Accuro Trust (Switzerland) SA v Revenue and Customs (NHERITANCE TAX - whether property in a settlement was excluded property) [2025] UKFTT 464 (TC) (28 April 2025)

	[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback] [DONATE]
	First-tier Tribunal (Tax)
You are here: BAILII >> Databases >> First-tier Tribunal (Tax) >> Accuro Trust (Switzerland) SA v Revenue and Customs (NHERITANCE TAX - whether property in a settlement was excluded property) [2025] UKFTT 464 (TC) (28 April 2025) URL: https://www.bailii.org/uk/cases/UKFTT/TC/2025/TC09501.html Cite as: [2025] UKFTT 464 (TC)

DECISION

Part A - Introduction

1. The appellant is the trustee of a settlement known as the Tiodab Trust ("the Trust"). The appellant has appealed against three notices of determination issued under s 221 of the Inheritance Tax Act 1984 ("IHTA") on 23 May 2019. All references in this decision to legislation are to provisions in IHTA as they were enacted and applicable at the relevant time unless it is expressly stated otherwise. The dispute relates to whether inheritance tax ("IHT") is payable under provisions in Chapter III of Part III IHTA relating to settlements ("the settlements regime").

Facts

2. The following facts are undisputed:

(1) On 14 September 1992, when he was neither domiciled nor deemed domiciled in the United Kingdom ("UK") for IHT purposes, the individual who set up the Trust ("the Settlor") settled $100 to create the Trust. The trustee of the Trust was Investec Trust (Switzerland) S.A. subsequently known as Salamanca Group Trust (Switzerland) SA and now known as Accuro Trust (Switzerland) SA ("the Trustee").

(2) Additional funds were added to the Trust on various dates from 2 November 1992 to 6 May 1993 and on 24 December 1998 and 8 March 2005.

(3) On 6 April 2005 the Settlor became deemed domiciled in the UK for IHT purposes. The issue is whether the appellant is subject to IHT as regards various events which took place after the Settlor became deemed domiciled.

(4) On 16 March 2006 additional substantial cash funds were added to the Trust.

(5) On 3 April 2006 the Trustee acquired 12,487 class "A" shares with a par value of €1.25 per share in Magnolia Investments & Partners SCA ("the A shares") from the Settlor for a total consideration of €250 and payment was made from the cash funds added to the Trust on 16 March 2006.

(6) On 25 October 2006 a capital distribution was made out of the Trust to the Settlor's wife.

(7) A ten-year anniversary of the Trust occurred on 14 September 2012. At that date, the property held by the Trustee included the property contributed to or representing property contributed to the Trust after the Settlor became deemed domiciled, the cash funds contributed on 16 March 2006 and the cash proceeds from the sale of the A shares ("the property").

As explained below, the settlements regime imposes a "ten-year charge" to IHT on the value of "relevant property" held in a "settlement" immediately before a ten-year anniversary. One of the main disputes in these proceedings is whether the property was "relevant property" for the purposes of the ten-year charge.

(8) The Settlor ceased to be resident in the UK on 5 April 2013 having become resident in Switzerland on 26 March 2013 but remained deemed to be domiciled in the UK for IHT purposes until 6 April 2016.

(9) The Trustee initially accounted for IHT on the basis that a ten-year charge was due in respect of the value of the property, on the basis it was relevant property, but later claimed a repayment of the sum paid:

(a) On 15 May 2013, the Trustee paid £1,702,142.73 to HMRC on account of an estimated ten-year charge.

(b) On 29 August 2013, the Trustee delivered a IHT100 form to HMRC in respect of the ten-year charge which included a revision to its estimated figures reducing the IHT due to £1,682,336 and interest of £2,212.

(c) On 16 November 2013, HMRC issued a calculation of IHT in respect of the ten-year charge in the sum of £1,680,658.59 and interest of £2,066.38. The notes in the calculation of IHT stated "Provisional Calculation". HMRC did not refund the overpayment of £19,417.76 made by the Trustee.

(d) On 25 September 2014, the Trustee made a repayment request in respect of the sum previously paid ("the disputed tax").

(10) The Trustees later made two capital distributions in respect of which HMRC have determined that, contrary to the appellant's stance, IHT is due:

(a) On 25 September 2014 a capital distribution was made from the Trust to the Settlor consisting of the proceeds of the sale of the A shares. On 20 May 2015, the Trustee, acting through Smith & Williamson ("SW"), wrote to HMRC enclosing a nil IHT100 return for this capital distribution. This was accompanied by the repayment request referred to above.

(b) On 31 January 2017 a further capital distribution was made out of the Trust to the Settlor in the form of shares, cash and a loan consisting of the proceeds of the sale of the A shares. On 28 July 2017 SW sent a further IHT100 in relation to this distribution made which again showed no IHT as due.

I refer to these distributions as "the capital distributions". Broadly, the issue is whether IHT is due on these capital distributions under provisions in the settlements regime which impose a charge to IHT when "relevant property" ceases to be comprised in a "settlement".

Repayment notice

3. In outline, in one of the notices of determination ("the repayment notice"), HMRC have denied the appellant's claim for repayment of the disputed tax. The appellant made the claim (a) under s 241 which provides that: "If it is proved to the satisfaction of the Board that too much tax has been paid" in respect of the relevant taxable event, "the Board shall repay the excess", (b) on the basis that the appellant wrongly paid the disputed sum on the correct interpretation of the following provisions in the settlements regime:

(1) Section 64(1) provides that "where immediately before a ten-year anniversary all or any part of the property comprised in a settlement is relevant property, tax shall be charged at the rate applicable under sections 66 and 67 below on the value of the property or part at that time". For this purpose (i) a "ten year anniversary" is the "anniversary of the date on which the settlement commenced and subsequent anniversaries at ten-yearly intervals" (s 61), and (ii) "references to the commencement of a settlement are references to the time when property first becomes comprised in it" (s 60).

(2) "Relevant property" is defined for the purposes of these provisions as "settled property in which no qualifying interest in possession subsists, other than...excluded property" (s 58(1)(f)) and "excluded property" is defined under s 48(3)(c) as follows:

"Where property comprised in a settlement is situated outside the United Kingdom-

(a) the property (but not a reversionary interest in the property) is excluded property unless the settlor was domiciled in the United Kingdom at the time the settlement was made,..." (Emphasis added.)

The dispute relates to the meaning of the highlighted wording.

(3) A person's domicile is ascertained for this purpose in accordance with the general law but under s 267(1) a person not domiciled in the UK at any relevant time is treated for the purposes of IHTA as domiciled in the UK (and not elsewhere) if: "(b) he was resident in the United Kingdom in not less than 17 of the 20 years of assessment ending with the year of assessment in which the relevant time falls".

(4) In this decision (a) all references to a person at any time being or becoming domiciled or not domiciled are to a person being or becoming domiciled or not domiciled in the UK for the purposes of IHT whether under general law or s 267(1), (b) I refer to property comprised in a settlement situated outside the UK as "foreign property", and (c) I refer to a settlement which is within the scope of the settlements regime as a "relevant property trust" and to a charge provided for under s 64 as a "ten-year charge".

4. What constitutes a "settlement" and "settled property" or "property comprised in a settlement" for IHTA purposes is set out in s 43:

"43 - Settlement and related expressions

(1) The following provisions of this section apply for determining what is to be taken for the purposes of this Act to be a settlement, and what property is, accordingly, referred to as property comprised in a settlement or as settled property.

(2) "Settlement" means any disposition or dispositions of property, whether effected by instrument, by parol or by operation of law, or partly in one way and partly in another, whereby the property is for the time being -

(a) ...

(b) held by trustees on trust to accumulate the whole or part of any income of the property or with power to make payments out of that income at the discretion of the trustees or some other person, with or without power to accumulate surplus income, or

(c)...,

or would be so held...if the disposition or dispositions were regulated by the law of any part of the United Kingdom; or whereby, under the law of any other country, the administration of the property is for the time being governed by provisions equivalent in effect to those which would apply if the property were so held...."

5. A "settlor" is defined for these purposes in s 44 as follows:

"(1) In this Act "settlor", in relation to a settlement, includes any person by whom the settlement was made directly or indirectly, and in particular (but without prejudice to the generality of the preceding words) includes any person who has provided funds directly or indirectly for the purpose of or in connection with the settlement or has made with any other person a reciprocal arrangement for that other person to make the settlement.

(2) Where more than one person is a settlor in relation to a settlement and the circumstances so require, this Part of this Act (except section 48(4) to (6)) shall have effect in relation to it as if the settled property were comprised in separate settlements."

6. It is common ground that, at all relevant times, (1) the Trust was a "settlement", and (2) for the purposes of ss 58 and 48(3), the property was situated outside the UK and was comprised in a "settlement" (as property comprised in the Trust) in which no qualifying interest in possession subsisted.

7. In claiming that a refund of the disputed tax is due under s 241, the appellant's stance is that no ten-year charge is due under s 64 by reference to the value of the property as, immediately before the relevant ten-year anniversary, the property was not relevant property as it qualified as excluded property under s 48(3). The appellant's view is that (1) "the time the settlement was made" in s 48(3), by reference to which time the Settlor's domicile status is to be tested, refers to the time at which the Trust was first created/constituted as a matter of trust law (14 September 1992) when, as is not disputed, the Settlor was not domiciled, and (2) that this is the correct interpretation of s 48(3) is established definitively in the decision of the Court of Appeal in Barclays Wealth Trustees (Jersey) Ltd v HMRC [2018] 1 WLR 2312; [2017] EWCA ("Barclays"). That case addressed precisely the question which arises here and the Court of Appeal rejected HMRC's argument to the contrary which is substantially the same as that they make here.

8. As set out in Parts C and D, in the repayment notice, acting through the relevant officer (Mr Ryder) HMRC rejected the appellant's claim for repayment of the disputed tax. That was on the basis that it was not proved to their satisfaction that a repayment was due under s 241 as:

(1) A ten-year charge is due by reference to the value of the property as, immediately before the ten-year anniversary, on the correct interpretation of s 48(3), the property was relevant property and not excluded property.

In HMRC's view (a) under s 48(3) "the time the settlement was created", by reference to which the Settlor's domicile status is to be tested, is when the property was transferred into the Trust (in March and April 2006). As is not disputed, the Settlor was domiciled at that time as a result of becoming deemed domiciled on 6 April 2005, (b) in Barclays the Court of Appeal deliberately left open the point which arises in this appeal, and (c) this interpretation is consistent with the underlying reasoning of the Court in that case.

(2) In any event, the property does not qualify as excluded property due to the operation of s 255. That provides as follows:

"Where any payment has been made and accepted in satisfaction of any liability for tax and on a view of the law then generally received or adopted in practice, any question whether too little or too much has been paid or what was the right amount of tax payable shall be determined on the same view, notwithstanding that it appears from a subsequent legal decision or otherwise that the view was or may have been wrong." (Emphasis added.)

In HMRC's view (a) the payment of the disputed tax was made and accepted in satisfaction of liability for the ten-year charge on the basis of HMRC's view of how s 48(3) operates as set out in Revenue Interpretation 166 ("RI 166") which essentially accords with HMRC's view in this case, (b) that view was generally received or adopted in practice, (c) therefore, the question whether too much tax has been paid under s 241 is to be determined on that view, as being correctly due, and (d) Mr Ryder decided in the repayment notice that s 255 applies and has that effect. The appellant disputes that the conditions for s 255 to apply are met. The tribunal received written and oral witness evidence on this issue as set out in Part D.

9. RI 166 (first published in 1997 in the Tax Bulletin), states this:

"In the light of the definitions of "settlement" and "settled property" in s 43, [HMRC's] view is that a settlement in relation to any particular asset is made at the time when that asset is transferred to the settlement trustees to hold on the declared trusts. Thus, assets added to a settlor's own settlement made at an earlier time when the settlor was domiciled abroad will not be "excluded", wherever they may be situated, if the settlor has a UK domicile at the time of making the addition."

10. I refer to this view of the law as "the stated view". It is common ground that at the relevant time the stated view was widely known. It was included in the Orange Book (1997/1998) as was referenced in the notes to s 48 (and was set out in the Yellow Book up to and after 2012/13). The same view of the law was also published in HMRC's Inheritance Tax Manual IHTM27220 (as set out in some of the leading textbooks):

"the legislation refers to the settlor's domicile at the time the settlement was made. You must proceed on the basis that, for any given item of property held in a settlement, the settlement was made when that property was put in the settlement. Consult TG or your Team Leader if this view is challenged."

The stated view is acknowledged and set out in the contemporary commentaries usually with an express reference to RI 166 and/or the Inheritance Tax Manual. The relevant main contemporary commentaries all refer to it as set out in the evidence set out in Part D.

11. As set out in Part C, the parties disagree as to the scope of the repayment notice and nature and scope of the tribunal's jurisdiction in relation to the appeal made in respect of the determinations set out in it, in particular, as regards the application of s 255: (1) HMRC argue that (a) the tribunal has a "supervisory" jurisdiction only, broadly, to determine if the relevant determinations were reasonably made by Mr Ryder, (b) Mr Ryder reasonably arrived at his decisions, including that s 255 applies but (c) if, as the appellant argues, the tribunal has full appellate jurisdiction, in any event, as a matter of law s 255 applies with the effect the disputed tax is to be taken to have been correctly paid on the basis of the stated view, and (2) the appellant disputes all of these points.

Notices dealing with the capital appointments

12. In the other two notices of determination, HMRC have imposed IHT on the capital appointments under s 65(1)(a). This applies as follows:

"Charge at other times

(1) There shall be a charge to tax under this section –

(a) where the property comprised in a settlement or any part of that property ceases to be relevant property (whether because it ceases to be comprised in the settlement or otherwise); ...

(7) Tax shall not be charged under this section by reason only that property comprised in a settlement ceases to be situated in the United Kingdom and thereby becomes excluded property by virtue of section 48(3)(a) above." (Emphasis added.)

The definition of "relevant property" set out above also applies for this purpose. I refer to a charge under this provision as an "exit charge".

13. The appellant did not account for IHT in respect of the capital appointments on the basis that the property appointed under the capital appointments was not relevant property as, at all material times, it qualified as excluded property for the same reasons as set out above. HMRC have determined in the two notices that the property appointed was at all material times relevant property for the same reasons as set out above.

Summary of conclusions

14. In summary, I have concluded that:

(1) If the tribunal's jurisdiction is appellate as regards the appeal made in respect of the repayment notice (a) as set out in Part B, on the correct interpretation of s 48(3), the property was excluded property at all relevant times, and (b) as set out in Part D, on the correct interpretation, s 255 does not apply. On that basis, the determinations made by HMRC in all of the notices, including the repayment notice, are not correct and (i) the disputed tax should be repaid to the appellant, and (ii) no IHT is due in respect of the capital appointments.

(2) If, as I consider to be the case (see Part C), the tribunal's jurisdiction is supervisory in nature as regards the appeal made in respect of the repayment notice, as set out in Part D, Mr Ryder's decision set out in that notice was not reasonably arrived at (principally as he erred in law as regards both the application of s 48(3) and s 255) and the overall result is the same as set out in (1).

Part B - Excluded Property Issue

The issue

15. The parties framed the specific question as being whether, in circumstances in which a settlor has initially established a settlement at a time when he is non-domiciled but then contributes additional property at a time when he is deemed domiciled, the additional property can be "excluded property" within the meaning of s 48(3). By way of shorthand I refer in this decision to such a set of circumstances as "additions of property" or "property additions". In summary and overview:

(1) HMRC's stance is that the words "at the time the settlement was made" in s 48(3) cover both (a) the time when the settlement was originally "constituted" as a matter of trust law and (b) the time when it is further constituted when any further property is transferred to it. Hence, whether any added property qualifies as excluded property is to be tested by reference to the settlor's domicile status when that property is transferred to the settlement. Mr Baldry noted that, on this interpretation, the corollary is that if the settlor establishes the settlement when domiciled, any additions of property made when he becomes non-domiciled are excluded property. As noted, HMRC consider that in Barclays the Court of Appeal deliberately left open the point which arises in this appeal and their interpretation of s 48(3) is consistent with the reasoning of the Court in that case.

(2) The appellant's view is that it is clear from the decision in Barclays that "the time the settlement was made" can only be when it is initially created or constituted for trust law purposes; in effect its status is fixed for the purposes of s 48(3) by reference to the settlor's domicile status as at that time. That case addressed precisely the question which arises here and the Court of Appeal rejected HMRC's argument which is substantially the same as that they make here. Mr Ewart submitted that (a) HMRC's argument is entirely circular: they are essentially saying that their interpretation matches the purpose they set out but, for the provisions to be viewed as having that purpose, one has to interpret the provisions as they argue for. As set out in further detail below, for HMRC's argument to succeed they would have to demonstrate that the tribunal can read words into s 48(3)(a) but that is not justified.

Law

Legislative framework

16. HMRC made a number of submissions regarding the overall policy of IHTA and of the settlements regime so it is helpful to have in mind the scheme of the legislation. In Barclays Henderson LJ set out a summary of how the settlements regime has evolved and how the relevant general provisions in IHTA operate and interact with that regime which I have summarised as follows:

(1) Under the IHT rules for settlements:

(a) Before 22 March 2006, the scope of the settlements regime was confined, broadly speaking, to discretionary trusts in which no interest in possession of any description subsisted. Since that date it also applies to settlements in which interests in possession subsist which are not "qualifying interests in possession". A "qualifying interest in possession" is defined in s 59 as being either an interest in possession (a) to which the individual became beneficially entitled before 22 March 2006, or (b) to which an individual became beneficially entitled on or after 22 March 2006, which is "an immediate post-death interest, a disabled person's interest or a transitional serial interest".

(b) Chapter II deals with interests in possession in settled property, although its scope was considerably curtailed by the changes made with effect from 22 March 2006 which brought many interest in possession trusts within the scope of the settlements regime. The key concept which underpins Chapter II, and which dates back to the introduction of capital transfer tax by the Finance Act 1975 (see schedule 5 paragraph 3(1)), is the equation of beneficial entitlement to an interest in possession in settled property with beneficial entitlement to the property in which the interest subsists. This is now provided for by s 49(1), which states that: "A person beneficially entitled to an interest in possession in settled property shall be treated for the purposes of this Act as beneficially entitled to the property in which the interest subsists." In other words, such an interest in possession in settled property is equated with beneficial ownership of the underlying trust property itself.

(c) By contrast, the settlements regime applicable to discretionary trusts and, since 22 March 2006, all other trusts except those in which a qualifying interest in possession subsists, although it has undergone various evolutions, has never sought to equate the interests of the beneficiaries with beneficial ownership of the whole or any specific part of the settled property. Instead, the general approach has been to impose a periodic charge to tax at 10-yearly intervals on "relevant property" which remains in the settlement, and further charges to tax whenever "relevant property" leaves the settlement (for example by distribution to a beneficiary) or otherwise ceases to be subject to the "relevant property" regime. The rate of tax payable at each 10-year anniversary is normally 30% of the full rate payable on a lifetime chargeable transfer (see below). As set out in Part A property comprised in a settlement which is within the settlements regime is not "relevant property" and so is not subject to ten-year or exit charges if it qualifies as excluded property under s 48(3).

In the remainder of this decision I refer to the circumstances where a person has an interest in possession or, since 22 March 2006, a qualifying interest in possession in settled property as that person having a "QIIP".

(2) Under the general IHT regime:

(a) By virtue of ss 1 and 2(1), tax is charged on the value transferred by "a chargeable transfer", which is a "transfer of value made by an individual which is not an exempt transfer". Section 2(3) provides that references to chargeable transfers include references to occasions on which tax is chargeable under Chapter III of Part III, and thus include ten-year charges and exit charges.

(b) Section 3 defines a transfer of value as a disposition as a result of which the value of the transferor's estate is reduced. A person's 'estate' is the aggregate of all property to which the person is beneficially entitled and is treated as including settled property in which the person has a QIIP (s 5). Section 3(2) provides, however, that "no account shall be taken of the value of excluded property which ceases to form part of a person's estate as a result of a disposition".

(c) Section 4(1) imposes a charge to tax on the death of a person as if, immediately before his death, he had made a transfer of value equal to the value of his estate at that time. Again, however, section 5(1)(b) provides that "the estate of a person immediately before his death does not include excluded property".

(d) The basic definition of "excluded property" for these purposes is contained in s 6(1), which states that: "Property situated outside the United Kingdom is excluded property if the person beneficially entitled to it is an individual domiciled outside the United Kingdom".

17. In Barclays the Court of Appeal noted that the excluded property rule in s 48(3) is analogous to that in s 6(1), in the sense that two similar conditions have to be satisfied if the property in question is to qualify as excluded property, one relating to the situs of the property and the other relating to the domicile of the settlor. Under s 48(3) the domicile condition only has to be satisfied by the settlor "at the time the settlement was made". The effect of that requirement is the issue here. It was common ground that where a settlor makes a settlement of foreign property at a time when he is not domiciled, and the property then remains comprised in the same settlement, its status as excluded property will not be lost merely because the settlor subsequently acquires an actual or deemed domicile.

18. The manner in which the precise rate of ten-year charge is computed is complex. In HMRC's IHT Manual it is stated that the underlying design of the charge is that IHT on relevant property trusts should be comparable to a charge of 40% once a generation. To achieve this there would normally be a 20% "entry charge" when property is transferred into the settlement and three ten year charges at 6% (3/10ths of 20%) on the trustees. The amount subject to the charge is the net value of the relevant property including any deemed relevant property after reliefs and exemptions. The rate of IHT on the amount subject to charge is based on a notional lifetime transfer (as if the trust funds were hypothetically transferred at the date of the ten-year charge). The components of that transfer have varied over time but the calculation method remains the same and is set out in s 66. The rate cannot exceed 6%. The amount of the IHT threshold, the nil rate band available against the hypothetical chargeable transfer, is reduced to account for previous cumulative transfers. A reduction is available for assets which have not been in the trust for the whole ten years.

Statutory interpretation

19. It is common ground that in carrying out the required exercise in statutory interpretation, the tribunal must have regard to the words of Lord Hodge in DPSC in R (Project for the Registration of Children as British Citizens) v Home Secretary [2022] 2 WLR 343 , which is the most recent and comprehensive statement of the principles to be applied in construing statutes. He said this at 353:

"29. The courts in conducting statutory interpretation are "seeking the meaning of the words which Parliament used": Black-Clawson International Ltd v Papierwerke Waldhof-Aschaffenburg AG [1975] AC 591, 613 per Lord Reid. More recently, Lord Nicholls of Birkenhead stated: "Statutory interpretation is an exercise which requires the court to identify the meaning borne by the words in question in the particular context." (R v Secretary of State for the Environment, Transport and the Regions, Ex p Spath Holme Ltd [2001] AC 349, 396). Words and passages in a statute derive their meaning from their context. A phrase or passage must be read in the context of the section as a whole and in the wider context of a relevant group of sections. Other provisions in a statute and the statute as a whole may provide the relevant context. They are the words which Parliament has chosen to enact as an expression of the purpose of the legislation and are therefore the primary source by which meaning is ascertained. There is an important constitutional reason for having regard primarily to the statutory context as Lord Nicholls explained in Spath Holme, p 397:

"Citizens, with the assistance of their advisers, are intended to be able to understand parliamentary enactments, so that they can regulate their conduct accordingly. They should be able to rely upon what they read in an Act of Parliament."

...

31. Statutory interpretation involves an objective assessment of the meaning which a reasonable legislature as a body would be seeking to convey in using the statutory words which are being considered. Lord Nicholls, again in Spath Holme [2001] 2 AC 349, 396, in an important passage stated:

"The task of the court is often said to be to ascertain the intention of Parliament expressed in the language under consideration. This is correct and may be helpful, so long as it is remembered that the 'intention of Parliament' is an objective concept, not subjective. The phrase is a shorthand reference to the intention which the court reasonably imputes to Parliament in respect of the language used. It is not the subjective intention of the minister or other persons who promoted the legislation. Nor is it the subjective intention of the draftsman, or of individual members or even of a majority of individual members of either House ... Thus, when courts say that such-and-such a meaning 'cannot be what Parliament intended', they are saying only that the words under consideration cannot reasonably be taken as used by Parliament with that meaning."

Barclays

20. The parties referred extensively to the decision in Barclays where the Court of Appeal considered the correct interpretation of s 48(3) in the following circumstances:

(1) The settlor, Mr Dreelan, settled £100 on the trusts of a settlement ("the 2001 settlement") on 21 June 2001 when he was not domiciled.

(2) He then settled further sums in 2001 and on 4 February 2003, shares in Qserv Limited, a UK company which the trustees then transferred to a wholly owned Jersey company.

(3) The settlor became domiciled on 6 April 2003.

(4) On 31 March 2008, the Jersey company was dissolved and its assets, including the Qserv shares, were distributed to the trustees of the 2001 settlement. On 4 April 2008, (a) another trust ("the DBJT") was created by the settlor and his three brothers, and (b) the trustees of the 2001 settlement appointed the Qserv shares to be held on the trusts of the DBJT. On 3 July 2008, the trustees of the DBJT sold the Qserv shares for cash and an earn out.

(5) On 2 June 2011, the cash which represented the Qserv shares was appointed to the trustees of the 2001 settlement ("the 2011 appointment"). It was held in a Jersey bank account on the ten year anniversary of the 2001 settlement on 21 June 2011. The issue was described, at [37], as being whether that money was or was not excluded property for the purposes of the ten-year charge. In the circumstances of that case, this depended upon the interaction of ss s 48(3)(a), s 81(1) and s 82(1) and (3).

(a) Section 81(1) provides

"Where property which ceases to be comprised in one settlement becomes comprised in another then, unless in the meantime any person becomes beneficially entitled to the property (and not merely to an interest in possession in the property), it shall for the purposes of this Chapter be treated as remaining comprised in the first settlement."

(b) Section 82(1) and (3) provide:

"(1) For the purposes of this Chapter ... property to which section 81 above applies shall not be taken to be excluded property by virtue of section 48(3)(a) unless the condition in sub-section (3) below is satisfied (in addition to the conditions in section 48(3) that the property is situated outside the United Kingdom and that the settlor was not domiciled there when the settlement was made).

...

(3) The condition referred to in subsection (3)(1) above is –

...

(b) ... That the person who is the settlor in relation to the second of the settlements mentioned in the sub-section concerned, was not domiciled in the United Kingdom when that settlement was made."

21. It was common ground that the Qserv shares were excluded property when held in the 2001 settlement but that they lost their character as such when they were held in the DBJT. HMRC argued that (1) the shares/cash could never get excluded property character back, and/or (2) the 2011 appointment was itself a settlement under s 43 which did not satisfy the requirements for the cash to be excluded property. So part of the debate in that case was about the deeming effects of ss 81 and 82 and part about the meaning of "settlement" and when it is made for the purposes of s 48(3)(a).

22. At [35] Henderson LJ said that in Rysaffe Trustee Co (CI) Ltd v Inland Commissioners [2002] STC ("Rysaffe") Park J made an important point with which he agreed that:

"the provisions of the 1984 Act relating to settlements are, in the absence of special provision, for the most part left to be interpreted in accordance with the general understanding of trust practitioners: see his judgment at paras 18-21."

23. He explained that (1) the issue in Rysaffe was whether the settlor had made five separate discretionary settlements by executing five separate trust instruments, or whether he should be treated as having made a single settlement, and (2) Park J concluded that there were no grounds on which to go behind the settlor's choice to execute five separate settlements, each of which had a separate identity for IHT purposes (see [17]) adopting the following approach:

"...After pointing out at [18] that the decision whether to make a single settlement, or several settlements, or to add property to an existing settlement, is one for the settlor to take, Park J continued:

"19. Examples of the sort in the previous paragraph can be multiplied many times over. They show that it is not necessary to have a statutory definition to determine whether there is one settlement or more, and if more than one, how many. Trust practitioners can recognise a separate settlement when they see one, and equally they can recognise a case where there is only one settlement, not several settlements, when that is what they see. Often, fiscal considerations apart, it will make little or no difference in the end whether, for example, a settlor chooses to make one large settlement, or instead to make several smaller settlements. But there is no doubt that as a matter of general principle the two courses are different and create analytically different structures. What kind of structure a particular settlor has created will depend on how he chooses to do it.

20. All of that assumes no detailed statutory definition. However, the Inheritance Tax Act 1984 has s 43, which begins by saying that the provisions of the section "apply for determining what is to be taken for the purposes of this Act to be a settlement". Notwithstanding those words of the statute, my opinion is that s 43 gives little or no guidance to answering the question of whether, in the many permutations of circumstances of which I gave a few illustrations in [18] above, there is one settlement or more than one. In my judgment the draftsman has for the most part left those questions to be answered in accordance with general principles - in accordance with what I have described above as the general understanding of trust practitioners.""

24. In Rysaffe Park J continued at [20] and [21] as follows:

"It appears that the distinguished editors of Dymond's Capital Taxes are of the same opinion. In para 16.232 they make the point that there are some inheritance tax provisions which enact that in specific circumstances a single settlement under the general law may be more than one settlement for inheritance tax, and (conversely), that there are other inheritance tax provisions which enact that in other specific circumstances a structure which the general law regards as separate settlements may be a single settlement for inheritance tax. But they go on to write: 'In the absence of special provisions, however, the general law must decide' They clearly do not regard s 43 as a 'special provision' in this sense. In the present case the Revenue do not suggest that any such special provision applies.

[21] ....I do not accept that the use of the plural means any more than that it is possible to have more than one disposition to the trustees of a single settlement. In particular, the use of the plural is not a positive enactment that, where there are two or more dispositions to different settlements, they are to constitute one settlement for inheritance tax purposes even if they would constitute two or more settlements under the general law."

25. At [36] Henderson LJ noted that the approach and reasoning of Park J in Rysaffe were approved by the Court of Appeal (see [2003] EWCA Civ 356, [2003] STC 536, at [25] to [26] per Mummery LJ, with whom Dyson and Schiemann LJ agreed).

26. At [38] Henderson LJ said that if the Qserv shares (and the proceeds of their later disposal) had throughout remained in the 2001 settlement, the conditions in s 48(3) would clearly be satisfied, because (1) the cash was foreign property, (2) the settlor was non-domiciled when he first established the 2001 settlement, and (3) (if it matters) he was also not domiciled when he transferred the shares to the 2001 settlement in February 2003 such that "even if the addition of the shares were properly to be regarded as the making by him of a fresh settlement within the meaning of section 43, of which he was the settlor pursuant to section 44, the "settlor" condition in section 48(3) would nevertheless still be satisfied".

27. At [39] he said that (1) the difficulty was caused by the involvement of the DBJT which engaged the special rules in ss 81 and 82, (2) as the settlor had a deemed domicile when the DBJT was established the property settled by him on its trusts could never qualify as excluded property while it remained property to which s 81 applied, and (3) finally, there was the question whether the 2011 appointment independently constituted the making of a further settlement made when the settlor was clearly UK-domiciled. At [43] to [45] he made the following main points:

(1) As under s 81 the shares (and, after the sale, their proceeds) are deemed to have remained throughout in the 2001 settlement, it must further follow, for the purposes of the settlements regime, that the 2011 appointment did not itself engage s 81; the property was already conclusively deemed to be comprised in the 2001 settlement, by the prior application of s 81(1), so there could be no scope for the section to apply again so as to deem the appointed property to remain in the DBJT" (see [43]).

(2) It is also clear that, while the shares and their proceeds remained in the DBJT, they were not excluded property (see [44]).

(3) Following the 2011 appointment, the deeming effect of s 81(1) in relation to the property restored to the 2001 settlement was "spent", and it was no longer "property to which section... 81... applies" for the purposes of section 82(1). There was no need to deem the cash derived from the sale of the shares to be comprised in the 2001 Settlement, since that was now the reality." (See [45]).

(4) The issue of whether the cash, once it was foreign property, when held in the Jersey bank account, was excluded property "must depend on section 48(3) and the answer to the question when "the settlement was made" within the meaning of subsection (3)(a)".

28. Henderson LJ then set out the divergent views of the parties:

(1) At [46], he explained that Mr Ewart submitted that (a) the settlement in which the property was comprised can only be the 2001 settlement, and it was made on 21 June 2001 when the settlor was non-domiciled. That is how any trust practitioner would view the matter, and on the authority of Rysaffe that is the correct approach, and (b) the 2011 appointment cannot be regarded as a "disposition ... of property" within the definition of "settlement" in s 43(2), because the appointed property was already conclusively deemed by s 81(1) to be comprised in the 2001 settlement when the 2011 appointment was executed.

(2) At [48] to [50] he set out that Mr Baldry argued to the contrary for the following main reasons:

(a) The general rule is that foreign property owned by a non-domiciled person is outside the scope of the tax, but if that person subsequently becomes domiciled all his non-settled foreign property immediately ceases to be excluded property within the meaning of s 6 and so falls within the scope of IHT. If he were then to transfer the foreign property into a settlement, this would give rise to a chargeable transfer and the settled property would not be excluded property within s 48(3). Thus, once an individual has become domiciled, the legislation does not permit him to take assets out of the scope of IHT by settling them offshore.

(b) The basic purpose of s 81 is to prevent avoidance of the periodic charge by treating settled property transferred to another settlement as remaining in the original settlement. However, this purpose does not extend to deeming excluded property to retain that status if it is transferred to a settlement created by a domiciled settlor: see s 82(3)(b). Against this background, it would be inconsistent with the scheme of the legislation if it were construed as permitting a settlor to settle assets after becoming domiciled in such a way that they became excluded property.

(c) The transfer of the property back to the 2001 settlement by the 2011 appointment was a non-event for the purposes of s 81, because the property was deemed to have remained in the 2001 settlement throughout, and s 81 therefore continued to apply to the property after it reverted to the 2001 settlement. If that is correct, s 82 remained in play, and the settlor remained unable to satisfy the condition in s 82(3)(b).

(d) With its focus on dispositions, the definition of settlement in s 43 "leads to the conclusion that a separate settlement is created for IHT purposes whenever a settlor adds property to an existing settlement". On that basis in 2001 the settlor made three separate settlements for IHT purposes when he settled the original £100 and then made two further transfers of property to it, even though any trust lawyer would say that he had made a single settlement and then added property to it. The time when "the settlement was made" must refer:

"to each act of settling property by a disposition, even though the opening words of the subsection ("Where property comprised in a settlement is situated outside the United Kingdom") are naturally read as referring to the ongoing settlement as it exists from time to time and as a trust lawyer would identify it, rather than to each dispositive act by which it was constituted. On the correct purposive approach, he submits, the court should not be reluctant to construe the word "settlement" as having two different meanings within section 48(3), although if necessary he would adopt the preferred view of the judge at [28] that both references to "settlement" in the subsection are to the individual dispositive acts whereby property is settled."

29. At [51] Henderson LJ concluded that the submissions for the taxpayers are correct and the property in question was excluded property immediately before the ten-year anniversary. His main reasons for reaching this conclusion, at [52] to [54], are as follows:

"52. First, I consider the better view to be that the 2001 settlement was a single settlement for inheritance tax purposes, constituted by a number of separate dispositions of property to be held on the trusts thereof. Those dispositions included the three transfers to the trustee made by Mr Dreelan in 2001, his transfer of the 25,000 Qserv shares to the trustee on 4 February 2003, and the transfer effected by the 2011 appointment. Not only is this how a trust lawyer or practitioner would view the matter, but it fits comfortably with the definition of "settlement" in section 43(2) which applies for all purposes of the 1984 Act. In particular, the express reference to "disposition or dispositions of property" in the definition is in my view naturally read as intended to cover the common situation where a settlement is first made, often with a small or nominal sum of money, and further assets are then added by the settlor. This was rightly recognised by Park J in the Rysaffe case, where he said, at para 21:

"Let me consider the implications of the plural 'dispositions'. In my view the use of the plural is merely a recognition that in a case where there is a settlement (i.e. only one settlement) it is possible for there to have been more dispositions to the trustees than one. A typical case is where a settlor creates his settlement with one disposition, and later adds more property to it by one or more other dispositions."

53. Secondly, I find it implausible to suppose that in section 48(3) the same word "settlement" was intended by Parliament to have two different meanings, or that it has a single meaning which requires one to focus separately on each occasion when property is added to a settlement. At least in cases of the type which I have described, the natural (and in my opinion correct) interpretation of the subsection is that it requires one to look at a single settlement as it is constituted from time to time, whether by one or a series of transfers into settlement, and provides that any foreign property comprised in it is excluded property unless the settlor was United Kingdom-domiciled "at the time the settlement was made". The time when the settlement was made will then be ascertained in accordance with the usual principles of trust law, and will normally be the occasion when the settlor first executed a trust instrument and constituted the trust by providing property to the trustee.

54. In the present case, there can be no doubt that the 2001 settlement was made by Mr Dreelan on 21 June 2001......". (Emphasis added.)

30. At [55] he observed that it does not make any difference that the Qserv shares (which were never foreign property) were appointed from the 2001 settlement to the DBJT, and the cash proceeds of their sale were later appointed back to the 2001 settlement by the 2011 appointment. These transactions engaged s 81, the effect of which was to deem the shares and their proceeds to remain comprised in the 2001 settlement. The force of that deeming was spent, in relation to the property appointed back to the 2001 settlement, once the 2011 appointment had been executed and the consequential transfers effected. He could not accept Mr Baldry's submission that the property somehow remained property to which s 81 applied, within the meaning of s 82(1), once those steps had been taken.

31. At [56] he said that even if it were somehow possible to regard the 2011 appointment as effecting a separate settlement within the meaning of s 43, he would accept the submissions of Mr Ewart that for the purposes of Chapter III of Part III, including in particular the periodic charge to tax, such an analysis is precluded by the deeming provision in s 81(1):

"Since the property subject to the 2011 Appointment is conclusively deemed to have remained throughout in the 2001 Settlement, it cannot consistently with that deeming be treated as the subject of a separate disposition into the 2001 Settlement at the same time. Furthermore, the purposes of Chapter III include the question whether the property subject to the 2011 Appointment became excluded property in the 2001 Settlement, because the periodic charge to tax under section 64 is on "relevant property", and "relevant property" is defined in section 58(1)(f) as not including "excluded property".

32. At [57] he said he did not see anything at all surprising or anomalous in reaching this conclusion:

" ...the Qserv shares were transferred to the Trustee at a time when Mr Dreelan was still non-UK domiciled, and if they had never left the 2001 Settlement their proceeds, if invested as foreign property, would undoubtedly have qualified as excluded property. I can see no mischief which might require the language of the 1984 Act to be interpreted so as to produce the opposite conclusion merely because there was an intermediate transfer of the Qserv shares between the two settlements. The facts of the present case are highly unusual, and there is no suggestion that they formed part of any scheme of tax avoidance. The consequences within Part III of a transfer of property between settlements are dealt with by section 81, and I see no occasion to strain the language of that section, or of section 48, in order to reach a different conclusion in this case."

33. At [58], he said this:

"For completeness, I should mention that different considerations may arise in cases where a settlor makes an offshore settlement when he is non-UK domiciled, later acquires a UK domicile, and then makes further substantial transfers of property into the settlement. It may arguably seem anomalous that, in such a case, the property in question, if it is or becomes foreign property, should qualify as excluded property in the settlement merely because the settlor was non-UK domiciled when the settlement was originally made. I emphasise that the present case is not of that character, because the Qserv shares originated in the 2001 Settlement, and were transferred to it at a time when Mr Dreelan was non-UK domiciled. I express no view on the question whether the same result as in the present case should be reached in cases of the other type which I have described, because wider policy considerations may then be engaged. For similar reasons, I prefer to express no view on some of the wider arguments which were addressed to us but are unnecessary for the resolution of this case."

34. He then turned to the judgment of Mann J in the High Court ([2015] EWHC 2878 (Ch)), who reached the opposite conclusion. At [60] his comments include that he could not "accept the judge's view that the word "settlement" may have two different meanings in section 48(3), or that the 2011 Appointment involved the making of a separate settlement of the appointed property by Mr Dreelan" and also he thought the judge was "over-influenced by policy considerations which do not arise on the facts of the present case: see in particular his judgment at [31] and [32]".

35. HMRC noted that (1) Mann J accepted HMRC's submission that the verb "made" is significant and that when the draftsman intended to refer to the initial creation of the overall settlement, he used different wording in ss 60 and 61 and that word "suggests an act not just an end result", and (2) he concluded that the property was not excluded property, as he considered was supported by the logic and plausibility in overall taxation terms at [31]:

"I can detect no logic, even in an artificially constructed tax world, for allowing excluded property status to after-contributed property by a domiciled settlor, merely because when he created the settlement in the first place he was non-domiciled. I do not consider it plausible that Parliament had that intention. It should be noted the structure of the Act would not allow that consequence where the subsequent contributor was a domiciled third party....."

36. At [62] Henderson LJ quoted the following passages from the decision of Mann J:

"39. ... The true construction of section 48(3) is one that requires one to look at the occasion of the settling of the property for the purposes of determining whether or not it is excluded property, and nothing else. It does not create a separate settlement for the other purposes of the Act, deemed or otherwise. The overall settlement for the purposes of section 64 remains the same. Rysaffe enables (and requires) one to look at basic trust law for determining that. [The judge then referred to sections 64, 61 and 60 of IHTA].

40. All those are (in the circumstances) references to the [2001 Settlement], implicitly acknowledging that property may have arrived in it at different times. That is what has happened here. The correct interpretation of section 48(3) has no impact on this. It involves considering the circumstances of the settlement of property at one point in time, and does not otherwise involve the creation of any "settlement" which differs from the overall [2001 Settlement].

41. Nor does Rysaffe itself compel a contrary conclusion on section 48(3). That case determines that the concept of what is a settlement, for the purposes of the provisions of the Act considered by Park J, is the same as that understood by the general law. Thus in his case there were 5 real world settlements, and not one, and therefore there were 5 settlements for the purposes of the Act. Park J was not considering every reference to the word "settlement" in the Act, and was not considering the detail of what was the making of the settlement for the purposes of section 48(3)(a). I therefore consider that Rysaffe does not assist Mr Ewart."

37. At [63] he said this favours a construction of s 48(3) which is "at odds with the normal conception of what constitutes a settlement, even though (as the judge recognised) that normal conception is reflected elsewhere in the 1984 Act, and notably in section 64 itself" and he preferred to construe it "in a way that accords as far as possible with the usual practice and understanding of trust lawyers and practitioners" and considered that such an approach provided strong support for Mr Ewart's submissions.

Submissions

38. In HMRC's view, that their reading of the provision is the correct interpretation follows from construing it purposively in its immediate context and in light of the general scheme of IHTA. Mr Baldry made the following main submissions:

(1) HMRC fully accept Henderson LJ's trust law analysis that a settlement is a continuing subsisting settlement, but rely on his observations that the settlement can be constituted from time to time. If one applies his reasoning to this case, the answer is that the settlement is made in relation to added property at the time that property is added. Moreover, the words of s 48(3)(a) must be read in light of the definition of "settlement" in s 43(2) and of "settlor" in s 44:

(a) Section 43(2) recognises that for IHT purposes a single continuing settlement for trust law purposes may be constituted by different dispositions of property made at different times and that property may become "property comprised in a settlement" by a disposition made at a later time than the initial disposition of property by which the settlement was constituted.

(b) Section 44 contemplates that a disposition of property consisting of the provision of funds to an existing settlement is an act which makes a person "a settlor", as "a person by whom the settlement was made", whether or not the person was otherwise a settlor. As Mann J said in Barclays it would be surprising if a separate settlor added funds to the same settlement at the same time as the original settlor with completely different IHT consequences, even if they were both domiciled at the time.

(c) The evident purpose of s 48(3)(a) is to determine the excluded property status of specific "property comprised in the settlement" by reference to the domicile of the settlor in relation to that property. Construed in light of its purpose and of the definitions in ss 43 and 44, s 48(3)(a) requires the testing of the settlor's domicile at the time when the particular property is transferred to the settlement. The natural thing is to ask when the settlement was made in relation to the particular asset.

(2) There are two further provisions which support this interpretation:

(a) Section 60 specifies that time runs for the purposes of the ten-year charge from the date on which the settlement "commenced". Hence, different statutory language is used where the legislature intends to refer to the particular time when the settlement was originally made/constituted. One would expect the legislature to use the same language to express the same concept in different provisions. The appellant's idea that a settlement can commence without any property in it is very difficult to arrive at and very obscure.

(b) When the deemed domicile rule was introduced in 1975 transitional rules in s 267(3)(a) provided that:

"In determining whether property comprised in the settlement which became so comprised before10 December 1974 is excluded property, section 45 of this Act [the deemed domicile rule] shall be disregarded."

Under this test, the event which triggers when one tests the settlor's domicile is the time at which particular property becomes comprised in the settlement. The appellant misapplies this rule in the example given below.

(3) Wider policy considerations support this interpretation. It is relevant that (a) the factors which determine whether under the general regime a transfer of value falls within the scope of IHT are the relevant individual's domicile status and the situs of the property transferred, and (b) the IHT consequences of any such transfer are generally assessed at the time of the transfer:

(a) It is notable that (i) an individual who is non-domiciled can change the situs of property, from UK to overseas, to remove it from the scope of IHT or make a transfer of foreign property free of IHT consequences and it remains excluded property unless the transferee is a domiciled individual at the time of the transfer. Similarly, where a non-domiciled settlor makes a settlement of foreign property, that property is excluded property and remains as such even if the settlor becomes domiciled. That is the case except that where the settlor or his spouse has a QIIP, when the property ceases to be subject to the QIIP, it is treated as becoming comprised in a separate settlement and is excluded property only if the settlor/ spouse is non-domiciled at that time (see s 80). The fact that the regime specifically provides for circumstances in which the domicile of the settlor is to be taken into account after the settlement was originally constituted supports HMRC's approach, (ii) once a non-domiciled individual becomes domiciled, any transfer of his foreign property is within the scope of IHT in the same way as for individuals who have always been domiciled and (ii) once an individual who was domiciled becomes non-domiciled, any transfers of property he makes from that time are treated consistently with those made by an individual who has always been non-domiciled.

(b) On HMRC's interpretation, s 48(3) ensures that the IHT treatment of added property is consistent with this; the excluded property status of foreign property transferred into a settlement prior to the settlor becoming domiciled is preserved, but foreign property transferred into the settlement after the settlor becomes domiciled is treated in the same way as other transfers by a domiciled individual.

(c) The consequence of the appellant's argument is that (i) where an individual happens to have established a settlement when he was non-domiciled, even with a nominal amount of capital, he could add foreign property to it once he is domiciled as excluded property, (ii) that foreign property would leave the scope of IHT and thereafter remain wholly outside the ten-year and exit charges, and (iii) if the settlement is an ordinary discretionary trust the property addition would be a chargeable transfer but there is no charge on transfers of property into settlements in which there is a QIIP and, prior to 22 March 2006, many qualified as QIIPs. It is very difficult to see any policy reason for the legislation to have created this anomalous and highly favourable treatment.

(d) HMRC's interpretation of s 48(3) is consistent with the policy of IHTA in respect of settled property more generally. The charges to IHT imposed under the settlements regime are intended to ensure broadly that, over the life of a settlement, the property comprised in it is charged to IHT at a level approximately equivalent to that which would be charged on the property in an individual's estate which is subject to the ordinary charge on death. The objective is to ensure that all property within the scope of IHT is subject to a similar tax burden. That is plainly not achieved on the appellant's interpretation of s 48(3) for the reasons already set out.

(4) The treatment of excluded property has remained the same in all material respects since IHTA was enacted, so the appellant has to establish that Parliament must have intended this arbitrary and anomalous treatment from the outset.

39. Mr Ewart made the following main submissions in relation to the points HMRC made in respect of ss 43, 44, 60 and the transitional rule:

(1) As regards ss 43 and 44: (a) The tribunal should note the number of technical terms from the law of trusts that are used in the IHT rules relating to settlements (as was acknowledged in Rysaffe (as cited in Barclays)), (b) the definition of a settlement in s 43 is very much linked to trust law concepts, and (c) s 44(1) simply provides a definition of settlor in relation to a settlement as constituted in a trust law sense. If someone other than the person who initially made the settlement, for example, added funds and became another settlor, then s 44(2) would apply. Nothing in s 44 indicates that a person who is a settlor because he made the settlement in a trusts law sense is more of a settlor than he was already on making any subsequent property additions. That section sets out alternative ways in which someone could become a settlor; they are not alternative ways the same person can become a settlor more than once.

(2) The point HMRC make about s 60 is irrelevant as it was decided in Barclays that IHTA refers to the time when the settlement was originally made in a trust law sense. Moreover (a) an argument that Parliament should be assumed not to vary its language through an Act is not borne out by experience and is not a sound guide to interpreting legislation, and (b) a settlement is not necessarily created/constituted when property first becomes comprised in it. That may generally be the case under English trust law but a foreign entity/structure may qualify as a settlement in other circumstances, and (c) for the purpose of determining when the ten-year charge is to apply it is reasonable to start from when there is property in the settlement. On the other hand s 48(3)(a) in effect gives a settlement a domicile for its lifetime and it is reasonable to expect that to happen when it is created, whether or not there is any property in the settlement.

(3) The transitional rule relating to the introduction of the deemed domicile rule is best considered by way of an example of how it operates:

(a) X created a settlement by settling overseas property on 29 April 1964 when X was non-domiciled. X added further overseas property prior to 9 December 1974. The first ten year charge arose on 29 April 1984. For the purposes of the excluded property rule, X's domicile at the time the settlement was made (on 29 April 1964) is to be assessed without regard to the deemed domicile rules. That was regarded as fair because X could not have known about the rules before they were announced in December 1974.

(b) X adds another further overseas property in 1986. On the next ten-year anniversary on 29 April 1994 for the purposes of the excluded property rule (i) as regards the property added prior to 9 December 1974 X's domicile status at the time the settlement was made is assessed leaving the deemed domicile rule out of account, but (ii), as regards the property added in 1986, X's domicile status at the time the settlement was made is assessed taking account of the deemed domicile rule.

It is notable that where the legislature wishes to divide a single settlement into parts depending on the domicile status of the settlor when the settlement was made, they do so by using specific words. The draftsman has used different terminology in different parts of even the same Act simply because it is more convenient to use a particular expression. One does not get any guidance from comparing different provisions, even within the same Act.

40. Mr Ewart added that there are a number of provisions which would be inoperable on HMRC's interpretation as they refer to a settlement being made before or on or after a specified date (see s 48(2), and s 81). It may be impossible to answer the relevant statutory question posed by these provisions on HMRC's approach if, for example, the settlor was not domiciled part of the time and domiciled for part of the time. The reporting provisions in s 218 would be very difficult to apply on HMRC's interpretation as they require a report to be made in certain circumstances "within three months of the making of the settlement".

41. Mr Ewart submitted that HMRC's alleged anomaly is no basis for giving s 48(3)(a) a meaning which it cannot bear as a matter of language.

(1) The appellant's interpretation of this provision does not lead to any more anomalous a result than these illustrations of how it operates in an arbitrary way:

(a) If the settlor was not domiciled when the settlement was initially made, foreign settled property remains excluded property even if the trustees and all of the beneficiaries are domiciled and resident in the UK and/or if the settlor becomes domiciled, even if that occurs on the day after the settlement was made and if he remains domiciled until death.

(b) On the other hand, if the settlor was domiciled when the settlement was made foreign settled property can never be excluded even if the trustees and all of the beneficiaries are not domiciled or resident in the UK and have no connection to the UK and/or if the settlor ceases to be domiciled, even if that occurs on the day after the settlement was made and if he remains non domiciled until death.

The situation in (a) and (b) could continue for many years (depending upon the perpetuity periods applicable to the settlement). It would not matter that the settlor has no interest in the settlement.

(c) The provision which states that there is no chargeable event as a result of property ceasing to be excluded property, is one of the many parts of this "patchwork" of provisions which are arbitrary in their operation. There is no particular reason why trustees should be able to turn property from relevant property into excluded property the day before a ten-year anniversary without any tax charge and then potentially convert it into relevant property just after that date.

(d) The operation of s 48(3)(a) could also operate in an arbitrary manner in less extreme circumstances of the kind described above.

(2) In any event, the alleged anomaly is greatly overstated. In practice, the addition of property was unlikely to occur certainly when s 48(3)(a) was originally enacted in the Finance Act 1975 (which introduced CTT) given that from 1975 to 1986 such additions were immediately chargeable transfers on which tax was charged at a rate of up to 75% in 1975 and up to 30% in 1984. The only exception was where property was transferred to a settlement which gave the settlor (or his spouse) a QIIP. In that case property ceased to be excluded property if the settlor or spouse was domiciled when their interest in possession terminated (whether on death or otherwise). In light of that, it is perfectly rational for Parliament to take the view in 1975 (and later) that (a) the addition of property was not likely to be a particular problem, and (b) it was better to have a rule which was simple and easy to operate in practice, rather than a more complex and cumbersome rule to cover the unlikely and rare occurrence of property additions. This approach obviously cannot be regarded as unreasonable, absurd or capricious.

(3) HMRC's interpretation requires words to be read into s 48(3)(a) to qualify the actual words impliedly but that is impermissible. The correct approach to interpretation where a party seeks to rely on an alleged anomaly is set out in Stock v Frank Jones (Tipton) Ltd [1978] 1 WLR 231 where the House of Lords rejected the argument that words could be read into trade union legislation to avoid anomalies. Mr Ewart referred extensively to the comments in this case including:

(a) Comments of Viscount Dilhorne and Lord Fraser of Tullybelton to the effect that when the language of a statute is plain, then although it may appear that it might have been better drafted or that amendment of it might be less productive of anomalies, it is not open to the court to remedy the defect (see 234G and 238E).

(b) Comments of Lord Edmund Davies that dislike of the effect of a statute is no reason for departing from its plain language (see 238C).

(d)

"If the words used by Parliament are plain, there is no room for the "anomalies" test, unless the consequences are so absurd that, without going outside the statute, one can see that Parliament must have made a drafting mistake. If words "have been inadvertently used," it is legitimate for the court to substitute what is apt to avoid the intention of the legislature being defeated: per MacKinnon L.J. in Sutherland Publishing Co. Ltd. v. Caxton Publishing Co. Ltd. [1938] Ch. 174, 201. This is an acceptable exception to the general rule that plain language excludes a consideration of anomalies, i.e. mischievous or absurd consequences. If a study of the statute as a whole leads inexorably to the conclusion that Parliament has erred in its choice of words, e.g. used "and" when "or" was clearly intended, the courts can, and must, eliminate the error by interpretation. But mere "manifest absurdity" is not enough: it must be an error (of commission or omission) which in its context defeats the intention of the Act."

(e) Most of the speech of Lord Simon of Glaisdale which includes the following comments at 237 C to F:

......(5) Parliament may well be prepared to tolerate some anomaly in the interest of an overriding objective; (6) what strikes the lawyer as an injustice may well have seemed to the legislature as no more than the correction of a now unjustifiable privilege or as a particular misfortune necessarily or acceptably involved in the vindication of some supervening general social benefit; (7) the parliamentary draftsmen knows what objective the legislative promoter wishes to attain, and he will normally and desirably try to achieve that objective by using language of the appropriate register in its natural, ordinary and primary sense: to reject such an approach on the grounds that it gives rise to an anomaly is liable to encourage complication and anfractuosity in drafting; (8) Parliament is nowadays in continuous session, so that an unlooked for and unsupportable injustice or anomaly can be readily rectified by legislation: this is far preferable to judicial contortion of the law to meet apparently hard cases with the result that ordinary citizens and their advisers hardly know where they stand.

All this is not to advocate judicial supineness: it is merely respectfully to commend a self-knowledge of judicial limitations, both personal and constitutional. To apply it to the argument on behalf of the appellant based on anomaly, a court would only be justified in departing from the plain words of the statute were it satisfied that: (1) there is clear and gross balance of anomaly; (2) Parliament, the legislative promoters and the draftsman could not have envisaged such anomaly, could not have been prepared to accept it in the interest of a supervening legislative objective; (3) the anomaly can be obviated without detriment to such legislative objective; (4) the language of the statute is susceptible of the modification required to obviate the anomaly." (Emphasis added.)

(f) Mr Ewart also referred to the case of Pearson v IR Commrs [1981] AC 753 as an example of how statutory construction has been approached in an IHT context. Viscount Dilhorne gave the following guidance on how statutory interpretation should be approached as regards the meaning of interest in possession which was not further defined in the relevant Act as at page 619:

"...In my view one should first seek to determine the ordinary and natural meaning of those words and then consider whether there is anything in the context in which they are used to lead to the conclusion that the proper interpretation of them involves a departure from the ordinary and natural meaning...In the light of these statements [in certain books], it appears that in the 19th century the words 'an interest in possession' would have been interpreted as ordinarily meaning the possession of a right to the present enjoyment of something...

Each side contended that the case put forward by the other side would give rise in a number of instances to anomalies and injustice. Time was spent in examining whether or not the alleged anomaly would in fact arise. I did not find this helpful for, as Buckley LJ said,...in the course of his judgment in this case:

'The ingenuity of counsel can almost always produce possible anomalies in either direction, and that has been the case here.'

In my opinion the words 'interest in possession' in schedule 45 should be given their ordinary and natural meaning, which I take to be the present right of present enjoyment..."

(g) Lord Russell and Lord Keith took a similar approach (see pages 626 and 634). Lord Russell said at 626:

".... Neither do I find it a useful exercise to compare anomalies and 'hard cases' asserted to arise on either solution: such are, I fear, only to be expected in the introduction of such a radical and complicated experiment in fiscal novelty....

(4) On the basis of this caselaw, the onus is on HMRC to identify an absurdity which Parliament cannot have intended to arise. However, clearly the language in question was carefully chosen and there is no such absurdity. If Parliament had turned its mind to this rather obscure issue, it may quite reasonably have taken a policy view that the addition of property was not a particular problem, for a number of reasons. Any perceived anomaly is simply a product of what was described in Pearson as the novel fiscal experiment introduced in 1975 which contains many complex and arbitrary rules. So it would not be surprising if there are at the edges some perceived anomalies which can be rectified by legislation. Many trusts are administered by lay trustees but even for professional trustees, there is a benefit in simplicity. Moreover, on HMRC's interpretation, the trustees would have to determine the domicile and perhaps the residence status of the settlor (due to the deemed domicile rule) each time property is added rather than only having to determine it once when the settlement is originally made. Determining domicile raises complex questions and so may determining if a person is resident under the pre-statutory residence test. It is reasonable for Parliament to take the view that a simple easily understood and administered rule was better than a complicated rule of the sort castigated by Lord Simon.

42. Mr Ewart submitted that HMRC are wrong to rely upon the position of an individual owning property directly as a guide to the policy behind s 48(3):

(1) The provisions concerning excluded property for individuals are based on the domicile status of the individual from time to time. Those relating to settlements operate, in effect, by fixing each settlement with a "domicile" once and for all at the time the settlement is made by reference to the domicile status of the settlor at that date.

(2) Individuals are taxed on transfers of value and the excluded property rule in s (6) operates by reference to the domicile status of the individual when the taxable event occurs (subject to exemptions). By contrast, under the settlements regime, (a) settlements are taxed on the ten-year anniversary and when property ceases to be relevant property, (b) the charge is calculated on a complex and to some degree an artificial basis, and (c) as set out above and as HMRC accept, the status of the property by which the settlement is originally constituted remains fixed according to the settlor's domicile status at the time of the settlement's creation. These are patently very different codes for taxing different situations.

(3) There are other arbitrary and anomalous situations in relation to individuals of note:

(a) If a non-domiciled individual transfers excluded property to a domiciled individual and reserves a benefit, the property remains in his estate and is not excluded property but if he transfers excluded property to a settlement and reserves a benefit, the property is excluded property.

(b) Before 2006 a non-domiciled individual could transfer foreign property to a settlement in which he had a QIIP without attracting a tax charge and the property would be excluded property whatever the subsequent domicile of the settlor. The settlor could be the trustee of the settlement and so could retain full control, both legal and beneficial, over the property or the settlement could be revocable.

(c) As set out in the evidence in Part C, HMRC accept that a settlor who creates a settlement when non-domiciled who then becomes domiciled, can transfer property to a foreign company owned by the settlement and thereby in effect remove the whole of its value from the scope of IHT. Even if the company is liquidated so that the property is owned by the trustees, it would still be excluded property. HMRC's case is one of form rather than substance. They are really saying that a settlor cannot make property excluded property by adding it directly to the settlement when domiciled but he can achieve the same thing by transferring it to a company owned by the trustees. HMRC's position is quite formulaic, a point of form rather than substance.

43. Mr Baldry said that:

(1) There is no difficulty in reading references to a settlement consistently throughout IHTA. The regular uses of the word "made" and "commenced" show that the draftsman had in mind two different statutory concepts.

(2) The appellant's view that the settlements regime is just an arbitrary patchwork which throws up anomalies is a fundamental mischaracterisation of the regime. It forms a coherent code which forms an integral part of a wider coherent scheme:

(a) Although the provisions concerning excluded property for individuals are different to those concerning settlements, (i) both regimes operate by reference to transfers of property and tax is charged by reference to the state of affairs at the time of the transfer, and (ii) the IHT treatment of both transfers to settlements and transfers to other persons is assessed by reference to the domicile status of the individual at the time of the transfer.

(b) Moreover, there is a very close relationship between the IHT treatment of property within a settlement and property within an individual's estate. This is most obvious in relation to property settled in QIIPs as it is treated as if it continued to belong to the settlor. The relationship is more indirect in relation to other settlements but again overall IHTA seeks to ensure that there is no great disparity of treatment. The entry charge and ten-year charges are broadly designed to ensure that settled property is charged to tax in a comparable way to property that is outside a settlement.

(c) The other key point that provides linkage is the territorial scope of IHT. In 1975 on introducing CTT, Parliament decided that the territorial scope would be limited by two key factors, the domicile position of the transferor and the situs of the property. The examples the appellant gives do not show that the provision is arbitrary:

(i) The fact that foreign property transferred into a settlement when the settlor is non-domiciled remains excluded property even if he later becomes domiciled is consistent with the fact that generally IHT applies by reference to the state of affairs at the time of the relevant transfer.

(ii) Different considerations arise in cases such as this. Once an individual becomes domiciled his entire estate is within the charge to IHT, and any transfer of property he then makes has IHT consequences. It is consistent with that that if a settlor transfers property into a settlement when he is domiciled it does not qualify as excluded property. The settlements code is not an arbitrary patchwork. There is a logic and a consistency throughout and the tax treatment of settlements mirrors the tax treatment of property outside the settlement.

(iii) That a non-domiciled individual can transfer UK property to an overseas company and thereby convert it into excluded property is not an anomaly; it is just a consequence of Parliament having decided to fix the territorial scope of this tax by relation to these two concepts.

(iv) The anomaly arises on the appellant's case; on that interpretation the excluded property status of added property is determined by reference to a previous transfer of different property.

(3) The assertion that additions of property would be "unlikely and rare" is unrealistic and it is unlikely that Parliament had that in mind. Any settlor could create a QIIP or vary the terms of an existing settlement to do so, so that foreign property could be put into it without an entry charge. It is clear that Parliament considered that a change in the settlor's domicile is a significant matter that needed to be provided for, as is demonstrated by the transitional rule. The provisions which Mr Ewart referred to show that Parliament had in mind that the benefit of property qualifying as excluded property is not obtained forever due to a settlor setting up a settlement when non-domiciled. It cannot be presumed that Parliament considered that someone who became domiciled would be able to get out of the regime introduced in 1975 simply because they had created a settlement when non-domiciled. The intention was to tax individuals on their capital transfers and that those capital transfers would have tax consequences.

(4) The assertion that Parliament must have chosen a once and for all approach because it is simple is based on an unrealistic view of the law. In any event, as the present case shows, there is nothing unworkable or inherently complicated about requiring trustees to assess the status of settled property by reference to the domicile status of the settlor at the time of the relevant transfer of property. Trustees are used to keeping records of property within their settlements given that, for example, (a) they may have to take account of the interest of different types of beneficiaries, and (b) for the purpose of the ten-year charge they need to work out what constitutes relevant property and excluded property. As HMRC said in interpretation RI 166 in 1997 ordinarily the trustee should keep a record of what property is excluded property and what property is non-excluded property.

(5) There is no plain and unambiguous provision against HMRC's interpretation such that it is inappropriate to take account of policy reasons or anomalies in interpreting the provision. This is a case where the words are capable of more than one meaning (as Lord Simon said in Stock at 517G) and "it is a perfectly legitimate intermediate step" in construction to choose between potential meanings by various tests. That is just the normal approach. The primary meaning of "made" in s 48(3)(a) is constituted and it is natural to speak of an ongoing settlement being constituted from time to time by the addition of further property and so being "made" from time to time by the additions. As Mr Justice Mann considered in Barclays, the circumstances HMRC have set out are not simply cases of anomalies. HMRC are asking the tribunal to look at the structure of IHTA and see whether there is any rational or logical policy reason why Parliament intended the settlements regime to operate as the appellant argues for. There is no such strong positive policy reason but there are strong policy reasons as to why the section should be construed in the way HMRC argue.

(6) The conversion of UK property into foreign property by transferring it to an overseas company is a different type of transaction raising different considerations. The use of that structure is not within the mischief or policy of s 48(3), which aims to ensure that relevant dispositions which result in property being held in trust are not excluded property. Inflating the value of excluded property, such as shares in an overseas company held in a trust, does not involve a disposition of property directly into trust. The company is not transparent for IHT purposes and companies are generally outside the scope of IHT (subject to special rules, for example, as regards close companies). In any event that does not demonstrate any great lacuna in the legislation, because if the settlor has a right to be appointed capital, it is likely that the reservation of benefit rules will apply. So the company example does not represent some sort of fundamental loophole.

Decision

44. For the reasons set out below (a) I accept HMRC's submission that in Barclays Henderson LJ left open the question of how s 48(3) is to be interpreted in the circumstances of this case, and (b) I do not accept Mr Ewart's submission that Henderson LJ's reasoning "leads inexorably" to the conclusion that HMRC's argument in this case is wrong on the basis that there is no material difference between their stance in Barclays and their stance in this case and/or that he can be taken to have been "well aware" that "policy considerations" alone would not be sufficient for it to be held that added property is excluded property:

(1) In Barclays, Henderson LJ was clear, at [53], that the natural interpretation of s 48(3) is that, (a) it requires one to look at a single settlement as it is constituted from time to time, whether by one or a series of transfers into settlement, and provides that any foreign property comprised in it is excluded property unless the settlor was domiciled "at the time the settlement was made", and (b) "the time the settlement was made" is then to be ascertained in accordance with the usual principles of trust law.

(2) However, Henderson LJ prefaced his comments at [53], on the basis that this was the natural and, in his view, correct interpretation "at least in cases of the type he had described". Mr Ewart submitted that by "cases of the type he had described" Henderson LJ meant the circumstances set out in [52], as described by Park J, whereas HMRC considered that he meant only the specific circumstances in Barclays. Reading the passage in context, it seems likely that Henderson LJ considered the natural interpretation of s 48(3) would usually apply to the circumstances he set out in [52] as well as the particular circumstances in Barclays.

(3) He then explained, at [57] and [58], in effect, that he expressed no view on whether the same interpretation and result would apply where, as here, substantial property is added to the settlement after its initial creation when the settlor is domiciled. He emphasised, at [58], that the circumstances in Barclays were not of that kind because the relevant asset, the cash proceeds held in Jersey bank accounts, represented the Qserv shares which themselves originated in the 2001 settlement, and were transferred to it originally at a time when the settlor was non-domiciled. At [57], he said that the conclusion in Barclays was not surprising or anomalous for the reasons he gave as set out above.

(4) Henderson LJ said that he expressed no view on whether the same result as in Barclays should be reached as regards added property and merely acknowledged that different considerations may arise, it "may arguably seem anomalous" for added property to qualify as excluded property and that "wider policy considerations may then be engaged" (emphasis added).

(5) HMRC accepted that, on the basis of Henderson LJ's analysis in Barclays, the Trust is a single settlement for IHT purposes, constituted by a number of separate dispositions of property to be held on the trusts thereof. Those dispositions include the initial transfers made to the Trust which were made when the Settlor was non-domiciled, and the later transfers of the property in question, the cash and the A shares, in 2006, which were made when the Settlor was domiciled. HMRC take the view that, in circumstances such as these, s 48(3) may be interpreted as meaning that the single continuing settlement is "made", in the sense of made or constituted in part, on each occasion when property is added. HMRC's contention is essentially that where the status of the settlor has changed between the date of the initial constitution of the settlement and a later transfer of property into it, the intention of the legislature is to test whether that property is excluded property for the purpose of ten-year and exit charges by reference to the domicile status of the settlor at the date of that transfer. In their view, therefore, the property in question here is not excluded property because the Settlor was domiciled when the Trust was made in part when that property was transferred into it. Whilst the difference between their argument in Barclays and in this case is a subtle one, it was not the precise argument before Henderson LJ as regards the application of s 48(3) to the facts of that case and, as noted, he expressly did not consider the application of s 48(3) as regards these circumstances.

45. For the reasons set out below, I have concluded that (1) s 48(3) does not operate on the basis that whether property is excluded property is to be tested by reference to a settlor's domicile status at the time that property was transferred to it. I cannot see that the term "at the time the settlement was made" can be taken to mean, in effect, each occasion when the settlement was constituted in part by the disposition of property to it, (2) the more natural, plain meaning of s 48(3), as set out by Henderson LJ in Barclays, applies in these circumstances also, and (3) on that basis, for the purposes of s 48(3), the settlement was made when the Trust was first constituted for trust law purposes (when the Settlor was not domiciled) and not when the property was transferred to the Trust (when the Settlor was domiciled) and at all relevant times the property qualifies as excluded property.

46. In my view, neither the specific provisions in the settlements regime which HMRC point to, nor a consideration of the way the IHT regime works generally or as regards the settlements regime, provides material support for HMRC's interpretation of s 48(3). Dealing first with HMRC's points on specific provisions in the settlements regime:

(1) As set out in Rysaffe and Barclays, s 43 acknowledges in the definition of the term "settlement" that a single settlement may be constituted by more than one disposition of property, seemingly in recognition that often a settlement is initially created with nominal funds followed by a transfer of more substantial assets. Section 44(1) includes as a settlor for IHT purposes not only persons who directly constitute the settlement for trust law purposes but also those who contribute to it in other ways such as by the provision of funds. It appears that in recognition that this broad definition may lead to more than one person being a settlor in relation to a single settlement as viewed for trust law purposes, s 44(2) states that where that is the case and the circumstances so require, the settlements regime is to have effect as if the settled property (by implication, as settled by different settlors) were comprised in separate settlements. The terms settlor and settlement used in s 48 must be interpreted in accordance with ss 43 and 44 but, in this case, that simply leads to the conclusion that the Trust is a single settlement constituted by the various dispositions of property made to it by the Settlor, as the settlor of that single settlement. I cannot see how the mere fact that these provisions recognise that a single settlement may be formed by more than one disposition of property and give a broad definition to the term settlor indicates that the legislature intended the term "at the time the settlement was made" in s 48(3), contrary to its natural meaning, to mean each time when the settlement was constituted in part by a particular disposition of property.

(2) Nor do I consider that material light is shed on how s 48(3) is to be interpreted by the fact that in s 60 the legislature has chosen to state that for the purposes of the ten-year charge time runs from the date on which the settlement "commenced", as specified to be when property first becomes comprised in it. It makes sense that, in that context, the legislature has chosen to specify that particular time given that the ten-year charge is imposed by reference to the value of relevant property comprised in the settlement with a reduction in the charge where that property is so comprised only for part of the 10 years. That raises the question of why the legislature has used different terminology in s 48(3), if the legislature meant to identify the term "at the time the settlement was made" with the time when it was first created for trust law purposes. However, it seems to me entirely possible that, as Mr Ewart submitted, the legislature preferred to use a more generalised term in s 48(3) on the basis that some settlements (such as those constituted under foreign laws) could be created otherwise than when property first becomes comprised in it. Certainly it is something of a leap to conclude from the use of different terminology in this different context that the legislature intended "at the time the settlement was made" in s 48(3) to have a meaning other than that indicated by its natural meaning.

(3) Nor can I see that support for HMRC's interpretation is provided by the operation of the transitional rule which introduced the deemed domicile rule as it applies to the excluded property rule. The effect of the transitional rule is that:

(a) If property became comprised in a settlement before the day on which the deemed domicile rule was introduced, whether it qualifies as excluded property under s 48(3) is to be assessed by reference to the settlor's domicile status, "at the time the settlement was made", as determined leaving the deemed domicile rule out of account.

(b) If property became comprised in the settlement after that date, whether it qualifies as excluded property under s 48(3) is to be assessed by reference to the settlor's domicile status "at the time the settlement was made", as determined taking the deemed domicile rule into account.

The legislature has chosen to apply s 48(3) differently, in terms of whether the deemed domicile rule applies or not in determining whether property is excluded property, according to when the property became comprised in the settlement. The plain purpose of the transitional rule is to provide a fair way of introducing the deemed domicile rule seemingly given that, under the settlements regime, IHT charges may apply by reference to the value of property which became comprised in the settlement before the deemed domicile rule was known about and introduced. I cannot see that of itself this provision gives any steer on how the term "at the time the settlement was made" in s 48(3) is to be interpreted and so when domicile status (whether determined under the general law or under the deemed domicile rule) is to be assessed.

(4) I have not found it useful to examine every instance where the settlements regime refers to the time a settlement is/was made. It may well be that HMRC's interpretation of that term could be applied consistently wherever used without too much difficulty. However, that does not detract from the conclusions set out above.

47. In my view, no clear policy in support of HMRC's interpretation of s 48(3) emerges from a comparison of the operation of the IHT regime as regards transfers of value by individuals otherwise than involving settlements with how it operates in relation to settlements. It must be borne in mind that the rules operate differently in these different contexts.

(1) To recap, in outline:

(a) The general regime essentially imposes IHT on the value transferred on the occurrence of specific events, namely, (i) any disposition made during a person's lifetime as a result of which the value of the transferor's estate is reduced (s 3), or (ii) a deemed disposition of a person's estate on his death (s 4). For ease of reference, I refer to both or either of those events as a transfer of value. In broad terms, the regime operates on the basis that only transfers of value made by domiciled individuals are to be captured in the IHT net. The effect of the excluded property rule in s 6(1), in combination with the charging rules (in ss 3, 4 and 5), is that there is no IHT charge on the value transferred in respect of foreign property which is the subject of a transfer of value if the transferor/deceased was not domiciled at the time of that transfer.

As regards settled property: (i) there may be a charge to IHT when a settlor transfers property into a settlement (broadly, except where the settlor/his spouse has a QIIP in the property) subject to the excluded property rule, (ii) for settlements which fall within the settlements regime (A) there is a charge on the value of relevant property comprised in the settlement every ten years, and (B) there are exit charges on the value of relevant property when it ceases to be such otherwise than by becoming excluded property. The ten-year charges and exit charges are imposed at a maximum of 6%, it appears with the aim of ensuring that the value of relevant property is subject to a total IHT charge over time broadly commensurate to that which would apply if the property were owned by a domiciled individual on death.

There is no ten-year charge or exit charge on property which qualifies as excluded property under s 48(3); similarly to the position under s 6, property qualifies if it is foreign property and the settlor was domiciled at the relevant time - the determination of that time is of course the issue here.

I note that the IHT charges set out above are subject to the application of any exemptions and reliefs which are not in point here.

(2) It is notable that, as HMRC accept, s 48(3) operates on the basis that the status of property transferred into a settlement which is within the settlements regime when it is first created is, in effect, fixed by reference to the settlor's domicile status at that time. The legislature has chosen not to take any subsequent change in domicile status into account in determining, in effect, if a ten-year or exit charge is due by reference to the value of such property: (a) if the settlor was non-domiciled at that time, the property in question can qualify as excluded property under s 48(3) if the settlor becomes domiciled, even where that occurs immediately after the creation of the settlement, (b) conversely, if the settlor was domiciled at that time, the property in question can never qualify as excluded property under s 48(3) even if the settlor were to become non-domiciled and remain so throughout the lifetime of the settlement. The legislature could have included specific provisions to take subsequent changes in domicile into account but has not done so. For example, the legislature could have provided for (i) a partial/apportioned ten-year charge and/or exit charge in respect of particular foreign property according to the amount of time for which the settlor was domiciled and that for which he was non-domiciled, whilst the property was comprised in the settlement, and/or (ii) whether property qualifies as excluded property under s 48(3) for the purposes of these charges to be determined by reference to the settlor's domicile status when the relevant event which triggers those charges occurs (namely, at the ten-year anniversary or when an event which triggers an exit charge occurs such as the property ceasing to be comprised in the settlement).

48. Hence the general regime and the settlements regime are not directly comparable in material respects:

(1) Under the general regime, there is a clear rationale for testing whether foreign property qualifies as excluded property by reference to the individual's domicile status at the date of the transfer of value. Given the intended scope of the regime, it seems natural and logical to test the domicile status of the transferor/deemed transferor when the one off event occurs which triggers a possible charge to IHT by reference to the value transferred by that event.

(2) I cannot see that it follows from that approach that, whether a ten-year charge or an exit charge in respect of foreign property which is or was comprised in the settlement is due, is to be determined by reference to the settlor's domicile status when a different potentially chargeable event took place in relation to that property - its earlier transfer into the settlement. The approach under the two regimes is not comparable, in particular, given that, as noted, unlike under the general regime (a) the legislature has chosen specifically not to determine whether property qualifies as excluded property under s 48(3) for the purposes of the ten-year and exit charges by reference to the settlor's domicile status when the event which triggers those charges occurs, and (b) those charges are intended to operate, in effect, in combination to capture the value of relevant property in the IHT net over a period of time.

49. With the points made in [47] and [48] in mind, I cannot see that a departure from the natural meaning of the legislation is justified on policy grounds and/or on the basis that the result of applying the appellant's interpretation of s 48(3) where property is added can be regarded, as it was put in the caselaw Mr Ewart referred to, as a clear and gross balance of anomaly, which the legislative promoters and the draftsman could not have envisaged, and which they could not have been prepared to accept in the interest of a supervening legislative objective:

(1) Once it is accepted that, for the purposes of the ten-year and exit charges, the legislature intended that whether property transferred into a settlement on its creation may qualify as excluded property (subject to it being foreign property at the relevant time) is to be determined, once and for all, by reference to the domicile status of the settlor at that time, it is difficult to see any rationale for the legislature to be taken to have had a different intention as regards those charges when applied to the value of property added to the settlement at a later time. As set out in [48], the acknowledged effect of s 48(3) is that, even if a non-domiciled settlor became deemed domiciled the day after the trust is created, the whole of the value of property by which it was originally constituted (including any increase in value) can qualify as excluded property and so be exempted from the ten-year charge and any exit charge. It seems to me that for property added later to qualify as excluded property and so also to be exempted from those charges, if the settlor was non-domiciled when the settlement was originally created, is no more anomalous (and see the comments below). As noted, if the legislature intended to take account of changes in the settlor's domicile status which take place after its creation in assessing whether and to what extent ten year and exit charges are due, they could have done so in a specific, clear way but they have not.

(2) I do not accept Mr Baldry's submissions that HMRC's interpretation of s 48(3) gives a position which is consistent with that for transferors/deemed transferors under the general regime, and gives a result which is consistent with the overall intended effect of the IHT rules applicable to settlements:

(a) Mr Baldry emphasised that, on the introduction of the CTT regime, the intention was that capital transfers would have tax consequences and it cannot have been intended that individuals could get around them by setting up a settlement when non-domiciled and adding property when domiciled. However, for the reasons already set out above, the comparison is not direct or straightforward given the different nature of the relevant charges under the settlements regime compared with the charges under the general regime. Also, this does not take account of the fact that, as set out below, the transfer of the property into a settlement at least in some instances is of itself a taxable event.

(b) As set out in Part B (see [16] to [18]), it appears that the intent is broadly, to impose a charge to IHT on the value of relevant property over time commensurate with the charge which would apply if it were owned by an individual who dies owning it. In cases where the settlor of a relevant property trust is either non-domiciled or domiciled at all relevant times, then the position would be comparable with that if he instead owned the property at the time of death. However, where the settlor's domicile status changes, I cannot see a meaningful comparison can be drawn. On either interpretation of s 48(3), whether foreign property can qualify as excluded property is fixed by reference to the settlor's domicile status at a point in time: on the natural interpretation, when the trust was created, and on HMRC's view, when the property was transferred to the trust. Therefore (a) on the natural interpretation changes in domicile after the trust was created are not taken into account at all in analysing whether any property transferred to the settlement qualifies as excluded property, (b) on HMRC's interpretation, changes in domicile status are taken into account but only any change when further property is added to the trust as regards the status of that property as excluded property. So on HMRC's interpretation (i) if the settlor was non-domiciled when the property was transferred to the trust, it can qualify as excluded property for the purposes of the ten-year and exit charges even if the settlor became domiciled the day after the transfer and remains as such, and (ii) if the settlor was domiciled when the property was transferred to the trust, it can never qualify as excluded property for the purposes of those charges, even if the settlor became non-domiciled and remained as such. Hence, HMRC's interpretation does not necessarily give a result which accords with the result which would apply if the property were held by an individual throughout any more than the natural interpretation does. That is because, under the general regime, whether IHT is due on foreign property owned on death depends simply on the deceased's domicile status at that point in time. If the individual was non-domiciled for some years but domiciled when he died owning foreign property, the property would not be excluded property. Whereas, on HMRC's interpretation, if he had transferred foreign property into a settlement when non-domiciled, it would retain its status as such on an indefinite on-going basis regardless of whether he later became domiciled.

(3) As Mr Ewart submitted, it is entirely possible that the legislature intended to have a simple rule for all property comprised in it, whether added initially when the settlement is first created or later, as the natural meaning of the provision indicates. In summary, when the provisions were originally enacted and on an ongoing basis, as set out in further detail in the submissions (a) there was a charge to tax on the transfer by a domiciled person of property into a settlement other than a QIIP. I refer to this as an "entry charge"; as Mr Ewart pointed out, when the relevant provisions were first introduced that was a very high charge, and (b) whilst there was no tax charge on the transfer of property into a settlement where the settlor/his spouse had a QIIP, there would be a charge to IHT when that interest ended, if the settlor/spouse was domiciled at that time. In light of that, the legislature may well have taken the view that the later addition of foreign property (or property which could be converted into foreign property) by domiciled persons was not a particular concern, and it was better to have a rule which was simple and easy to operate in practice. In my view, that is the clear implication of the legislature choosing, as HMRC accept, to have such a simple rule as regards property transferred into a settlement when it is first created. Such an approach cannot be regarded as unreasonable, absurd or capricious.

50. I accept that (1) HMRC's interpretation of s 48(3) does not necessarily lead to an unworkable result, and (2) there is no material guidance to be drawn from the fact that the same result as applies on the appellant's interpretation of s 48(3) may come about by parties using a different structure involving the transfer of UK property to an overseas company owned by a settlement. However, that does not detract from the points made above.

Part C - tribunal's jurisdiction in respect of the repayment notice

Notice of repayment and right to appeal

51. This part and Part D of the decision relates only to the repayment notice which states the following:

"The Commissioners for Her Majesty's Revenue and Customs have determined -

In relation to –

A. A settlement made on 14 September 1992 by [the Settlor] ("the Settlor") and called the Tiodab Trust ("the Settlement");

B. A claim for repayment of inheritance tax made by Salamanca Group Trust (Switzerland) SA ("former name of Accuro Trust (Switzerland) SA") ("the Trustee") on 20 May 2015 ("the Claim") in relation to inheritance tax and interest paid on 15 May 2013 in respect of the ten year anniversary charge occurring on 14 September 2012.

That -

1. It has not been proved to the satisfaction of the Commissioners for HM Revenue and Customs that too much inheritance tax and interest has been paid having regard to section 241 and 255 Inheritance Tax Act 1984 ("IHTA").

2. Immediately before the ten year anniversary occurring on 14 September 2012 the value of the relevant property comprised in the Settlement was £43,397,593.00.

3. A charge to inheritance tax arises on the occasion of the ten year anniversary in respect of all of the property comprised in the Settlement which was relevant property immediately before that anniversary, in accordance with sections 64 and 66 IHTA.

4. The rate of inheritance tax calculated in accordance with section 66 is 5.958% and the inheritance tax due in respect of the charge is £1,680,658.59 all of which has been paid."

52. HMRC issued the notice under s 221 which provides as follows:

"Section 221 - Notices of determination

(1) Where it appears to the Board that a transfer of value has been made or where a claim under this Act is made to the Board in connection with a transfer of value, the Board may give notice in writing to any person who appears to the Board to be the transferor or the claimant or to be liable for any of the tax chargeable on the value transferred, stating that they have determined the matters specified in the notice.

(2) The matters that may be specified in a notice under this section in relation to any transfer of value are all or any of the following –

(a) the date of the transfer;

(b) the value transferred and the value of any property to which the value transferred is wholly or partly attributable;

(d) the tax chargeable (if any) and the persons who are liable for the whole or part of it;

(e) the amount of any payment made in excess of the tax for which a person is liable and the date from which and the rate at which tax or any repayment of tax overpaid carries interest; and

(f) any other matter that appears to the Board to be relevant for the purposes of this Act."

53. Where the appellant appeals against a determination such as the notice (under s 222), and the appeal is subsequently notified to the tribunal, the tribunal must determine the "matter in question" (under s 223D) which is defined as the "matter to which an appeal relates" (s 223I) On appeal, under s 224, the tribunal must confirm the determination appealed against unless the tribunal is satisfied that it ought to be varied (or further varied) or quashed.

Overview of the issues

54. In HMRC's view, the tribunal has a "supervisory" jurisdiction only as regards the appeal made against the relevant matters in the repayment notice including the conclusion that s 255 applies. In their view the question for the tribunal is confined to being, broadly, whether the relevant HMRC officer, Mr Ryder, made the relevant decisions reasonably or unreasonably. The appellant considers that the tribunal has a full "appellate" jurisdiction to decide the relevant issues for itself.

55. There have been a number of cases which have established that, where an appeal is made to the tribunal in respect of HMRC's exercise of discretionary powers conferred on them by statute, the tribunal's jurisdiction is limited, as explained, for example by Lord Lane in Customs and Excise Commissioners v J H Corbitt (Numismatists) Ltd [1980] STC 231 at 239:

"The jurisdiction of the tribunal in cases such as this where the Commissioners are exercising discretionary powers has been clearly established in previous cases. It is, for instance, clear that the tribunal cannot substitute its own discretion for that of the Commissioners for the tribunal has no discretion in these matters. If it is alleged that the Commissioners have reached a wrong decision then there can be a question of law but only of a limited character. The question would be whether their decision was unreasonable in the sense that no reasonable panel of Commissioners properly directing themselves could reasonably reach that decision. To enable the tribunal to interfere with the Commissioners' decision it would have to be shown that they took into account some irrelevant matter or had disregarded something to which they should have given weight."

56. In John Dee Limited v Customs and Excise Commissioners [1995] STC 941 ("John Dee") the Court of Appeal decided this was the correct approach to follow as regards an appeal against a decision of HMRC that it appeared to them that it was requisite to require security for VAT purposes. Lord Justice Neill said this in his conclusions:

"In furtherance of his argument that, once the tribunal had decided that the decision of the Commissioners was flawed, it could substitute its own discretion, counsel for the company was constrained to submit that it was for the Tribunal to decide whether it appeared to it "requisite for the protection of the revenue" to require a taxable person to give security. I am quite unable to accept this submission. It seems to me that the "statutory condition".... which the Tribunal has to examine in an appeal under s 40(1)(n) is whether it appeared to the Commissioners requisite to require security. In examining whether that statutory condition is satisfied the Tribunal will, to adopt the language of Lord Lane, consider whether the Commissioners had acted in a way in which no reasonable panel of Commissioners could have acted or whether they had taken into account some irrelevant matter or had disregarded something to which they should have given weight. The Tribunal may also have to consider whether the Commissioners have erred on a point of law. I am quite satisfied, however, that the Tribunal cannot exercise a fresh discretion on the lines indicated by Lord Diplock in Hadmor. The protection of the revenue is not a responsibility of the Tribunal or of a court.

I do not consider that it is necessary or would be appropriate in this case to give guidance as to other categories of appeal under section 40(1), other than to say that in my view the function and powers of a Tribunal in each case will depend in large measure on the nature of the decision appealed against and of course on any special statutory provisions.....

....It was conceded by Mr. Engelhart, in my view rightly, that where it is shown that, had the additional material been taken into account, the decision would inevitably have been the same, a Tribunal can dismiss an appeal....."

57. HMRC's reasoning that this approach applies here is as follows:

(1) Whilst the tribunal's powers are prescribed by s 224, that section provides no indication of the nature of the tribunal's jurisdiction on appeal. The "matter in question", which the tribunal must determine comprises matters within (a) s 221(2)(e) as regards the statement in the repayment notice that, under s 241, it was not proved to HMRC's satisfaction that too much IHT was paid as a matter of law, and (b) 221(2)(f) as, by the reference to s 255, HMRC determined that, under s 241, it was not proved to their satisfaction that too much IHT was paid as they considered that s 255 applied with the effect that the correct sum was paid.

(2) Parliament has chosen in s 241 to specify that repayment is only due when HMRC are satisfied of that state of affairs. That confirms that it is the opinion of HMRC which is the decision that is the subject of challenge on appeal as the "matter in question". The jurisdiction of the tribunal must necessarily be supervisory in nature; otherwise the fact that it is HMRC's opinion which is the subject matter of the appeal would be immaterial.

(3) This is consistent with the tribunal's approach in other cases albeit in a different context: see Currie v HMRC [2017] UKFTT 539 (TC) and Hymanson v HMRC [2018] UKFTT 667 (TC). In each case the tribunal held that it had supervisory jurisdiction (a) in the first case in relation to the statutory condition (for special relief) whether "in the opinion of the Commissioners it would be unconscionable for the Commissioners to seek to recover the amount" (see [29]), and (b) in the second case, in relation to a statutory condition which stated "if HMRC have reason to believe" that a specified event had occurred, they were entitled to issue a certificate (see [48] to [56]).

58. HMRC further submitted that (1) on the basis that HMRC are correct in relation to the jurisdiction issue, the decision of Mr Ryder that s 255 applies is eminently reasonable and therefore the appeal against the repayment notice falls to be dismissed, and (2) if the tribunal decides that it has full appellate jurisdiction, as a matter of law s 255 applies with the effect that the disputed tax was correctly paid for all the reasons set out below.

59. The appellant disputed all of these points. Mr Ewart submitted that:

(1) HMRC's argument that the tribunal only has supervisory jurisdiction is misconceived:

(a) The appellant (a) has a statutory right to appeal against any determination in the repayment notice under s 222(1) and was, therefore, entitled to appeal against the determination of (i) the amount of relevant property immediately before the ten-year anniversary, and (ii) the amount of tax due, as set out in paras 2 and 4 of the repayment notice, and (b) has exercised that right to appeal.

(b) It is notable that in John Dee there was a right of appeal expressly in respect of the statutory condition in question. That is not the case here. The jurisdiction of the tribunal on an appeal against the repayment notice is set out in s 224. It is for the tribunal to decide whether the repayment notice is correct in its determination of the amount of relevant property at the ten-year anniversary and the amount of tax due. That is a full merits appeal in which the onus is on the appellant to satisfy the tribunal that the notice is wrong. If the appellant succeeds then it is entitled to the repayment which it claimed.

(2) If the tribunal's jurisdiction is supervisory only, the decisions in the repayment notice were not reasonably arrived at for the reasons set out below. If the tribunal agrees with that, the tribunal must allow the appeal against the whole of the notice and reduce the value of the relevant property and the amount of IHT due to zero. If the tribunal does not do that the effect of s 221(v) is that the notice will remain as conclusive for the purposes of IHTA as regards the determinations in paras 2 and 4. Either this is a quasi-judicial review appeal which determines the entire matter or it is an orthodox appeal against paragraphs 2 and 4 of the repayment notice.

(3) If it is held that the tribunal has full appellate jurisdiction, the requirements of s 255 are not met for the reasons set out below.

60. Mr Baldry replied that (1) the fact that IHTA is generally far less prescriptive than the VAT regime does not affect the analysis of the tribunal's jurisdiction in this case, and (2) there are no procedural difficulties. If HMRC's decisions are found to be correct, the tribunal simply has to uphold the determination and, if they are wrong, the tribunal can reduce the determination to nil, and can direct repayment.

Decision on the scope of the tribunal's jurisdiction

61. My conclusions on the scope of the tribunal's jurisdiction as regards the appeal made against the determinations in the repayment notice are as follows:

(1) The first step must be to decide what the "matter in question" is which depends on what HMRC have determined in the repayment notice. On a reasonable reading of the notice, in light of the introductory wording and the reference in para 1 to both s 241 and 255, HMRC determined in paras 1 to 4 that for the purposes of s 241, it was not proved to their satisfaction that too much tax was paid in respect of the relevant ten-year anniversary, so that no repayment of the disputed tax is due to the appellant, on the alternative basis that:

(a) HMRC considered that, as a matter of law, the property was relevant property (not excluded property) with the value stated in para 3 with the consequence that the disputed tax due is as stated in para 4 (reading paras 2, 3 and 4 in context).

(b) HMRC considered that s 255 applied with the consequence that (even if HMRC's position is not correct as a matter of law), the property is to be taken to be relevant property with the value stated in para 3 with the consequence that the disputed tax due is as stated in para 4 (reading paras 2, 3 and 4 in context).

(2) Hence, the matter in question, in respect of which the appellant has appealed to this tribunal, encompasses all the decisions described in (1). I cannot see, as Mr Ewart seemed to suggest, that the appellant can be taken to have appealed only in respect of the statement in the repayment notice of the value of the relevant property and the amount of tax (in paras 2 and 4). That would involve somehow divorcing those statements from the overall context in which they are made whereby they are plainly integral to and a consequence of the decision made by HMRC under s 241 refusing the appellant's claim for repayment of the disputed tax.

(3) As in other cases where discretion is conferred on HMRC, the wording of s 241 (that a repayment is due only where it is proved to the satisfaction of HMRC that too much tax has been paid) indicates that the tribunal's jurisdiction is supervisory in nature; otherwise the discretion conferred on HMRC would be deprived of meaning. However, in the particular circumstances of this case, it does not confer on HMRC meaningful discretion, in that it is undisputed that, if Mr Ryder is found not to have applied the law correctly, the decisions cannot stand under the principles set out above:

(a) As regards the decision set out in (1)(a), (i) whether too much tax has been paid depends on the correct application of s 48(3) as a matter of law, and (ii) I have decided, as set out in Part B, that HMRC's interpretation of s 48(3) is incorrect as a matter of law. It follows that Mr Ryder's decision on that was not reasonably arrived at.

(b) As regards the decision set out in (1)(b), (i) whether too much tax has been paid depends on whether s 255 applies as a matter of law, and (ii) for all the reasons set out below in Part D, I have found that, as a matter of law, s 255 does not apply in this case, and Mr Ryder did not apply the proper legal test under s 255 such that that decision cannot stand for that reason alone.

Part D - Application of s 255 and whether Mr Ryder's decision was reasonable

Facts and evidence

62. In this Part D, (1) I have considered whether s 255 applies as a matter of law and concluded that it does not, and (2) I have considered whether Mr Ryder's decision was reasonably arrived at and concluded that it was not, in particular, as he failed to apply the legal tests set out in s 255 correctly.

63. I have determined the factual issues relevant to Mr Ryder's decision in the repayment notice and whether, as a matter of law, the requirements of s 255 are met on the basis of the documents produced and the evidence of (1) Ms Lorraine Jweinat who was employed by the Trustee from July 2008 to 2016, (2) Mr Ryder of HMRC, and (3) for HMRC, Ms Summers and Mr Goldstone, and for the appellant, Ms Chamberlain, all of whom are practitioners with considerable experience in this area of tax law. All of the witnesses attended the hearing and were cross-examined. I found them to be honest and credible.

Evidence of Ms Jweinat

64. I consider it apparent from Ms Jweinat's evidence and that of Mr Ryder that the disputed payment of tax was made and accepted in satisfaction of a liability to tax on the basis of the stated view. Ms Jweinat explained that she had day to day responsibility for the day to day administration of the Trust when she was the Trustee's employee. She gave the following evidence:

(1) The Settlor was a senior international executive of the Macsteel Group ("Macsteel") from 1981 until his retirement in December 2011. Macsteel operated a share incentive plan, which provided executives with the opportunity to participate indirectly in the cumulative profits of the group. The Settlor was awarded A and C shares under that plan. The A shares were acquired as set out above and the C shares were acquired as funded by sums contributed to the Trust before 6 April 2005 when the Settlor became UK domiciled.

(2) Following his retirement from Macsteel, the Settlor was required to terminate his participation in this plan and to dispose of the A and C shares. On 21 August 2012, the proceeds of the disposal of the A and C shares were paid to the appellant. The Trustee subsequently made the two capital distributions to the Settlor (in 2014 and 2017) from the proceeds of the A shares.

(3) Following the disposal of the A shares, Ms Jweinat carried out a review of the original transfers and of the trust records generally and found that cash had been added in mid-March 2006. She later became aware that the Settlor had become deemed domiciled in April 2005; there was a misunderstanding over the date of domicile. She undertook a tracing exercise from which it transpired that the cash contributed to the Trust after that date was used to pay the purchase price of the A shares. The Trustee sought advice and in April 2012 she and a director of the trust met with Ms Stephanie Jarrett of Baker & McKenzie. They then consulted Mr Robert Ham QC and a valuation of the shares was obtained. He was approached so they could understand the implications of the misunderstanding. He confirmed that the C shares did not form part of the relevant property, and noted that to ensure that the correct amount of tax is recorded with HMRC and the division between relevant and excluded property was clear going forward, the Trustee instructed KPMG to value both the A shares and the C shares for IHT purposes.

(4) Ms Jweinat was aware of the Liechtenstein Disclosure Facility ("LDF") as a means of regularising the tax affairs of the Trust in the period before 2009. She discussed this with Mr Jeff Millington of SW and he was formally engaged to provide advice on that on 28 January 2013. The Trustee thought it was prudent to engage SW to provide tax computations for the ten-year charge which arose on 14 September 2012. The appellant paid £1,702,142.73 in satisfaction of this charge and the accrued interest on 15 May 2013. On 29 August 2013, the appellant delivered an account to HMRC in which it accounted for the 2012 ten-year charge on the basis that the proceeds of the A shares were relevant property for the purposes of IHT but that the proceeds of the C shares were not relevant property. The advice from SW was that the assets in the form which were originally contributed to the Trust in 1992 to 2005 represent property that was originally contributed to the Trust before the Settlor became domiciled and therefore were excluded property, but as the Settlor had become deemed domiciled when the A shares were contributed to the Trust, the proceeds were not excluded property.

(5) On 21 November 2013, the appellant made an application under the LDF in relation to IHT. The IHT issues Ms Jweinat had in mind that might be resolved by the facility were the potential entry charges on the contributions that the Settlor had made to the Trust when he was deemed domiciled.

(6) The disclosure made by the appellant under the LDF included the following statements:

(a) It had recently come to light that inadvertently relevant property was transferred to the Trust by the Settlor in March 2006:

(i) On 16 March 2006, on the understanding that this was the final year in which he could still settle overseas assets into the Trust while he was not deemed to be domiciled, the Settlor transferred cash to be held as additions to the Trust fund; unfortunately, this was a misunderstanding.

(ii) On 30 March 2006, the Settlor sold the A shares to the Trustee at nominal consideration based on a value determined to be fair by the employee incentive scheme administrator given the shares were subject to certain conditions and restrictions. The consideration was ultimately paid for with the cash.

(b) The appellant also acquired the C shares on 27 March 2006 but it acquired them directly from the employee share incentive scheme and funded the purchase from funds already held within the Trust (ie excluded property) and not from the cash transferred to the Trust in March 2006.

(c) Both the A and C shares related to two awards in Macsteel granted to the Settlor, in January 2000 and January 2004. The Settlor intended that his interest in these would immediately vest in the Trust. However, due to an exceptionally protracted implementation period, the legal structuring of the non-UK employee incentive scheme, which held the Macsteel BV Global interest, the purchase of the A and C shares was not completed until much later than the Settlor understood it would be.

(d) When the Settlor retired from Macsteel in 2011 it came to light that the Settlor may well have been deemed to be domiciled in the UK for IHT purposes from the 2005/06 tax year.

(e) Although Queen's Counsel had confirmed that in his opinion the C shares do not form part of the relevant property, to ensure that the correct amount of tax was recorded with HMRC and the division between relevant and excluded property was clear going forward, the appellant instructed KPMG to value the A shares and the C shares for IHT purposes.

(f) Following the sale of both sets of shares, the consideration for the A shares was received in August 2012. The 2012 ten-year charge was calculated and the form IHT100 was submitted and a charge of £1,682,336.19 paid.

(g) "Provided that a settlor is non-domiciled and settles foreign situs property on an offshore trust, the funds will be treated as excluded property [under 48(3)]." The point was made that the Settlor was non-domiciled at the time he settled the trust on 14 September and any additions to the trust fund until 5 April 2005 were of foreign property and therefore excluded property. "At the time of the transfers to the trust in 2006, the settlor had become deemed domiciled and therefore IHT was due at that time on any assets settled, upon any distributions from relevant property, and upon the relevant property at the ten-year anniversary date." Ms Jweinat agreed that the position that was adopted here in the LDF is completely consistent with the position adopted in the IHT100 as to the treatment of the property within the Trust.

(h) The Trustee had elected to settle the IHT liability under the composite rate option provided for under the terms of the LDF. As a result, the only IHT due is the tax that is due after the LDF disclosure period. IHT is therefore limited to the tax years commencing 6 April 2009 onwards (other than the single rate charge for 2010/11, which should not be applicable here as no IHT resulted in that tax year).

(i) There was no tax to pay under the LDF as the Trustee has elected for the composite rate option calculation of liabilities. The liability under the LDF was £nil.

(7) On 9 May 2014, HMRC accepted the appellant's application under the LDF. Ms Jweinat's understanding was that the composite rate option involved applying a flat rate to income gains and IHT. The benefit of the LDF was that she could say to HMRC that since 2009 all the IHT has been paid properly, but because the liabilities are historic and occurred before 6 April 2009 the overall effect was that the slate is wiped clean for the 2006 IHT liabilities.

(8) When the Settlor received a distribution in 2014, Ms Jweinat approached Mr Millington to attend to any required UK filings. Mr Millington gave his opinion that the Trust was not in fact in scope as relevant property and wished to approach HMRC on that basis. This was agreed and SW set out their technical argument in a letter to HMRC of 20 May 2015 and filed a nil return in respect of the 2014 distribution. HMRC responded on 18 January 2016 disagreeing with this. In January 2017 there was another distribution and further returns were made as set out in Part A. Ms Jweinat confirmed that the change of view as regards the IHT position came as a bit of a surprise to her.

Evidence of Mr Ryder

65. Mr Ryder gave evidence as the HMRC officer who made the decisions in the repayment notice. Mr Ryder set out that he became substantially involved in this case in the summer of 2018 and explained the process as follows:

(1) When the Trustees made a tax account and paid tax, the data in the account was captured on the computerised system on 16 November 2013 and the tax and interest was calculated on that day as being a total of £1,682,724.97 and the words "PROVISIONAL CALCULATION" were inserted in the Notes panel on the calculation.

(2) The file for the account seems to have gone missing at that point. Otherwise it would have been referred to the compliance risk team given the large value of the assets. The relevant sum was accepted in satisfaction of the ten-year charge.

He did not know how the IHT account got lost. It may have been electronically available but there is a retention period attached to each item of post; HMRC are required not to keep things unnecessarily long. So it is possible it was scanned and then deleted.

(3) On 20 May 2015 SW wrote to HMRC to report the distribution which took place on 25 September 2014. The Trustees submitted an IHT100 account showing no chargeable property and no IHT due.

(4) Ms Boraster of HMRC replied to SW on 18 January 2016 and maintained HMRC's position that each disposition of property was a settlement for the purposes of s 48(3). SW responded on 24 March 2016 disputing that. It was agreed to await the outcome of the Barclays case which was released on 13 October 2017. On 28 July 2017 SW sent a further IHT100 in relation to the distribution made on 31 January 2017 which again showed no tax as due.

(5) On 3 November 2017 SW sent a revised IHT100 for the ten-year charge. They said that Jersey counsel was being instructed with a view to possible action in Jersey to set aside dispositions made by the Settlor to the Trust. Ms Boraster acknowledged receipt on 30 November 2017 and said that the Barclays case might be appealed.

(6) On 29 March 2018 Ms Boraster informed SW that the Barclays case was now final and that she would be seeking further advice.

(7) Later in 2018 Mr Ryder reviewed the papers and made the decision on the application of s 255. Having sought advice internally he concluded that:

(a) Although no formal claim for repayment was made, the letters from SW dated 15 May 2015 and 24 May 2016 amounted to a claim.

(b) Section 255 could be engaged in relation to the ten-year charge as there had been a payment brought to account made by the Trustees on a calculation of tax. That process meant that the payment had been accepted in satisfaction of a liability to tax. Marking the payment as provisional does not affect the position as that was done simply to protect HMRC's ability to open a compliance check and impose additional tax at a later date.

(c) HMRC had a well-established view stated in RI 166, and whilst it appears that HMRC's view was not universally accepted to be correct, overall it was generally received and adopted in practice.

66. He said the following about the processes relating to accounting for and paying IHT in this case and more generally and how HMRC identify risks and his decision:

(1) Since November 2005 he has been involved in almost all IHT cases which have entered litigation in the UK or the EU courts and was aware of all cases approaching litigation where the taxpayer contests HMRC's view. As part of his role of working on improvements to taxpayers' experience of interacting with HMRC, he looks at the processes for handling IHT accounts on receipt:

(a) When HMRC receive an IHT100 relating to settlements, the paper form and attachments are scanned into a document management system and channelled to the Customer Services Group who create a case reference number. They then data capture information into the computerised assessing system which is used to generate a calculation of the IHT due. It has been their practice to mark these calculations as provisional to avoid the taxpayer arguing they are final as HMRC do not review the full detail at that point, so it is possible the account would throw up risks that would need to be looked at in more detail and that HRMC may need to collect more tax. Where the account meets certain criteria it may then be referred to the Inheritance Tax Risk Assessing Team within that group, who decide if there are any risks that require a compliance check to be carried out and, if so, it is assigned to a compliance officer. The proportion of cases referred for a compliance check is around 5% of the returns received and roughly two thirds of the checks result in additional IHT being imposed.

(b) The system has remained broadly the same for a number of years. HMRC used paper filing until a few years ago. The document management system came in the 2010s and before that everything was paper driven and someone had to read every single account. When the papers come in HMRC data capture them if there is tax due and do so for some of the non-tax paying cases for statistical purposes. They are briefly evaluated as to whether they meet any quick criteria for obvious errors (such as mathematical errors and incorrect signatures), to assess if HMRC can accept the return into the computerised system or need to reject it.

(c) Once the return is put through to the Risk Assessing Team they check through the detail and read every single one looking for risks on whether the tax is correctly declared. Non-domiciled settlors are certainly higher risk. Not all are selected for an assessment/check. Changes in domicile may be more likely to throw up a risk that needs investigation. As regards settlement returns, there are considerations regarding the valuation of property (offshore property may be a higher risk) and issues around the domicile of settlors. It was not necessarily the case that a trust return with a non-domiciled settlor and excluded property would be assessed as a risk. Big numbers can excite interest and if HMRC have other information about the trust, they would try and synthesise that with what they were seeing.

(d) The Risk Assessing Team normally decide which returns will be subject to a compliance check. He did not know if it was commonly discovered that trusts had accounted for property as excluded property in circumstances in which HMRC took the view that it was not. That was not his area. He understood that trustees of a discretionary trust with a non-domiciled settlor that thought that all the settled property was excluded property had no obligation to report on a ten-year anniversary or on property exiting from the trust. He thought if the view was taken that property was excluded property there was no obligation to render an account or make a report if there was a termination of an interest in possession trust before 2006 during the lifetime of the life tenant or on death. If the settlor was non-domiciled and had no UK property there was no obligation on them to take out a grant of representation on death.

(2) He was not involved when the claim for repayment was made and Ms Boraster sent her letter in 2016. The technical team would have assisted with the drafting of this. The letter sets out quite a detailed explanation of HMRC's technical view on the law in relation to the legal issue and reflects HMRC's view at the time. It was not usual for there to be a nil return showing no tax payable because all the property is excluded property.

(3) Usually decisions in repayment claim cases would be recorded with reasons on the case papers as part of the internal governance procedures but he thought that this was done orally in this case. He worked in the same office as the case worker. He sought advice internally, in discussion with technicians and lawyers around what that decision should be and what form it should take and there would have been a record of the communications between them. There may have been an email but it would not have been detailed in terms of reasoning.

(4) The actual decision is the repayment notice; that sets out the reason in headline terms and the correspondence would set out more detail around how HMRC had reached that. He confirmed that there was no correspondence with the taxpayer about the s 255 issue at all and that the taxpayer was never consulted or approached or asked by HMRC about s 255 and there is no reasoning in the repayment notice at all. He said that should have been in the correspondence. He did not think the letter of 26 September 2018 reflects the full reasoning, because that would have been developed up to the issuing of the notice but it does reflect some of the reasons.

(5) The discussion they had with the technicians and lawyers was about what format the decision should take. His understanding was that the repayment notice was appealable and could be challenged in the normal way. He did not know if he was fully clear that it was his decision that could be challenged. He thought that is why they had quite a lot of discussion with the technicians and the lawyers.

(6) He was familiar with how the system works in respect of notices of determination, appeals and then what happens. He has dealt with a fair few. He agreed that if the notice determines the amount of tax that is due it is conclusive, subject to an appeal and, if the taxpayer succeeds on appeal and the amount is reduced, that would also be conclusive. He thought that the repayment notice does not just say that HMRC are not satisfied but states the amount of tax due and the amount of relevant property follows the advice, probably because this is the first type of this case.

67. He also made the following points in his witness statement:

(1) The focus in the period before 2006 was very much on the status of the settlement on the death of the life tenant or settlor as discretionary beneficiary which would potentially be chargeable at a rate of up to 40%. Following the changes introduced by FA 2006 there was also the increased risk of the 20% charge (25% where the settlor paid the tax) on the value of added property where the settlor was domiciled.

(2) He referred to correspondence between the Chairman of the Capital Taxes Sub-Committee of the Chartered Institute of Tax ("CIOT") and the IHT policy team at HMRC. In summary:

(a) On 25 June 1998 Mr Matthew Hutton wrote to HMRC and Mr Lakhanpaul of HMRC responded on 3 July 1998. Mr Hutton raised that one of the members had asked how the stated view can be correct as it does not seem to fit with the rest of the legislation. In his response Mr Lakhanpaul said:

"There seems to be some misunderstanding here. Our comments...do not state, and were not intended to suggest, that an addition of assets by a settlor... is itself treated as constituting a separate settlement, made at the time of the addition. As explained in the article, our view is that, in relation to any particular asset comprised in a given settlement, the settlement was made at the time when the asset (or any other asset...) became subject to the trusts of the settlement. Put another way, new assets added by a settlor to his/her existing trust count as assets freshly settled So, it is the settlor's domicile at the time of the addition, ...which is relevant in determining whether the asset is 'excluded property'. As I am sure you will appreciate, sections 80 and 82 ... may also have a bearing on the application ...of the provisions on 'excluded property'."

(b) On 12 October 1998 Mr Hutton asked for clarification:

"Are you saying that, although there is a single settlement, that settlement is made in relation to each asset comprised in the settlement at the time when that asset becomes so comprised."

Then he set out an example and asked what the position was.

"I am sorry if the comments in our letter of 3 July lack adequate clarity. It is indeed our view (subject to section 44(2))...that there would be a single settlement which was made in relation to each asset comprised in the settlement at the time when that asset (or any previous asset which the current asset now represents) became so comprised."

(d) On 20 May 1999 Mr Zigmond, then the Chairman of the Sub-Committee wrote to HMRC and a response was sent by Mr Peter Twiddy of HMRC on 28 September 2001. Mr Zigmond's comments included the following:

"What is the Inland Revenue's view in circumstances...where out of the assets 1.5 million a proportion (0.3 million ...) is situated in the UK and the remaining 1.2 million is situated outside the UK, but it is not possible to say on the ten-year anniversary whether the 0.3 million derives from the initial settlement of 1 million or the later 0.5 million addition? In these circumstances would you agree the trustees can properly regard the UK situated assets with a value of 0.3 million as forming part of the 0.5 million later additions to trust, so that in the Inland Revenue's view liability arises on 0.5 million (but no greater sum).

...Please could you clarify further the Inland Revenue's view on this, and in particular how the Inland Revenue's view is supported by the words of the legislation...A more detailed explanation, as well as being most helpful to me personally, may be of general interest to our members (and may save the Inland Revenue correspondence in specific cases)."

(e) In his reply Mr Twiddy said:

"...There can be no single response to this question, we would have look at each case on merits...certainly we would expect trust records to be kept scrupulously...In my experience the Revenue does take a pragmatic and sympathetic view whenever this can be warranted...I do not see why the reference to the settlement in these provisions should be read differently from the settlement in section 48(3). On the interpretation of 4(c) above one need not draw that inference, indeed to do so would result in manifest absurdity."

Mr Ryder commented that Mr Twiddy made clear that HMRC's view was as set out in RI166 and in the IHT Manual at IHT27220. He said these comments confirmed HMRC's view of the law and it is that view which HMRC officers would have applied in practice.

(3) He is only aware of 2 instances where taxpayers have taken issue with HMRC's stated view. One case is Barclays and the other was resolved without the need to issue a notice of determination. He is not aware since 1997 of any other instances where a taxpayer took a different approach to that of HMRC. Had a taxpayer delivered an IHT account expressing a different view, the account would have been referred to the Risk Assessing Team and on to a compliance case worker to carry out a compliance check and had a taxpayer persisted in disputing HMRC's stated view, the matter would have proceeded to litigation which he would have known about. The absence of any decision on this issue confirms that no taxpayer has ever challenged HMRC's position.

(4) The textbooks fall into 2 categories and his comments on them were as follows:

(a) Works where the author reports HMRC's stated view without qualification or any substantive criticism: (i) Eastway and Richards Tax Advisers' Guide to Trusts (2002) ("Eastway"), at 6.166, (ii) Tolley's Inheritance Tax 2015/16 (2015) (Gunn) at 258 ("Tolley") observes that HMRC's interpretation is open to question although it has not so far been challenged on any appeal, (iii) Hutton on Estate Planning - Practical Solutions to Today's Problems ("Hutton") (6^th edition, 2013). The authors disassociate themselves from views of some advisers that a pilot trust can safely be established with the substantive funds added later when a settlor has become domiciled observing this is in direct opposition to the stated view. They express disappointment that the legislation taken as a whole is not entirely consistent, (iv) Fosters Inheritance Tax (November 2015) ("Fosters") at J3.32 reports the logical consequence of the stated view that additions of property to a settlement created when the settlor was domiciled made when he had become non-domiciled are excluded property, (v) Tiley and Collison UK Tax Guide (2007/08) ("Tiley") at 47.5 reports the consequences of HMRC's stated view. That passage remains the same in the 2010/11 edition at 48.15 and in the 2011/12 edition at 56.15 when the authors included the barristers Keith Gordon and Ximena Montes Manzano, and (vi) Summers, The Offshore Trustees UK Tax Handbook (2011) ("Summers") at 3.4 where the author reports HMRC's view set out in the IHT Manual and in a section headed "Traps and Tips" advises: "To ensure the trust is an excluded property settlement, all funds/assets must be added at a time when the settlor is not UK domiciled or deemed domiciled."

(b) Works which provide a critical analysis of HMRC's stated view and express varying degrees of doubt as to its accuracy: (i) McCutcheon on Inheritance Tax ("McCutcheon") (5^th edition 2009). The authors express doubts as to the correctness of the stated view and recommend "all things being equal" that settlors in this position place property in a separate settlement. The authors draw parallels to the situation arising with additions made after the changes in 2006 to settlements created before FA 2006 came into force, (ii) Trust Taxation (Chamberlain and Whitehouse) (2006) ("Chamberlain"). The authors refer to the stated view as being controversial and point to the parts of the legislation that do not support the stated view and also refer to the parallel mentioned above. They suggest it is common sense to follow the stated view where practical whilst maintaining there is a strong argument that property additions are excluded property on the wording of s 48(3), (iii) Offshore Tax Planning Clarke ("Clarke") (14^th edition, 2007) observes that the stated view may be wrong but urges caution by avoiding property additions as the argument is not worth having, and (iv) Taxation of Foreign Domiciliaries, Kessler ("Kessler") (4^th edition, 2005) considers that the stated view leads to the conclusion that separate settlements are made which he accepts is not conceptually impossible but he maintains that as a matter of fact separate settlements are not created. He observes that it may take litigation before HMRC amend the stated view. Until the point is resolved he advises that trustees should follow the advice in the stated view about keeping adequate records where there is added property.

(5) Mr Ryder commented that (a) the authors acknowledge HMRC's view and essentially advise that in practice the stated view should be adopted or at least caution in adopting a different position, and (b) the doubts expressed have translated into very few contested cases for HMRC. That leads him to the view that taxpayers have taken a much more cautious approach to the matter, acknowledging that the stated view was correct or, in any event, should be adopted in practice.

At the hearing he clarified he had looked at all the above works at the time he made his decision apart from Chamberlain.

(6) He had checked postings on the Trust Discussion Forum, a publicly accessible moderated forum for discussion by practitioners hosted by the Society of Trust and Estate Practitioners ("STEP"). He appended two sets of correspondence from June 2015 and October 2017 which he considered showed that the practitioner accepted the stated view. At the hearing he confirmed that he did not look at these postings when he made his decision. He looked at them subsequently for the purposes of this hearing.

(7) Some commentators suggest that there may be a way around HMRC's approach by transferring property to a company owned by a trust. This can work as regards QIIPs. HMRC accept that this works provided it can be shown that no person other than the transferor life tenant can benefit. The fact that taxpayers were seeking to add value to the trust in this way confirms that taxpayers were acting in accordance with the stated view rather than challenging it. Commentators suggest that the changes in 2006 are an inhibitor to people making transfers to existing trusts as a result of which there is an increased risk to exposure to an immediate charge to IHT at 20%. He does not disagree but observes that it is possible to avoid that by transferring property that attracts relief and by using the nil rate band. It is interesting that whilst some commentators draw attention to the 20% charge it is the spectre of the ten-year charge which remains the focus of attention.

(8) Following the 2006 changes HMRC received questions from STEP and CIOT. The questions and answers were published on 4 April 2007. In the course of some of the discussion HMRC made a statement which in his view is consistent with the stated view.

68. At the hearing he was asked to expand on his comments in his statement about the basis for his decision. From his comments, it is clear that he based his decision as regards s 255 on the view that the "adopted in practice" test was met as follows:

(1) It was put to him that he did not take account of what the appellant's view of the law was when the Trustee made the return and paid the disputed tax. He said he took their position to be what they said in their letters of May. He said he was not sure he was thinking that he had to address whether the taxpayer specifically had adopted HMRC's stated view in practice, rather whether it was adopted by practitioners, taxpayers in public.

(2) As regards his statement that overall the stated view was generally received and adopted in practice, he said:

"The generally received point and adopted in practice point, I understood to be two separate points. So that in looking at section 255, only one of them needed to be satisfied for it to apply......

Now what is meant by "generally received" was a bit unclear. We did obviously look at a case I think which is mentioned in Ms Boraster's letter, is it Murray's Trustees...Which I don't think we found terribly helpful, given that it was on something that was different.

But where we probably thought we had a stronger case was on the adopted in practice. And in doing that, we looked at what was being said publicly, and also what our experience in this area was in terms of the approach taxpayers had taken in relation to HMRC. So in that regard we looked at what was in the publications in the ones that I've mentioned earlier on. And we reviewed what we knew about how taxpayers had behaved, certainly since 1997. And our view was that there had been virtually no challenge on this at all. I can't recall if - I think we knew at the time there was a potential issue with [Barclays], because I think the litigation may have started or been close to starting so we would have been aware that there was some issue there."

(3) On his interpretation "adopted in practice is a stand-alone" provision without generally in front of it. The view he took was to ask "how do people behave, what do they do?...In terms of: do they act in a way which is compliant with HMRC's view or do they not". It was put to him that people could act as he described because they agreed the stated view of the law is correct or simply for other reasons, knowing what the stated view is, they do not go down that path because of it. He thought it could be either of those things but that it does not make any difference to whether the stated view is "adopted in practice". It was put to him, in effect, that in his view the "adopted" test is satisfied where people take account of the stated view in deciding what course of action to take. He said:

"they need obviously to take it into account...And it may be that they do as HMRC suggest and keep adequate records or else they take steps which don't put them into direct contravention and they might take a different course of action which might achieve their objectives."

(4) He confirmed that in coming to his decision he took into account that there had not been many challenges by taxpayers to HMRC on this point. There are no other factors that he took into account that do not appear in his statement: "the principal ones were our experience and what was being said publicly". In referring to a challenge he meant a case where there was an actual dispute with HMRC, and it was either heading to or went to litigation. The case he referred to in addition to Barclays was resolved by agreement through correspondence. He could not clarify the amount of tax at stake in that case. His understanding was that initially the taxpayer took the opposite view and then they did not press it. It was another case where he could not find the file. Under the litigation settlement strategy the amount of tax is one of the factors HMRC need to take into account.

(5) It is plain he did not consider at the time whether there were any factors affecting how likely HMRC were to come across taxpayers taking a view contrary to the stated view:

(a) He said, in effect, that he did not have any feel for how common it is for the added property issue to have arisen in practice. It was put to him that it was quite unlikely a settlor would add property to a discretionary settlement in the pre-2006 period due to the substantial entry charge. He thought that was an inhibitor, but noted "there are obviously allowances and ways round that" and that it is possible to contribute property which would attract relief. However he had very little if any experience of that actually happening. It was put to him that if property qualified for relief if the individual held onto it until death he would get a capital gains tax uplift and have no IHT. He said that is certainly a factor. It was put to him that assuming the property is not subject to relief, it would be unlikely that a taxpayer would want to incur a dry tax charge of 20% or as much as 75% in the earlier times. He said if it were him it would be a strong inhibitor.

(b) It was put to him that the fact that he had not seen many challenges could be because property additions do not happen very often and, when they do, taxpayers may take the view that they did not have to make a return on the basis the property qualifies as excluded property. He said that was possible, but he does not know whether that is what happened. When pressed to agree that HMRC would not know about taxpayers taking a contrary view as no IHT account would be required if they did, he said that is possible but he can only go on the experience HMRC had, which was the lack of any great number of challenges. When it was put to him that may be easily explained by the fact that it just would not arise, he said he did not know and "I think I'm taking the view that if there was a disagreement and it was material we would know about it". When pressed he said he "can only go on the cases that get referred to us if it's not reported to us on account we wouldn't be aware of it" and he thought that the fact that there were no challenges was a relevant factor in looking at adopted in practice along with the commentaries he referred to.

(6) As regards the books he referred to:

(a) As regards Fosters, he accepted that in that book there is no express endorsement of the stated view as being correct. He added it is a reporting of the view and he thought it took a fairly neutral stance.

(b) He confirmed that Dymonds, Clarke and Kessler do not express the view that they agree with HMRC's position. He said "quite the opposite". He agreed that what he took account of in his decision was the lack of challenge and these books.

(c) In re-examination, when asked if now he had looked at Chamberlain, or at any other material that came to his attention before or during this appeal he may have reached a different view on the practice generally prevailing, or what was generally received, he said:

"No, I don't think I would have reached a different decision at all. The Trust Taxation book I think sets out in quite detail the technical arguments, but none of those are dissimilar to the ones in some of the other sources. I have looked at some of the more neutral ones... and I think my view remains exactly the same, that although there is quite widespread criticism in the professional press, it just remains a fact that taxpayers in general are not sort of willing to put their money where their advisers' mouths are...I don't think it would have altered my view at all."

(7) His understanding of HMRC's position on the issue is that "throughout that...you look at the domicile at the point the property is added, and assess its status. The argument that there is a new settlement each time something is added is not a point that HMRC have taken since 1997." He understood the point made in Rysaffe and Barclays that there was a separate settlement each time the settlor transferred property to the trust to be made on the facts of those cases but not in a wider context. So far as he knew HMRC's position has always been exactly the same on that issue. He said: "The guidance we work to says we are obliged to adopt that position [in the stated view]". So he had to come to that conclusion if the circumstances fitted, which they did and "The guidance we've had all along is if it's contested to refer to our technical team. The implication being that we have to maintain that, even when challenged." He accepted that meant that in a sense he was not making a decision on the legal issue - that was something that he had to follow, he may have made a decision on the s 255 issue but not on the other issue.

(8) He was questioned about the correspondence with the CIOT:

(a) He confirmed that he did not actually take this correspondence into account when he made the relevant decisions.

(b) He could not shed light on why HMRC do not appear to have given guidance on the issue raised as regards the mixing of funds. He thought it is probably not a very common occurrence, but Mr Twiddy's letter makes clear that HMRC take a pragmatic view and the stated view recommend people keep adequate records. His experience of seeing these cases is very limited and he had no actual experience of having to deal with that kind of situation. In re-examination he said he was not aware of guidance on how to deal with mixed funds in other areas. His experience of multiple settlors is that there tends to be very clear allocation of funds to individuals that are kept separate. In order to work out the ten-year charge trustees would have to carry out some kind of reasonable tracing of assets and there tends to be good bookkeeping.

(c) He thought this correspondence was relevant, given that it related directly to the stated view and was a more or less contemporaneous exchange with quite an important body around what it actually meant and it did, to some extent, he thought, clarify HMRC's view. He accepted it does not demonstrate acceptance of the stated view by those corresponding.

(d) In re-examination he said (i) when HMRC announce a formal interpretation, there is quite often interplay between HMRC and bodies, such as CIOT and STEP, usually around clarifying specific issues, and (ii) it seems Mr Hutton was persuaded of the stated view in his comments in his book.

(9) It was put to him that the example he gives of a transfer of the property to a company could achieve exactly the same result HMRC are arguing against in this case. He said that such a transfer "sets them outside the terms of RI 166 because they are transferring to the company not to the trustee" and it may not have the consequence HMRC are arguing for in this case. It was put to him that where the settlor transfers property to a settlement when domiciled and then the trustees transfer it into an overseas company held in the settlement, the relevant property would then disappear, because it would then be owned by the company, which was itself excluded property. He said he does not know the answer to that without thinking long and hard about the technical implications of that. He had thought about a transfer to a company. He confirmed that what he said in his statement about putting assets into a company is not published anywhere.

(10) He accepted, in effect, that taxpayers are likely to behave in the same way in relation to all published HMRC guidance. He thought it is not just the 20% charge that is a factor that taxpayers take into account in settling property: "Because if there is the potential for a ten-year anniversary [charge], that's also going to weigh in their decision making". That was his perception from "what I see in sort of the general traffic in commentaries in magazines and suchlike". It was put to him that it is not very plausible that anyone would want to pay 20% charge upfront in order potentially to avoid much smaller charges over a long period of time. It does not seem very commercial. He said he understood the point.

(11) In re-examination, he confirmed that (a) the 40% charge on death would be a relevant factor if there was a QIIP whereas for a relevant property trust there is the 20% charge plus the ten-year anniversary and exit charge, (b) if there was a chargeable transfer into a settlement which gives rise to tax, then there is an obligation to account for that, (b) if the entry charge arises when a domiciled individual transfers assets into a settlement, that would alert HMRC that the settlement exists if the account identifies the source of the destination of the contribution and he believes the form specifies who the transferee is, (c) if HMRC were made aware of a transfer into a settlement that they were not previously aware of, that would obviously place the settlement itself on their radar, and it is something that they would then pay attention to in relation to any other potential charges.

Evidence of Ms Emma Chamberlain

69. Ms Chamberlain's evidence and that of the other advisers is relevant to the question of whether as a factual matter the stated view was at the relevant time generally received or generally adopted in practice. Ms Chamberlain explained that:

(1) She was a solicitor originally and it was only when she became a barrister and after 1998 that, because of the boom and a lot of non-domiciled persons coming to London, there was a lot more of that work around. She wrote a number of books, including Dymonds and Chamberlain. The aim of Chamberlain was not to be an academic book but to address the practical problems that the authors saw come up. She was instructed by a quite wide range of solicitors, such as the top City firms with very large sums of money but also very small firms, who might have had one very large client, who was a non-domiciled person.

(2) As an adviser one is always trying to balance a number of factors and ask what are the downsides, the risks and the upside. One can only put those options to the client as clearly as possible. That is how she saw her practice both as a solicitor and as a barrister.

70. Ms Chamberlain set out the following in her witness statement:

(1) Prior to the publication of the stated view (as also set out in IHTM4272), she believes the official view of HMRC on this issue was unknown. The view taken by Park J, in 2002, that one applies general trust law principles when dealing with additions (an approach followed in Barclays), strongly influenced her interpretation of s 48 (she referred to some of the passages in Rysaffe set out in Part B). In the light of this, as well as the various other technical arguments (see below), until s48(3)(a) was amended in 2020, her view was:

"IR 166 was not technically correct and that additions made after the settlor became deemed domiciled to a settlement made when the settlor was foreign domiciled could become excluded property. I had understood from the publications written by other barristers I respected such as Barry McCutcheon, Giles Clarke and James Kessler QC that this was also their view, expressed with varying degrees of caution. I attach extracts below."

(2) Chris Whitehouse (now deceased) a colleague at 5 Stone Buildings and co-author with her on publications including Dymonds, took the same view as she did which they discussed when writing Chamberlain and Dymonds. She cannot say what the general view of most solicitors was at that time, because it was not in practice a point that arose very often. However, solicitors from Withers LLP and Charles Russell Speechlys LLP were contributors to McCutcheon and Clarke which criticised the stated view.

71. In her witness statement she said that her view appeared to be well-supported by other specialist writers and practitioners, both in 2005 and in 2013. She commented on the relevant publications as follows:

(1) The Third Edition of McCutcheon published in 1988 was at that time the leading book on IHT written by Mr Barry McCutcheon and Mr Chris Whitehouse, barristers specialising in IHT. They considered the point about additions before IR166 was published and at 15-25 said that the words in s 48:

"raise an intriguing question, namely what is the position where a UK domiciliary adds property to a settlement he made when he was domiciled abroad. Since such an addition forms part of the original settlement it should be excluded property if it is situated abroad. Whether the Revenue would seek to resist this contention is not known - after all, on this view property added by a foreign domiciliary to a settlement he made when he was a UK domiciliary is not capable of being excluded property. Even if the Revenue do argue the point, it appears that a foreign domiciliary can lose nothing by establishing a small pilot settlement in anticipation of subsequently becoming domiciled in the UK."

Mr McCutcheon probably was the leading expert on IHT.

(2) The Fourth Edition published in 2005 discussed the stated view at paras 21-47 and the authors, Mr McCutcheon and Withers LLP (a leading firm of London solicitors) note:

"While the authors have some sympathy with the Revenue's attempt to prevent individuals who have become domiciled in the UK from converting non-excluded property to excluded property by adding it to a settlement made when they were not so domiciled, they are far from convinced that the Revenue's approach, which would seem strange to the mind of a Chancery lawyer, is correct. If the Revenue's view is correct, or perhaps more pertinently, if one is concerned that it may be correct, consideration could be given in a case in which the abovementioned settlor had an interest in possession in the settlement to his gifting property to an underlying company wholly owned by the settlement....

....the corollary to the Revenue's view is presumably that property added by a person to a settlement when he is not domiciled in the UK is excluded property even though he was domiciled in the UK when he created the settlement. Since the authors are not convinced that the Revenue's view is correct they would advise emigrants from the UK to create new settlements rather than add property to existing settlements."

(3) This seems to be advising practitioners on the basis that HMRC's view was wrong. The above paragraphs remained unchanged in subsequent editions through to the seventh edition published in 2017 when the authors were Aparna Nathan (now KC) and Marika Lemos both well respected barristers at Devereux Chambers writing with Withers LLP. That edition at 32-59 contained the same passage as above and the authors also referred to the High Court decision in Barclays, noting that despite their view stated above that decision provides support for HMRC's position but they still concluded that HMRC were not correct.

(4) The 4th edition of Kessler published in 2005 set out at length (at 26-11 onwards) the reasons why HMRC's view was not correct (citing Rysaffe) and concluded that it may take litigation before HMRC would amend their published stance and that if Kessler's view was right then HMRC would have the worst of all worlds. He also notes at 26.12.1 that where a trust was created when the settlor was non-domiciled but the settlor became domiciled and then gave property to a company owned by the trust "then HMRC's argument does not run at all. The shares in the company (if not UK situate) must be and remain excluded property." The 11th edition published in 2012/13 is almost identical except it adds: "it might take litigation before HMRC amend their published stance on this issue but I think they would be advised not to fight. Until the point is clear, trustees should follow this advice in RI 166" [to keep additions separate when made after the settlor became domiciled].

(5) Chamberlain in 2007 quote extensively from the HMRC Manual from 22.20 onwards, discuss the HMRC examples given and the merits of the stated view and then address practical tax planning ideas:

(a) The authors note at footnote 49 that Dymond expressed the view that additions should be regarded as part of a single settlement and that the contrary view is not compatible with FA 1977 which, unlike s 48, clearly distinguished between the time when a settlement was made and the time when the added property was settled, at least in the context of excluded property. One of the consultant editors of Dymond at that time was a former member of the Capital Taxes Office, Mr Roy Greenfield.

(b) It is pointed out at para 22.26 that although it is common sense to follow the HMRC guidance where possible, "in the light of the Rysaffe decision and the legislation there is a strong argument that additions made after the settlor is domiciled become excluded property". It is noted that additions after 21 March 2006 to existing trusts would generally be subject to a 20% entry charge even if the property thereafter became excluded property. From 22 March 2006 it was no longer possible to convert a discretionary trust into a QIIP and then add to that trust, thus avoiding an entry charge.

(c) They also discussed how to add value to an existing interest in possession trust after 21 March 2006 without incurring an entry charge but ensuring that such further added value becomes excluded property. Paras 22.28 onwards reflect the Kessler view that such additions could be done by additions to a company held on QIIPs.

(d) The 4th edition was published in 2014 and (like the intervening editions) contains almost identical commentary from paras 37.43 to 37.54. Ms Chamberlain saw advice from leading counsel on two separate occasions sometime after 2010 that additions could be made to an existing company held on QIIPs for the settlor without triggering an entry charge. The advice was that the shares of the company would remain excluded property and she concurred in that advice although the other disadvantages of this course of action are noted in Chamberlain.

(6) In Clarke 13th edition published in 2006 he notes the technical difficulties in the HMRC view at para 45.8 albeit conceding that it is more logical quoting Rysaffe as authority. The 22nd edition was published in 2015 by Mr Clarke and Mr Dominic Lawrance and Ms Alice Wilne who were both partners in Charles Russell Speechlys LLP. At para 60.6 the same arguments are reiterated (albeit noting that the Courts had found against the taxpayer on the point in the High Court decision of Barclays which was reversed later on appeal).

(7) She had not been able to locate a textbook which specifically endorsed the HMRC view at that time except that Foster discusses the point in a more general way citing HMRC's view and noting that: "these points are not determined and there is no specific provision requiring additions to a settlement to be treated as being comprised in either a separate settlement or the original settlement."

72. As regards her views and matters she had advised on, she said the following in her witness statement:

(1) Clarke and Kessler were clearly read and respected as they were often quoted to her by practitioners when they sent her instructions requesting advice on offshore matters: in some such instructions, planning based on the Kessler view had already been done on the advice of other counsel and she was asked to advise on the status of the settled property. At no time was she asked to advise about ten-year charges in the context of added property.

(2) In the period after 1999, she recalls (a) two sets of instructions about the status of additions that were already made before 2006, on the advice of other barristers, to existing QIIPs for the settlor. The issue was whether the added property would be excluded property on the settlor's death on the assumption that he might die deemed domiciled. She advised that although HMRC took a different view and the position could be challenged, the addition was excluded property and therefore not chargeable on the settlor's death, and (b) two sets of instructions about the status of additions already made after March 2006, on the advice of leading counsel, to companies owned by existing QIIPs. The advice given by those counsel was that the added property was excluded property, and (c) on one occasion in January 2006 a client put into effect her own advice that he could add to an existing QIIP and the settled property would be excluded property.

(3) If she had been asked about whether additions to discretionary trusts were subject to ten-year charges if foreign situs and made to an existing excluded property settlement she would have given the same advice that they were excluded property.

(4) At no time did she receive instructions, or see practitioners advising clients, to add property to an existing discretionary trust after they had become deemed domiciled. This was not because the stated view was generally adopted but because any such addition was subject to an immediate 20% entry charge unless the property qualified for relief. The discussion from March 2006 was not about the stated view but about avoiding the entry charge.

(5) She recalls also advising a client who wanted to add to a settlement set up when he was domiciled not do to so, as he had by then become foreign domiciled. She therefore took the view that additions made to an existing settlement set up when the client was domiciled could never be excluded property and he should set up a completely new trust because the stated view was wrong. At the hearing she said that she could not recall coming across that situation in practice.

73. She remained throughout of the view that the words in s 48(3) should be given their natural meaning. She also considered that there were other reasons why the stated view was likely to be wrong if litigated:

(1) The requirements as to the settlor's domicile and the commencement of the settlement were specifically altered in relation to ss 80, 81 and 82 by amendments made in the Finance Act 1982 which were designed to circumvent devices for obtaining excluded property status for property not otherwise qualifying. The fact that similar alterations were not expressly made in the context of s 48 suggested that one should take the words "when the settlement was made" at face value.

(2) Under s 80 where a settlor or his spouse was beneficially entitled to an interest in possession immediately after the property became comprised in a settlement the property was treated as not having become comprised in the settlement on that occasion but only on termination of the last interest in possession. For the settled property thereafter to qualify as excluded property, the conditions as to domicile had to be satisfied both by the settlor when the settlement was actually made in the first place and by the original settlor or his spouse at the time that the interest of the settlor or his spouse ceased and the property was deemed to have been resettled. Although the remaking of the settlement at the date when the last interest in possession ended is a deeming provision it is instructive that the draftsman specifically saw the need to refer to the person's domicile at that time.

(3) Under ss 81 and 82 where property ceased to be comprised in one settlement and became comprised in another without any person having become beneficially absolutely entitled to the property, the property is treated as remaining comprised in the first settlement and for the settled property to be excluded property the person who is the settlor in relation to the second settlement must have been non-domiciled at the time the second settlement was made (ss 81 and 82).

74. Ms Chamberlain explained that, prior to 22 March 2006, entry charges could be avoided simply by ensuring that when any additions were made, the fund to which such property was added was held on a QIIP for the settlor. Such an addition prior to 22 March 2006 would have been a non-event for IHT purposes and it would not have been subject to ten-year charges while the settlor's QIIP continued. As noted, she saw advice on such planning by others and she also advised one client in January 2006 that they could add to an existing QIIP for the settlor and such addition would be excluded property on his death. It was accepted by the solicitors and clients that HMRC might litigate the excluded property position on the settlor's death but it was considered that the transfer was worthwhile provided the addition resulted in no immediate entry charge. No IHT or CGT reporting was necessary. She is not aware of cases where this point was actually tested; as far as she is aware the settlor did not die before the changes introduced in the Finance Act 2020. It is likely that in such cases the settlor would eventually leave the UK, lose deemed domicile and the trust might not exist any longer.

75. As regards the position from 22 March 2006, she said the following:

(1) It was no longer possible to add property to existing QIIPs without this being treated as an immediately chargeable transfer if the settlor was domiciled or to convert an existing discretionary trust to a QIIP. Therefore, any addition to an existing excluded property trust was highly unlikely in practice once the settlor had become deemed domiciled given that it would trigger an immediate 20% entry charge. There was no easy way of avoiding this charge. She was never asked to advise on the point. The immediate downsides of any such addition made the stated view largely irrelevant.

(2) Although she saw instructions about adding property after March 2006 to a company owned by a pre-March 2006 QIIP for the settlor, this addition had certain other disadvantages and could only apply in limited circumstances. She saw two sets of instructions where leading counsel had advised it could be done and it was done. She agreed that the addition to the company did not prevent the entire value of the company shares from being excluded property.

(3) However, from March 2006 the benefits of obtaining excluded property status after the addition were rarely worthwhile if one had to pay an immediate charge of 20% because the trust was not a QIIP, particularly as there was no certainty that HMRC would accept the excluded property status of such added property later without litigation.

(4) She recalls seeing some instructions relating to different foreign domiciliaries who had become deemed domiciled but intended to leave the UK before they died with whom this option was discussed; for them it was preferable to wait until they had left and had lost their deemed domicile later and add to existing trusts at that point if they so wanted without incurring an entry charge. They would also then have certainty on excluded property status during the continuation of the trust and on death. Hence she saw no additions to existing excluded property discretionary trusts made by any deemed domiciled person after 21 March 2006.

(5) She had assumed prior to the Barclays case that it was the presence of the entry charge that meant HMRC saw no point in amending the legislation to put the matter beyond doubt: clients saw little advantage in possibly saving IHT later if they incurred a 20% entry charge now. However, this did not mean that the stated view was accepted as being correct in law. It simply meant that there was a different impediment, particularly after 2006, in making additions, as new trusts could not be set up or converted into QIIPs. Hence the HMRC position on additions was in practice not likely to be tested very often (unless perhaps a mistake was made over when the settlor became deemed domiciled).

(6) In her own experience within the profession and within her own practice (which she accepts is limited by the fact that she receives somewhat specialised instructions on IHT) she saw some discussion over the stated view. However:

"to the best of my knowledge as cited above from discussions with other barristers such as Chris Whitehouse, that interpretation was not generally considered to be correct. However, in practical terms once the settlor had become deemed domiciled an entry charge could not easily be avoided particularly after March 2006 and therefore such additions were only made in limited circumstances."

76. At the hearing she gave the following evidence:

(1) She had not seen examples of "pilot" settlements in practice and said it would have been odd to have seen them:

" why would you wait until after you became deemed domiciled to settle your assets when you knew that if you put them in before you became deemed domiciled you had this incredible tax break, where you could benefit from the assets forevermore and never be taxed on your death and...it would be unusual to think I will just put a pilot trust up and then I will wait until I am deemed domiciled before I am settling the assets. You would want to get as much as possible in before you were deemed domiciled."

(2) After 1997 one would have been unwise to advise a client to do that, because it was going against the stated view. It did not mean that one necessarily agreed with their view, but one would obviously always do the safest thing for the client. There is a very large difference between the position before and after 2006; after 2006 the entry charge really is quite an issue although it was not just tax considerations which lead to clients choosing a trust or what type of trust to have. People do not set up trusts and incur an entry charge; they will probably end up doing other sorts of planning, like family investment companies or will wait and leave assets to their spouse in the will. She had come across only one client who was prepared to incur an entry charge who was not a foreign domiciled person.

(3) Before 2006 interest in possession trusts were much more popular for UK people. For non-domiciled persons, the choice of type of trust depended on a number of factors. When advising clients one would balance a number of factors. If, before 2006, a client had a QIIP and had become deemed domiciled and was going to leave the country but wanted IHT protection in the meantime there was no reason not to add property. There would be no downside in doing that, potentially quite a lot of upside and no reporting is required. That indeed is what she would have and did advise a client to do. After 2006 it was simply not possible to do that due to the entry charge. She agreed that a settlor could add property to an overseas company which is in a QIIP which would in effect then become excluded property. However, after 2006 if a client did not have an already existing QIIP, short of using the company route which had its own problems (as set out in Trust Taxation), in effect there was nothing the client could do; people would not have wanted to incur up to a 25% charge. Most important non-domiciled persons do not have foreign assets that qualify for reliefs such as business property relief. If it is UK property, they could sell it and take the proceeds offshore, and argue it is excluded property but she never came across a client who had such a property that they wanted to settle. People often want to hang on to such property until death, because they get the CGT uplift then. Advising clients is a mixture of balancing upsides and downsides.

(4) It was put to her that in the 4th Edition of McCutcheon and in Kessler it was stated that consideration could be given to using the company route. She said:

(a) As regards the position before 2006, she only advised one client to put his assets directly into an existing QIIP. She did not use a company device partly because she thought it was artificial. She did not know that structure was accepted by HMRC. She did not think it was necessary to use that route because her view was that added property became excluded property and there was no entry charge prior to 2006. The company route is quite complicated. Technically there is a reservation of benefit whilst the property is in the company and one cannot dividend the money out because it goes back to the settlor as the life tenant. So one has to extract it by way of a buyback or a liquidation of the company and that may not suit the trustees for other reasons. So that was not something she advised people to do before 2006.

(b) After 2006 one used a company for a slightly different reason. The legislation then said that there was no QIIP if one newly settled property. However, if a settlor had a QIIP and there was a company already in the trust, the settlor could add value to the company without affecting the QIIP status and triggering an entry charge. That was as much discussed in the context of domiciled persons as non-domiciled persons. She did not know that HMRC basically accepted that anomaly.

(c) She thought Kessler must be referring to a company held in an existing QIIP because a transfer to a company owned by any other trust would be a chargeable transfer. So in that case there would be the same problem of an entry charge. So it would not be a very common scenario after 2006, but it would and did occur.

(d) She added that the idea that a settlor can add value to a QIIP without prejudicing its status as such was something that is discussed at great length in Chamberlain and was something:

"we had many debates with the Revenue about and we do talk a lot in the Trust Taxation book about preserving your qualifying interest in possession status, because obviously it comes up in a very practical context...how to include property, is it still subject to a qualifying interest in possession trust? We were trying to solve those problems in 2006 when we were talking about this."

(5) She was asked if she shared Kessler's view that the stated view leads to a more sensible result. She said she thought that the whole excluded property regime is absolutely crazy and there are lots of anomalies in it. The changes in the new legislation show that it is quite difficult to get to a sensible result; that still does not really solve the problem. The sensible result is not to have a non-domiciled excluded property regime, but if there is one, there are always going to be anomalies in it and this is not a particularly striking anomaly - there are lots more worse anomalies in the foreign domiciled regime than this one. She did not really think the stated view gives a more sensible result.

(6) It was put to her that Kessler also posits the settlor creating a trust when domiciled and adding property to it when non-domiciled and stated:

"On my view, none of the property is excluded property. However, HMRC must abide by their statement (at least until it is officially and publicly withdrawn with appropriate transitional relief) and accept the added property may be excluded property! Thus, the consequence of their statement (if my view is right) is that HMRC have the worst of both worlds. Of course, a well advised settlor will not find himself in this situation, but it does arise from time to time by accident."

She agreed that (a) Kessler seemed to mean that taxpayers would challenge the stated view in the tribunal/courts and the Kessler view would prevail, (b) taxpayers in the opposite circumstances would bring judicial review proceedings and demand that HMRC must apply the stated view which the tribunals/courts found to be wrong.

(7) As regards the examples in her statement of her giving advice on added property, she said the following:

(a) There was nothing for her to consider reporting as regards the pre-2006 cases.

(b) The one she advised on in January 2006 involved a deemed domiciled client who had to delay his departure from the UK. He had a very long-standing QIIP set up when he was non-domiciled, he had then become deemed domiciled and was planning to leave and return to his home country. However, his departure was delayed and he was worried that if he died whilst deemed domiciled he would have a large IHT charge. He did not want to give his assets away to his children so was looking for a form of insurance policy to tide him over until he left the UK. Hence he wanted to add property into his trust in case he died unexpectedly while he was still in the UK. She advised him to add the property and noted that HMRC do not accept that is excluded property, but it was better than doing nothing at all because otherwise there would definitely be IHT on death. There was also the issue of capital gains tax and the remittance regime before 2008 was very favourable, so he could put the assets into trust and the assets would be re-based. She thought there was no downside in doing that; it did not need to be reported because there was no immediately chargeable transfer. If he had died after March 2006 the position would have been quite different but in fact he is living in his home country and the trust has long since been wound up and distributed to his relations. That was the only occasion she advised a person who had become deemed domiciled on adding property to a QIIP before 2006.

(c) She did not know what happened in the other examples. She recalled that one of them in 2008 was about creating a transitional serial interest and that in itself did not require any reporting because it was a potentially exempt transfer and she did not know the status of that in terms of excluded property because it was not relevant.

(d) She thought that is not that surprising that she had not advised on the issue more as regards the pre-2006 situation, because most non-domiciled persons have good professional advisers and would generally want to settle assets before they become deemed domiciled. She agreed that was to avoid an argument with HMRC but noted that it was also to avoid an entry charge if the settlor had a discretionary trust. She confirmed that she was asked to advise there were no reporting obligations because it was all in a QIIP.

(e) Her advice in each of the other examples was that on the death of the life tenant, accepting that HMRC did not agree with this view, the entire settled property, if it was foreign situs at that point, would be excluded property. The concern was that obviously if that life tenant, settlor or spouse died deemed domiciled, one would then have to worry about s 80; that is why people asked about transitional interests because they wanted to bypass that. Her advice was that it was excluded property, but in a sense they were then going to be into a different regime because of s 80. She said she would have advised the clients to make a disclosure if they had died and her understanding is that practitioners would make that recommendation if clients were taking a position that was directly opposite to HMRC's official stated view. In re-examination she clarified that (a) in the examples on the additions pre-2006 the other counsel she referred to had advised that HMRC were simply wrong on the stated view and a settlement is made when property first becomes comprised in it so that the additions were excluded property, provided the settlement was originally when the settlor was non-domiciled, and (b) the post-2006 ones were based on the idea that one had not added any settled property at all after the settlor became deemed domiciled but rather added value to an existing settlement.

(f) It was put to her that if people thought adding property was a viable option all settlors who had become deemed domiciled with a QIIP would have done so as regards all their assets before they died but that does not seem to have happened. She noted that it only worked for foreign assets. It was put to her that it might be possible to transfer UK assets into a trust and then convert them into foreign assets by putting them into a company. She said there would have been a capital gains tax charge on that. She said people may well have done this and HMRC would not know about it, because there is no reporting in any of it. In her view, it is all about balance. A lot of non-domiciled persons are married. So for them spouse exemption is good enough. They do not need to pay her to advise them on complicated excluded property trusts. If they are going to leave the UK, in many ways it is just easier to rely on spouse exemption, leave and then they will lose their deemed domicile. Obviously the nature of being a foreign domiciled person is that they do leave and go back to whatever country is their home. So she was not totally surprised that people did not do this, because most people would be thinking right it is time to go or there was a spouse and they would do the planning when they leave.

(g) She confirmed that if people had added property and it had resulted in a taxable event she would expect the advisers or the trustees to tell HMRC about it. She said that whether they did a nil return or put something on the return, "they would obviously want to put some sensible disclosure, because you are taking a view that is clearly contrary to what the Revenue are saying".

(h) It was put to her that as regards the position since 2006 she said in her statement this issue never really arose but that is surprising given how much has been written about it. She said that may be because people were adding to interest in possession trusts and so they would be interested in the question. She confirmed that apart from the examples she referred to there are no others that she has seen in practice. It was put to her that as a result of all the guidance cautioning against adding property perhaps that was the view adopted in practice and that is more consistent with the fact that HMRC have not heard anything about it. She said it is all about balancing options and this is one option to consider bearing in mind HMRC's view. But there is rarely just one option available to someone. People might think they want to do a family investment company because it suits them better in other ways. An adviser does not just look at IHT in isolation but has to look at all the taxes and after 2008 there were some quite significant issues about the remittance rules. So it was not a neutral ground where advice was given to just add to an interest in possession trust; one has to balance things and HMRC's stated view would clearly be a factor in how one would advise clients.

(i) She said that there was no need or indeed ability in a way to put in a return when one had got an interest in possession trust and whether one was put in on the death of a life tenant is not something she would have visibility on, unless the client came back and instructed her and asked her what to do but they did not.

Evidence of Ms Summers

77. Ms Jo Summers of Jurit LLP gave the following evidence:

(1) She would first have been aware of the stated view in about 1998 or thereabouts She would have advised a client against adding property on the basis that the added property would not be excluded property. Her view was/is that if the settlor added property to a discretionary trust (a) a charge to IHT would arise, and (b) the added property would be relevant property within the settlement. That is advice that she gave back in 2013 and would still give now. The law has changed since, so it has made it a bit clearer perhaps.

(2) In forming her opinion at the relevant time, she would have started with the stated view and her view always was that on the balance of probabilities the stated view was more likely than not to be correct. She took into account the definition of settlement in s 43(2) and was quite comfortable with the idea that IHT had its own definitions and that the settlement definition was not the same as the trust settlement definition and any disposition of property is a settlement under this definition. She had had an example of how tax law and trust law differ in other scenarios and she noted that in s 44, the definition of settlor is anybody who has put money into the settlement. She considered that Rysaffe was correctly decided. In her view the settlement definition in s 43 is quite clearly trying to get at something other than a settlement for trust law purposes. She also had "a general policy feel about this, that it would strike me as remarkably odd if HMRC had allowed legislation to go through that allowed an individual who was non-domiciled to set up a trust with £1, £10 and then become domiciled and then put in all their assets". There may or may not be an entry charge, depending on what assets are added, but for added property to become excluded property the instant it hit the trust seems counterintuitive and it did not sound right.

(3) It was put to her that everyone agrees that the result she considers anomalous can be achieved through putting property into an overseas company owned by a trust. She said that is very different because the relevant provisions are concerned with transfers to trustees not transfers to companies. She commented that using that route shows that people were trying to get around the stated view and that proves people thought the stated view was at least correct enough not to ignore it by doing that. I do not put any weight on this comment as Ms Summers was here giving an opinion and she had no experience of this in practice. She did not recall ever advising clients that they could add property to a company owned by a trust to convert it into excluded property. Due to the reservation of benefit problem, she could not see how in the long term that achieved the clients' objectives of avoiding IHT. She had actually considered that and came to the view that there was an issue.

(4) She gave advice on added property in or around 2013 to a client and she worked with a trust company in Guernsey to review their files on a project which started in 2011 and lasted over 2 years:

(a) The client did not appreciate that he had already become deemed domiciled when he was considering transferring assets to an offshore discretionary settlement. She was aware there were some schemes going around on methods to get assets into trust without the entry charge arising. She told clients advice could be obtained on that but she did not advise on that. He was prepared to look into not paying the 20% charge but when he was told that, even if he succeeded on that, there would then be relevant property charges on the trust, he decided against it.

(b) The review of files was prompted by trustees wanting to check the advice they had cause to doubt. It was a health check on 11 or 12 files to see if there was anything untoward. She thought there were two cases where the settlor was non-domiciled but became domiciled. There was one that was deemed domiciled and one that had been domiciled right from the start but mistakenly thought he had a foreign domicile. The one who had become deemed domiciled after the settlement was made paid tax in the region of £13,000/£14,000.

(5) She asserted there are numerous professionals who accepted the stated view, she did not know any private client practitioner who regarded the stated view as incorrect or, if they were unsure, who would advise a client to ignore it. As set out below, Ms Summers later explained precisely who she had spoken to about the stated view. I have not put material weight on these general imprecise comments.

(6) In her experience, private clients do not like being the guinea pig who challenges HMRC's view at their own expense in a public forum. She did not think that practitioners should have considered the stated view to be incorrect given it was repeated in all books and commentaries covering this topic. She referred to McLaughlin (10^th edition) from 2012 and Tiley 2011/12 and an article by Mr Goldstone and Natalie Quail in the Tax Journal on 15 November 2017 and her own book. As regards the textbooks:

(a) She thought there were maybe 2, 3 or 4 authors who were all barristers who disagreed or doubted the stated view. She thinks it is easier for barristers to be more technical and esoteric in their advice than for solicitors who are at the coalface. The client rarely has the appetite for being the guinea pig in her experience. In one case a counsel's opinion said categorically HMRC were wrong but the client still did not want to take it on for the cost, expense and the publicity of it all. It would not have made any difference if she had told the client that she thought HMRC were wrong, as she would still have had to caveat that with "but you can expect them to challenge this, you can expect them to fight it". In her view, if a person wins at tribunal HMRC are more likely than not to appeal it so the client may possibly have to go higher and the costs are huge. She could not recall ever advising a client to challenge HMRC's published view on anything.

(b) It was put to her that the textbooks she referred to, including her own book, do not say that the stated view is correct. She pointed to comments in McLaughlin as an example of a textbook writer accepting the stated view. The authors state that if there were joint settlors, they would be treated as making separate settlements under s 44(2) and then state: "For similar reasons, a settlor who subsequently becomes UK domiciled should avoid adding assets to the settlement." Ms Summers agreed that this statement is perhaps not entirely accurate; the analysis as regards added property does not depend on s 44(2). She said that "they are definitely saying a settlor who subsequently becomes UK domiciled should avoid adding assets to the settlement". One could say possibly that they were incorrect in their rationale for saying that, but they "are definitely saying that that is the case...it's pretty categorical".

(c) Her book was aimed at offshore trustees as a practical guidance, not as an analysis of the tax legislation or of HMRC's views on anything. She did not agree that there is no express endorsement of the stated view by anyone. The example above seems to her to be an express endorsement and the other ones that were critical of HMRC said it is not worth arguing.

(d) It was put to her that Tiley sets out their view that "an addition to an existing settlement is, by itself, to be treated as a settlement in its own right". She said that she does not think it creates a separate settlement; it is the disposition to the settlement that is the trigger for the IHT charge. So she did not entirely agree with that but this comment was written in 2011/2012, before Barclays. She would class this as one of the neutral comments.

(e) It was put to her that a reasonable reader may interpret the following comment in her book as meaning she did not agree with the stated view: "The second test at the time the settlement was made might suggest an individual could set up a trust whilst non-domiciled but add further funds at the time when he is UK domiciled. This is not HMRC's interpretation". She said she was aiming what she said in her book at trustees to give them practical guidance. She would have said at the time that, on the balance of probabilities, the stated view was correct. But she was not certain, because she does not know what any tribunal would decide on the matter. At best she thought it more likely than not that the stated view was correct.

(7) She deals with wealthy foreigners who do not like publicity. None of the clients who she and her colleagues have dealt with, and who the accountants and trustees she knows have dealt with, have a massive appetite for being the guinea pig. She is not aware of any client who would actively go against the stated practice of HMRC, knowing that that would mean being in public view potentially for a long time.

(8) She could not think of any examples in an IHT context where she had dealt with a taxpayer considering taking a filing position that is contrary to a published view of HMRC. In relation to other taxes, in that case (a) her advice is always to disclose to HMRC and put additional wording in the tax return to explain the view taken in order to protect the client on penalties, and (b) she would always tell the client that even if they had a King's Counsel's opinion saying that HMRC are wrong, no one knows what is going to happen until the court or tribunal decides and so there is a risk of having more tax to pay than they thought. If one puts disclosure in the relevant tax return one protects the client as far as one can on penalties as one will hopefully show that reasonable care has been taken. Another reason for making disclosure is to protect the firm and herself from being sued and to protect their reputation. She had not come across clients filing in this context except in limited scenarios not of relevance here.

(9) She had spoken to other practitioners and the consensus is that it would not be advisable to add property and she had spoken to trustees who said they would not accept an addition after the settlor had become domiciled. The practitioners who share her view on the stated view are (a) a colleague, Michael Jepson, (b) two accountants: Mark Giddens, who is head of UHY Hacker Young's Private Wealth and Probate department and Viraj Mehta of Bourner Bullock, a firm that does a lot of international work, and (c) two trust companies that she works with, one in Guernsey and one in the Isle of Man. She asked them whether in practice they thought added property would be excluded property or relevant property. One of the trustees said that they would avoid that because of the amount of tracing they would have to do to work out what was relevant property and what was excluded property and said how tricky it would be for them to keep the accounts separate. They focused more on that rather than anything technical. She agreed that they would not want to have an addition to the trust fund if it could cause them a lot of work and it would be a particular problem where there is a single settlor, because there would be no reason other than this for the trustees to segregate the assets. She thought it is interesting that none of them made any reference to Barclays. She did not raise that in her question to them. She made it clear to them that she was talking about the position before the change in the law.

(10) She was asked whether the ongoing taxation of the settlement would necessarily be the most important factor in advising clients who had become domiciled not to add property. She said many clients did not want their excluded property settlement to have to file IHT returns even if it was on a small amount of the trust; that would then bring the whole settlement within the scope of HMRC's gaze. If they had just excluded property and never any relevant property they would not need to file. That was a massive factor for clients. She agreed that on making a chargeable transfer on transferring property into the trust a client would come within HMRC's gaze to begin with as a return would have to be filed for that and for both reasons - that there would be an immediate chargeable transfer filing and, as it would be relevant property, a ten-year filing and exit charge filing, "they would have been reluctant". She said that, if the added property was not subject to any relief on being added, very few clients would have wanted to pay an upfront 20% charge only then to find that they had not actually saved IHT in the long run because it was then also caught by the relevant property regime. There would have been a few, possibly, who might have thought it worthwhile because a 20% charge is better than a 40% one if the client kept the assets and died. But she had never been in that situation where a client has wanted to do that, so she was speculating. She did not think it was the chargeable transfer alone that put clients off but rather a combination of the chargeable transfer and having assets within the relevant property regime after the chargeable transfer.

I note that it was clear that Ms Summers evidence on this point largely reflected her views of what people would do based on her own view of the stated view and not on anything that she had actually experienced in practice.

Evidence of Mr Goldstone

78. Mr Andrew Goldstone of Mischon de Reya, a partner and head of the Tax and Wealth Planning Group gave the following evidence:

(1) He would have advised a client not to add property but it would have been rare for him to advise on that. He generally and routinely advised clients to add any personally owned offshore assets to their excluded property trusts whilst they were non-domiciled on the basis that would be free of IHT and would constitute excluded property, whereas if it was added when the individual was deemed domiciled it would constitute a lifetime chargeable transfer attracting tax at 20% and HMRC would not accept it was excluded property. Where the client owned assets with a UK situs he would advise on how they may be able to convert them to overseas assets with a view to settling them on the existing trusts. He thought this would also have been the generally accepted advice of other tax advisers and in his view it still is.

(2) Whilst before 1986 a client could add property without an immediate charge to tax or periodic or exit charges provided the trust conferred a life interest on the settlor, he would not have advised clients to do so when domiciled due to the risk of it not being excluded property. He did not recall other advisers advising clients to do so.

(3) He did not recall a case on whether added property would be excluded property. He thought there would have been a case had it been the practice for individuals to add property when domiciled. His recollection was that it was not the practice.

(4) In summary his advice was not to add property and to look at other options; that was the case both before and after 2006. He would have advised that (a) although the relevant legislation could be construed as meaning that only domicile at the time the settlement was set up was relevant - the point was uncertain. That construction produces a highly anomalous result and he thought it was inevitable that HMRC would challenge it and that it was more likely than not that a court would decide the issue in their favour. His view was partly informed by the prevailing tendency at the time for judges to find against the taxpayer where there was perceived to be a tax avoidance motive and he thought that view was likely to continue and increase by the time the case was litigated, (b) HMRC's long held public view was that added property did not qualify as excluded property, as was also set out in the IHT manual which he and other professional advisers would have considered once the manuals were publicly available from 1996 onwards. Leading textbooks referred to this as HMRC's stance specifically by reference to the bulletin or more generally, and (c) there would be an entry charge effectively due at 25% because of the required grossing-up, after 2006, even for life interest trusts unless the transfer fell within the nil rate band or qualified for a relief such as business property relief. The uncertainty of the status of the property as excluded property combined with the entry charge would together be a strong reason not to add property.

(5) Even after the Barclays decision he was concerned about adding to an existing excluded property trust as he had written in an article for tax journal.

(6) He said in his witness statement that (a) he thought at the time other professional advisers took the same view as he did, (b) every professional adviser he had spoken to said they would have given the same advice. That was the view of all seven of his partners in his department. Only one of them worked at that firm in 2013, the rest worked elsewhere, and (c) he asked a number of other well respected and experienced professional advisers and the consensus was that, in almost all circumstances, they would not have advised clients to add property for the reasons he had already given. They all said they would have taken into account the stated view. That was their practice notwithstanding the reservations expressed by some commentators and text book authors as to the correctness of the stated view. I note that Mr Goldstone was not so certain in his comments on this when cross-examined as set out below.

(7) He was aware that all of the specialist textbooks acknowledged and referred to the stated view and some of them considered the stated view to be open to question or incorrect but some of those cautioned against acting against HMRC's view. He did not recall any professional advisers advising their clients to take a different view. Nor did he recall any articles in the professional press before 2013 or before Barclays espousing a contrary view to the stated view. In his view, the vast majority if not all tax practitioners de facto adopted HMRC's view for pragmatic reasons. Even after Barclays some advisers remained concerned about it. For example MacFarlanes' commentary of November 2017 cautioned that such a transfer was likely to be subject to IHT. I note that it is apparent from cross-examination that the view expressed in the highlighted passage appears to be speculation on Mr Goldstone's part.

(8) He had advised a client only once to take a view different to a published view of HMRC. That was in the context of double trust schemes for IHT on the family home and they fully disclosed to HMRC. If he/the client were to take a view that is contrary to the known view of HMRC he would always advise the client to disclose fully for similar reasons to that given by Ms Summers.

(9) When asked if, in practice he had dealt with added property, he said the managing partner of his firm had asked him if a client could add property and he had said he could not as it was likely to be relevant property. He did not think he was paid a lot of money to spend a lot of time looking at it. I do not take this to be an example of Mr Goldstone advising on the stated view in practice given he only appeared to have had a discussion with his colleague. The technical aspects would have been relevant but certainly his view would not have just been based on the legislation because that is not really how solicitors work when they do not have a particular client paying them to spend time to do the analysis.

(10) When advising a client on a course of action HMRC's view of the law is an important factor and if someone were to go against that view they would be likely to be challenged. He said that does not mean HMRC's view always prevails but it is a factor, as are the numbers involved, the robustness of the client, "the willingness to have a row and all of those factors will come into giving advice". He agreed that another factor might be considering what else could be done to achieve the objectives and if there were other simpler methods that could be used. He said that the "existence of an entry charge clearly is a factor and will deter lots of clients" but he would not have advised a client to add property even where there was no entry charge, such as (a) on the addition of property before 2006 to an existing life interest trust, or (b) where property qualifies for business property relief. He did not advise anyone to do that before 2006 because he was very concerned that the added assets would be relevant property. He did not think he ever advised in practice on cases in (b) but he would not have recommended it. Whilst the added assets could be sold and turned into cash held overseas there is, in his view, a high risk that cash would be relevant property. If the client retained then sold them and sat on the cash personally, he might shed his deemed domicile and if he died when he was no longer domiciled the cash is clearly excluded from IHT. He was talking hypothetically as he had not advised on this. Whilst a 20% entry charge is usually a deterrent there are some clients who are set on having a trust for their children, for example because they do not trust them, and particularly if they are elderly they may well pay the entry charge in the hope that they would survive seven years; that gives a better result than adding the assets on death to a trust in their will and paying IHT at 40%. If a client had an excluded property trust, became deemed domiciled, added assets on paying the entry charge, the risk is that the assets that are added are relevant property and if they survive seven years and then they die, if they were a beneficiary, there would be a further 40 % charge which makes a 60 % charge.

(11) There is no doubt the provisions could be construed in different ways, as is demonstrated by the textbooks and principally the barristers take a view that "is just different from those who advise on the ground". The barristers take a slightly, or in some cases a, clearly more, robust view about the technical meaning and the correctness of the stated view.

(12) The normal course of action was to advise clients to make all their settlements whilst they were non-domiciled but it is not always practical. Clients often come for advice very late in the day when they are about to become domiciled and some of them do not have an existing trust. In his entire career he never said to any of those clients: "look, create a pilot trust, because then when you are in a position to or if you become wealthy in the future, which you hope to do, then you can stick your assets in at that stage and they will be excluded property". He does not think he would have given a thought to the entry charge; it was based on the fact that he did not believe the added assets would get excluded property status.

(13) He agreed that it is rare to have someone who created a trust, became deemed domiciled, and who then asked whether they should add property. He had not considered the route of adding property to a company owned by a trust; he thought he first saw that in the other witness statements prepared for this hearing. He agreed that if that structure produced the result counsel said it did, it would produce an anomalous result.

(14) He was probably not distinguishing IHT from other taxes in his comments on the views of the courts. He thinks from the early 2010s onwards there was a very clear move, both in legislation and in the courts, away from what might be deemed to be contrived tax arrangements. That would have informed his views about advising clients not to add property.

(15) He was asked if he interpreted any of the commentary in the various textbooks as endorsing HMRC's view. He said that a number of them said that in their view it was wrong or could be argued to be wrong. In his article he set out his view that it is risky to assume that the decision in Barclays meant that property could be added safely, whether or not there was an entry charge and assume that it would be treated as excluded property. He thought there was a risk that it would continue to be relevant property.

(16) As regards the other advisers Mr Goldstone spoke to for the purposes of these proceedings he gave the following evidence:

(a) He initially said (i) in his own experience and that of everybody he spoke to, nobody would have advised a client to add property, (ii) that was for number of reasons, but one of them absolutely was the status of the added property, and (iii) he did not know how often in practice those he had asked about this had come across this point. He just asked them the hypothetical question he was asked by HMRC.

(b) He later said he thought he asked the question in a way that was not purely hypothetical as regards the partners in his own firm; they absolutely would have told him if they had had a case they had specifically advised on or if they had dealt with HMRC on this. He asked a number of people outside of the firm by email as a hypothetical question.

(c) None of his partners said "Yes, I had this situation and I advised in such and such a way". He initially said they all essentially said: "We wouldn't have advised the client to add, because we don't think it would have been excluded property". They all said they would have advised their clients before they were deemed domiciled to add property. He had discussions face to face with his colleagues on this and none of them said anything about the entry charge. It was always focused on the status of the added property. Later he modified this and said he was sure they said "No, we wouldn't do it" and he was pretty sure he might have asked and they would have said "Because it won't be excluded property". It appears from this that whilst he was sure these persons said they would advise against adding property he was not sure of the reason for that advice.

(d) He was certain that none of them said "Oh, it's because there would be an entry charge and they don't want to pay the 20%". He thought the entry charge was not a point they would have thought of immediately when he asked the question, so he thought their replies were all based on the status of the added property. He added "I mean it is pure speculation as to what was going through their mind....in the response to the way that I asked the question". In my view this comment confirms that Mr Goldstone was not sure what rationale those he spoke to had for saying they would not have advised settlors to add property.

(e) The advice was to get property into a trust before the client becomes deemed domiciled because if property was added after that occurred there was every likelihood that the assets would not be excluded property. He thought that all advisers would recommend to their non-domiciled clients to do so because of the risk of the added property being relevant property afterwards. This comment appears to be based on his view/speculation that others would take the same view as he did as regards the stated view rather than on any actual experience of what other advisers advised. As he had said earlier, he had never advised a client to set up a pilot trust with a view to adding to it after they had become deemed domiciled.

79. I note that (1) in a number of instances, Mr Goldstone appeared to be speculating that other advisers both now and over the years agreed with his view that the stated view is likely to be correct, (2) (a) his colleagues and other persons he put the issue to for the purposes of these proceedings were asked a hypothetical question. Whilst he said he thought it was not purely such as regards his colleagues he appeared to mean only that he expected them to tell him if they had dealt with the issue in practice and none of them did so, and (b) it is unclear from his evidence whether any of these persons agreed with the stated view or considered it was likely to be correct or simply considered that it was something they should take into account in advising clients on the safest course of action (to settle property whilst non-domiciled) to avoid conflict with HMRC over the stated view.

Submissions and decision on the application of section 255

Payment in satisfaction of any liability to tax

80. I consider it clear that, as HMRC submitted, on the evidence of Ms Jweinat and Mr Ryder, for the purposes of s 255, the payment of the disputed tax has been made and accepted in satisfaction of "any liability for tax", namely, the ten-year charge relating to the property, on the basis of the stated view. Mr Ewart disputed this but I do not accept the point he made detracts from this conclusion:

(1) He said that (a) HMRC's stance is fundamentally inconsistent with the fact that, earlier in these proceedings, HMRC argued that the disputed tax was paid on a wrong view of the law and should be a higher sum. They made an application to amend their Statement of Case to include an argument that the C shares were not excluded property. That was refused on case management grounds but HMRC indicated that they would revisit the point in the future, depending on the result of this appeal, (b) the stated view is not only a narrow view of the body of principle; RI 166 is about how one actually deals with the situation more generally.

(2) I consider that whether the ten-year charge is also due by reference to the value of the C shares is not relevant to the current proceedings. I note that s 255 refers to the payment of tax being made and accepted in satisfaction of "any liability for tax" made on the relevant view of the law. In my view, this requirement is satisfied as regards a payment of tax which was plainly paid by the appellant and accepted by HMRC in satisfaction of the ten-year charge so far as it relates to the particular property which the appellant accounted for as subject to that charge, on the basis that that property was not excluded property on the stated view.

General points on the application of s 255

81. HMRC made the following points which Mr Ewart did not contest and, subject to my comments below, I accept them:

(1) As regards the statutory question of whether the stated view represents "a view of the law then generally received or adopted in practice":

(a) This must be examined by reference to the point in time ("then") at which the payment of tax was "made and accepted in satisfaction" of the liability (which in this case was in 2013). There is no previous case considering the application of s 255. The decision in this case was before the first case considering a similar issue - Barclays (which was only heard in the High Court in July 2015).

(b) The legislation does not require that the reasoning or analysis underlying the "view of the law" must be generally received or adopted in practice, only the view of the law itself.

(c) On an ordinary reading, the word "generally" qualifies both "received" and "adopted in practice". Moreover, the use of "or" indicates that s 255 will be engaged where either: (a) The view of the law is "generally received"; or (b) The view of the law is "generally...adopted in practice".

(d) The word "generally" indicates that there is no need for the view of the law to be universally received or adopted, or for there to be an established consensus. Where HMRC have stated a view of the law and the majority of practitioners either receive that view or adopt it in practice, s 255 will be engaged. The fact that a particular view of the law may have a limited application does not mean that it cannot be "generally received or adopted in practice" if, in the circumstances in which the view of the law is relevant, the majority of practitioners either receive it or act in accordance with it.

82. The main dispute was over how the terms "generally received" and "generally adopted in practice" are to be interpreted and precisely how these tests are to be applied in these circumstances. I refer to these tests as the "received test" and the "adopted test" and their application is considered in full below. The use of the term "generally", on its natural meaning, suggests that these tests are satisfied only if usually or, in most cases, relevant persons receive the relevant view of the law or adopt it in practice. I refer to "relevant persons" as it seems to me that the use of the term "generally" leaves open the question of whose views or conduct is relevant in this regard. I consider that, where the view of the law is a view published by HMRC as regards a specific area of tax law, the sensible view is that this must require an assessment of the views/conduct of those likely to be concerned with the stated view, namely practitioners in the relevant field of tax law and their clients, as well as HMRC (although due to their status and employment by HMRC, HMRC officers can be taken to accept and act on the stated view, as was the case here, as Mr Ryder confirmed). I use the term "generally" in the rest of this decision in this sense.

83. Mr Ewart made the point that it must be borne in mind (1) that the application of s 255 has the effect that either (a) HMRC cannot collect what is known to be the correct amount of tax as the law is correctly to be interpreted, or (b) the taxpayers suffer more tax than is due on the correct interpretation of the law, and (2) hence, one would expect the hurdle for s 255 to apply to be a high one and it is a high hurdle. I have commented on this point below.

Submissions on the received and adopted tests

84. As regards the received test:

(1) Mr Baldry submitted that (a) on the natural meaning a view is (i) received when it is communicated to the recipient, and (ii) generally received if it is publicised in such a way that taxpayers and advisers would generally be aware of it such that it is available for all taxpayers to operate on the basis of. It must be set out in such a way that it has become embedded in the consciousness of the relevant taxpayers and advisers such that it is "part of the furniture", and (b) it does not require that the view is agreed or accepted as correct; it does not require an enquiry into the subjective minds of all taxpayers to ascertain whether each taxpayer personally agreed with it or not.

(2) Mr Ewart submitted that (a) the definition in the shorter Oxford English Dictionary is of assistance as Parliament has used the term received in a sense that is not the most common. This gives a meaning of "accept as authoritative or true, chiefly as received; past participle". It is used in the same sense as one uses it in the phrase "the received wisdom". Hence the dictionary gives examples of: "The received wisdom held the sector was poised to fall" and "Teachers remained locked in on received ideas about poetry". In this context authoritative must be taken to mean as a textbook writer might be authoritative or correct. One would expect that, if a view of the law was accepted as authoritative or correct, it would appear as a statement of the law in at least a huge majority of the textbooks/commentaries. That seems broadly to be the view Mr Ryder took; he thought HMRC had a stronger position on the adopted test. HMRC's interpretation would mean that HMRC could effectively make the law by just issuing statements of practice. Inevitably, practitioners have to take account of it in giving their advice but in doing so they are not accepting (or adopting it). They are simply taking account of it as any competent professional would. There is not a single textbook writer or commentator who expresses support for HMRC's view of the law. There are varying degrees of disagreement or scepticism and doubt about it. These are not simply unreasoned assertions by textbook writers, they are reasoned remarks.

(3) Mr Baldry responded that: (a) it is unlikely that "received" is intended to mean that the relevant view of the law must be positively agreed or even accepted as authoritative because that would involve considering people's subjective views, which is unlikely to have been Parliament's intention in applying a broad test, which should be tested by objective evidence. In any event, if that is the meaning, the test is met as set out below, (b) moreover, if the draftsman had intended this test to mean that the relevant view of the law must be accepted as true, he would not have used the word "received" but rather a term such as "accepted", and (c) it is not a requirement for this test to be met that, when people are taking decisions on how they should act, they must have in their mind that they agree with the stated view. It suffices that they take a course that adopts the stated view, such as by advising a settlor to take action in order to avoid the application of the stated view by settling property before they become domiciled.

85. As regards the "adopted in practice" test:

(1) Mr Baldry submitted that this test is focussed on whether HMRC and taxpayers/practitioners chose to take action in practice on the basis of the stated view, regardless of whether or not they agreed with it. On its ordinary meaning "adopted" has the flavour of a person taking a course of action as a matter of choice. As well as evidence of HMRC's and practitioners' experience, it is highly relevant to consider what the commentators advised taxpayers to do, in particular, including advice to act on a basis which is consistent with the stated view (as opposed to advise to challenge or go against it); undertaking a transaction to avoid contradicting the stated view is a means of adopting it.

(2) Mr Ewart agreed that this test requires consideration of the way that practitioners act when advising clients and of the course of action people took. However, in his view, this test is met only if relevant persons viewed the stated view as correct. It is not enough that they take a course of action to be safe from a conflict with HMRC; simply taking account of the stated view in one's actions does not amount to adopting it.

Decision on the meaning of the received and adopted tests

86. In my view:

(1) It is likely that, in choosing to use the term "received" the legislature intended s 255 to apply where the relevant view of the law is viewed generally (by relevant persons) as being correct or, at least, likely to be correct, and not merely where the view of the law is well publicised and generally known to relevant persons. That is a well-known and understood meaning of "received" when applied, in effect, to a state of knowledge or way of doing something. I note HMRC's contention that the legislature would have used the term "accepted" (or something similar) if this test was meant to have the meaning I ascribe to it. However, I consider that this interpretation reflects a natural, plain meaning of "received" (as in "received wisdom") whereas HMRC's interpretation does not sit well with any natural meaning of the term. It would be an odd use of language to refer to a well-publicised view of the law which is accessible and known to relevant persons by using the one word reference to a "received" view of the law.

(2) (a) I do not accept that applying the received test in this way involves any particular difficulty as it requires consideration of the relevant persons' subjective views of the relevant view of the law; that can be determined by an objective assessment of the available evidence such as commentary in the textbooks and other published materials and the actions of relevant persons, and (b) As Mr Ewart said, it is reasonable to suppose that the legislature intended to set the bar somewhat higher than would apply on HMRC's interpretation, given that the effect of s 255 applying may be that tax is due at a higher or lesser amount than would be due on the correct view of the applicable law. It would be very surprising if the effect of this provision were that HMRC could, in effect, override the law as correctly interpreted, if their official view is not in accordance with that, simply where that official view is published and widely known, even if it were thought generally by relevant persons to be incorrect.

(3) As the parties were agreed, the adopted test requires consideration of what course of action people took, in real life situations, as regards the relevant view of the law. On its natural meaning, the term carries the connotation that the view of the law must be taken on board in a positive sense that it is used, followed, or acted in accordance with in a way which shows it was viewed as correct or highly likely to be correct. To my mind, as Mr Ewart submitted, taking a course of action which it is thought would achieve the desired result without falling within the parameters of the relevant view of the law, thereby avoiding a potential conflict with HMRC, does not amount to adopting that view in the required positive sense the use of that term implies. The same point made in (2)(a) applies here also and similarly to the point made in 2(b), given the nature and effect of the provision, I consider it reasonable to suppose that the legislature did not intend HMRC's view of the law to override the actual law unless relevant persons generally acted on it on the basis it is correct/highly likely to be correct.

(4) Overall, whilst there may be some overlap between the two tests, essentially the difference appears to be that (a) the focus under the received test is on whether the relevant view of the law is generally accepted by relevant persons as correct/highly likely to be correct as a theoretical matter of intellectual analysis, and (b) the focus under the adopted test is on whether the relevant view of the law is accepted as correct/highly likely to be correct by relevant persons as demonstrated by the actions they take in practice.

Conclusion on application of s 255

87. I have concluded that, on the basis of the evidence, the stated view was not at the relevant time generally received or generally adopted in practice, as I consider those tests are to be applied, such that, as a matter of law, s 255 does not apply in this case.

88. The evidence does not demonstrate that the stated view was generally received by relevant persons. Rather it demonstrates that, on balance, the stated view was not generally received by relevant persons.

89. I accept that the evidence establishes that (1) practitioners in this field were aware of the stated view as the official view of HMRC at all relevant times, (2) the practical advice given by commentators and the witnesses, even from those who took a view that the stated view was not or might not be correct, was to avoid HMRC seeking to apply the stated view by taking steps to avoid being in the circumstances to which it is applicable. HMRC noted, in particular that Chamberlain stated: "...although it is common sense to follow HMRC guidance where practical and never add to a settlement after a settlor has become deemed UK domiciled" and in Clarke where the authors advise that although the stated view may be wrong "additions in such circumstances should be avoided as the argument is not worth having", and (3) practitioners routinely advised non-domiciled clients to the extent possible to add property to their settlements before they became domiciled.

90. I do not accept, however, that the evidence demonstrates, as HMRC submitted, that by 2013 the majority of practitioners agreed with the stated view albeit that they evidently gave it consideration when advising taxpayers how to conduct their affairs. There is little evidence that the stated view was accepted as correct or likely to be correct and/or as an authoritative statement of the law. Rather, the weight of opinion appears to have been that the correctness of the stated view was doubtful.

91. As regards the textbook writers:

(1) Mr Ewart submitted that the textbook writers either make no comment, are not on point/take a mistaken view or show scepticism that the stated view is correct. He noted the following:

(a) (i) Eastway quotes the stated view with no comment, (ii) the comments in McLaughlin to which Ms Summers referred, as she agreed, are not correct. In making these incorrect comments the authors do not endorse the stated view; they appear to have misunderstood how the provisions work, (iii) Tiley states in 2011/2012 that the stated view has the effect that the addition to an existing settlement is by itself to be treated as a settlement in its own right but it is clear that is not HMRC's position, and there is no comment on the issue.

(b) (i) Ms Summers' comments suggest to the reasonable reader that she is sceptical about the stated view because she says "the second test might suggest" and comments this is not HMRC's interpretation. At the least, this is a neutral view, (ii) Foster states: "there is no specific provision requiring additions to a settlement to be treated as being comprised in either a separate settlement or the original settlement", (iii) Tolley records the stated view and says: "This interpretation is open to question although it has not so far been challenged in any appeal", (iv) see the comments of McCutcheon set out in Ms Chamberlain's evidence, (v) Chamberlain makes the comments HMRC cite above and also note in the summary that, in light of Rysaffe and the legislation, there is a strong argument that additions of property may be excluded property, (vi) Clarke states that although "HMRC's view has the attraction of logic, it is very difficult to support as a matter of construction" and points to Rysaffe as pointing against the stated view and notes two further arguments against it, (vii) Hutton states the following: "To gain the protection of section 48(3) it is of course essential that the trust is made before the settlor becomes actually or deemed UK domiciled. Some advisers, but not the authors, take the view that a pilot trust can safely be established with the substantive funds added later even at a time when the settlors become deemed or actual UK domiciled." He then states that the stated view conflicts with other provisions of IHTA in particular s 67 which assumes that property can be added to an existing settlement. On a proper reading of his comments, Hutton expresses doubt about the correctness of the stated view but states that one cannot safely take the stated action because it will undoubtedly be challenged by HMRC because they take an opposite view, (viii) as set out in Ms Chamberlain's evidence, Kessler expressly states the stated view is incorrect and he notes that "the legislation is so clearly inconsistent with the HMRC view that even a purposive construction cannot assist". See also the passages put to Ms Chamberlain.

(2) Mr Ewart added that the fact that the advisers advised people to do the safest thing does not demonstrate that they accepted the stated view; they did so as naturally one would do the thing that is most obvious and safe in the circumstances.

(3) Mr Baldry responded that the commentary in the textbooks cannot be described as sceptical of the stated view. The commentators were sensibly pragmatic in their advice. Only Mr Kessler is categoric in stating the stated view is incorrect. If a work simply states the stated view, it seems the author accepted the stated view as the law. Many of the commentators put forward HMRC's view and recognised there is a theoretical possibility of challenge, but said unless and until someone challenges it, one should operate on the basis that the stated view should be followed. That is the common-sense position to follow.

92. Overall the textbook writers either made no comment on the stated view, in a couple of instances, as Mr Ewart noted, seemingly misinterpreted it or expressed doubt or reservation about whether it is a correct statement of the law. None of them expressly stated they agreed with it and/or endorsed it as a correct statement. I can see no reason to assume, as HMRC suggested, that if the textbook writers made no comment on the stated view, they are to be taken to accept it. Nor do I consider that, reading the relevant comments in context, any positive endorsement or acceptance can be taken from the fact that commentators advised that it was best to adopt courses of action, which would avoid HMRC seeking to apply the stated view. This appears simply to be practical advice given on the basis that, where there is more than one course of possible action which it is thought would achieve the desired result, it is advisable to take the course which avoids the taxpayer coming into conflict with or, facing a challenge from, HMRC.

93. As regards the other evidence, overall the picture is one of divided opinion: (1) the stated view was accepted as authoritative or correct by Ms Summers, Mr Goldstone and several people who Ms Summers discussed this with. It is not clear from his evidence that the persons who Mr Goldstone spoke to for the purposes of these proceedings agreed with the stated view or regarded it as authoritative. I make no finding that they did so, and (2) those who did not agree with the stated view are Ms Chamberlain, the two counsel she referred to and Mr Millington, who when working for SW advised the Trustee. For the reasons set out in [99] below, the evidence on the conduct of practitioners/their clients in practice also does not support a finding that generally the stated view was viewed as correct or likely to be correct.

94. HMRC also referred to comments made in the Trusts Discussion Forum as showing that practitioners advised on the basis of the stated view: (a) In June 2015 Francesca Gandolfi (a technical specialist in tax and estate planning at Cananda Life) said that (i) the simple way to avoid IHT applying to non-UK assets, before moving from a non-UK domicile to a deemed domicile status, is to transfer the assets to an excluded property trust, and (ii) if assets are settled into this Trust before a person becomes deemed domicile the assets within the trust will be classed as excluded property, even after the individual becomes deemed domiciled, (b) in November 2017 Mr Andrew Goodman of Osborne Clarke LLP described assets as excluded property provided the settlor "was not UK domiciled when the trust was established and funded....and the assets were located outside the UK on his death. - s.48 IHTA" and Mr Magill (of Shipleys LLP) added comments in which he seemed to agree: "it is property that is excluded, not the trust, provided the settlor was non-domiciled when he or she added the property...". I note that these comments were made some considerable time after the period in question and therefore I consider that no material weight can be attached to them.

95. There is insufficient evidence that the stated view was "generally adopted in practice" for me to conclude that the adopted test is met. If anything the evidence establishes that it was not so adopted.

96. I accept that HMRC applied the stated view and challenged any taxpayer who to their knowledge took a different view. As regards examples of the behaviour of taxpayers/practitioners:

(1) Ms Summers gave two examples where the stated view was adopted. There is no evidence that any of the other persons whom she spoke to had applied the stated view in practice. Mr Goldstone's view and discussions with his colleagues and others were on a hypothetical basis and he was clear his colleagues would have told him if they had advised on this in practice.

(2) On the other hand there are five examples from the evidence of Ms Chamberlain of a view contrary to the stated view being applied in practice. I do not accept HMRC's submission that these are not relevant examples. HMRC noted that (a) two concerned transfers to a company - as a means by which taxpayers got around the stated view, and (b) in two cases no (potentially) chargeable event had occurred. Given the evidence above about the importance of disclosure (particularly on death), it is unlikely that the taxpayers have adopted a position contrary to the stated view as otherwise HMRC would know of it, and (c) in the case in which Ms Chamberlain advised in 2006, the taxpayer did not encounter any potentially chargeable events. In my view, a professional advising on a view of the law contrary to the stated view, as regards the consequences of a proposed transaction, is an application of that contrary view in practice. I consider little can be taken from the fact that initially in this case the appellant acted in accordance with the stated view given the appellant then changed stance on the advice of SW acting through Mr Millington.

97. HMRC submitted that the fact that practitioners advised their clients to undertake transactions which did not contravene the stated view is a form of adopting that view in practice and that there is an absence of evidence of taxpayers adopting an inconsistent view should also be taken as indicating that the stated view was adopted in practice:

(1) Mr Baldry emphasised that (a) the practice was for non-domiciled settlors to contribute property to trusts before they became deemed domiciled, and (b) some advisers advised taxpayers to transfer assets to a company held by the trustees. Mr Baldry noted that Chamberlain described this as a "way round" HMRC's position. In taking such action, the taxpayers chose to take a course of action on the basis that the stated view was correct; in effect, they acted to ensure that they complied with the stated view, and (c) even advisers who considered that the stated view of the law was or may be incorrect advised it was unwise to go against it and that the client should take the safest course of acting in compliance with it.

(2) Examples of the type of advice referred to above in the commentaries include (a) the sections in Clarke, Chamberlain and Summers referred to above and the statement in Kessler put to Ms Chamberlain, and (b) McCutcheon suggested that HMRC's view of the law was incorrect but still advised that, where a settlor has become domiciled, they should add property to a separate settlement to avoid disputes - by putting the assets into a separate settlement the settlor forfeited the argument counter to the stated view.

(3) Even after Barclays, commentators suggested that taxpayers should still assume that property additions would be relevant property. See an article by Andrew Goldstone and Natalie Quail (Mishcon de Reya) dated 15 November 2017 and an article by Macfarlanes (Nicholas Harries, Robin Vos and Kynaston) dated 3 November 2017. I note that these articles were published some years after the period in question and so are not of material relevance.

(4) Mr Ryder only has one example of a taxpayer initially adopting a position which was contrary to the stated view which was settled by the taxpayer. HMRC's system was more likely than not to have picked up activity counter to the stated view if it was being generally adopted. The witnesses confirmed that, if a taxpayer adopted a position inconsistent with a clear statement of HMRC, they would make disclosure, as the Trustees, in effect, did in this case. The evidence that taxpayers would have disclosed to HMRC occasions when they had departed from the stated view and that HMRC did not receive such disclosures demonstrates that taxpayers did not generally adopt a practice different from the stated view. For the reasons already set out I do not accept HMRC's further contention that Ms Chamberlain's examples are not relevant or that the fact that Mr Millington advised the appellant to take a view contrary to the stated view cannot be taken into account as he did not attend to give evidence. There is documentary evidence in the bundles from which it is clear what Mr Millington advised.

98. Mr Ewart responded that:

(1) The set of facts giving rise to the issue in this case are very rare and it is even rarer for there to be a chargeable event for IHT purposes as a result of it. It is extremely unlikely that anyone, except by accident, would add property to a discretionary trust after becoming domiciled as that would result in a tax charge at 20% at a minimum (and in the past it would have been greater). A property addition may occur in relation to a QIIP as there is no such entry charge; but a chargeable event would occur in that case when the settlor/his spouse died. As Ms Chamberlain said, typically such people who were non-domiciled but become deemed domiciled intend to leave the UK (see the example she gave). The only reason it has arisen in the cases that have come before the courts is because, as in this case, there was a mistake about when the settlor became domiciled, and in Barclays where there were transfers between settlements.

(2) It is not an adoption of the stated view for an adviser to advise a client to take the safest and most straightforward course of action, such as to settle property before becoming deemed domiciled. The advice professional advisers would have given in 1976 would be exactly the same as they would have given in 1996 or in 2006. The reason for such advice is not the stated view; it is simply because any professional can see that the best course of action is to create a settlement when persons are non-domiciled and there are other tax reasons for doing that. That avoids an entry charge and ensures that the property is excluded property without any possible argument. That is simply what any professional would do, even if they were convinced that the stated view was wrong. No competent professional would ever advise a non-domiciled client to wait until they became deemed domiciled to settle property. Moreover, there is no actual evidence of situations where the stated view was applied in practice where a non-domiciled person was about to be deemed domiciled.

(3) The reason why people transferred assets to a company after the changes to the settlement code in 2006 was not particularly to do with s 48(3), as Ms Chamberlain explained. Even if they did that simply to avoid an argument with HMRC about the stated view or because they knew that HMRC agreed that the company route worked, it would be an abuse of language to say they adopted the stated view. That does not mean they necessarily agreed with the stated view.

99. Mr Baldry accepted that for the issue in this case to arise appears to be rare and accepted there is not a great deal of evidence as to what people actually did but submitted (1) that is because non-domiciled settlors were advised, on the basis of the stated view, to add property to their settlements before they became deemed domiciled, and (2) in doing so, people adopted the stated view by taking action on the basis that it applies; plainly one can adopt a practice by refraining from doing something and it is irrelevant whether or not they agreed with the stated view.

100. Mr Baldry added that (1) it is notable that there was a significant period of time after RI 166 was published and before the 2006 changes came into force, where settlors were entitled to add property to many settlements without any such charge as many settlements qualified as QIIPs. However, it appears it was rare for that to occur. So the potential charges under the settlement regime were a significant further disincentive and this tends to show that the stated view was adopted in practice. Ms Chamberlain accepted that there was extensive discussion about the stated view because people were thinking about adding property to their QIIPs before the 2006 changes and that continued after the changes, (2) it is also relevant that the commentators advised settlors not to add property; it is unlikely that people would disregard such advice, and (3) is not clear that the company route came about to deal with changes in the law in 2006. McCutcheon and Kessler had introduced the idea of using a company in the context of excluded property settlements. The planning assumed that the stated view was right and the origins of the idea was as a workaround to avoid the application of the stated view.

101. I do not consider that generally practitioners can be viewed as adopting the stated view as a result of advising settlors (a) to settle property whilst non-domiciled, and/or (b) to transfer assets to an overseas company held in a settlement and/or that their clients can be viewed as doing so in taking that action:

(1) First, it is not clear to what extent the use of these courses of action was a reaction to the stated view. Mr Ewart made the valid point that it was sensible for practitioners to advise clients to add property to a settlement whilst non-domiciled even before the stated view was published and regardless of it. I accept that the company route appears to have been under consideration/in use before the changes in the law in 2006. However, on the evidence of Ms Chamberlain, after the changes to the settlements code in 2006, the reason for using the company route was not particularly to do with s 48(3); it was because of those changes. None of the other witnesses had advised clients on the company route.

(2) Even if it is assumed that relevant persons generally advised on and took these courses of action in response to the stated view, that does not, in my view, demonstrate that they generally adopted the stated view in the required sense. The evidence does not establish that most or even a majority of practitioners accepted that the stated view was correct or likely to be correct in advising clients to take such action or that clients who took the advice did so. Overall the view appears to have been that where more than one course of action is available to achieve a particular result, taking the action which would avoid falling within the scope of the stated view and hence any conflict with HMRC as regards the stated view was simply a practical course of action to take (see the comments above regarding the commentary in the textbooks).

102. It is not reasonable to make an assumption that the stated view was adopted in practice from the fact that there is little other evidence of taxpayers adopting an inconsistent view:

(1) I note that Mr Ryder only has one example, other than the Barclays case, of a taxpayer initially adopting a position which was contrary to the stated view, and I accept that the evidence establishes that (a) practitioners would advise taxpayers to make disclosure to HMRC if they took a course of action contrary to a published view, such as the stated view, if that occurred in relation to an event requiring an account to HMRC, (b) HMRC would be likely to pick up any instances of a contrary view being taken when a report was made to them for IHT purposes, and (c) due to the nature of his position Mr Ryder would be likely to be aware of relevant challenges by taxpayers.

(2) However, overall the evidence of the witnesses is that (a) the addition of property was a rare occurrence. Non-domiciled settlors were routinely advised to contribute property to their settlements before they became deemed domiciled and the "entry charge" which would apply in many instances was a disincentive, and (b) even if there was an addition of property, that would not necessarily require an account to be made for IHT purposes if the view was taken that, contrary to the stated view, it was relevant property. On that basis, it cannot be assumed that the absence of known challenges to the stated view demonstrates that it was being adopted in practice.

(3) I note that before 2006 settlors could add property into many settlements without an entry charge and there is some evidence that this was not a frequent occurrence. However, I do not consider that it can be assumed that the stated view was, as HMRC submitted, the main or a significant factor in influencing individuals not to add property before 2006, given that the evidence is that (a) there are other reasons influencing individuals' relevant choices as to how to deal with their valuable assets (see the evidence of Ms Chamberlain in particular), (b) the precise extent to which property was added is simply not known, and (c) the rarity of additions of property, whether before or after 2006, was at least to some extent because settlors were advised to and did add property whilst they were non-domiciled. For the reasons already given, on the correct interpretation of s 255, advising on and/or taking that course of action of itself that does not show that the stated view was adopted in practice.

(4) I accept that, as HMRC submitted, it is reasonable to assume that advisers/taxpayers would act on the basis of advice given in the leading textbooks which discussed the stated view. However, for the reasons already set out, the comments made in the textbooks do not demonstrate that the majority of authors regarded the stated view as correct/likely to be correct and to the extent authors gave advice on action to be taken, such action does not amount to the adoption of the stated view in practice.

Decision of Mr Ryder

Submissions

103. Mr Baldry submitted that Mr Ryder's decision was reasonably made:

(1) Mr Ryder took into account the correct factors in reaching his decision: (a) He reviewed the various commentaries and, in particular, those which criticised HMRC's view of the law except Chamberlain but he looked at Dymonds, which was written by the same authors. There is nothing now produced in evidence which he did not look at which would have affected his decision materially, and (b) he considered the lack of examples known to HMRC of taxpayers adopting positions contrary to the stated view. The evidence now produced simply provides further strong support for the view he took, as the evidence is that taxpayers would have made disclosure had they taken a contrary view to the stated view. The weight to be given to these factors is essentially one for the decision-maker, Mr Ryder, but there is no sense he has given undue weight to any one matter. He looked at the appropriate matters to take into account and he formed a view which was entirely reasonable. He certainly did not take into account any irrelevant consideration and there is no category of evidence he failed to take into account which makes his decision unreasonable.

(2) There is a final cross-check. He has now seen all the other evidence, including Ms Chamberlain's witness statement and all the other books, and he confirmed that, in light of all this additional material, he would not alter his view at all. So the evidence is that even if he had reviewed this additional material the result would inevitably have been the same.

104. Mr Ewart submitted that on the correct approach Mr Ryder made a number of errors which vitiate his decision:

(1) It is highly unusual that there is no decision in writing. Mr Ryder erred in his decision on the substantive legal points: (a) he improperly fettered his discretion by considering that he had no choice but to follow the official guidance in the stated view. It is only guidance and any officer exercising the powers of the board to decide whether they are satisfied that they should make a repayment must make up his own mind on all matters relevant to the decision including the substantive legal issue, (b) his decision on the law (which he felt he had to come on the basis of the stated view) is wrong as a matter of law.

(2) Mr Ryder also made a number of errors in deciding that s 255 applied:

(a) He made errors of law which of themselves vitiate his decisions: (i) He did not take any account of the basis on which the tax was paid. He believed that the only question which he had to address was whether the stated view was adopted in general by practitioners. However, the starting point of s 255 is that the officer has to address on what view of the law the payment was made, and (ii) he took the view that adopting in practice is a stand-alone phrase which is not qualified by the word "generally". The parties are agreed that the correct interpretation is that generally applies to that term.

(b) He failed to take into account a relevant matter. He took into account that there had been no challenges, other than the two cases he mentioned, and what was said in the books and articles. He failed to take account of and should have sought evidence on how practitioners actually advise clients. A pertinent factor is what view of the law practitioners adopted in advising clients in adding property. HMRC had no difficulty in obtaining evidence for this tribunal on this matter, so they could have obtained it before making the decision. In failing to obtain that evidence and to take it into account, Mr Ryder clearly failed to take account of a critical consideration.

(c) No reasonable officer could have come to the conclusion Mr Ryder came to on the basis of the factors he did take into account. He gave undue weight to the lack of challenges to the stated view given that (a) he had no idea how common it was for the situation to arise in practice, and (b) there was no reason to think cases where taxpayers took a contrary view to the stated view would come to the attention of HMRC. He agreed that in many situations, there would be no account filed. Even if an account was filed, although there were procedures for selecting accounts for compliance checks, there was no guarantee that any of the small number of cases where this issue might arise would be selected for compliance check.

(d) It is simply impossible to conclude that he would have inevitably come to the same conclusion if he had not made all of those errors. Whether he would have come to the same conclusion is not a matter for Mr Ryder to determine. It is a matter for the court to determine objectively.

105. Mr Baldry replied as follows:

(1) There was a decision in writing - the notice itself.

(2) The Court of Appeal held in Rowe, 2017 EWCA Civ 2105 at 226 to 229 that for an HMRC officer to follow HMRC guidance in reaching a decision he is required to reach is not a fetter on his discretion in a judicial review sense.

(3) Mr Ryder did not otherwise err in law: (i) HMRC did invite the taxpayer to put forward their response to the proposed notice and Mr Ryder did take into account their position as set out in their letters. In any event, their view of the law was set out in the IHT100 and in light of that it would not have made any difference to the decision, (ii) as regards Mr Ryder's view that the term "generally" does not apply to the adopted test, from his overall evidence it is clear that he was taking a broad view in terms of assessing whether mostly this view was adopted in practice. That is consistent with his conclusions: he recognised that the stated view was not universally accepted, but nevertheless having looked at the practice, he concludes that the provisions of s 255 were satisfied.

(4) Mr Ryder did not fail to take into account a relevant factor. Looking at the various statements in the books and the commentaries was precisely what enabled Mr Ryder to assess what people were being advised to do in practice and so what view the taxpayers in practice adopted. The advice in these books is as good as evidence as there can be on what taxpayers were doing in practice.

(5) Mr Ryder's decision was reasonably arrived at. It was entirely appropriate for him to take account of the absence of challenges. It is plainly a relevant factor albeit there is a question for him as to how much weight to put on to the relevant evidence in light of HMRC's processes of which he was well aware. The fact that if taxpayers had been adding property they would have notified HMRC just means that now, with hindsight, we can see that the view Mr Ryder took was completely reasonable.

(6) Mr Ewart was not really able to suggest any particular aspect of the case which might possibly lead to a different conclusion. Even if the tribunal did think that there had been a possibility of an error of law, the decision inevitably would have been the same.

Decision that Mr Ryder's decision was not reasonably arrived at

106. I have concluded that Mr Ryder's decision was not reasonably arrived at:

(1) In summary, (a) Mr Ryder appears to have made his decision that s 255 applies on the basis that the adopted test was satisfied (i) as he stated, on the assumption that the term generally did not apply to that test, (ii) taking account only of (A) that there had been only two challenges to the stated view that HMRC were aware of - one of which was the Barclays case, and (B) his interpretation of the commentaries in the textbooks, and (b) he approached this by asking "how do people behave, what do they do?...In terms of: do they act in a way which is compliant with HMRC's view or do they not" and appeared to consider that it was immaterial to the test whether people agreed with the stated view or simply took it into account in deciding on a course of action. He thought the test could be either of those things but that it does not make any difference to whether the stated view is "adopted in practice".

(2) I accept that (a), on the basis of the decision in Rowe, Mr Ryder did not improperly fetter his discretion by considering that he had no choice but to follow the official HMRC stance set out in the stated view, (b) it was not unreasonable to take the view that the appellant paid the disputed tax on the basis of the stated view on the basis of the correspondence.

(3) However, Mr Ryder erred in law (a) in taking the view that the adopted test is not qualified by the term "generally". As he did not have the correct test in mind he cannot have considered what was required for the stated view to be regarded as "generally" adopted in practice, and (b) in misinterpreting the term "adopted in practice" as not requiring the form of positive action and acceptance of the stated view as set out above.

(4) Whilst the absence of challenges and what was said in the books/articles are pertinent matters, (a) in order to assess what actually happens in practice, Mr Ryder failed to take account of and should have given consideration to/sought evidence on how practitioners advised clients in practice and what taxpayers did in practice, and (b) in deciding what weight to give to the lack of challenges, Mr Ryder failed to give any consideration at all to the factors affecting whether challenges were likely to occur and/or come to HMRC's attention. His evidence was that he did not have experience of this issue arising and he did not take steps to investigate what happened in practice other than looking at the textbooks. It is not possible to conclude that had Mr Ryder taken these matters into account he would have reached the same conclusion - and in any event, as noted, he applied the wrong legal test.

PART D - CONCLUSION AND RIGHT TO APPLY FOR PERMISSION TO APPEAL

107. For all the reasons set out above, the appeal is allowed.

108. This document contains full findings of fact and reasons for the 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.

HARRIET MORGAN

TRIBUNAL JUDGE

Release date: 28^th APRIL 2025

First-tier Tribunal (Tax)