Protected settlements: tainted love
The Finance (No2) Act 2017 came into effect over three years ago, and with it the concepts of deemed domicile under the 15 out of 20 year rule, protected foreign source income (‘PFSI’), and tainting, among others. It might be assumed that advisers would have got used to these rules by now and that the legislation would have ‘bedded down’. However, the legislation continues to be challenging. Complexities, anomalies and drafting defects are still coming to light. One area where such issues are particularly rife is protected settlements and tainting.
Protected settlements and tainting
As is well known, non-UK resident trusts settled by UK resident non-domiciliaries, who were not deemed UK domiciled at the time, by default benefit from ‘protected settlement’ status.
The overall effect of the protected settlement regime is to prevent the taxable attribution of foreign income and gains arising within the trust structure to the settlor. Absent this protection, once the settlor has become deemed UK domiciled, the structure effectively becomes transparent vis a vis the settlor for income tax and CGT purposes.
The great threat is of course posed by tainting, which causes protected settlement status to be lost. In very broad outline, a protected settlement becomes tainted where value is provided, directly or indirectly, at trust or underlying entity level, by the settlor or another settlement settled by him or of which he is a beneficiary.
This article considers what is meant by ‘settlor’ in the context of the protected settlement and tainting rules. As explained below, in some circumstances ‘settlor’ potentially has a wider meaning than might be expected, which, if correct, could have serious implications for the availability of protected settlement status. The issue arises because of a deeming provision in the Transfer of Assets Abroad Code (ITA 2007, Part 13, Chapter 2; ‘the ToAA Code’).
The attribution rules
Generally, and very broadly, the ToAA Code attributes income of a non-UK resident entity (termed a ‘person abroad’) to which assets have been transferred, either to the UK resident transferor or to other UK resident individuals who benefit from the transfer.
Where the person abroad is a non-UK resident trust or a non-UK resident company in which such a trust is a participator, the attribution of income to a UK resident transferor is subject to an exemption for PFSI. Broadly, this exemption applies to non-UK source income of the person abroad. If the transferor is deemed domiciled, the exemption is conditional on the settlement not having been tainted. Where income qualifies as PFSI, it is left out of account in calculating the income that is attributed to the transferor.
There is a degree of overlap between the ToAA Code and the set of income tax rules regarding settlor-interested settlements (ITTOIA 2005, Part 5, Chapter 5: ‘the Settlements Code’). Where the Settlements Code has effect, the rules attribute trust level income to the settlor. As with the ToAA Code, there is no attribution of income that is PFSI under the Settlements Code. Again, this exemption broadly applies to non-UK source income. And again, if the settlor is deemed domiciled, the exemption is conditional on the trust not having been tainted.
For trust level income, the Settlements Code applies in priority to (and effectively displaces) the ToAA Code. The reason for this is that where income is caught by the Settlements Code, it is deemed to be income of the settlor ‘and of the settlor alone’. This results in the income being deemed not to be income of a person abroad. For income arising to underlying non-UK resident companies, however, the Settlements Code does not apply, and therefore the ToAA Code is the relevant legislation.
Gains arising within settlor-interested trusts are also attributed to the settlor, under TCGA 1992, s 86 (‘Section 86’). This attribution rule generally applies to gains realised by underlying non-UK resident companies as well as to gains realised by the trustees directly. However, Section 86 can only apply to a foreign domiciled settlor if he has become deemed domiciled and the trust has been tainted.
Settlors and transferors
There is a difference in terminology between the three sets of attribution rules. Where the Settlements Code or Section 86 apply, they cause income / gains to be attributed to the ‘settlor’ of the settlement. The ToAA Code, on the other hand, creates scope for income to be attributed to any ‘individual’ who is a transferor. The difference is because the Settlements Code and Section 86 are restricted in scope to income of settlements (which generally means trusts, although in principle the meaning is wider than this for income tax purposes), whereas the ToAA Code applies to income of non-resident entities of all kinds.
A further distinction between the codes, which is important in practice, is that in the Settlements Code and Section 86 the term ‘settlor’ means just that – the person who has provided property for the purposes of the settlement. In the ToAA Code, though, there is a deeming provision which means that references to a particular individual who is a transferor may also include a ‘significant other’. Section 714(4) ITA 2007 provides that ‘In this Chapter references to individuals include their spouses or civil partners’. (For brevity, references below to spouses include civil partners).
What might this mean?
Taken literally, the deeming provision in s714(4) has the effect of requiring all references to ‘the individual’ anywhere in the ToAA Code to be read as ‘the individual or his/her spouse or civil partner’. If correct, this would have very significant implications for the protected settlements regime.
It would mean that a transaction between a protected trust/underlying company and the spouse of the settlor would be capable of tainting the trust, and depriving income of the trust/company of PFSI status for the purposes of the ToAA Code. This would apparently be so even if, at the time of the transaction, the settlor was not yet deemed domiciled, but the spouse was either actually or deemed domiciled. There is no obvious reason why, if an action of the spouse is effectively imputed to the settlor, his or her domicile status shouldn’t also be so imputed, however strange the outcome.
If the deeming in s 714(4) is taken to its logical conclusion, the consequences are even more bizarre and drastic than this. The deeming would appear to mean that a settlor with a UK domiciled or deemed domiciled spouse could not create a protected settlement for the purposes of the ToAA Code in the first place. Income of a settlement can only qualify as PFSI under the ToAA Code where ‘the individual’ was neither domiciled nor deemed domiciled in the UK when the settlement was created. If s 714(4) is read literally, and is taken as being applicable wherever there is a reference to ‘the individual’ in the ToAA Code, this domicile requirement would apply not only to the settlor at the inception of the settlement but also to his/her spouse. This would be an extraordinary result.
Conjugal complications
If the s 714(4) deeming provision does apply as described above, various logistical issues and inconsistences arise.
Neither the Settlements Code nor Section 86 have an equivalent provision to s 714(4), and so it is only the settlor and his own domicile status that are relevant to the protected settlement regime.
As the Settlements Code applies in priority to the ToAA Code, at first glance, this suggests that tainting by a spouse could only cause income within underlying companies to lose its protected status. For the purposes of the Settlements Code, an addition of value to a settlement by a spouse cannot on any view result in tainting and so cannot cause trust level income to lose PFSI status. However, where trust level income is not attributed to the settlor under the Settlements Code, it will be income of a person abroad (the trustees) within the scope of the ToAA Code. If it is correct that the spouse’s addition of value will have tainted the settlement for the purposes of the ToAA Code (due to the s 714(4) deeming), then the settlor will be taxable on trust level income under that code. The same dissonance between the application of the rules would arise if the settlor’s spouse was UK domiciled or deemed domiciled when the settlement was created.
This could produce the bizarre overall result of a partially protected settlement, where gains of the structure could continue to roll-up tax-free indefinitely, but income of the structure would become immediately taxable.
There would also be complications in determining who is taxable by reference to what income. For the purposes of the ToAA tainting provisions, it is thought that ‘settlement’ bears its natural trust law meaning, so that multiple individuals may transfer assets to a single settlement. However, for the purposes of the ToAA charging provisions, it is generally accepted that it is the transferor alone who is taxable, or potentially taxable, on the income attributable to his or her transfer of assets. So it will be the spouse who is taxable by reference to income arising from his or her addition of value, not the original settlor. This could mean that part of the income arising within the trust is taxable (eg on a UK resident and deemed domiciled settlor, following the tainting by the spouse) and part of the income is not (eg because it is attributable to an addition of value made by the spouse, who is non-UK resident).
Unintended consequences
It would be surprising, to say the least, if it had been intended for spouses of settlors to be capable of precluding the availability of protected settlement status in this way. It is clear from discussions with HMRC at the time and subsequent guidance that the focus of the protected settlement regime and the concept of tainting concept is on settlors and the interaction between them and their trusts. The s 714(4) deeming provision significantly pre-dates the introduction of the tainting rules and, as noted above, there has been no attempt to replicate it for the purposes of the equivalent Settlements Code or CGT tainting rules.
It can be inferred that s 714(4) and its potential implications were simply overlooked by the draftsman of the tainting provisions. However, as with the other drafting glitches in the 2017 and 2018 legislative changes, there seems to be little prospect of the issues being addressed by corrective legislation.
Thus far shall you come, and no farther …
Having established the potential effect of s 714(4) on the protected settlement regime, the question arises as to whether it is possible to construe the provision purposively to limit its scope.
The difficulties posed by deeming provisions in tax legislation have been addressed in a long line of cases, the latest of which is Fowler v Revenue and Customs ([2020] UKSC 22). It has been clearly established that, in interpreting the relevant legislation, a deeming provision should be applied only so far as is required to give effect to the purpose for which the statutory fiction is created. Deeming ‘should not be applied so far as to produce unjust, absurd or anomalous results, unless the court is compelled to do so by clear language’.
It is generally accepted that the s 714(4) deeming provision applies in determining whether the ‘power to enjoy’ and ‘capital receipt’ conditions are met, but should arguably not have effect for the purposes of the ToAA transferor charging provisions. In other words, s 714(4) should not be read as giving HMRC a discretion to impose tax either on the transferor or his spouse. HMRC appear to accept this, noting in their draft guidance on the ToAA Code that they will not ‘generally’ seek to tax the spouse of a non-taxable transferor, such as where the transferor is non-UK resident but the spouse is UK resident.
It could therefore cogently be argued that the purpose of s 714(4) is to ensure that where the spouse, but not the transferor, has power to enjoy or has received a capital receipt, the transferor is nonetheless within the charge to tax. On that basis, s 714(4) should not be applied any further, and should not have effect for the purposes of the PFSI and tainting provisions. As is clear from the discussion above, if the deeming is not so restricted, it could produce absurd and anomalous results that seem to contradict Parliament’s intentions.
Nonetheless, judicial decisions about the extent of deeming provisions in tax legislation can be unpredictable – there is a distinct lottery element here. A court or tribunal might refuse to look beyond the ‘clear language’ of s714(4), despite the absurdities created by the provision. Where possible, the safer course must be to proceed on the basis that spouses can cause tainting and can prevent a settlement from having protected status from the outset.
Implications
It would undoubtedly have been sensible for s 714(4) to have been expressly disapplied for the purposes of the protected settlement rules. Clearly, particularly from a policy perspective, it would behove HMRC to resolve this ambiguity and the many other drafting ‘glitches’ found in F(No2)A 2017. But it seems unlikely that this will be top of the legislative agenda for quite some time.
The first line of defence for potentially affected settlors must be to file their tax returns on the basis that the deeming in s 714(4) must be limited in the manner suggested above, including a suitable ‘white space’ disclosure. It must be hoped that HMRC take a pragmatic and reasonable approach to such filings and do not seek to exploit the apparent inconsistences created by the provision.
However, it is worth noting that HMRC have so far declined to take such an approach in the context of the inadvertent exclusion of ‘offshore income gains’ from the PFSI definition. In the event that HMRC take a similar approach here, then where tainting has already (theoretically) occurred as a result of a transaction between the settlor’s spouse and the (previously) protected settlement / underlying entity, consideration could be given to making an application to set aside the offending transaction under the law of mistake. But this is only likely to be successful if there has been an outright addition of funds by the spouse – inadvertently falling foul of the loan tainting rules, for example, is unlikely to be capable of rescission.
For trusts that remain protected settlements, pending legislative change or (perhaps) judicial intervention, the solution must be to ensure that transactions between the trust / an underlying entity and the spouse are carefully reviewed to ensure that they do not offend the tainting rules.
Given all the complexities and problems with the protected settlements legislation, and the significant risks of things going disastrously wrong, a well-informed settlor of such a trust might well ask himself, with some justification, what he has got himself into. He might even regret his decision to create such a trust; although the failure to create a protected settlement before the onset of deemed domicile is also something that many foreign domiciliaries ending up rueing. Like marriage (and various other things) in Søren Kierkegaard’s famous aphorism, it may be that if you create a protected settlement, you will regret it; if you don’t create one, you will regret it; you will regret it either way.
This article was first published in The Tax Journal on 22 June 2020. Please access it here.
For more information please contact Sophie Dworetzsky at sophie.dworetzsky@crsblaw.com or contact Catrin Harrison at catrin.harrison@crsblaw.com.
