The First-tier Tribunal has applied the Ramsay principle to reject the highly artificial ‘blue box’ tax planning scheme in Ferguson v HMRC. The scheme aimed to create tax relief for gifts to charity, even though 99% of the value flowed through the charity back to a donor’s family trust. The case gives a useful summary of the current position on Ramsay, including a recapitulation of the ‘14 canons’ of the principle first set out in Berry v HMRC.
A week is famously a long time in politics. And ten years is an age in tax planning: in 2014, neither the public nor the courts view tax sheltering, mitigation or avoidance in the way they did in 2004. In 2004, the tax avoidance industry was still buoyant, schemes were ten a penny and it seemed possible to believe that even the most egregious of them, taking advantage of the most questionable of interpretations might work. O tempora o mores.
It is to some extent a testimony to the efficacy with which HMRC has managed to turn round the opinion of the public and the courts that some, perhaps most, schemes now coming before the courts have advisers shaking their heads and wondering how it was ever thought that they could have worked. And to some extent one must feel some sympathy for taxpayers who find their attempts at avoidance examined by the courts in an atmosphere which is so very different from that in which they were devised.
And so to the First-tier Tribunal decision in William Ferguson v HMRC [2014] UKFTT 433 (TC). This was the lead case for a scheme which was known, for reasons neither explained to the court nor known to the writer, as the ‘blue box’ scheme.
The basic premise of the scheme was that the taxpayer would donate a qualifying asset to a charity and would claim tax relief under what is now ITA 2007 s 431 in respect of the donation. The asset donated was undoubtedly a qualifying asset (in fact, a gilt-edged security); and the charity was a genuine charity which apparently applied or intended to apply all its income and gains (which seem to have comprised substantially the £670,000 it is said to have made from its involvement with users of the ‘blue box’ scheme) to genuine charitable objectives.
The part of the scheme to which HMRC understandably took exception was that, by dint of fairly simple devices, the arrangements as a whole ensured that 99% of the value which was ostensibly given to the charity passed straight through it and out the other side into a trust for the benefit of the ostensible donor. Thus the fact that £670,000 of income is said to have been generated for the charity implies that in aggregate the scheme purported to create some £60m–£70m of loss relief for its users: this was an important case for HMRC.
In comparison with the tangled complexity of some other schemes, the mechanics were essentially very simple.
The first stage of the plan was the grant of options that would in practice ensure that only 1% of the value of any donation would stick with the charity: that was in substance the charity’s fee for lending its name and charitable status to the scheme (a fact the relevance of which will later become apparent). This first stage was achieved by the charity granting options to the trustees of Mr Ferguson’s family trusts under which one or other of them was certain to have the right to acquire from the charity a specified gilt for a price equal to 1% of its market value should it come to pass that the gilt in question should happen by chance to be gifted to the charity by Mr Ferguson in the 60 days following the grant of the options. Which of the two trusts would have the right to exercise the option would depend on how the FTSE100 index moved: all that was important to the planning was that one of them would.
The second stage involved Mr Ferguson buying the gilt with borrowed money and generously donating it to the charity (secure, of course, in the knowledge that one or other of his family trusts would end up with 99% of the value). It was this gift that was central to the scheme and it was this gift, viewed in isolation, which counsel for Mr Ferguson aimed to persuade the tribunal fell within what is now s 431.
The third stage had the relevant trust exercise the option and acquire the gilt for 1% of its value, thus ensuring that the necessary 99% of the value duly flowed through the charity.
The final stage was to tidy up the loose ends by having the trust sell the gilt and loan the proceeds to Mr Ferguson so that he could repay the loan he had taken to acquire it at stage 1.
The tribunal had to consider both the ‘black-letter’ law and the application of the Ramsay doctrine (after the House of Lords decision in WT Ramsay v IRC [1981] STC 174). In effect (though the tribunal did not put it in quite these terms), the questions were: whether the gift, viewed in isolation, qualified for relief; whether it should be viewed in isolation; and if not, whether – viewed as a whole – the transaction met the statutory test.
One of the requirements which must be fulfilled if tax relief is to be obtained in respect of any transfer of a qualifying asset to a charity is that the transfer must be ‘otherwise than by way of a bargain made at arm’s length’. One might have thought that the disposal in the blue box case was fairly evidently not such a bargain – after all, Mr Ferguson’s ‘gift’ to the charity would not have happened if the other transactions had not also happened – and the tribunal so held. It was plain to the tribunal that the transfer ‘was made very much by way of a bargain at arm’s length, on tightly agreed terms as to the option and charge arrangements that would apply to the gilts in question and as to the turn that [the charity] would obtain for participating’. Indeed, the tribunal pointed out that the transaction between Mr Ferguson and the charity ‘far from involving any kind of gift or other gratuitous transfer of assets, was a fiercely-negotiated arm’s length transaction’.
That would have been enough on its own to dispose of the appeal. But it was in fact only the secondary reason given by the tribunal for the failure of the scheme.
The principal part of the tribunal’s judgment was concerned with the application of the Ramsay doctrine. On that front, the tribunal adopted the well-known summary of the present state of play on the principle enumerated by Lewison J in Berry v HMRC [2011] UKUT 81 TCC, setting out in full the ‘14 canons’ of that summary. The first canon is that Ramsay is no more and no less than a general principle of statutory construction; and this was indeed accepted by both sides in blue box. The battleground was largely in the application of the sixth canon – namely that ‘the more comprehensively Parliament sets out the scope of a statutory provision or description, the less room there will be for an appeal to a purpose which is not the literal meaning of the words’ – and specifically in the application of that canon to the statutory requirement that the donor should dispose ‘of the whole of the beneficial interest … to a charity’. Counsel for Mr Ferguson argued that the term ‘beneficial interest’ had a particular and specific legal meaning and that as a matter of law the options did not ‘have the effect of taking anything away from the totality of the beneficial interest that was transferred to and vested in’ the charity. In other words, Mr Ferguson had as a matter of law given away to the charity the whole of his beneficial interest in the gilt and that was an end of it.
The tribunal rejected the argument – the requirement in question was not a ‘highly prescriptive provision’ nor a ‘detailed and prescriptive code’. It was thus open to the tribunal to interpret the phrase in the context of the wider transaction and to consider the ‘composite transaction’ rather than any individual step in it. Thus regarded, the tribunal thought it clear that 99% of the disposal was made to the family trust via the charity and not to the charity itself. There had thus been no disposal of ‘the whole of the beneficial interest … to a charity’ and no relief was due. Even the 1% of the value that had undoubtedly passed to the charity did not meet that statutory test: so in consequence even that 1% was denied relief. This, the tribunal pointed out, was in effect simply a fee paid to the charity ‘for its participation in the attempt to avoid tax’ and the lack of tax relief for it was not a surprising result.
Probably few advisers will be surprised at the decision: the scheme belongs to another age. As counsel for HMRC put it:
‘It was always intended that 99% of the £500,000 odd value of the gilts would end up, as it did, in a private trust of which the taxpayer and his family were beneficiaries … yet it is sought to be argued by Mr Ferguson that he can obtain a £200,000 odd reduction in his income tax bill on the basis of a gift to the charity of £500,000 of gilts. However cleverly the argument may be dressed up, that is what it boils down to; and it is submitted that it only has to be stated in those terms for its falsity to be apparent.’
The case pre-dates the GAAR, of course. There can be very little doubt that, even had it succeeded under the pre-GAAR law, it would have been caught had the GAAR been in place. But that, in a sense, is the most striking point about the case. Blue box was an egregiously abusive tax scheme and it was readily countered by existing law and practice. The vast majority of pre-GAAR schemes now coming before the courts are meeting the same fate.
Some might be tempted to say that the present tidal wave of HMRC successes before the courts is lending credence to the view that the GAAR adds nothing of substance to HMRC’s armoury and is an expensive, unnecessary legislative irrelevance.
