]This, however, involves a complete re-writing of the transactions, and it would involve treating the Appellant as having a debt liability that it was plainly not liable for, and the Appellant to have made a £200 million capital contribution to LMPI which, as a resident company, the Appellant would never have dreamt of doing. If we contrast this summary of what seems to us to be the only alternative set of transactions that could have occasioned the end result that cannot be disputed, with the transactions contended for by the Appellant, we have no hesitation in saying that the actual transactions effected represent the only realistic analysis of the facts. On the Appellant’s version of the facts, the parties might well have appreciated that Canmore’s funding of LMPI would be short-term, and it would require Canmore to have a put option to dispose of the shares (and recover its cash, plus its banking-style return), and the Appellant plainly needed a call option to acquire Canmore’s shares once the properties had been acquired. But all those features were present, and those features, coupled with the disposal of the 9 shares to Canmore, Canmore’s capital contribution, and the re-acquisition of the 9 shares, remain the indisputable realistic appraisal of the facts of these transactions.

[82]We should make two further observations.

[83]Firstly we agree with Mr. Ghosh that he succeeded in demolishing most of the commercial advantages that Mr. Wood had claimed for the scheme, such for instance that the form of the transaction secured financing in an attractive manner from a balance sheet perspective, or that it secured any advantages from the perspective of the rating agencies. We even accept that Mr. Wood said to Mr. Ghosh that when the properties were acquired in about mid-August, there was a desire to “get the properties into LMPI” because that would enhance the appearance of the tax scheme. Whilst that is so, it would be a completely wrong summary to say that these transactions were just a tax scheme, and no more, and that the properties were only injected for window-dressing purposes. The initial driver was that the Land Securities group wanted to make the property acquisitions, and it was keen to look at any financing structures that would facilitate those acquisitions. When the tax scheme was suggested, that was undoubtedly attractive, and the main benefit of the Morgan Stanley transaction. None of the transactions with Canmore would have occurred, tax hopes apart. But the feature that the tax scheme blended conveniently with what was Land Securities’ first objective (to acquire the properties, and to fund that acquisition) was the reality. To suggest that this scheme was just a tax scheme, and that the property transactions were “window dressing” was wrong.

[84]Secondly, we agree with Mr. Gardiner that it simply cannot be denied, as Mr. Ghosh sought to do, that the Morgan Stanley group did provide funding of £200 million. It may in the event have been only for a six-week period, and it would doubtless have been for marginally longer had there not been delays and difficulties in effecting the originally planned property acquisitions. It is simply not tenable, however, for HMRC to say that “the Appellant funded LMPI; Canmore did not”.

[85]Our conclusion, accordingly, on the Ramsay point, and on the suggestion that section 106 was irrelevant because there was no disposal, or because on a purposive basis section 106 did not apply to certain transactions directed at achieving particular benefits, is that the Appellant was right to say that there was a disposal and that section 106 applied in relation to it.

[86]It will be noted that we have made no reference to the facts and the rationalisation of the decision in Furniss v. Dawson. That case appears to have been the most extreme case in which the rationalisation of the decision was based on a set of rules that were said to justify disregarding certain steps, and to be the case least consistent with the approach of “realistic interpretation of the facts” that it is now clear that we should adopt in analysing what has occurred in a case that is manifestly a tax scheme. We consider that we have undertaken that analysis in the manner that the House of Lords and the Supreme Court have both advocated that we should in more recent decisions and we consider that it would simply confuse matters to follow a different approach than the one now recommended. The section 30(9) issue, disregarding the argument that the acquisition preceded the acquisition because of the operation of section 106

[87]The question as to whether sub-section 30(9) brought section 30 into operation, whilst disregarding the feature that section 106 had deemed the Appellant’s disposal to have been of the shares acquired in September, revolved around two points of interpretation, one of them concentrating on just one word, and the other, concentrating at most on three words. 87. To repeat it, the wording of sub-section 30(9) provides that: “(9) In relation to a case in which the disposal of an asset precedes its acquisition the reference in subsections (1)(a) and (2) above to a reduction shall be read as including a reference to an increase”.

[88]It was agreed by the parties that if this wording applied, then all the other pre-conditions to the application of section 30 were satisfied. The question would of course remain as to whether any increase to the disposal consideration might be “just and reasonable”, and if so how much, but that apart, section 30 would apply. The contentions on behalf of the Appellant

[89]It was contended on behalf of the Appellant that the reference to “the disposal of an asset [preceding] its acquisition” made it clear that the section applied only to genuine bear transactions where the disposal was of an asset that the taxpayer had yet to acquire.

[90]The Appellant criticised HMRC’s counsel for paraphrasing sub-section (9) to refer to “a disposal preceding an acquisition” or to “a disposal preceding a re-acquisition”, either of which might have lent slightly greater credibility to HMRC’s case that the disposal should be treated as the disposal of the later acquired shares. . Adopting the first less strained re-wording of the sub-section, it might then be said that there was a disposal, and there were two acquisitions. By then stressing the way in which the scheme involved the disposal and the later, rather than the earlier, acquisition, the argument might just get off the ground that the disposal should be matched with the later acquisition. Where, however, the indisputable wording of sub-section 30(9) referred to a “disposal of an asset preceding its acquisition” , the circumstances contemplated by the sub-section simply could not be satisfied when the disposal was of an asset, or the asset, which was actually, and manifestly, owned by the Appellant prior to its disposal. The disposal on 31 March was plainly of an asset that the Appellant owned and had acquired years before 31 March, and so its disposal followed its acquisition. The disposal of the asset simply did not precede its acquisition.

[91]The Appellant sought to support its argument by contending that were HMRC’s construction correct, it could then be invoked in cases where the acquisition or re-acquisition might be years after the disposal, and that could not conceivably be right. The contentions on behalf of the Respondents

[92]The Respondents’ argument was to the effect that everything was clarified by the words “In a case…”. Whether one looked at those words in isolation, or whether one read sub-section 30(9) back into sub-section 30(1) to pick up the notion that the disposal in question must have occurred in the context of a “scheme that had been effected or arrangements that had been made”, the point remained the same. That point was that by virtue of the words “In the case”, or “in the scheme”, one was really addressing the transactions just involved “in the case”, or just the transactions involved “in the scheme”. On either approach there was only one acquisition that was part of the scheme, so that the disposal had to be treated as the disposal of that asset that had been acquired in the course of the scheme, or “in the case”, and therefore the identified acquisition was the later one in September.

[93]HMRC then made the point, which we agreed with, that the Appellant’s objection, referred to in paragraph 91 above was not correct in that HMRC’s contention would only apply to disposals and acquisitions embraced in one scheme, and not to utterly remote later acquisitions that would have nothing to do with the disposal. In any event, it was unlikely in the case of the acquisition years later that the Appellant postulated at paragraph 91 that the other pre-conditions to adjustment in section 30 would be in point. Thus the Appellant’s argument that HMRC’s construction might have unintended consequences was incorrect. Our decision in relation to the section 30(9) issue, disregarding the section 106 argument

[94]We agree with the Appellant’s natural interpretation of sub-section 30(9). It seems to us to be reasonably plain that, ignoring the application of section 106, the wording of section 30(9) can only apply to the facts of a bear transaction, where in fact the asset disposed of is simply not acquired until after the disposal.

[95]If one eliminates from one’s mind the fairly familiar feature that section 106 can often match disposals of assets with later acquisitions, then, ignoring the bear transaction, it is going to be very unnatural to match a disposal, where the asset disposed of was already owned at the time of disposal with some different acquisition. Not only does that appear to be in complete conflict with the natural interpretation of section 30(9) that the Appellant contends for, but it is a fairly unusual and extraordinary proposition.

[96]HMRC’s argument would of course be correct if it were permissible to paraphrase section 30(9) to contain the opening words that we advanced at the end of paragraph 31 above. We suggested there that the wording that would sustain HMRC’s case was roughly along the lines:

“Where a situation or a scheme involves particular transactions, and amongst those transactions there is a disposal and an acquisition of the same asset and the disposal precedes the acquisition ……”

We advanced that wording at paragraph 31 above simply in order to explain and clarify the interpretation that HMRC were contending for, since we considered that in fact it was so unnatural an interpretation that we needed to paraphrase it in this way so that the point contended for would be understood. The reason why we now repeat the suggested wording is that we consider that wording of that nature would indeed have been required in order to make HMRC’s case credible. Plainly our suggested wording goes far further than the simple words “In a case….”.

[97]We consider that sub-section 30(9) is a self-standing sub-section, and until its wording is satisfied, nothing is read back into sub-section 30(1). We are far from convinced that HMRC’s argument would be materially stronger even if it could be advanced in relation to sub-section 30(1)’s wording which refers to “a scheme or arrangement”. Nevertheless in the self-standing sub-section 30(9), the question is whether “In a case in which the disposal of an asset precedes its acquisition”, can refer to, and then effectively only to, the later acquisition, when there was an obvious earlier acquisition. We agree with the Appellant that the words “In a case……” mean “In a case or in a situation where ……”. On that basis, we conclude that the Appellant’s argument on the interpretation of sub-section 30(9) must be correct. This, ignoring section 106, is simply not a case in which the disposal of an asset preceded its acquisition. The section 30(9) issue, paying regard to section 106

[98]As we have already said, both parties accepted that one could not rely on the provisions of section 106 to treat the disposal on 31 March as being identified with the later acquisition in September, when considering whether the situation in section 30(9) was in point. It was suggested to us that this was not only a point governed, and ruled out, by the plain logic of the earlier decision in Davies v. Hicks, but that since Park J. had affirmed the decision of the Special Commissioners on appeal, and Park J’s decision was binding on us, the issue was closed. We indicated on several occasions that we were far from clear that the decision in Davies v. Hicks was relevant to the very limited application of section 106 that we considered to be more than tenable in this case. It therefore seemed highly likely that if we felt unable to accept HMRC’s Ramsay argument, and HMRC’s interpretation argument that we have just dismissed, we would wish to consider whether both parties had been right to suggest that section 106 had to be altogether ignored when considering the possible application of section 30(9). At the end of the hearing, we indicated to the parties that we would add this issue to our considerations and we asked both parties to send us written submission on the point, which they duly did.

[99]We note, with some regret, that the parties will have appreciated, from the overall decision that we mentioned in the last sentence of paragraph 7, and the feature that we have dismissed HMRC’s contentions up to this point that it will be obvious that the basis of our decision in this case is that the Appeal fails, and fails only, because we do consider it appropriate to read section 30(9) by taking into account the identification rule in section 106 (5) (b) that applies in this case. We thus regret that we will disappoint HMRC, which wishes to win this case on other grounds, and we will disappoint the Appellant far more markedly because they would have been content to win the case on any basis.

[100]We deal first with Davies v. Hicks. In that case, section 106A identified the disposal by the UK trustees with the re-acquisition by the Mauritius resident trustees, such that the disposal occasioned no gain. This was not in dispute. The question in issue was whether, beyond making that identification that governed the calculation of gain or loss on the disposal by the UK trustees, the deeming notion of section 106 went further, and deemed the trust to hold the shares at the point when it became non-resident, such that there was then a further disposal of the shares under a quite different charging provision, in which the gain would then be traced from the original low acquisition cost, and matched with the value of the shares that would then be treated as disposed of under the exit charge “deemed disposal”. This argument was quite a stretch on any basis because, quite apart from the fact that, on the true facts, the trust held no shares (but cash or indebtedness from the broker) at the time the trust became non-resident, the argument still had to rely on the logic that if the UK resident trustees’ market disposal had been identified with the Mauritius trustees’ re-acquisition, this meant that the trust’s original acquisition had not been matched with anything, so that it was reasonable to infer (albeit that this was certainly neither stated, nor particularly implicit in the language of section 106A) that the trust must be treated as still owning the shares, resulting from that un-matched acquisition.

[101]Not surprisingly, it was held that section 106A did not have that further deeming effect. It was said to be a computation section and, once the computation, resulting from its application had been made, its effect was spent. It certainly did not have the much broader effect of deeming the trust to hold shares when the trust in fact held no shares at the point of emigration. The Special Commissioners also observed that it was not the effect of the section to make the disposal by the UK trustees (at the point when they sold shares in the market disposal prior to the change of residence) a bear transaction. We agree. At the point of that disposal, the trustees did plainly own shares that they had acquired prior to the disposal. The only effect of section 106A was to identify the disposal with the shares acquired very shortly afterwards by the Mauritius trustees. But nothing changed either of the true facts, namely that: · at the point of the market disposal, the UK trustees did hold shares that they had acquired prior to the disposal, such that there was no “bear” transaction; and that · the result of the market disposal was that the trust held no shares when it became non-resident.

[102]It seemed to us that in this case, when we address the possible application of section 30(9) by reference to the earlier application of section 106, we are not straining section 106 to give it any wider application than it manifestly had. As a matter of description, we would say that section 106 was not so much “a computation section”, as an “identification section”, the effect of which was of course to govern the calculation that would then be made. The wording however does plainly refer to the fact that “For the purposes of corporation tax on chargeable gains, shares disposed of shall be identified” in accordance with various rules, the effect of which in this case was plainly to identify the shares disposed of on 31 March 2003 with those re-acquired in September. When thus addressing sub-section 30(9) it seemed at least tenable, if not manifestly obvious, that when the artificial loss resulted from this identification, it would be permissible to treat the disposal of the asset as having “ preceded its acquisition”. That analysis, after all, had been the prevailing analysis for capital gains purposes, and to consider sub-section 30(9) by matching the disposal with an acquisition (the original acquisition) that had in fact been rendered irrelevant by the identification rule seemed perverse in the extreme. The contentions on behalf of the Respondents

[103]It is convenient to take the Respondents’ contentions first, because they are simpler. The Respondents first observed that the consequence of treating section 30(9) as being satisfied by reliance on the operation of section 106 was entirely consistent with their preferred analysis, and raised no different or problematic point. They secondly conceded that “having considered this point further, HMRC accept that the point made by the Tribunal is perfectly tenable ……. HMRC further accept that the “computational” nature of section 106 does not affect this analysis. Section 30 is a “computational” section in that it adjusts a computation. …. There is nothing in the text of section 106 or section 30 to suggest that the computation effect in circumstances in which section 106 applies excludes the application of section 30. Neither is there any reason in principle to exclude the application of section 30 where section 106 applies.

[104]HMRC then confirmed that they accepted that there was nothing in the Davies v. Hicks decision that conflicted with this analysis, nor indeed was the observation in Davies v. Hicks that section 106 did not occasion a transaction to be regarded as a bear transaction, when it was not, in any way inconsistent with the analysis that we wished to consider.

[105]HMRC concluded by saying that they still advanced, as their preferred analysis, the point of interpretation that we have already dismissed, but they accepted that “the application of Section 106 to the transactions in this case which caused a scheme or arrangement to be effected in the first place also caused Section 30(9) to be engaged as a matter of principle, where the scheme or arrangement causes an asset to be inflated in value”. The contentions on behalf of the Appellant

[106]The Appellant has raised various arguments, and it will be best to comment on each as we summarise them.

[107]The first was a contention that section 106 was just a qualification to the pooling rule of section 104, and that it cannot have any more general purpose.

[108]We accept that section 106 qualifies the pooling rule, but it does of course positively identify which particular acquisition within the prescribed periods before or after the disposal shall be identified with the assets disposed of. This effect plainly leads to a computation, and the question for us is whether precisely the identification provided for by section 106 can also govern the identification which is in issue under section 30(9). What is absolutely clear to us is that we are not seeking to apply section 106 for any purpose other than the identification that plainly it governs.

[109]It was then suggested that section 30(9) raised a “factual threshold question”, and the suggestion was that because the shares originally acquired constituted the pool and were held prior to the disposal, section 30(9) could not apply.

[110]Since however the entirety of the shares in the pool were excluded from it by the rule in section 106 in this case, and the disposal was then identified with the later acquisition of shares, we still need to address the issue of whether section 30(9) applies by reference to the actual facts, or the notional facts that result from the identification made in this case by section 106.

[111]It was next suggested that the construction that we are ourselves considering could lead to “nonsensical” results, in particular because it might be suggested that section 106 could have nullified the effect of deemed disposal provisions in the legislation, such as the provision in section 179. In other words, because on a deemed disposal and re-acquisition at market value, where plainly the intention was that the gain should be calculated from original cost (or, at any rate, not from the re-acquisition figure because of section 106) all the deemed disposal rules would be undermined by section 106. Indeed, much the same point was noted in the Davies v. Hicks case, in that on the occasion of the deemed disposal under the exit charge, there is also a deemed re-acquisition, and if the two were matched, and that was the right analysis, then the exit charge, just like section 179, would be undermined. The Special Commissioners ignored this point in Davies v. Hicks, because, whether there was no exit charge because in fact no shares were owned, or no exit charge because the deemed disposal was matched with the re-acquisition (all at the same value) there was no charge in either event.

[112]Our view is that the theoretical point that all deemed disposals might have been affected by the identification rules is a point that: · has been noted as an oddity for years; · understandably, no taxpayer has ever chosen to advance in seeking to undermine the effect of a deemed disposal; and · this point anyway results simply from the unhappy interplay between the deemed disposal and re-acquisition provisions and the identification rules, and is not influenced by the point that is presently in issue, namely whether section 106’s identification consequence is to be ignored when addressing sub-section 30(9).

[113]The final point advanced was that there was a logical reason why section 30(9) should only be brought into play in the case of a bear transaction, where there was only one acquisition, because were it to be applied in this case, and were the loss to be reduced or eliminated, this would be unfair, and contrary to principle. This would be because the Appellant would not only forfeit its loss, but would be left with a latent gain in respect of the re-acquired shares, since they would inevitably be matched with the £9 base cost on any later disposal. We will have to deal with this point in considering the final questions concerning the measure of the “just and reasonable” adjustment, but for present purposes we simply make the following observations. This “downside” risk of the scheme was noted by Morgan Stanley and Land Securities during the planning stages. It was considered to be of little significance, either because the shares in LMPI could be retained indefinitely, and the latent gain never realised, or alternatively a tax-free dividend could be extracted to reverse the capital contributions, and the same amount (if desired) re-injected as the subscription price for new shares. Were there a concern that such a dividend would be attacked under Part 15 Corporation Tax Act, 2010 (“old section 703”) then if no prior clearance could be obtained, “normal return” dividends might be extracted over time. For present purposes, we simply note that this risk had always been foreseen, and never treated as a particularly material point. Our decision in relation to the section 30(9) and section 106 issue

[114]The question that we must address is whether, when the disposal of the 9 shares of LMPI on 31 March 2003 was matched with those acquired on their later re-acquisition, this identification under section 106 should be ignored, or alternatively addressed, when seeking to apply sub-section 30(9).

[115]The first point that we note is that in our view nothing concluded in Davies v. Hicks has any bearing on this issue. We are not seeking to give any effect to section 106 other than to respect the very consequence specifically provided for by the section, which is that, for the purposes of corporation tax on capital gains, the shares disposed of were to be identified with the later acquired shares. We are not remotely concerned with any wider effect that the section might have, such as whether it would deem shares not held to be treated as being trust property at the point when a settlement became non-resident.

[116]The next point that we observe is that, when it is section 106 that has occasioned the loss, that might or might not be adjusted according to whether sub-section 30(9) is satisfied or not, and when the Appellant’s whole scheme, and every step of it has been occasioned by the identification rule of section 106, it would appear rather odd to address the meaning of section 30(9) by linking the disposal with the historic earlier disposal, when that disposal has had nothing whatsoever to do with the calculation that the section might adjust, and nothing whatever to do with the Appellant’s scheme.

[117]Our next observation is that it showed some foresight on the part of the draftsman of section 30 that sub-section 30(9) was inserted, and we now ask the question of whether there can have been any obvious circumstance in which real bear transactions could have occasioned the type of value increases and tax-free benefits that would make it obvious that that was all that the sub-section was targeted at. This was not discussed during the hearing, and we may have failed to note situations where there could have been obvious schemes involving genuine bear transactions, where but for sub-section 30(9), tax free benefits, mated with value increases could have flourished. We find it difficult to discern what those transactions would have been however, and certainly none are obvious to us.

[118]That thus leaves us considering the words of sub-section 30(9) which refer to “ a case in which the disposal of an asset precedes its acquisition” and needing to decide what “its acquisition” is meant to refer to. Having regard to the facts that: · “the disposal” to which reference is made is obviously a disposal of an asset that is material for the purposes of corporation tax on capital gains; · “the acquisition” or “its acquisition” to which reference is then made in the context of the acquisition material to the disposal referred to seems most obviously to be a reference to the acquisition that was the relevant one in relation to the disposal for capital gains purposes; · in precise terms, section 106 provided in this case that “For the purposes of corporation tax on chargeable gains, shares disposed of shall be identified ……[with] shares acquired after the disposal” (tying together the only words relevant to this case from section 106(1) and section 106(5)(b); and · the Appellant’s entire planning, and the very existence of the large, unrealistic and contentious loss were all geared to the very identification that the Appellant now seeks to brush aside, we reach the conclusion that it is by virtue of the operation of section 106 that the terms of sub-section 30(9) are satisfied in this case. The “just and reasonable” adjustment to the disposal consideration

[119]The starting point to the enquiry as to what adjustment and increase to the disposal consideration is appropriate in this case is obviously that the most obvious answer is that the loss should be entirely eliminated. After all, the whole loss was artificial; there was in reality no loss; and the tax-free benefit received by LMPI was also equal to the entire loss.

[120]We acknowledge that in many situations where adjustments are made under section 30, and gains are re-created, there are matching provisions to prevent a double charge when it would be fair for the value of some other asset to be treated as reduced to avoid a double charge. And because, in this case, we acknowledge that the total denial of the loss would of course leave the Appellant with a latent chargeable gain in respect of the re-acquired shares of roughly £200 million (subject to later value movements), Mr. Gardiner contended that this feature meant that no adjustment should be made. We find that suggestion to be completely untenable, because if the whole loss was then conceded, the Appellant would plainly avoid the realisation of the gain indefinitely, even if it found it difficult to reverse the gain with tax-free dividends, and the result would in practice be precisely as if we had allowed, rather than dismissed, the Appeal.

[121]We are also influenced, in rejecting any suggestion that the reduction of the loss should be diminished on account of the latent gain, by the fact that the risk of this ultimate outcome was one that had been acknowledged by both Morgan Stanley and Land Securities, and neither had counted this risk as particularly material, because the gain could either be frozen and deferred indefinitely, or diminished by dividends. We would certainly confirm that, if this decision was accepted by the parties, or sustained on any appeal that there might be, such that the adjustment that we propose that totally eliminates the loss eventually stands, then we would consider it appropriate and fair that a clearance should be sought and granted under the Part 15 (old section 703 provisions) that we have referred to. We agree in other words that the result of the denial of the loss should, in fairness, be matched by there being no risk of the realisation of the matching latent gain. We consider it inappropriate to go further than that, and thus inappropriate to diminish the reduction in the loss at this stage.

[122]Our decision is accordingly that, on account of the combined application of section 106 and sub-section 30(9) TCGAct, the loss realised by the Appellant in its accounting period ended 31 March 2003 referable to the transactions with which we have been concerned in this Appeal should be reduced to nil. Costs

[123]We were informed that both parties made an application for reasonable costs to be awarded if they were the successful party, and we accordingly award HMRC their reasonable costs. Right of Appeal

[124]This document contains full findings of fact and the reasons for our decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) Tax Chamber Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice. HOWARD M. NOWLAN (Tribunal Judge) Released: 14 September 2011