Mr A has gains in his GIA portfolio, has no remaining CGT exemption, and is a higher rate taxpayer. He gives £10k to Mrs A, who has her full exemption. She sells the assets without a capital gains tax liability and then gives the proceeds back to her husband for him to fund his ISA.
Ignoring the CGT aspect, how does IHT apply to this situation?
I don’t think Mr A’s transfer is exempt, as it is conditionally made to the spouse (whether expressed, implied, or inferred). I believe this would be a PET of £10k (less any available annual exemption). But is her transfer back to him also caught within IHT, or is it a true gift, or is the entire series of transfers simply ignored?
I suggest you look at the example in GAAR D19. If it were not written down in your actual writing you might have thought that HMRC’s attitude would be that Mr and Mrs Jones were 'aving a giraffe. Note that the Will’s gift back to Mrs Jones is not qualified so presumably it could even be a specific gift as opposed to residuary. Also it can apparently be inserted in the Will before the lifetime gift is made or later. Despite the tax advantage purpose found in 19.6.1 the conclusion in 19.7.1 is that it passes the double reasonable test. The conclusion in 19.8.1 is then surprising, as HMRC are not there to help us normally.
Note that to work the planning also needs to attract the s18 exemption to the lifetime gift. So can we read across this CGT outcome to IHT generally in the context of the question?
1 Before we get to the GAAR we have to ask whether the first gift is effective in law and for IHT under s18. There are no specific conditions in s18 (compare s23 and gifts to charity) save that the gifted property must become comprised in the estate of the transferee spouse. That means the full equitable interest must pass. What if there is an express understanding that it will pass back to the donor? What if there is not but it is later alleged by anyone?
You might ask how any third party including HMRC could meet the evidential burden of such a challenge. Spouses have been known to fall out (Oh yes they have!) and might later seek to disown one or both of the gifts. A deed executed by both parties and acted upon stating their common intent to pass the equitable title unconditionally will be hard for the parties to resile from or for any one else with standing to successfully challenge. In that event it may not be totally essential for the transferee to put the sale proceeds of the asset into a solely owned account for the legendary tax planner’s decent interval, ideally bearing interest which is retained. If convenient however so much the better. If the amount of the sale proceeds exceeds that required for the ISA, only give back the latter. If the sale proceeds do not need to be returned apart, from a s8C(2) TMA statement, return them anyway in the SA return because the taxpayer is entitled to HMRC’s agreement of the computation to obtain finality under s29. It is a formal assertion that the disponor is the beneficial owner for CGT.
Another deed for the cash gift back should be executed for the same purposes. I assume both spouses have full capacity. Documentation of the first gift will avoid running the gauntlet of whether the (anyway rebuttable) presumption of advance is still part of English Law or applicable in the circumstances.https://www.repository.cam.ac.uk/bitstreams/0fe2a9ea-d98b-4cc6-bde6-e1a6e19ad8fb/download
2 I assume that the transferee spouse either has a maximum ISA in the year or they both intend that the transferor should use his. For GAAR it is hard to then see how a tax advantage is not the main purpose of the arrangement. It is equally hard to see why the same conclusion, of passing the double reasonableness test as in D 19.7.1, would not apply.
3 The two deeds should also settle the point that neither gift is caught by ITTOIA s626, as not being “outright”, if it is possible for ISA income to be taxable thereunder in the hands of a settlor, either at all or perhaps (which I doubt) only if the settlor has an unused ISA. entitlement. The answer to that, and the applicability of the presumption of advancement, may only be found ultimately in the Summa Theologica of Thomas Aquinas, provided judicial cognisance of that source can be properly prayed in aid.
I’ll have a go at writing something in a different way (i.e. put words into jack’s mouth). The “will” bit just comes from a tax avoidance scheme example that HMRC refer to in their GAAR examples.
So, one set of facts might be:
Mr A gifts to Mrs A without any conditions - no gain, no loss disposal for CGT and not a transfer of value because of the exemption in s18 IHTA.
Mrs A sells the shares, realising a gain - it’s just Mrs A’s capital gain
Mrs A watches tv one evening a couple of months later, spots an advert for an ISA and gives some cash to Mr A - not a transfer of value because of s18 again
Another might be:
Mr A is a bit busy in the shed, asks Mrs A to sell the shares and use the cash to fund his new ISA and signs a declaration of trust - this is a nothing.
The shares are transferred into Mrs A’s name - this is a nothing, the shares are merely in her name but aren’t beneficially hers
Mrs A sells the shares - this is a disposal of the beneficial interest by Mr A and CGT is due as normal.
There’s obviously lots of other possible sets of facts (e.g. no declaration of trust, Mr A just does it all online as he has access to Mrs A’s GIA password and knows nothing about this, etc).
I think what jack is saying is how on earth would anyone (HMRC) know which has been done and so it would be important to evidence things in the right way.
Now some people would say this looks and feels like tax avoidance. Some people might go as far as to say that this whole thing consisted of a number of pre-arranged steps which had no purpose other than the avoidance of tax and so these intermediate steps should be ignored. I couldn’t see that jack mentions that bit but he does mention GAAR. However, the principles of statutory interpretation and GAAR are quite different.
Now someone might (if HMRC can’t or doesn’t want to show that it wasn’t tax effective because the true facts were whatever or because the principles of statutory interpretation mean it doesn’t work) say that GAAR applies. I think jack is saying that there is a tax avoidance scheme mentioned in D19 of the GAAR examples that involved:
a partner giving their spouse, who is on their death bed, some shares that are pregnant with gain (no gain, no loss so no CGT and s18 applies), and
the original owner inheriting those assets from their spouse (no IHT but now with a market value base cost).
Now you might say that’s absolutely fine or it looks a bit dodgy. But HMRC’s GAAR examples say: “Do the tax arrangements accord with established practice and has HMRC indicated its acceptance of that practice? Yes”. This is even though HMRC say: "The main purpose of the arrangement is to obtain a tax advantage. The gift of shares was made by [Mr A] in the hope of washing out the gains on the understanding that his wife would leave them back to him.
What I think jack is then implying is that if the D19 example scheme is ok from a GAAR perspective, then what the OP mentions is probably ok too from a GAAR perspective.
But what I think jack makes clear that it is still helpful to have contemporaneous evidence as to what actually happened.
I agree with your last couple of paragraphs and Jack’s prior conclusion - the GAAR example cited does indeed look dodgy but HMRC says it is fine.
To directly answer the point about “how on earth would anyone (HMRC) know which has been done” - because in this (and many similar) situations, a regulated financial adviser is telling them to do it like this and providing a written trail of evidence that the transfer to and fro was inserted with the explicit intention of avoiding CGT.
My reading of it is that the intended action is open to challenge and would in fact be treated as if it were your second example. Only in this case neither Mr or Mrs is busy in the shed, but their adviser is administering the entire transaction.
For me, I don’t like it. Clients are open to challenge from HMRC, and adviser is very much in the middle.
Tigger’s two contrasting examples are exactly to the point and recognise that there may be a plethora of factual circumstances in between. Benjamin Fabi is closer to the facts as he either knows or strongly suspects them to be. If they get anywhere near Tigger’s Scenario 2 he should indeed counsel his clients that they do not pass the sniff test. Even within Scenario 1 his clients need to be advised as de rigueur of all the downsides if an arrangement is ultimately found not to work.
1 I agree of course that you don’t get into the GAAR if in limine the arrangements fall foul of the Ramsay doctrine. We must assume that those in D19 did not and so that it is in theory possible that like ones would not either. The fact that the return gift is made by Will may be a significant factor. It is commonplace for a wife to provide for her husband and if the Will pre-dates his lifetime gift it may be juridically difficult to find it had any intention when signed as to that later gift… *"*Mrs Jones has full capacity at the time of the gift…Mrs Jones has not executed a new Will since the gift." These are important caveats in D19. Her consequent ability to change her will and defeat her husband’s purpose indicates her freedom to exercise her independent beneficial ownership of the asset up to her death. It also encapsulates a key element of much good tax planning which purports to transfer ownership of an asset: a genuine risk that the transferee will not do what the transferor expects.
In general one must be wary of direct pronouncements by HMRC about their established practice, let alone as here an indirect one. They have been known to cavil, distinguish and even disown statements in Manuals etc. And the first words of every newborn tax officer are not Mum or Dad but “it all depends on the facts”. Out of the mouths of babes… So a read across of D19 really only illustrates that the bare and different facts in the question may not be regarded by them as utterly doomed. It cannot be taken as indicating that a return gift by the donee in her lifetime is Gaar-proof.
2 Scenario 2 may well be regarded as a sham. Ineffectiveness may be the least of the participants’ problems. My Code 8 and 9 clients were often amazed when they found out, if they ever did, how they were rumbled by HMRC’s ingenuity, resourceefulness, and sheer bloody luck.
HMRC detest arrangements in which advisers are involved in design and especially in implementation. I know there are many good financial adviers out there. I have one such myself. But too many are ignorant, cavalier or naive, for whom the “I” in IFA stands for “I know”. It is madness to embark on Scenario 2 via some misguided calculation of the percentages of HMRC ever finding out or then applying some de minimis scope test. No proper adviser should become involved though there are some scheme merchants whom I will not name as they are highly litigious SLAPPERs
3 To emphasise the need for true and proper legal analysis of proposed arrangements involving gifts I would cite some planning around offshore trusts. To avoid tax on stockpiled gains the ploy has been to make a distribution to one or more beneficiaries who were ND or NR or both, for them to resettle or make onward gifts. In the past the amount of the distribution needed only to be enough to wash out the gains. Now the beneficiary must be a non-temporary NR and a limited washout does not work… This was also countered by s87I TCGA 1992. Note that a mere intention to make an onward gift is enough to be caught. But a total decanting still works within that section. There are no remaining funds to match with beneficiaries who are resident…
Fortunately there is a safe harbour for gifts made after 3 years from receipt. This means that during that period the donee must exert genuine and visible characteristics of beneficial ownership, invariably by spearheading the investment and dealing with the assets and any interim income as such owner. There must be a genuine risk at the outset that the donee will use the assets in a way no one else envisaged or approves. Even so the distribution to him must not be trammeled by conditions restricting that usage and probative evidence of that must be retained. Trustees will any way wish to take separate advice that this drastic action is within their powers, especially their duty to consider other beneficiaries the objects of their discretion, and will obtain the consent of living sui juris class members to being left out completely. HMRC only have standing to challenge an exercise of a power which is void e.g by fraud on it or its being a sham or otherwise legally defective but its validity, even if still voidable by a beneficiary in theory, and the vesting of full beneficial ownership must be demonstrable. Even so clients have to be aware of the enormous capacity of HMRC to cause trouble by both lawful and scurrilous means, bit I doubt the latter would apply to Tigger’s Scenario 1 if well documented.