
Case Number: TC09706
CAPITAL GAINS TAX – whether the appellants have made a valid claim for relief from CGT on the disposal of shares made by Trusts of which the appellants were settlors and beneficiaries under article 14(4) of UK/New Zealand double tax agreement – yes – whether the appellants are entitled to relief – no
TC/2019/05088; TC/2019/05090;
TC/2019/05108; TC/2019/05109;
TC/2019/05111
Judgment date: 1 December 2025
Before
TRIBUNAL JUDGE HARRIET MORGAN
TRIBUNAL MEMBER SONIA GABLE
Between
1.TERENCE MURPHY
2. ANGUS MACSWEEN
3. DAVID HARRIS
4. MATTHEW MURPHY
5. CHERYL MURPHY
AS EXECUTRIX OF EDWARD ERNEST MURPHY
Appellants
and
THE COMMISSIONERS FOR HIS MAJESTY’S REVENUE AND CUSTOMS
Respondents
Representation:
For the Appellant: Mr Quinlan Windle and Ms Laura Ruxandu, of counsel
For the Respondents: Mr Christopher Stone KC and Harry Sheehan, of counsel, instructed by the General Counsel and Solicitor to HM Revenue and Customs (“HMRC”)
DECISION
Part A - Overview
Each of the appellants was the settlor of a trust (together “the Trusts”) which in the tax year 2000/01 or 2001/02 was involved in tax motivated planning which was intended to allow them to dispose of shares held in the Trust (“the disposal”) without incurring a UK capital gains tax (“CGT”) liability for the trustees or the settlors. The appellants have appealed against “discovery assessments” issued to each of them by HMRC (pursuant to s 29 of the Taxes Management Act 1970 (“TMA”)) in which HMRC have sought to impose CGT on the gains realised on the disposals. HMRC issued the assessments (1) on 19 October 2007 to Mr Terence Murphy (“TM”), Mr Angus MacSween (“AM”), Mr Matthew Murphy (“MM”) and Mr Edward Ernest Murphy (“EM”) (together “the Murphys”), and (2) on 6 January 2004, to Mr David Harris (“DH”). DH disputes that the assessment issued to him meets the conditions to be validly issued under s 29 TMA. That issue is addressed in Part B.
The Murphys and DH were advised on the planning by Mr Terrance Morris and Mr Graham Blower of Lansburys International Limited (“Lansburys”). They devised the plan and liaised on its implementation with Mr Andrew Paul, a long term resident of Hong Kong, and Mr Gordon Richards, a chartered accountant and resident of New Zealand (“NZ”). Mr Morris, Mr Paul and Mr Richards all gave evidence for the appellants.
Under the planning, trustees who were resident in NZ were put in place as trustees of each of the Trusts for a short period to make the disposals and then replaced with UK trustees solely to enable the appellants to avoid CGT in respect of the gains for which they would otherwise be liable on the disposals. This is known as “round the world” planning (“RTW planning”) and has been considered in a number of cases. It hinges on the taxpayers being able successfully to claim relief from CGT which would otherwise arise under the terms of an applicable double tax agreement, in this case, that in place between the United Kingdom (“UK”) and NZ (“the treaty”). HMRC dispute (1) in the first place, that a valid claim for relief from CGT was made in respect of any of the disposals made by the Trusts; it was common ground that relief has to be claimed, and (2) in any event that the conditions for relief to apply are met.
Overview of the facts
In summary, the relevant facts as regards the Murphy Trusts are as follows:
The Murphys were all involved in a family business and had indirect interests in Aerohawk Group Limited (“Aerohawk”) which they each settled into a trust (together “the Murphy Trusts”): (a) TM was the settlor of the Terry Murphy Family settlement of 6 December 1996 (“the TM Trust”), (b) AM was the settlor of the Angus McSween Family Settlement of 6 December 1996 (“the AM Trust”), (c) MM was the settlor of the Matthew Murphy Family settlement of 6 December 1996 (“the MM Trust”), and (d) EM was the settlor of the Edward Murphy 1998 Family Settlement of 26 May 1998 (“the EM Trust”).
At all material times, the Murphys were clients of Lansburys and were advised by Mr Morris and Mr Blower. The predecessor to Lansburys was established in the early 1980s. Mr Morris had known the Murphys since the early 1990s. The Murphy Trusts were established with the intention of using tax planning arrangements to avoid CGT on the eventual sale of the shares in Aerohawk. That particular planning did not take place due to a change in legislation. Mr Morris explained that (a) in 1999 the Murphys sold the business/entities underlying Aerohawk to Enron, and (b) after that sale, Aerohawk was a cash rich company without an active business, and (c) sometime after the sale, Mr Morris showed TM “how the settlements could sell the underlying companies without incurring CGT” and explained the RTW planning “only in broad lines” as TM “had complete trust in [Mr Morris] and Lansburys”. He thought his clients did not completely understand the planning and the steps that had to be taken by the trustees at various times but were happy to put their trust in Lansburys.
The Trusts all operated in the same way:
The beneficiaries of each of the Murphy Trusts included the settlor of that trust. They were subject to Jersey law. In each case the deed under which each trust was settled (a) granted the settlor power to remove one or more of the trustees and to appoint one or more persons or companies to be a trustee in place of any trustee who was removed, and (b) granted any trustee the right to resign after giving no less than fourteen days’ notice.
The TM, AM and MM Trusts were originally constituted with a trustee resident in Monaco: Republic International Trust Company Limited (“the Monaco Trustees”). The EM Trust was originally constituted with Altech Limited (“Altech”) as trustee, a company incorporated in Liberia, which was wholly owned by Mr Paul, who was also its sole director.
Each of the Murphy Trusts held shares in a different company incorporated in the British Virgin Islands (“BVI”) which each indirectly owned shares in Aerohawk: Cowbit Holdings Limited for the TM Trust; Neythis Limited for the AM Trust; Superb Returns Limited for the EM Trust; and Best Cambridge Limited for the MM Trust.
Mr Morris said that from the outset the plan was that (a) the Murphy Trusts would sell their shares in the BVI companies before 1 April 2002, (b) the sale would be made by trustees resident in NZ to a buyer unconnected with the Murphys, and (c) before the end of the 2001/02 tax year, the trustees would be replaced by trustees resident in the UK.
By Deeds of Removal and Appointment of Trustees dated 1 March 2002, the settlors of the Murphy Trusts each exercised their power to remove the existing trustees and replace them with Oxford Trustees Limited (“Oxford Trustees”), who were resident in NZ. Mr Richards and his wife were directors of Oxford Trustees. He incorporated this company in July 2000 and was the sole owner of it. This entity and the company incorporated to act as trustee for DH’s trust (see below) provided trustee services for a number of Lansburys’ clients before the RTW planning in these cases was implemented and until the 2005/06 tax year. Four minutes of meetings of the directors of Oxford Trustees dated 1 March 2002 record that the Deeds in respect of each of the Murphy Trusts were tabled and considered. Oxford Trustees accepted the appointment as the new trustee of the Murphy Trusts.
Four minutes of meetings of the directors of Oxford Trustees dated 20 March 2002 record that (a) the trustees considered offers made by No Boundary Limited (“No Boundary”) for the purchase of the shares held by the Murphy Trusts in the BVI companies for consideration in cash, and (b) it was agreed that the “contract presented to the meeting for the sale of the shares be executed”. Mr Paul was the sole director of No Boundary. Pursuant to four sale and purchase agreements between No Boundary and Oxford Trustees, each of which is dated 20 March 2002, No Boundary purchased (a) 100 shares in Best Cambridge Limited for £339,090 from the MM Trust, (b) 8,250 shares in Superb Returns Limited for £1,276,985 from the EM Trust, (c) 100 shares in Neythis Limited for £612,072 from the AM Trust, and (d) 100 shares in Cowbit Holdings Limited for £423,835 from the TM Trust. In total, the sale and purchase agreements provided for the payment of £2,651,901 in exchange for the shares disposed of.
On 21 March 2002, by letter to the settlor of each Murphy Trust of that date, Oxford Trustees gave notice of their resignation to take effect on 3 April 2002.
By deeds of 3 April 2002, the settlor of each Murphy Trust appointed Funcode Limited (“Funcode”) as trustee and removed Oxford Trustees. Funcode was resident in the UK by incorporation. The sole director of Funcode was Mr Paul.
On 11 July 2002, the settlor of each Murphy Trust removed Funcode as a trustee. TM appointed himself and John Edward Murphy as trustees of the TM Trust. AM appointed himself and TM as trustees of the TM Trust. EM appointed himself and MM as trustees of the EM Trust. MM appointed himself and John Edward Murphy as trustees of the MM Trust. The Deeds were sent to Mr Paul by Mr Graham Blower on 13 June 2002 and were sent to the settlors of the Murphy Trusts by TM on 20 June 2002.
As regards DH and his trust:
DH was the settlor of the DJ Harris settlement of 6 March 1991 (“the Harris Trust”). On 28 November 2002, the settlement had trustees resident in Monaco: the Monaco Trustees. The settled property included 249 shares in GED Technology Group Ltd.
On 28 November 2000, Oxford Trustees and Avon Trustee Services Limited (“Avon Trustees”), were appointed trustees in addition to (and not instead of) the Monaco Trustees. The appointment was effected by DH. Avon Trustees was set up by Mr Richards at the same time as he set up Oxford Trustees and on the same basis as set out above. We refer to the Oxford Trustees and the Avon Trustees as “the NZ Trustees”. The deed of appointment provided that the Monaco Trustees were to hold the trust property “to the order of” the Monaco Trustees and the NZ Trustees.
The appellants submitted that by way of an agreement for sale on 30 January 2001 the trustees of the Harris Trust disposed of the 249 shares in GED Technology Group Ltd for consideration of £840,000 and of 45,427 shares in SITEC. There is a lack of documentation relating to this and HMRC put the appellants to proof of demonstrating this took place. This is discussed in Part C.
On 20 March 2001, the three existing trustees of the Harris Trust were removed by DH and were replaced by a UK resident trustee, Allglory Limited (“Allglory”). Allglory was UK resident by incorporation. The directors of Allglory were long-time residents of Monaco.
In March 2001 Allglory appointed the 45,427 SITEC shares to the Harris Trust and in June 2002 the shares were appointed to DH.
Overview of the issues
In summary, NZ was chosen as the location for the Trusts’ administration whilst the disposals were made because the treaty contains a provision which the appellants consider applies to exempt the capital gains arising on the disposals from CGT as a result of the steps taken under the planning. The treaty was implemented by the Double Taxation Relief (Taxes on Income) (New Zealand) Order 1984 under powers granted in s 497 Income and Corporation Taxes Act 1970 (and subsequently reenacted in s 788 of the Income and Corporation Taxes Act 1988 (“ICTA”)).
Article 1 of the treaty provides that it shall apply to persons who are residents of one or both of the UK and NZ. Article 2 provides that the taxes that are the subject of the convention include CGT.
The appellants rely on article 14(4) as applying to relieve them from CGT on the gains arising on the disposals:
“(1) Income or gains derived by a resident of a Contracting State from the alienation of immovable property referred to in Article 7 and situated in the other Contracting State or from the alienation of shares in a company the assets of which consist wholly or principally of such property may be taxed in that other State.
(2) Income or gains from the alienation of movable property forming part of the business property of a permanent establishment which an enterprise of a Contracting State has in the other Contracting State or of movable property pertaining to a fixed base available to a resident of a Contracting State in the other Contracting State for the purpose of performing independent personal services, including such income or gains from the alienation of such a permanent establishment (alone or with the whole enterprise) or of such fixed base, may be taxed in that other State.
(3) Income or gains derived by an enterprise of a Contracting State from the alienation of ships or aircraft operated in international traffic or movable property pertaining to the operation of such ships or aircraft, shall be taxable only in that State.
(4) Income or gains from the alienation of any property other than that referred to in paragraphs (1), (2) and (3) of this Article, shall be taxable in the Contracting State of which the alienator is a resident.” (Emphasis added.)
Article 4 of the treaty defines “resident of a Contracting State” as follows:
“(1) For the purposes of this Convention, the term “resident of a Contracting State” means, as the context requires:
(a) any person who is resident in the United Kingdom for the purposes of United Kingdom tax; or
(b) any person who is resident in New Zealand for the purposes of New Zealand tax.
…
(3) Where by reason of the provisions of paragraph (1) of this Article a person other than an individual is a resident of both Contracting States, then it shall be deemed to be a resident of the State in which its place of effective management is situated.
The test in article 4(3) is commonly referred to as the “tie breaker”. It was common ground that, for the purposes of article 14(4), the Contracting State of which the alienator is a resident is to be determined under the provisions of article 4 and (a) if the alienator is resident only in one Contracting State under the test in article 4(1), that determines the alienator’s residence for the purposes of article 14(4) but (b) if the alienator is a resident of both Contracting States under article 4(1), the “tie breaker” applies to deem the alienator to be a resident only of the Contracting State in which its place of effective management (“POEM”) is situated. We note that in neither article 4 not article 14 is there any reference to the time at or, period during which, the alienator’s residence under domestic law is to be determined or of the period over which its POEM is to be assessed under article 4(3).
The appellants consider that the gains realised on the disposals are relieved from CGT under article 14(4) of the treaty on the basis that the Trusts, as the “alienators” of the shares on which the gains arose, were residents only of NZ at the relevant time under article 4(1). There is no dispute that (1) for the period that the trustees of the relevant Trusts were the NZ Trustees, the Trusts were resident in NZ for the purposes of NZ tax and were not resident in the UK for the purposes of UK tax, (2) within the same UK tax year as that in which the disposals took place (2000/01, as regards the Harris Trust and 2001/02, as regards the Murphy Trusts), upon the appointment of the UK Trustees, the Trusts were resident in the UK for the purposes of UK tax and from that time onwards were not resident in NZ for the purposes of NZ tax, (3) as explained in further detail below, as the trustees were resident in the UK for part of the relevant tax year, they were liable to CGT on chargeable gains accruing to them in that year under s 2 TCGA, and (4) the Trusts were not subject to any tax charge in NZ on the capital gains arising on the disposals.
The appellants’ view is that (1) article 4 of the treaty is concerned with “factual residence”, and (2) the right to taxation in article 14(4) is allocated to the Contracting State in which the alienator is resident for treaty purposes, as determined wholly under article 4(1), by reference to the position at the time of alienation, and (3) accordingly, article 14(4) applies with the effect that the gains arising on the disposals are taxable only in NZ and not in the UK.
In HMRC’s view, (1) the fact that, under the test in article 4(1), the Trusts were resident in both the UK and NZ at some point in the UK tax years in which the disposals took place, means that it is necessary to apply the tie-breaker in article 4(3) to determine where the Trusts were resident for the purposes of article 14(4), (2) under the tie-breaker in article 4(3), POEM of each of the Trusts was in the UK whilst the NZ Trustees were in place so that the Trusts were residents only of the UK for the purposes of article 14(4), and (3) on that basis, article 14(4) applies with the effect that the gains arising on the disposals are taxable in the UK. HMRC argue that the appellants’ interpretation of article 14 is not tenable, in particular, in light of the purpose of provisions such as those in article 14(4) as explained by the Court of Appeal in HM Revenue and Customs v Smallwood & Another [2010] EWCA Civ 778, [2010] STC 2045 (“Smallwood CA”).
Smallwood CA concerns RTW planning of the kind implemented here except that the trustees appointed to make the disposal of valuable assets held in the taxpayers’ trusts were resident in Mauritius and accordingly the taxpayers relied on article 13(4) of the UK/Mauritius double tax agreement (“the Mauritius treaty) to relieve the gains arising on the disposals. Article 13(4) contains very similar provisions to those in article 14(4) of the treaty. We refer to the provisions in those articles as “the gains provisions”. Article 4(1) of the Mauritius treaty contains a different definition of “resident of a Contracting State”. As accords with the Model Tax Convention in place at the time, the Mauritius treaty defines this as “any person who, under the law of that State, is liable to taxation therein by reason of his domicile, residence, place of management or any other criterion of a similar nature” (emphasis added). The Court of Appeal held that the Special Commissioners (“the Commissioners”) had not erred in their decision (see [2008] UKSPC 00669 [2008] STC (SCD) 629 (“Smallwood SpC”)) that the planning failed, and CGT was due on the relevant gains. It was held that no relief was available under article 13(4) of the Mauritius treaty on the basis that the alienators of the shares, the trusts/trustees, were residents of the UK at the relevant time. The analysis was that they were resident in both Contracting States for the purposes of article 4(1) such that their treaty residence had be determined under the tie-breaker and, on the correct interpretation of the POEM test (as to which there was much debate) the POEM of the trusts was in the UK.
A similar approach was taken in the later decision in Haworth & Ors v Revenue and Customs [2024] UKUT 58 (TCC), [2024] STC 436 (“Haworth UT”). The Upper Tribunal upheld the decision of the tribunal ([2019] UKFTT 149 (TC) (“Haworth FTT”)) that the RTW planning, under which the taxpayers also sought relief under article 13(4) of the Mauritius treaty, failed. Since the hearing, the Court of Appeal has also upheld that decision in Haworth & Ors v Commissioners for His Majesty's Revenue and Customs [2025] EWCA Civ 822, [2025] WLR(D) 342 (“Haworth CA”). The parties made written submissions on the effect of the judgment in that case which have been taken into account in this decision.
HMRC argued that the analysis set out in Smallwood CA applies in this case also to give the same result, namely, that the tie-breaker is in point to determine the residence of the Trusts and, applying the POEM test on the approach set out in that case, their POEM was in the UK. The appellants submitted that (1) Smallwood gives no material guidance on the correct analysis in this case on the basis that the different definition of “resident of a Contracting State” in the Mauritius treaty is highly material, (2) if the tie breaker is in point, POEM of each of the Trusts was in NZ and not in the UK. In their view, in the treaty POEM means “practical day to day management, irrespective of where the overriding control is exercised” as this is the interpretation set out in an observation by NZ on the OECD Commentary on the 1977 Model Tax Convention (as the Model in place at the relevant time). Alternatively, if this is found not to be correct, the second most sensible construction of the POEM test is the central management and control test (“CMC”) set out in De Beers Consolidated Mines Ltd v Howe [1906] AC 455 (“De Beers”) and further explained in Wood v Holden [2006] EWCA Civ 26; [2006] 1 WLR 1393 (“Wood v Holden”) given that this test was laid down by the House of Lords at a time when litigants in NZ had a right of appeal to the Privy Council. HMRC did not agree that either of these approaches can be taken. In their view, the correct approach is that taken in Smallwood SpC, as approved in Smallwood CA and, as has now been endorsed in Haworth CA. The Court of Appeal decided in Haworth CA that the tribunal took the correct approach in that case in interpreting the decision in Smallwood CA as requiring a broader test than the CMC test to be applied for determining POEM under article 4(3) of the Mauritius treaty.
The thinking behind the planning from a UK tax law perspective is as follows:
For CGT purposes, the Trusts are “settlements” under TCGA on the basis that they comprise “settled property”, as defined under s 68, as “any property held in trust” other than property held (a) by a person as nominee for another person, or (b) as trustee for “another person absolutely entitled as against the trustee”; namely, where the other person has the exclusive right to direct how that property shall be dealt with, subject only to satisfying any outstanding charge, lien or other right of the trustees to resort to the property for payment of duty, taxes, costs or other outgoings. There is no further definition of a “trust” in TCGA but the meaning under English law is well known. Broadly, under English law a trust is a legal relationship created when a person (a settlor) puts assets into the ownership of one or more persons (the trustee(s)) to be held and managed by the trustee for the benefit of another person or persons (the beneficiary) (or for a specified purpose) usually subject to terms set out in trust deed and subject to duties imposed by law.
The Trusts, as “settlements”, are treated as follows under TCGA as regards the tax year in which the gains on the disposals arose:
The trustees of the Trusts from time to time are treated as “a single and continuing body of persons” (a “deemed trustee body”) under s 69. This provides that:
“the trustees of a settlement shall for the purposes of the TCGA [i] be treated as being a single and continuing body of persons (distinct from the persons who may from time to time be the trustees) and [ii] that body shall be treated as being resident and ordinarily resident in the United Kingdom unless the general administration of the trusts is ordinarily carried on outside the United Kingdom and the trustees or a majority of them for the time being are not resident or not ordinarily resident in the United Kingdom.” (Emphasis and numbering added.)
Under s 69[ii], as regards each Trust, each deemed trustee body is treated as being resident in the UK from the time of the UK trustees’ appointment on the basis that, from that time, the UK trustees were all resident in the UK.
The trustees of each Trust are, as a deemed trustee body, chargeable to CGT in respect of gains accruing to them in the relevant tax year under s 2 TCGA. This provides that:
“...a person shall be chargeable to capital gains tax in respect of chargeable gains accruing to him in a year of assessment during any part of which he is resident in the United Kingdom...” (Emphasis added.)
The trustees of each Trust are accountable for CGT arising on the gains under s 65. This provides that: CGT “chargeable in respect of chargeable gains accruing to the trustees of a settlement...may be assessed and charged on and in the name of any one or more of the relevant trustees...” (under s 65(1)) (subject to specific provisions as to which trustees are liable when a “settlement” migrates (under s 65(3)).
Subject to certain exceptions and express provision to the contrary, “chargeable gains accruing to the trustees of a settlement...and [CGT] chargeable on or in the name of such trustees.., shall not be regarded for the purposes of this Act as accruing to, or chargeable on, any other person, nor shall any trustee...be regarded for the purposes of this Act as an individual”.
The intention was that the gains accruing to the trustees of the Trusts would not be taxable in the tax year in which they arose under s 86 TCGA and would be relieved from the tax that would otherwise be due under s 77 TCGA under article 14(4) of the treaty. Broadly, both of s 86 and s 77 subject a “settlor” of a “settlement” to CGT on chargeable gains which accrue to the trustees of the “settlement” from the disposal of certain property in a particular tax year where, as is the case here, at any time during the year the “settlor” has an “interest” in the “settlement”.
Section 77 only applies where, the “settlor” and the trustees are resident in the UK “during any part of the year” (or ordinarily resident in the UK during the year) (under s 77(7)). On the other hand, one of the conditions for s 86 to apply is that the trustees are not resident or ordinarily resident in the UK during any part of the year (under s 86 (2)).
Hence, it seems to have been considered to be essential that the trustees of the Trusts (in each case, as a deemed trustee body) were resident in the UK at some point during the relevant tax year. As noted, it is common ground that the Trusts became resident in the UK part way through the relevant tax years due to the appointment of the UK trustees.
If the conditions for s 77 to apply are met
sub-s (1) provides that it applies where:
“(a) in a year of assessment chargeable gains accrue to the trustees of a settlement from the disposal of any or all of the settled property,
(b) after making any deductions provided for by section 2(2) in respect of disposals of the settled property there remains an amount on which the trustees would, disregarding section 3 (and apart from this section), be chargeable to tax for the year in respect of those gains, and
(c) at any time during the year the settlor has an interest in the settlement,”
sub-s (2) provides that “the trustees shall not be chargeable to tax in respect of the gains concerned but instead chargeable gains of an amount equal to that referred to in subsection (1)(b) above shall be treated as accruing to the settlor in the year”.
If the appellants are correct that article 14(4) applies, it is given effect as follows:
Section 277 TCGA provides that s 788 ICTA, which gives relief from UK income tax (and corporation tax) where applicable under a double tax agreement, applies to arrangements for the purposes of giving relief from CGT on capital gains in the same manner as it applies to arrangements for the purposes of giving relief from income tax.
Section 788(3) ICTA provides that tax relief arrangements in double tax agreements “shall, notwithstanding anything in any enactment, have effect in relation to income tax and corporation tax in so far as they provide– (a) for relief from income tax, or from corporation tax in respect of income or chargeable gains…”.
Section 788(6) ICTA provides that “except in the case of a claim for an allowance by way of credit…a claim for relief under subsection (3)(a) above shall be made to the Board.”
It was common ground, therefore, that, pursuant to s 277 TCGA and s 788(6) ICTA, on the making of a valid claim for relief from CGT under article 14(4) (a “claim for treaty relief”), if the appellants’ analysis is correct article 14(4) would apply to prevent there being any such “remaining amount of the chargeable gains...” within the meaning of s 77(1)(b) TCGA so that there is nothing which s 77(1) TCGA can treat as accruing to the appellants, as settlors of the Trusts. The appellants contend that valid claims for relief were made in the correspondence they had with HMRC and that, even if they were not, HMRC are estopped by convention from contending that no valid claim for relief under the treaty has been made. HMRC dispute both of these contentions.
In summary we have concluded as follows:
For the reasons set out in Part B, HMRC did not issue a valid discovery assessment in respect of DH on the basis that they have not demonstrated that the requirements set out in s 29(5) TMA were satisfied.
For the reasons set out in Part B, the appellants made valid claims for treaty relief from CGT on the gains arising on the disposals. If we are wrong on that, HMRC are estopped by convention from contending that valid claims for treaty relief have not been made.
For the reasons set out in Part C, (a) whether the Trusts were residents of NZ or the UK for the purposes of article 14(4) of the treaty is to be determined under the tie-breaker in article 4(3) of the treaty and, (b) the POEM of the Trusts is to be determined under the test set out in Smallwood CA and Haworth CA, and (c) the POEM of the Trusts was in the UK when the NZ Trustees were in place. On that basis, article 14(4) does not apply to relieve the appellants from the CGT charge arising in respect of the disposals.
Part B – Discovery issue and claims issue
Law – discovery issue
The assessments were issued under the power granted by s 29 TMA which at the relevant time provided as follows:
“(1) If an officer of the Board or the Board discover, as regards any person (the taxpayer) and a year of assessment -
(a) that any income which ought to have been assessed to income tax, or chargeable gains which ought to have been assessed to capital gains tax have not been assessed, or
(b) that an assessment to tax is or has become insufficient, or
(c) that any relief which has been given is or has become excessive,
the officer or, as the case may be, the Board may, subject to subsections (2) and (3) below, make an assessment in the amount, or the further amount, which ought in his or their opinion to be charged in order to make good to the Crown the loss of tax.
…
(3) Where the taxpayer has made and delivered a return under section 8 or 8A of this Act in respect of the relevant year of assessment, he shall not be assessed under subsection (1) above -
(a) in respect of the year of assessment mentioned in that subsection; and
(b) in the same capacity as that in which he made and delivered the return, unless one of the two conditions mentioned below is fulfilled.
(4) The first condition is that the situation mentioned in subsection (1) above is attributable to fraudulent or negligent conduct on the part of the taxpayer or a person acting on his behalf.
(5) The second condition is that at the time when an officer of the Board -
(a) ceased to be entitled to give notice of his intention to enquire into the taxpayer's return under section 8 or 8A of this Act in respect of the relevant year of assessment; or
(b) informed the taxpayer that he had completed his enquiries into that return,
the officer could not have been reasonably expected, on the basis of the information made available to him before that time, to be aware of the situation mentioned in subsection (1) above.
(6) For the purposes of subsection (5) above, information is made available to an officer of the Board if -
(a) it is contained in the taxpayer's return under section 8 or 8A of this Act in respect of the relevant year of assessment (the return), or in any accounts, statements or documents accompanying the return;
(b) it is contained in any claim made as regards the relevant year of assessment by the taxpayer acting in the same capacity as that in which he made the return, or in any accounts, statements or documents accompanying any such claim;
(c) it is contained in any documents, accounts or particulars which, for the purposes of any enquires into the return or any such claim by an officer of the Board, are produced or furnished by the taxpayer to the officer, whether in pursuance of a notice under section 19A of this Act or otherwise; or
(d) it is information the existence of which, and the relevance of which as regards the situation mentioned in subsection (1) above -
(i) could reasonably be expected to be inferred by an officer of the Board from information falling within paragraphs (a) to (c) above; or
(ii) are notified in writing by the taxpayer to an officer of the Board.
(7) In subsection (6) above -
(a) any reference to the taxpayer’s return under section 8 or 8A of this Act in respect of the relevant year of assessment includes -
(i) a reference to any return of his under that section for either of the two immediately preceding years of assessments; and
(ii) where the return is under section 8 and the taxpayer carries on a trade, profession or business in partnership, a reference to any partnership return with respect to the partnership for the relevant year of assessment or either of those periods; and
(b) any reference in paragraphs (b) to (d) to the taxpayer includes a reference to a person acting on his behalf…”
It was common ground that:
It is for HMRC to satisfy the tribunal that the relevant conditions have been met for a discovery assessment to have been validly made (see Burgess v HMRC [2016] STC 579). DH made and delivered a return under s 8 TMA. HMRC must therefore show that a discovery was made and that one of the conditions in s 29(4) or (5) was met. They did not argue that s 29(4) was in point but rather that the condition in s 29(5) is met.
The correct test for whether a discovery has been made is as set out in Charlton & ors v HMRC [2013] STC 866 at [37]:
“In our judgment, no new information, of fact or law, is required for there to be a discovery. All that is required is that it has newly appeared to an officer, acting honestly and reasonably, that there is an insufficiency in an assessment. That can be for any reason, including a change of view, change of opinion, or correction of an oversight. The requirement for newness does not relate to the reason for the conclusion reached by the officer, but to the conclusion itself.”
This comment was approved by the Supreme Court in HMRC v Tooth [2021] UKSC 17, [2021] 1 WLR 2811, but the point was not in dispute (see the decision at [64] and [65]). A similar formulation of the test appears in Cenlon Finance Co Ltd v Ellwood [1962] AC 782 at 794, which was endorsed by the Supreme Court in Tooth at [73].
The discovery must be something more than a suspicion of an insufficiency, but need not go as far as a conclusion that an insufficiency of tax is more probable than not: Anderson v HMRC [2018] UKUT 159 (TCC), [2018] 4 WLR 90. The same discovery that is made by one officer can be made for a second time by a different officer (see Tooth at [78]). There is no possibility of a discovery becoming stale once made (see Tooth at [76]).
The appellants referred to the decision in Strategic Branding v HMRC [2021] UKFTT 474 (TC) at [182] to [193] as showing the level of evidence required to demonstrate that the discovery test is satisfied. The appellants noted, in particular, that in that case the tribunal rejected the proposition that the mere issue of an assessment was sufficient to demonstrate that a discovery was made but, as the relevant officers were not available to give evidence, relied on the extensive documentary evidence recording the officers’ thought processes. In their view, HMRC have failed to provide sufficient evidence to show that there was a discovery. HMRC dispute this as set out below.
It was also common ground that the burden of establishing that the s 29(5) TMA condition is satisfied is upon HMRC. HMRC referred to the comments in the caselaw on those provisions (1) that the burden of proof is “unlikely to be of much practical significance because the nature of the enquiry is an objective one and the return and accompanying documents which have been submitted to HMRC should always be available”: HMRC v Household Estate Agents [2007] EWCH 1684 (Ch), [2008] STC 2045 at [48], and (2) if the level of disclosure is to prevent HMRC from issuing an assessment, it must be such that the hypothetical officer must be aware of the actual insufficiency, not merely such information as would prompt the officer to ask questions or open an enquiry: Beagles v HMRC [2018] UKUT 380 (TCC), [2019] STC 54 at [100(5)]. HMRC also submitted that s 29(6) TMA contains an exhaustive list of the information that is made available for the purposes of s 29(5) and in order to be relevant for the purposes of s 29(5), the information has to have been provided in the appellants’ tax returns. Information contained in the trustees’ tax returns and any information provided to HMRC for the purposes of an enquiry into the trustees’ return is not relevant: Trustees of the Bessie Taube Discretionary Settlement Trust and ors v HMRC [2010] UKFTT 473 (TC) at [74] to [77].
The appellants submitted that (1) this is one of the rare cases where the burden of proof is relevant, and (2) as explained below, it appears that information existed and its existence and relevance could reasonably be expected to have been inferred by an officer of the Board from information provided by DH.
Law and submissions - Claim Issue
Legislative requirements
Other than as set out above, neither TCGA nor ICTA provided any details about how a claim for relief under article 14(4) of the treaty should be made or what it should include. It was common ground that, therefore, the general provisions in s 42 TMA apply. This provides as follows:
“(1) Where any provision of the Taxes Acts provides for relief to be given, or any other thing to be done, on the making of a claim, this section shall, unless otherwise provided, have effect in relation to the claim.
(1A) Subject to subsection (3) below, a claim for a relief, an allowance or a repayment of tax shall be for an amount which is quantified at the time when the claim is made.
(2) Subject to subsections (3) and (3A) below, where notice has been given under section 8, 8A or 12AA of this Act, a claim shall not at any time be made otherwise than by being included in a return under that section if it could, at that or any subsequent time, be made by being so included.
…
(5) The reference in this section to a claim being included in a return include references to a claim being so included by virtue of an amendment of the return.
…
(9) Where a claim has been made (whether by being included in a return under section 8, 8A or 12AA of this Act or otherwise) and the claimant subsequently discovers that an error or mistake has been made in the claim, the claimant may make a supplementary claim within the time allowed for making the original claim.
…
(11) Schedule 1A to this Act shall apply as respects any claim or election which -
(a) is made otherwise than by being included in a return under section 8, 8A or 12AA of this Act,
…”
Paragraph 2 of schedule 1A TMA (which by s 42(11) TMA applies to claims made outside of a return) relevantly provided:
“(1) Subject to any provision in the Taxes Acts for a claim to be made to the Board, every claim shall be made to an officer of the Board.
…
(3) A claim shall be made in such form as the Board may determine.
(4) The form of claim shall provide for a declaration to the effect that all the particulars given in the form are correctly stated to the best of the information and belief of the person making the claim.
(5) The form of claim may require -
(a) a statement of the amount of tax which will be required to be discharged or repaid in order to give effect to the claim;
(b) such information as is reasonably required for the purpose of determining whether and, if so, the extent to which the claim is correct;
(bb) the delivery with the claim of such accounts, statements and documents, relating to information contained in the claim, as are reasonably required for the purpose mentioned in paragraph (b) above;
and
(c) any such particulars of assets acquired as may be required in a return by virtue of section 12 of this Act or paragraph 13 of Schedule 18 to the Finance Act 1998. …” (Emphasis added.)
It was common ground that as for all legislation, the provisions set out above must be construed purposively and applied to the facts viewed realistically (Hurstwood Properties (A) Ltd v Rossendale BC [2021] UKSC 16; [2022] AC 690). We note that at the time s 42 TMA applied to all claims for relief from income, corporation and CGT except those that had different claims regimes defined elsewhere in the tax legislation. It appears, therefore, it was intended to provide a simple way to meet requirements for making a claim, which HMRC have the power to supplement (under para 2(3) of schedule 1A TMA (and previously in s 42(5) TMA)).At the relevant time HMRC had not exercised that power as regards claims such for the relief under a treaty which the appellants seek to benefit from in these cases.
The parties referred to the case law in which the operation of these provisions has been considered albeit in a different context, in particular, the decisions of the House of Lords in Gallic Leasing Ltd. v Coburn (Inspector of Taxes) [1991] 1 WLR 1399 (“Gallic Leasing”) and of the Court of Appeal in R v Inland Revenue Commissioners ex parte Unilever [1996] STC 681 (“Unilever”). In Gallic Leasing and Unilever the House of Lords considered the operation of s 42 TMA in the context of respectively (1) claims for group relief for losses of one or more companies in a group against the profits of one or more companies in the same group, and (2) claims for relief for trading losses.
Overview of submissions on the claims issue
No specified claim for relief was made in any tax return submitted by the appellants. It appears that a tax return was submitted by the trustees of the Harris Trust for the tax year 2000/01 but it did not contain a claim for relief and no tax returns were submitted by the trustees of the Murphy Trusts for the 2001/02 tax year at all.
Section 9ZA TMA provided that a person could amend a personal or trustee return within twelve months of the filing date by notice to an officer of the Board. This meant that the trustees of the Harris Trust could amend the tax return for the 2000/01 tax year to include a claim for relief under the treaty until 31 January 2003. The TMA provided no further rules about what was required to amend a return but HMRC’s Self-Assessment Claims Manual says that if a claim to which s 42(2) applies is received after the return is filed but within the amendment window of twelve months from the statutory filing date, HMRC will treat this as an amendment to the return (at SACM3030).
Section 43(1) TMA provided that a claim for relief could be made no more than “five years after the 31st January next following the year of assessment to which it relates”. Alternatively, s 43A TMA permits consequential claims to be made within one year of the end of the year of assessment when an assessment was made (s43A(2)(a)).
The assessments for the Murphys related to the 2001/02 tax year and were issued in October 2007. The time limit to make a claim under s 43 was 31 January 2008 (being 5 years after 31 January next following the next year of assessment) and to make a consequential claim, under s 43A, 5 April 2009. The assessment which HMRC assert is a valid discovery assessment for Mr Harris related to the 2000/01 tax year and was issued on 6 January 2004. The time limit for making a claim under s 43 expired on 31 January 2007.
In summary, the appellants submitted that, having regard to the statutory requirements as interpreted in the case law:
As regards DH and the Harris Trust (a) the correspondence shows that, the 2000/01 return of the Harris Trust was amended, in accordance with HMRC’s approach to claims made within the period for amending a return (set out above), to include a claim for relief under the treaty, or (b) alternatively, that the trustees of the Harris Trust made a valid claim for relief under the treaty outside of the trust return after the time for amending that return had expired.
As regards the Murphys and the Murphy Trusts the correspondence shows that claims were made outside of a return, in particular a letter of 9 August 2007 and the notices of appeal made on 12 or 14 November 2007.
HMRC submitted that no valid claim was made in any of those cases, in DH’s case whether by amendment to his tax return for 200/01or otherwise. In particular:
The correspondence was between the appellants’ agents and specific officers of the Board, which does not suffice for a claim to be made to the Board itself (s 788 ICTA).
There is no quantification of any claim in any of the correspondence.
Any correspondence that was on behalf of the appellants, rather than the trustees, cannot amount to a valid claim under the treaty (such as the letter of 9 August 2007 and the appellants’ notices of appeal). A claim can only be made by the trustees.
The understanding of the appellants’ representative was that no claim needed to be made. A claim cannot have been made accidentally, in passing, without any intention to make a claim or understanding that the claim was required.
No document ever sent by the appellants was accompanied by the declaration of accuracy that is required by para 2(4) of schedule 1A TMA.
The notices of appeal are not a claim for relief, they are grounds of appeal which are fundamentally different. An officer of HMRC reading these letters of appeal may quite reasonably conclude that the taxpayers intended to challenge an assessment made by HMRC and not that they intended to make a claim for relief under the treaty. These letters were not intended to amount to claims for relief and are not drafted as such. They are made on behalf of the settlors and not the trustees and fail the requirements set out above. Similar comments apply in respect of the other documents the appellants rely on.
The appellants disputed all of HMRC’s points as set out below.
Disagreement on meaning of specific provisions in schedule 1A TMA
As noted, HMRC said that the effect of sub-para (4) is that any claim, whether to be made in a form prescribed by the Board or otherwise, must contain a declaration that all particulars given in the claim are correctly stated to the best of the information and belief of the person making the claim. However, in our view, as the appellant submitted it is clear, reading that provision in context, that it requires the Board, if it prescribes a form, to include in the form a requirement for that declaration. The immediately preceding provision in (3) sets out that a claim shall be made in such form as the Board may determine and then the provisions in (4) and (5) set out what any such form shall or may require. We note that this provision was in place when Gallic Leasing was decided and there was no suggestion in that case that where, as in that case, there was no prescribed form, there was any requirement of this declaration to be made for there to be a valid claim (in that case, for group relief).
The parties also disagreed over who a claim for relief needs to be made to. This hinges on how the requirement for a claim of this type to be notified to the Board may be satisfied. At the relevant time s 832 ICTA provided that “the Board” means the Commissioners of Inland Revenue. The appellants submitted that the relevant HMRC officers who were in correspondence with the appellants had actual or apparent authority to accept claims under s 788(6) on behalf of the Board:
In support of this view, they noted that with effect from 7 April 2005 (1) s 50 of the Commissioners of Revenue and Customs Act 2005 states that: “A reference to the Commissioners of the Inland Revenue shall be taken to mean a reference to the Commissioners for Revenue and Customs, HMRC” and (2) s 13(1) of that Act provides that: “An officer of Revenue and Customs may exercise any function of the Commissioners.” Mr Windle submitted that hence from that time a claim had to be made to HMRC and any officer of HMRC could exercise the function of receiving that claim.
The relevant provisions in ICTA were replaced by provisions in the Taxation (International and Other Provisions) Act 2010 and the explanatory notes make clear that the function of accepting such claims by an officer of HMRC was the accepted practice; it records that all functions affected by the change to s 788(6) were in fact exercised by officers of HMRC before that time.
Mr Stone said that this argument is not viable on the basis that (1) the provisions in para 2 of schedule 1A TMA specifically draw a distinction between a claim made to an officer of the Board and a claim to the Board, (2) changes to legislation made long after the tax years in question, cannot provide any assistance as to how the legislation in place in those tax years is to be interpreted, and (3) even if the Commissioners had a practice of delegating functions to officers, that does not mean that, in effect, it is a taxpayer’s choice to write to an officer. The taxpayer cannot force the delegation upon the Commissioners; it is for them to delegate the functions if they so desire. HMRC confirmed at the hearing that in their view this requirement may be satisfied simply by the relevant document being stated to be “to the Board”. They did not suggest that the intention is that the claim must actually be considered by one or more of the Commissioners of the Board or that it needs to be sent to a specified address.
Mr Windle responded that (1) the provisions referred to are plainly in point as regards the Murphys as the documents which they rely on as constituting claims for relief were submitted to HMRC after those provisions came into force in 2005. It must be the case that the officer who was in correspondence with the Murphys/their agents, Mr Wood, was authorised to receive a claim on behalf of the Board, (2) as regards DH, in effect the position is the same notwithstanding that the documents relied on as constituting claims pre-date the provisions referred to. Once it is accepted that the claim may be sent to an officer, there can be no meaningful distinction according to whether the relevant document is addressed “to the Board” or not, (3) there was a time when the distinction between “officers of the Board” and “the Board” was meaningful in some way but over time, it appears as HMRC became bigger and more professionalised, more and more responsibilities were delegated. By the time in question the powers to receive claims had been delegated and without, for the avoidance of doubt, a particular requirement for the words “to the Board” to feature at the top of the claim.
It is not readily apparent why the relevant provisions draw a distinction between the making of claims to the Board and to an officer of the Board. It does seem that, from 7 April 2005 onwards, the distinction is one without a material difference, given that the Board could delegate the ability to receive claims to an officer of HMRC and it appears that was the practice. On HMRC’s own case, there is no apparent purpose to a claim being made to the Board as opposed to an officer given that there was no suggestion by them that the claim actually has to be placed in front of one or more officers of the Board or addressed to them otherwise than stating in the relevant tax return or other document “to the Board”. Moreover in all the circumstances of these cases, it is reasonable to suppose that the function of delegating the ability to accept claims for treaty relief on behalf of the Board was delegated to the officers who were in correspondence with the appellants. We do not accept, therefore, that the documents which the appellants rely on as showing that the appellants made claims are prevented from being such because they were not specifically addressed “to the Board”.
Caselaw on requirements for a claim when there is no prescribed form
The appellants submitted that:
The purpose of requiring a taxpayer to make a claim for relief under a double tax agreement is simply to alert an inspector to the fact that the relief is being sought by the taxpayer. Further, at the relevant time, as set out above, s 42 TMA was intended to provide a simple and easy way to meet requirements for making a claim, which HMRC have always had the power to supplement. An overly formalistic approach to this legislation, as contended for by HMRC, would frustrate these purposes and risk doing damage to legislation of very wide application.
It suffices for there to be a valid claim for treaty relief that HMRC could have been under no misapprehension from the relevant correspondence that the taxpayers relied on and wished to benefit from treaty relief, albeit that the correspondence did not contain any formal claim for relief as such and that the appellants/their agents did not consider it was necessary for them to make a formal claim in order to benefit from the relief. That follows from the decision of House of Lords in Gallic Leasing and Unilever.
From Gallic Leasing it is clear that there is no requirement for a particular form of words or for the claim to be in a single document or for the taxpayer to state a claim is being made. The subjective intention of the taxpayer and/or their agent is irrelevant to whether a claim has been made for relief; if a reasonable reader would understand from the correspondence that a taxpayer intends to benefit from a relief that suffices for there to be a claim. It would be unduly onerous for HMRC to have to be concerned with what the taxpayer actually thought or meant. Moreover where the actual officer dealing with the matter plainly understands that relief is being sought, there must be a high bar on HMRC being able to say that the claim has not been adequately communicated. The requirements for there to be a claim are not very high (see the decision in Unilever). The purpose is to ensure HMRC understand and what is being claimed, but not to impose an excessive burden or set out traps for taxpayers. As the House of Lords in Gallic Leasing identified that is not surprising as, when HMRC consider that higher requirements are needed, they have the power to impose them as they see fit.
Strong support for this conclusion can be drawn from the Upper Tribunal decision in X-Wind Power Ltd v HMRC [2017] UKUT 290 (TCC). Which held that X-Wind Power Ltd had by submitting the wrong form made a claim for relief under the Enterprise Investment Scheme (“EIS”) even though its subjective intention had always been to claim SEIS relief (at [17] and [18].
HMRC submitted that the appellants’ stance is too simplistic and is not justified on their reading of Gallic Leasing and Unilever. They did not dispute that the relevant correspondence between the appellants/their agents and HMRC proceeded on the basis that the appellants sought to benefit from treaty relief and/or that the various HMRC inspectors involved knew that the appellants wished to benefit from treaty relief. However, in their view, that does not suffice for there to be a valid claim for relief. Their approach was essentially to assess individually each letter or other document the appellants relied on, to see if, of itself, it satisfied the various requirements they considered it needs to meet to constitute a claim. They considered that none of it did because, (a) in each case it does not demonstrate that there was a conscious decision by the relevant person, in their view, an identified trustee, to make a claim whereby that person was actively and unambiguously seeking relief from tax specifically pursuant to articles 4 and 14 of the treaty, and/or (b) it did not adequately identify the specific disposal for which relief was sought, and/or it was not addressed to the Board, and/or it did not contain the required quantification or the required declaration of accuracy.
HMRC emphasise that, in their view, further to the requirements imposed by statute and by case law, there are requirements which are implicit in the concept of making a claim or which are necessary for a system of claiming for relief to function in the real world:
A taxpayer must actively seek relief from tax. It cannot be HMRC’s role to consider a tax return and decide what relief would be in the taxpayer’s best interests, or what form of relief the taxpayer should be seeking. It is not HMRC’s role to act as a taxpayer’s agent and perfect an imperfect tax return. Claiming relief from tax is not always unambiguously positive. There are various occasions on which a taxpayer may quite reasonably choose not to claim a relief to which he or she is entitled.
A claim must be unambiguous. One of the key purposes of making a claim is to alert HMRC to the fact that a specific form of relief from tax is sought. A claim cannot leave any room for doubt that relief from tax is sought or which specific form of relief from tax is sought. Moreover, the right to claim relief from tax is subject to time limits and the making of a claim itself triggers time limits (such as to enquire into the claim). There must therefore be no doubt about when a claim for relief is made.
The appellants did not dispute these points in principle albeit that the parties views differed on whether these requirements were satisfied in this case.
A claim must be deliberate, in that it should follow from a conscious decision to claim a form of relief. Whilst administrative errors are always possible, any system whereby relief could be claimed unintentionally or accidentally would lead to real injustice and uncertainty. The decision in X-Wind Power Ltd v HMRC [2017] UKUT 290 (TCC) does not support the proposition that a taxpayer’s subjective intention is irrelevant in considering whether a claim for relief has been made. The case concerned the effect of a compliance statement for EIS purposes. As Arnold J said, at [22], a compliance statement is not a claim. In any event, this is simply authority that a compliance statement for EIS does not become a compliance statement for the Seed Enterprise Investment Scheme simply because that is what the taxpayer intended.
The relevant requirements are not onerous. They are easily met by the inclusion of a single sentence in a trustee’s tax return saying, “Relief is sought pursuant to article 14(4) of the [treaty] in relation to the disposal of [X asset] on [Y date] such that the trustee should be granted relief from liability to capital gains tax in the sum of [Z amount]” or words to similar effect.
We note that the appellants dispute HMRC’s view that a valid claim for treaty relief could only be made by the trustees and not by the appellants and take a different view of what is required to satisfy the quantification requirement. We first consider the Gallic Leasing and Unilever and the submissions made above before turning to those specific issues.
In Gallic Leasing the General Commissioners upheld the Inspector’s decision that the references to group relief in a notice of appeal against assessments to tax, in the note on the accounts of the relevant companies and in the accountants’ computation of assessable income did not, either severally or collectively, constitute a valid claim to group relief, but amounted to no more than an intimation that group relief would or might be claimed in the future. On appeal the High Court ([1989] STC 354) held that in order to make a valid claim to group relief all that was required was for the claimant company to make it clear that a claim was being made and that it was unnecessary to identify either the surrendering companies or the amount of the relief to be surrendered by each company. The Court of Appeal [1991] STC 151 reversed this but it was then upheld by the House of Lords.
Lord Oliver who delivered the leading judgment set out that the making of a claim was governed by the general rule in s 42 TMA as no specific form had been prescribed under s 42(5) TMA which is now contained in para 2(3) of schedule 1A TMA. He held that ultimately one was “thrown back to whatever assistance one can derive from” the provision setting out that reliefs can be surrendered and enables another company in the group “on the making of a claim” to set those reliefs off. He noted that provision does not indicate any order in which the events leading to such allowance (i.e. surrender, claim and consent) are to take place and there appears to be no reason why a claimant company should not make its claim to group relief before it knows the extent of any available reliefs or whether the company to which they are available is willing to surrender them: “On this analysis the making of a claim serves no other purpose than that of alerting the Inspector to the fact that reliefs are to be sought by the claimant.” He said on that basis the documents the appellants relied on sufficiently constituted a claim for the purposes of the section.
He noted that the Inspector, who received the accounts with the references to group relief at the same time as he received the accounts of other companies in the group in which the relationships were clearly set out:
“cannot have been under any misapprehension as to the existence of a group or that the Appellant was claiming reliefs anticipated to be available within the group to an extent necessary to extinguish the tax liability on its profits. Indeed the correspondence shows that he so understood the position and, as Vinelott J. observed in the course of his judgment ([1989] STC 354, 362), the combined effect of the documents sent to the Inspector was precisely the same as if the accounts had been accompanied by a letter saying: “We hereby claim group relief under s 258(1) in respect of the full amount of the profits of the taxpayer company for the accounting period to 31 March 1982, being the amount shown in the profit and loss account”.”
He rejected HMRCs arguments that a claim must include all the information required for a claim to be valid (1404D-H). In considering the alternative submission for HMRC Lord Oliver said, at 1406E-H;
“It cannot, it is said- and, indeed this must I think be obvious - be sufficient simply to say “I claim,” which is meaningless. By necessary implication a claim must be made by an identified claimant and must relate to the corporation tax liability and thus to the profits of that claimant. Section 259, which specifies the reliefs available, necessarily ties those reliefs, and thus group relief, to a particular accounting period. Thus, a “claim” by definition has to be the assertion by an identified company of a right to set against its profits for a defined accounting period the reliefs available to another group company. A "claim" cannot, for instance, consist of a general assertion that for all future years group relief against whatever profits are made is claimed. Nor equally can a "claim" be made by an annual assertion on 1 April in some such form as "we do not know what (if any) profit we shall make during the year now current but, if and so far as we do make any, we hereby claim group relief." It is strictly unnecessary for the purposes of this appeal to decide the point, but, for my part, I do feel able to derive from the Act of 1970 that a claim to have any meaning at all must at least be a claim by an identified claimant to relief against identified or identifiable profits for an identified accounting period.” (Emphasis added.)
In Unilever the Court of Appeal essentially set out a useful summary of what they took from Gallic Leasing. Sir Thomas Bingham referred to Gallic Leasing, noted that it concerned group relief but that the point of principle is the same and it was held that “no more was required than a general and unparticularised intimation of an intention to claim” (at 689b). As Simon Brown LJ explained (at 692j), Sir Thomas Bingham held, in effect, that the taxpayers annual estimates in their profit schedules took account of the losses they wished to claim but not in such a way as to indicate to HMRC whether in any given year a particular company had incurred such a loss which it was setting off against profits. So although the taxpayers’ estimates reflected both their right to claim and their intention to claim, and in quantum, subsumed the value of their claims, they did not in fact alert HMRC to these matters. Simon Brown LJ continued, at 693a and b, that Gallic Leasing:
“construes the equivalent statutory time provision in respect of group relief claims as favourably as conceivable to the taxpayer but suggests there is required “at least…a claim by an identified claimant company to relief against identified or identifiable profits for an identified accounting period….The taxpayers’ estimates did not achieve that here: a claim is not made at least until the Revenue is able to recognise it as such.”
Although Gallic Leasing relates specifically to a claim for group relief, as the appellants submitted it gives clear guidance that, where HMRC have not exercised their power to prescribe the form of a claim for a particular relief, documents may amount to a claim for the relief even if on their face they do not purport to be such and even if they can only be taken to alert HMRC to the fact the relief is sought when assessed in combination. As recognised in Unilever the key point was that, within the applicable time limit, HMRC could not be under any misapprehension as regards the tax relief sought by the particular taxpayer in relation to an identified subject matter. The requirements that Lord Oliver identified cannot simply be transposed to any other relief; the precise position depends on what the relief is but it is notable that in that case the level of detail required was not high. There was no requirement, for example, to identify the company from which group relief was being surrendered.
The appellants submitted that it is notable that (1) the documents which the House of Lords considered founded a claim in Gallic Leasing included an appeal against an assessment that included an application to postpone payment of the full amount of tax claimed on the basis that the profits would be covered by group relief. This is similar to the position in the Murphy cases. There was no suggestion that the postponement application was not relevant, and (2) the House of Lords did not at all suggest that it was relevant whether the taxpayers considered they were somehow making a claim for group relief in any technical sense. Rather the position was that the relevant documents taken together conveyed to HMRC the nature of what was being sought, in terms of group relief, with sufficient particularity.
The parties also referred to various comments in the Upper Tribunal decision in The Commissioners for His Majesty's Revenue and Customs v Applicants in the Post Prudential Closure Notice Applications Group Litigation & Anor [2024] UKUT 23 (TCC), [2024] BTC 503 (“Post-Prudential UT”). Since the hearing that case has been heard on appeal in the Court of Appeal [2025] EWCA Civ 166 (“Post-Prudential CA”). The parties made written submissions on that judgment which have been taken into account in this decision.
That case relates to the position for the making of claims under the general provisions set out above after the requirement for quantification for the claim was introduced following the decision in Gallic Leasing. One issue was whether a particular letter could constitute a claim for credit for overseas tax paid under a double tax agreement although the letter stated the taxpayers were seeking repayment of tax on the grounds that foreign dividends are exempt. HMRC submitted that the Upper Tribunal give useful guidance on the making of claims in this context, in their conclusion on this issue, in particular, at [99], [103] and [110] of their decision:
“99.…It was common ground that we should give the letter the meaning it would convey to a reasonable HMRC officer having all the background knowledge which would reasonably have been available.”
“103…We accept that labels are not determinative…”
“110. It is also relevant that there is no such thing as a claim for exemption. The income is either exempt under section 208 ICTA 1988 or it is chargeable. It is then included or omitted from the return and the computation of tax accordingly. If it is included, an appropriate claim for relief by way of credit must be made. The purpose of a claim is to give notice to HMRC of what is being claimed and in what amount. If a taxpayer does not state what is being claimed, then it cannot be said to have made a claim. Unless the taxpayer indicates that it is claiming credit, it is not entitled to credit. In our view, the Taxpayer cannot say one thing and be treated as saying another merely because it produces the same result. That was the point being made by Lewison LJ at the end of [30] in BTPS. We cannot see how a document which seeks repayment of tax on the grounds that foreign dividends are exempt can be construed as a claim for credit on the basis that they are taxable.” (Emphasis added.)
The Court of Appeal upheld the Upper Tribunal’s decision on this issue and, in doing so, referred to Gallic Leasing as follows at [100] of their decision:
“…whilst there are no particular formalities for a claim to a DTR credit, it must as a minimum convey to HMRC the nature of what is being claimed (compare, in a different context, Lord Oliver's comments about the then limited requirements for a valid group relief claim in Gallic Leasing Ltd v Coburn (Inspector of Taxes) [1991] 1 WLR 1399, [1991] STC 699, especially at p.1406E-G). As the UT indicated at [103], it is necessary to consider how a reasonable officer of HMRC would understand the letter. The answer to that is obvious. It would be read as a claim to repayment of tax on the basis of an exemption, not a claim to a DTR credit which would (if validly made) entitle the claimant to a repayment of tax. The fact that the practical effect of an exemption in terms of the tax repayable could well be the same as a credit at the FNR makes no difference to the nature of the claim.” (Emphasis added.)
The appellants also referred to a further aspect of the decision in the Upper Tribunal, which the Court of Appeal has now reversed. One point of dispute in that case concerned dividends originally returned as taxable but with a claim for a double tax relief credit in respect of withholding tax, for a period for which an enquiry was opened. The question was whether HMRC should also have allowed a credit for tax on the underlying profits out of which the dividend was paid when closing the enquiry. The taxpayer’s case was that the claim it made in its return was sufficient to extend to the credit to which it later understood that it was entitled in respect of the dividends. The Upper Tribunal decided that it did but the Court of Appeal has now decided that was not correct. In a passage the appellants referred to in the decision of the Upper Tribunal, at [154], they said this:
“We accept HMRC’s submission that they are entitled to know that a claim to DTR is being made and the amount which is being claimed. However, in circumstances where the provisions do not identify any additional level of detail to be included in a claim, we cannot read any further requirement into the provisions. There are various forms of DTR described in Part XVIII ICTA 1988, and they are all reliefs which mitigate the effects of double taxation, whether it is juridical double taxation or economic double taxation. We agree with Mr Bremner that a claim for DTR encompasses all types of double taxation. Hence, if a company claimed treaty relief when it ought to have claimed unilateral relief, or vice versa, we cannot see anything to suggest it should be denied relief because of that error. Similarly, if a company limits its claim to WHT when it could have claimed relief for underlying tax, there is no reason it should not make a supplementary claim which is within time. It would not need to make a new, separate claim.”
In considering this issue in the Court of Appeal, Falk LJ, who gave the leading judgment with which the rest of the panel agreed, noted, at [123], that the legislation “says relatively little about the mechanics and content of claims” to double tax relief, and “the claim must be quantified, but the quantification need not be accurate: Class 8 at [98]-[101]”. She concluded, at [124], that “what is required is a claim for the relief in question. A claim for a credit, is a fundamentally different kind of claim to a claim for a withholding tax credit…Rather, the point is about what relief is being claimed….” and, at [127]:
“while it need not be quantified correctly, a claim to HMRC must indicate what is being claimed, not least so that HMRC are able to determine whether or not to accept it without enquiry. If a taxpayer chooses to claim credit for withholding tax, and as in that case makes it clear that that (alone) is what it is claiming, that cannot sensibly be treated as a claim that extends to anything else.” (Emphasis added.)
At [129] she noted that the taxpayers were not arguing only that if credit for withholding tax is claimed in respect of a particular dividend, then a credit for tax on the underlying profits should be treated as claimed in relation to that dividend but to a general claim for a credit that extends to all foreign income (or at least foreign dividend income) on an accounting period by accounting period basis. She said: “HMRC would then be obliged, on closing an enquiry, to allow credit for all DTR that could legitimately have been claimed on any income, even though the actual claim was not in fact contemplated by either party, and could not reasonably be understood, as extending beyond a claim to withholding tax credits on specified dividends” (emphasis added). She said, at [130], that this argument is wrong in its more limited form in respect of a particular dividend, but “the more obvious incorrectness of the broader argument helps to illustrate the point that the relief claimed must be articulated” (emphasis added).,
The appellants submitted that (1) these comments are consistent with the appellants’ stance that what is required to make a claim for a particular relief is for the taxpayer to communicate to HMRC that they are seeking to rely on that relief albeit that there is guidance on the level of specificity that is required, and (2) there is nothing in this decision to support HMRC’s stance that making a claim requires subjective intention of the kind they refer to. What is required is that a reasonable reader, in the position of HMRC, would understand that the taxpayer is relying upon a particular relief.
HMRC referred in particular Falk LJ’s comments at [124] to [130]. They noted the reference, at [127], to a taxpayer choosing to make a claim and that the taxpayer must then communicate that claim to HMRC. They submitted that it is not for HMRC to review all communications from a taxpayer to determine what claim a taxpayer could have made. They considered that Falk LJ’s comments at [129] and [130] are inconsistent with the appellants’ stance which, as they see it, is that “a taxpayer can make a claim that it did not intend to make or understand that it was making”. They submitted that (1) a taxpayer cannot indicate to HMRC that it is making a claim if it does not itself realise that it is making that claim, and (2) as set out in further detail below, on the facts of this case, (a) the appellants’ advisor considered it was not necessary to make a claim, because he considered that the relief was automatic, and (b) consistent with the fact that the appellants did not intend to make a claim, in none of the correspondence relied upon did the appellants communicate that they were in fact making a claim.
We can see no basis in the statutory requirements or in the case law for HMRC’s view that, in order to make a valid claim for treaty relief, the claimants must have consciously thought that there was a procedural requirement to make a claim in a some technical sense with which, in submitting a particular document or documents to HMRC, they were consciously seeking to comply. There is no suggestion in Gallic Leasing, Unilever or Post-Prudential that that is a requirement for there to be a valid claim albeit those decisions were made in the context of claims for different relief to that sought here. In the absence of specific requirements/formalities, the focus in each case was on whether the taxpayer had brought to the attention of HMRC or clearly articulated to HMRC, as Falk LJ put it, that it wished to obtain the benefit of a specific relief. In her comments at [124] to [130] of Post-Prudential CA Falk LJ was saying that the taxpayers had articulated to HMRC that they sought credit for withholding tax on dividends but had not articulated that they were seeking to rely on credit for tax on the underlying profits from which the dividends were paid. In her view these were two different kinds of relief. Nothing in her comments suggest that there was some additional requirement, as HMRC argue for here, that the taxpayers had in mind the need for the relevant reliefs to be claimed in a formal sense.
Similarly in this case there are no particular formalities specified for a taxpayer to be able to benefit from relief under article 14(4). Treaty relief of this nature in effect operates as an exemption from the CGT which would otherwise arise on a relevant disposal. It is reasonable to suppose that the purpose of the requirement for the taxpayer to make a claim for treaty relief to apply is simply so that HMRC is made aware, within the prescribed time limit, of the relief/exemption sought from the CGT charge which would otherwise apply. Moreover, we can see force in the appellants’ submission that it would be contrary to the purpose of the UK entering into double tax agreements which provide such reliefs/exemptions to create significant obstacles for persons to obtain the benefit of those reliefs/exemptions. There is certainly nothing to suggest that a taxpayer may not benefit from treaty relief unless he consciously has in mind the need to make a formal claim of some kind as opposed to having in mind and bringing to HMRC’s attention that it wishes to rely on a specific relief in respect of an identified tax charge.
Quantification
The requirement for quantification of a claim at the time the claim is made was introduced following and it seems in response to Gallic Leasing. The parties disagreed as to what is required for this requirement to be satisfied in these cases. The appellants submitted that, (1) as set out in HMRC’s Self-Assessment Claims Manual at SACM3025, the purpose of this requirement for quantification in s 42(1A) is to provide sufficient information to HMRC so that they can give effect to the claim without checking it at that point, and (2) consistent with that purpose, where the claim is for full relief for the whole of the tax liability, saying that full relief is claimed is sufficient to meet the purpose of subsection (1A). This relief functionally operates like an exemption, in effect, to relieve all the tax on a particular event, a disposal, but whether it applies or not has no tax effect on any other event/any of the taxpayer’s other tax affairs. Another example in a CGT context is private residence relief. This provision is not intended to require taxpayers to undertake the pointless but potentially difficult, costly and time-consuming task, of calculating a figure for a tax liability that never arises because of the relief. That would have no identifiable benefit for HMRC (as they would be in a position to determine the CGT charge that would otherwise apply) but could be onerous for taxpayers. The appellants considered that this approach is supported by the decisions of the High Court in Claimants Listed in Class 8 of the CFC and Dividend GLO v HMRC [2019] EWHC 338 (Ch); [2009] 1 WLR 5097 (“Class 8”) and of the Upper Tribunal in Post-Prudential UT. As set out above, since the hearing the Court of Appeal has decided the relevant issues as set out above.
In very broad terms the Class 8 litigation concerns claims by UK taxpayers that certain tax rules were incompatible with European Union law and that sums paid under them should be repaid with interest. One of the issues was whether the UK rules provided an effective remedy for the UK taxpayers under the rules dealing with the making of claims. As set out at [14] one of HMRC’s arguments was that certain taxpayers could have made valid claims for double taxation relief under ss 788 and 790 ICTA (for the reasons set out in full) even though at the time when the claim could have been made within the applicable time limits there was continuing uncertainty as to what might be decided in the courts as regards the correct quantum of the claims. The rules for the making of claims, equivalent to those under consideration in this case, required the claims to be quantified when made. The argument was that this requirement was met if the claimants estimated the quantum of their claims; the quantum claimed did not have to be legally correct.
At [98] this issue was described as being whether a taxpayer could have an effective remedy for relief under s 790 before any taxpayer could know how much to claim under the provisions on how claims can be made. The taxpayers argued that such certainty was not achieved until, at the earliest, 2012, when a decision of the Court of Justice of the European Union decided that the claim for relief should be at the foreign nominal rate. HMRC’s response was that “it was sufficient to make a claim that any figure was put in, provided it was quantified, because the legislation did not say that the claim had to be for the correct amount in order to be valid as a matter of law”. The High Court essentially accepted HMRC’s arguments as follows at [99] to [101]:
“99. In my judgment, it is necessary to look closely at the statutory provisions to resolve this question. Paragraph 54 provided, as set out above, that a claim “under any provision … for a relief … must be for an amount which is quantified at the time when the claim is made”. By itself, that provision makes no reference to the amount being accurate, only to it being quantified. And one can quite see why quantification is necessary in a self-assessment system, where paragraph 7 of schedule 18 provides for taxpayers to make a self-assessment tax return “of the amount of tax which is payable by the company for that period … (a) on the basis of the information contained in the return, and (b) taking into account any relief … for which a claim is included in the return”. Moreover, paragraph 56 allows for a supplementary claim to be made to correct an original claim.
100. In these circumstances, it is, in my view, clear that the taxpayer could have had an effective remedy for the available reliefs by claiming under section 790 for any quantified amount, even if it did not know at the time of claiming whether the correct claim was for the foreign tax actually paid on the dividends or the tax that would have been paid at the foreign nominal rate. The lack of knowledge of the precise rate at which the claim should be made may make it harder to make an effective claim, but it does not make it impossible in practice, as is required for the EU law principle of effectiveness to be violated.
101. Accordingly, the relief allowed by section 790 is not, I think, prevented from being an effective remedy because the taxpayers could not have been certain how much to claim under paragraph 54.”
HMRC submitted that (1) the quantification requirement is not met as the appellants did not in any relevant document specifically compute the tax otherwise due and hence the relief sought. The mandatory statutory requirement is for the claim to be quantified and not simply capable of quantification, (2) this is consistent with how the self-assessment regime operates. That system was introduced following the decision in Gallic Leasing: under it taxpayers have to identify the tax that they have to pay and, in doing so, identify any claims they wish to make and the quantum of the claims in determining the tax that they have to pay, and (2) this is supported by the view taken in two cases in the tribunal: Robins v Revenue and Customs [2013] UKFTT 514 (TC) (see [5] to [9]) and Buxton v Revenue and Customs Commissioners [2012] UKFTT 506 (TC) (at [79] to [83]). In these cases the tribunal held, in effect, that it did not suffice for this requirement to be met that the relevant claims were capable of quantification. HMRC also said that the Class 8 case simply demonstrate that the figure for the relief claimed can be estimated; a document is not prevented from being a claim because one does not know the final figure at the time, or that it is not the correct figure when put in a claim.
Mr Windle responded that (1) the cases HMRC refer to do not deal with the claiming of reliefs of this type which, in effect, operate as exemptions, (2) in Robins neither side was represented by Counsel, Gallic Leasing was not cited and that case and Buxton have been superseded by the decision in Class 8, and (3) in Class 8 it was not simply stated that the figure for the claim could be estimated. The court accepted that it is sufficient to make a claim that any figure is put in. It is possible that they meant that a best estimate would suffice, but that is not what they said.
Essentially we accept the appellants’ submissions on this point. It appears that the purpose of the requirement for quantification in s 42(1A) is to enable HMRC to be clear as to the particular tax charge for which relief is sought, as the tax which would be otherwise due if the claim is not successful. As noted, a claim for relief such as this, in effect operates to provide an exemption from a tax charge which, but for a successful claim for relief, would be due on a specified event; whether the claim is successful or not, the precise sum of the relief claimed does not otherwise affect the taxpayer’s overall tax position. We consider that in such circumstances, as is consistent with the approach taken in Class 8, in principle, it suffices for this requirement to be met that the taxpayer makes plain that relief is sought for the full amount of the CGT charge which would otherwise be due from the taxpayer on the specified disposal which gives rise to it.
Who can make a claim?
HMRC said that a claim for treaty relief must be made by the trustees of the Trusts whereas in the appellants’ view it can be made by the settlors. The appellants argued that, in any event, when the claims were made the settlors of the Trusts were also the trustees of them so that they can be taken to have made the claims in either or both capacities.
HMRC submitted that:
The claim: (a) is premised on a disposal that was made by the trustees and (b) concerns the residence of the trustees. The facts which are the subject matter of the claim all relate directly to the trustees and not to the settlors. It is not possible for a claim premised on the residence of the trustees to be made by any person other than the trustees themselves. The treaty applies to the trustees and it is not suggested that it applies to the settlors.
In order to calculate the amount of chargeable gain treated as accruing to the settlor, it is necessary first to calculate the chargeable gain that would accrue to the trustees after taking into account applicable deductions (see s 77(1)(b) and 2(2) TCGA). Any relief which is available to the trustees will therefore operate so as to reduce the amount of chargeable gain which is treated as accruing to the settlor. Therefore, it is necessary for any reliefs that could reduce the gain to be claimed by the trustee as part of the calculation. That includes any relief claimed under a double tax treaty.
The purpose of a trustees’ tax return is set out by s 8A TMA which, at the relevant time, said, “for the purpose of establishing the amounts in which the relevant trustees of a settlement, and the settlors and beneficiaries, are chargeable to income tax and capital gains tax for a year of assessment…an officer of the Board may…require the trustee…to make and deliver to the officer…a return…”.
The appellants said that:
The better construction is that s 77(1)(b) TCGA requires a hypothetical approach to the computation of the chargeable gain on which the trustees of the relevant trust would otherwise be chargeable. It is for the settlor of the relevant trust, as the person who is actually chargeable on the relevant gain and who therefore actually benefits from relief under the treaty, to inform HMRC that the treaty applies. That approach fits much better with the scheme of the legislation. Rather than requiring trustees to claim a relief, which makes no difference to their actual CGT liability, one simply considers what their CGT liability would be in the abstract and allows the person who is actually affected by that liability to make a claim: any relief favourable to the trustees will therefore operate so as to reduce the amount of chargeable gain which is treated as accruing to the settlor.
This approach is consistent with the decision in Eyretel Unapproved Pension Scheme & Ors v Revenue & Customs [2008] UKSPC SPC00718, [2009] STC (SCD) 17:
The settlors of an unapproved pension scheme were assessed to CGT under s 77 in respect of chargeable gains accruing to the trustees of the scheme. The trustees undertook a scheme intended to create a capital loss. HMRC ruled there had been no capital loss, albeit the trustees had claimed the benefit of it in their tax returns, and assessed the settlors to CGT without giving relief for any loss. HMRC had not opened an enquiry into the trustees’ return.
The settlors argued that in s 77(1)(b) the “amount on which the trustees would…be chargeable to tax” is the gain less the loss claimed in the return arising from the Transactions. This is the amount on which they would be chargeable in the real world, not in the abstract. HMRC did not enquire into the return and so they cannot now go behind the return to ignore the existence of the loss in taxing the settlors. The trustees made a proper disclosure of the transactions in their return (see paragraph 3(15) above) so as to enable the HMRC to understand them (see [20]).
HMRC contended that the trustees’ information return does not determine liability at all, and the trustees’ self-assessment return will not include the gain since the trustees are not chargeable in respect of it. The important return for HMRC to enquire into is the settlor’s self-assessment return. There was no need for HMRC to open an enquiry into the trustees' return to no effect (see [21]).
The decision was made in favour of HMRC, at [25], following these comments at [23] and [24]:
“Section 8A of TMA requires a return by the trustees of certain information “for the purpose of establishing the amounts in which the relevant trustees of a settlement, and the settlors and beneficiaries, are chargeable.” That is merely a means of the Revenue obtaining information. Section 9 requires a self-assessment of the “amounts in which… the person making the return is chargeable.” The effect of s 77 is that “the trustees shall not be chargeable to tax in respect of those [gains].” The trustees are therefore not obliged to return the gains and losses in their self-assessment return. Since the settlor is chargeable on the trustees' gains, s 9 will apply to the settlor, and the trustees will not make any self-assessment of the gains.
The procedures relating to self-assessment therefore apply to the settlor and not to the trustees. The Trustees are merely providing information about the gains and in so far as they are being self-assessed the gains will not be included as they are not chargeable in respect of them. I can therefore see no reason why the Revenue need open an enquiry into the trustees' return when it will not enable them to collect any tax. I see Mr Prosser’s point that the trustees were trying to fix the amount on which the settlor had a right of recovery but s 78 gives them that entitlement “Where any tax becomes chargeable on and is paid by a person in respect of gains treated as accruing to him under section 77…”. Actions by the trustees do not affect the amount the settlor actually pays.”
In Duke v HMRC [2010] UKFTT 306 (TC), the same issue arose and the tribunal followed the same approach as was taken in Eyretel.
Moreover there is no basis offered by HMRC for contending that no other person can make the claim. Indeed, given the variety of entities which are partially transparent (such as limited liability partnerships) it would be surprising in the context of double tax relief if there was a hard and fast rule against a person making a claim in respect of someone else. Until May 2020 it appears that HMRC believed that claims for double tax relief could be made by the settlor as that is what their own guidance appears to say. The CGT manual stated that: “A settlor may be able to claim exemption on some or all of the attributed trust gains, but this depends on the terms of the particular double taxation agreement.”
Mr Stone submitted that the appellants overlook the fact that this issue has to be assessed in the specific context of a claim for relief under article 14(4) of the treaty; in effect, such a claim is based on the residence of the alienator, the trustees of the Trusts. The settlor of each Trust was not the alienator and so 14(4) can provide the settlor with no protection. It cannot be right as a fundamental principle that a claim can be made by a person as regards another person’s residence status. The trustees must have a status under the treaty and it must be for them to determine how they want to be taxed and to make a claim accordingly. Neither of the cases the appellants refer to assist them as they do not deal with claims under a double tax agreement. The conclusion in those cases was effectively that s 77 TCGA enables HMRC to go straight for the settlor; they do not need to open enquiries into the trustees’ tax return. That does not answer the question of what is required for a claim for relief under a double tax agreement. As regards the CGT manual, the passages cited are not about who can make a claim for double taxation relief under section 788(6) ICTA; they are saying that a settlor may claim an exemption on some or all of the attributed trust gain.
Mr Windle replied that the treaty is drafted to apply automatically and the claim is made under UK domestic law (as set out above) for relief from CGT for the benefit of the settlor as the person otherwise taxable on the relevant gains. He said it is also wrong to say that a person cannot make assertions about another person’s residence in seeking to benefit from reliefs/exemptions provided for in the treaty (as illustrated by article 11(3) of the treaty).
We can see no bar to the settlors being the persons who make the claim for treaty relief rather than the trustees. In our view it is highly relevant that the effect of s 77(1)(b) TCGA is that (a) the appellants/settlors are the persons who are chargeable to CGT on the gains realised on the disposals made by the trustees of the Trusts, after any deductions, and thereby stand to benefit from the availability of treaty relief, and (b) the trustees are specifically not chargeable to CGT on the gains realised by them on the disposals, after deductions, such that, as was held in similar circumstances in Eyretel, although they may be required to include details of the gains in the return required to be made under s 8A for information purposes they would not be required to make a self-assessment of the CGT due, as it was not chargeable on them.
Of course whether the appellants/settlors can in fact obtain treaty relief depends on them demonstrating that the trustees of their Trusts satisfy the requirements of article 14(4) in the treaty, in particular as regards the residence test in article 4. However, we cannot see that the terms of article 14(4) have any bearing on this issue. The procedure for charging CGT on gains realised by the trustees and for making a claim for relief from that tax charge and obtaining the relief is governed entirely by UK domestic law. The treaty is simply not concerned with how the CGT charge is dealt with under UK law or how relief under the treaty is given effect under UK law.
Law on estoppel issue
On the estoppel issue, it was common ground that the principles governing estoppel by convention, to which the tribunal must have regard, are those set out by the Supreme Court in Tinkler v HMRC [2021] 3 WLR 697 (“Tinkler”). Lord Burrows, who gave the leading judgment with which the rest of the panel agreed, approved the following statement of principle by Briggs J in HMRC v Benchdollar Ltd [2009] EWHC 1310 (Ch); [2010] 1 All ER 174 (“Benchdollar”) at [52] (Tinkler at [45]) subject to clarification of point (i) which he went on to make:
“(i) It is not enough that the common assumption upon which the estoppel is based is merely understood by the parties in the same way. It must be expressly shared between them. (ii) The expression of the common assumption by the party alleged to be estopped must be such that he may properly be said to have assumed some element of responsibility for it, in the sense of conveying to the other party an understanding that he expected the other party to rely upon it. (iii) The person alleging the estoppel must in fact have relied upon the common assumption, to a sufficient extent, rather than merely upon his own independent view of the matter. (iv) That reliance must have occurred in connection with some subsequent mutual dealing between the parties. (v) Some detriment must thereby have been suffered by the person alleging the estoppel, or benefit thereby have been conferred upon the person alleged to be estopped, sufficient to make it unjust or unconscionable for the latter to assert the true legal (or factual) position.”
Prior to approving that passage from Benchdollar Lord Burrows set out a comprehensive survey of the cases pre-dating Benchdollar. As the appellants submitted the principles to be derived from that review include the following:
An estoppel by convention can be established through correspondence with a person with actual or apparent authority to represent the principal (see [27]).
An estoppel by convention can apply to a common assumption of facts or law (see [31]).
It is no bar to an estoppel by convention that the mistake of the party raising the estoppel was not instigated by the other party (see [32] and [56]).
The common mistaken assumption must be one that each party has communicated to the other. This is the requirement that there must be a statement or conduct between the parties that “crosses the line”. In other words there must be some mutually manifest conduct which is based on the common but mistaken assumption. That is, the party which is alleged to be estopped is required to manifest its acceptance or sharing of the assumption to the party which is alleging the estoppel. What was in mind, therefore, was communication (whether by speech or conduct, and whether express or implied) crossing an imaginary line, drawn between the parties to the effect that the estopped accepted or shared the assumption. (See [34]).
A key factor in whether it would be unfair, unjust or unconscionable for a party to assert the true legal or factual position is whether there was still time for the party seeking to raise the estoppel to take corrective action when they become aware of the true legal or factual position (see [44] summarising Benchdollar).
As Lord Burrows set out, at [39], the first case in which the House of Lords clearly confirmed the existence of estoppel by convention was Republic of India v India Steamship Co Ltd (No 2) (“The Indian Endurance”) [1998] AC 878 (cf Lord Goff’s speech in the earlier case of Kenneth Allison Ltd v AE Limehouse & Co [1992] 2 AC 105 in which he alone relied on estoppel by convention). He described Lord Steyn as setting out the elements of the doctrine in a clear and simple form in the following way, at p 913 which he set out:
“It is settled that an estoppel by convention may arise where parties to a transaction act on an assumed state of facts or law, the assumption being either shared by them both or made by one and acquiesced in by the other. The effect of an estoppel by convention is to preclude a party from denying the assumed facts or law if it would be unjust to allow him to go back on the assumption…It is not enough that each of the two parties acts on an assumption not communicated to the other. But it was rightly accepted by counsel for both parties that a concluded agreement is not a requirement for an estoppel by convention.”
Lord Burrows said, at [48], that the Benchdollar principles brought a welcome degree of clarity to the law on estoppel by convention, at least in the context of non-contractual dealings. He added, at [49], that however it was unfortunate that Briggs J’s first principle made no reference to the need for conduct to have “crossed the line” but soon after Benchdollar, Briggs J was presented with the opportunity to make that refinement in Stena Line Ltd v Merchant Navy Ratings Pension Fund Trustees Ltd (“Stena Line”) [2010] EWHC 1805 (Ch); [2010] Pens LR 411 (upheld on appeal without discussing this point at [2011] EWCA Civ 543; [2011] Pens LR 233). In that case, Briggs J accepted the submission of counsel that his first principle should be amended to include that “the crossing of the line between the parties may consist either of words, or conduct from which the necessary sharing can properly be inferred” (at para 137). He continued to explain, at [50], that in Blindley Heath Investments Ltd v Bass [2017] Ch 389 per Hildyard J at [92] the same point was made by the Court of Appeal as follows:
“On the facts of that case, the parties to a share sale agreement had conducted themselves on the incorrect assumption that there was no earlier shareholder’s agreement by which any sale of the shares first had to be offered to existing shareholders. The parties had forgotten about an earlier shareholders’ agreement conferring pre-emption rights. It was held that estoppel by convention applied. The parties had conducted themselves on the basis of a common assumption that there were no valid rights of pre-emption and it would be unconscionable to allow the directors to go back on that assumption. While citing Briggs J’s principles with apparent approval, Hildyard J, giving the judgment of the court, at para 92, made clear in relation to the first principle that “something must be shown to have ‘crossed the line’ sufficient to manifest an assent to the assumption.”
Lord Burrows added the following commentary at [51] and [52] which explains his view of the concept of “crossing the line”:
“[51] It may be helpful if I explain in my own words the important ideas that lie behind the first three principles of Benchdollar. Those ideas are as follows. The person raising the estoppel (who I shall refer to as “C”) must know that the person against whom the estoppel is raised (who I shall refer to as “D”) shares the common assumption and must be strengthened, or influenced, in its reliance on that common assumption by that knowledge; and D must (objectively) intend, or expect, that that will be the effect on C of its conduct crossing the line so that one can say that D has assumed some element of responsibility for C’s reliance on the common assumption.
[52] It will be apparent from that explanation of the ideas underpinning the first three Benchdollar principles that C must rely to some extent on D’s affirmation of the common assumption and D must (objectively) intend or expect that reliance. This is in line with the paragraph from Spencer Bower, The Law Relating to Estoppel by Representation, 4th ed (2004) p 189, which was cited by Briggs J just before his statement of principles:
“In the context of estoppel by convention, the question here is whether the party estopped actually (or as reasonably understood by the estoppel raiser) intended the estoppel raiser to rely on the subscription of the party estopped to their common view (as opposed to each, keeping his own counsel, being responsible for his own view).”
… But this is not to suggest that C must be relying solely on D’s affirmation of, or subscription to, the common assumption as opposed to C relying on its own mistaken assumption. It is sufficient that, as D intended or expected, D’s affirmation of, or subscription to, the common assumption strengthened, or influenced, C in thereafter relying on the common assumption.”
The appellants also referred to Benchdollar as further explaining the requirement in (v), in particular, once the parties become aware that the relevant shared common assumption is not correct. In that case, HMRC had sought, in effect, to extend the limitation period for them to seek recovery of national insurance contributions (“NICs”) from certain taxpayers whilst proceedings were brought to establish liability to NICs. However, it transpired that the method they used was ineffective as a matter of law. One of the arguments they raised, to support the view that they could nevertheless seek to recover the sums despite the expiration of the limitation period was that the relevant exchanges between them and the taxpayers manifested a mutual understanding, expressed by each of the parties to the other, that certain signed acknowledgements or part payments (“the scheme”) postponed the running of time of the limitation period and that, coupled with their decision not thereafter to bring proceedings within the primary limitation period, and with the obtaining by the relevant taxpayers of the benefit of not being subjected to civil proceedings while their statutory appeals remained to be determined, was sufficient to give rise to an estoppel by convention (see [26]).
As set out at [30] and [31], (1) HMRC became aware on 9 August 2001 that the scheme was ineffective, (2) by that date the primary limitation period in respect of claims arising from the 1994/95 tax year had expired, but numerous claims arising under the subsequent two tax years could still have been preserved from becoming statute barred by the prompt issue of protective claims in relation to them, and (3) the result was that all the relevant claims were instituted well outside the primary limitation periods relating to them, so that they are all statute barred unless the taxpayers were disabled from relying upon the statutory limitation period, either by contract or by estoppel.
Having decided that the other elements for an estoppel by convention were established, Briggs J turned to the question of injustice or unconscionability, and to the question as to the effect upon any estoppel of HMRC’s discovery on 9 August 2001 that the scheme they used was ineffective (see [57]):
In relation to claims for NICs arising from the 1994/95 tax year, he explained that HMRC had in reliance upon the shared understanding that the scheme was effective to postpone the running of time in fact allowed the primary limitation period for those claims to expire by, at the latest, the end of May 2001. By then, HMRC had been given reason to think that the efficacy of the scheme was in doubt, but did not know, prior to obtaining advice to that effect on 9 August, that it was in fact ineffective. HMRC therefore suffered a detriment in reliance upon the shared assumption, in relation to those claims from which it could not extricate itself in or after August 2001. At the same time, the taxpayers subject to disputed claims for that tax year had enjoyed the full anticipated benefit of HMRC’s reliance upon the assumed effectiveness of the acknowledgements and part payments, in the sense that they had not been visited with protective recovery proceedings in the meantime and, subject to certain isolated possible exceptions, continued to enjoy that benefit until the resolution of the statutory appeals. He concluded at [59] that the combination of a detrimental reliance by HMRC coupled with the obtaining by the relevant taxpayers of the anticipated benefit of HMRC’s reliance upon the shared assumption is sufficient to render it unfair and unjust for those taxpayers now to advance a limitation defence in relation to NICs arising out of that tax year.
He then addressed NICs arising in the 1995/96 tax year where time for the pursuit of those claims had already passed by 9 August 2001 or between that date and 10/11 September 2001, when HMRC decided not to take steps to protect claims from becoming statute barred. He noted that the effect of a party becoming aware of the untruth of the shared assumption “is not necessarily to kill the estoppel stone dead there and then. The reliant party is commonly afforded a limited time in which to protect itself from the consequences of a discovery of the true legal or factual position” (see Arden LJ’s analysis in London Borough of Hillingdon v ARC Ltd [2000] EWCA Civ 191, at [64]). He said that on the facts of the present cases, a reasonable time for HMRC to react included the period between 9 August and 10/11 September 2001. There was no need artificially to extend the period beyond that date, since by then HMRC had made a conscious decision not to take steps to protect itself in relation to any claims for which the primary limitation period was still running at that later date. (See [60] and [61].)
At [62] onwards he considered the position in relation to claims which only became statute barred after 10/11 September 2001, and said that:
“the analysis of the question of injustice or unconscionability assumes an altogether different aspect. The Revenue could have taken steps to protect those claims from becoming statute barred either by issuing protective claim forms or by seeking to make effective tolling agreements with the relevant employers. Those NIC claims did not therefore become statute barred by virtue of the Revenue’s reliance upon the shared assumption.”
At [63] he noted that HMRC submitted that (a) the fact that HMRC’s decision in September 2001 to take no further steps meant that taxpayers who had participated in the scheme continued to enjoy in full the benefit which they anticipated that they would receive, namely not being made the subject of protective recovery proceedings while liability to pay the NICs remained in issue in the statutory appeals process, and (b) in cases where the receipt of the anticipated benefit made it unjust for the recipient to resile from the shared assumption, the fact that a case in detrimental reliance could not also be established was neither here nor there. He concluded that it was not on balance unfair, unjust or unconscionable for taxpayers in these circumstances to assert a limitation defence to those claims, arising out of HMRC’s decision not to protect them once aware of the true legal position for the following reasons:
In my judgment, questions of injustice or unconscionability must be addressed in the round, and due weight given to all relevant factors. In that context the following observations of Neuberger J in the PW & Co v. Milton Gate Investments case, (supra) at paragraph 222 are of real force:
“In almost all cases, such unconscionability must be based on the prejudice which would be caused to the claimant if the strict legal position applied. As I see it, the claimant must also establish that that prejudice arises from its reliance upon the convention. In other words, the court generally must be satisfied that (a) the claimant will suffer real prejudice, and (b) the prejudice arises from its reliance (upon) the convention. It should be emphasised that, even if the claimant satisfies these criteria, there may be no estoppel, because there may be other, more powerful, factors pointing the other way.”
In the present cases, in relation to claims which became statute barred only after September 2001, the prejudice occasioned to the Revenue by the loss of the ability to pursue those claims was in no sense reliant upon the convention. The Revenue had been advised in terms that the acknowledgements and part payments did not have the effect assumed by the convention, and knew that, nonetheless, those claims could still be saved by the taking of prompt protective steps….
While I recognise that the continued enjoyment by employers of the benefit (i.e. not be sued) which they sought to obtain from providing the acknowledgements and part payments on the shared assumption as to their effectiveness is a relevant element to be considered on the question whether it would now be unjust, unfair or unconscionable for them to rely upon the true legal position under section 29(5), it is in my judgment a relatively modest factor, in particular when compared with the Revenue's decision not to protect claims which it knew could have been protected by other means. In relation to those claims, the Revenue was, quite simply, the author of its own misfortune.” (Emphasis added.)
The parties’ full submissions on this issue are set out below. The appellants submitted essentially that the correspondence set out below demonstrates that the requirements set out in the caselaw are met for HMRC to be estopped from raising that no valid claims for treaty relief were made.
Evidence and facts relating to the procedural issue
General early correspondence
We have set out below the evidence in the documents/correspondence between the appellants/Lansburys and HMRC, on which appellants primarily relied, and the evidence given by Mr Morris and Mr John Bentley, an officer of HMRC who dealt with these cases from 2017 onwards. Mr Bentley confirmed that he was not the decision-maker in respect of the decisions to issue the assessments, the decision-maker has retired, he had not spoken to the officers involved in the correspondence with the appellants before his involvement in 2017 and his views on what happened are based on the documents that are referred to in his statement and in the bundle. We have found Mr Bentley’s evidence to be of limited assistance given that he was not involved in these cases at the relevant time and much of the commentary in his witness statement was simply his opinion on the meaning of the correspondence he reviewed. We have taken account of his evidence to the extent it relates to his experience of HMRC’s practices and procedures of relevance.
In his statement Mr Bentley said that his predecessors were based in teams that had specialist knowledge of non-resident trusts and the tax consequences for settlors and beneficiaries in relation to the trust. He did not think that specialist knowledge would necessarily have included non-specialist knowledge of double tax conventions as ultimately the teams that those individuals were on and the team that he joined were trust compliance teams. They would have had someone they could speak to about double taxation conventions in HMRC.
Correspondence in relation to Lansburys’ other clients in 2000 to 2004
Lansburys advised several other individuals and the trusts settled by them about RTW planning involving NZ. HMRC opened enquiries into many of these individuals and trusts. It appears the earliest enquiry into an individual or trust advised by Lansburys was into a trust we refer to as the “O trust”. From late 2005, it was agreed between HMRC and Lansburys that they would focus on HMRC’s enquiries into the trusts we refer to as the “E trusts” as an informal lead case with the appeals of other individuals and trusts advised by Lansburys waiting behind. The appellants placed some emphasis on the history of HMRC’s enquiries into the O and E Trusts on the basis it gives context to the correspondence relating to the appellants:
On 11 May 2001, Mr John Roberston an Inspector of Taxes at HMRC wrote to Funcode, the trustee of the O trust, opening an enquiry into the 1999-2000 trust tax return of that trust:
He wrote: “I infer that gains were made by the Trustees in the period [when the trust was not resident in the UK] and that the Trustees believe that any such gains are exempt from UK Capital Gains tax under the provisions of Article 14(4) of the [treaty] on the basis that the Trustees were resident in [NZ] when the gains were realised.”
He also explained that he believed the residency tie-break provision in Article 4(3) would “come into play” and asked for documents and information to allow him “to determine where the Trustees [sic] place of effective management was during 1999/00”.
This letter proceeded on the basis that the application of the treaty was in point. Mr Morris’ evidence is that he understood this letter to mean that there were no procedural issues between the parties, such as whether claims for treaty relief had been made.
On 13 June 2001, Funcode replied stating that it did not agree that article 4(3) applied because there was “no provision of UK domestic capital gains tax legislation which deems the Trustees to be resident throughout 1999/00” .
On 29 June 2001, Mr Robertson replied to Funcode:
“As the trustees were resident during part of 1999/2000 they are chargeable on all the gains realised by them in that year subject to any successful claim they may have to DTA relief and subject to the provisions of [s 77]. I need the information I have requested to determine whether treaty relief is available to the trustees. But I also need to check whether the trustees did in fact cease to be resident in the UK as they have claimed in order to check whether Section 80 TCGA 1992 applies to the trustees.”
On 6 August 2001, Mr Robertson sent an information notice to Funcode (under s 19A TMA) seeking much of the information that had been requested in the letter of 11 May 2001 which Funcode provided on 29 August 2001.
On 13 September 2001, Mr Robertson wrote to Funcode saying:
“The No. 2 trustees have not returned any gains in their 1999/00 Self Assessment Return, presumably on the basis that the taxing rights for the gains are allocated to [NZ] under the [treaty]. The No. 2 trustees were residents of the UK for the purposes of the DTA in 1999/2000 and I need to clarify for the purpose of considering any claim under the [treaty] whether or not the No. 2 Trust was a resident of [NZ] for the purposes of the [treaty] at the time the disposals were made.”
This letter proceeded on the basis that the treaty applied. Mr Morris considered that this request for information as to the residence of the trustees could only be justified if Mr Robertson took the view that there were no procedural issues between the parties, such as whether treaty claims had been made.
He said that he understood that there was nothing further that needed to be done to rely on treaty relief because of HMRC’s actions. If HMRC had said at the time that a formal claim needed to be made, he would have advised the trustees to do so. He accepted that even if a claim had been made or could be treated as having been made in respect of the O trust that does not mean that HMRC had received and/or accepted a valid claim had been made by the appellants.
Funcode and Mr Robertson then exchanged letters discussing whether the O Trust was resident in NZ during the period when it had NZ resident trustees. On 20 December 2001, Funcode sent to Mr Robertson a letter from the NZ Inland Revenue stating that it had “classified the trust as resident for taxation purposes” .
On 30 January 2002, Mr Robertson replied to Funcode saying that he was:
“seeking clarification on the contents of the letter from the Greymouth Branch Office of the [NZ] Revenue as it appears to conflict with other guidance received from the [NZ] Revenue concerning the application of their domestic tax code to trusts with [NZ] resident trustees.”
On 6 February 2002, Funcode wrote to Mr Robertson asking for a copy of the communication with the NZ Inland Revenue.
On 20 February 2002, Mr Robertson replied saying that:
“Article 25 of the [treaty] allows for exchange of information between the competent authorities of each contracting State for the purposes of carrying out the provisions of the Agreement”.
Following further correspondence, on 26 April 2002, Mr Robertson wrote to Funcode saying:
“As I explained in my letter of 20 February 2002, I am waiting for clarification from the [NZ] competent authority of the tax status of trusts in [NZ] as I need that information to decide on your client’s claim for relief under the terms of the [treaty]. This is because the letter written by the Gremouth [sic] District Office appears to be at variance with our understanding of the tax treatment of trusts in [NZ].”
On 21 July 2003, Mr Robertson wrote to Funcode saying:
“I apologise again for the delay but I can now confirm that I accept that the [O] Trust was a resident of [NZ] for the purposes of the [treaty] at the time of chargeable disposals by the trustees. The Trust was also a resident of the UK for the purposes of the Treaty place of effective management.”
Mr Robertson sent letters in near identical terms to the trustees of other settlements in August 2003. Since then, HMRC have accepted that the trusts advised by Lansburys that appointed NZ resident trustees were resident in NZ during the trustees’ appointment.
On 5 August 2003, Funcode wrote to Mr Robertson noting that he “now accept[ed] the Trust was a resident of [NZ] for the purposes of the [treaty] at the time of disposals by the Trustees”.
On 28 August 2003, Mr Robertson replied to Funcode saying:
“In order to obtain the benefit of the Treaty the Trust has to show firstly that it was a resident of [NZ] for the purposes of the [treaty]. Secondly if it is a resident of [NZ] it has to show that if as I contend it is also a resident of the UK for the purposes of the Treaty the tie-break clause awards taxing rights to [NZ]. If the Trust was not accepted by [NZ] as a resident of [NZ] for the purposes of the Treaty then any gains realised by the trustees in 1999/2000 would have been chargeable gains of the trust and a like amount would have been assessable on the settlers under Section 77 TCGA 1992. That was why it was necessary for me to establish first of all whether or not the Trust was a resident of [NZ] for the purpose of the Treaty and then when that was established to consider the application of the tiebreaker.”
Throughout the rest of 2003 and most of 2004, Funcode and Mr Robertson continued to correspond about the effect of the treaty on the O Trust.
On 16 January 2004 Funcode requested that Mr Robertson set out all his arguments He agreed to do so. This was followed by a chaser from Funcode. Mr Robertson then apologised for the delay in providing “a comprehensive explanation of the Revenue’s views on the facts of this case and the issues raised by them”. As Mr Bentley accepted, the comprehensive response from HMRC on 18 June 2004 proceeded on the basis that the only relevant concern was the interpretation and application of the treaty (leaving to one side the options issue). He also accepted that it was possible that similar correspondence was sent in other of the around 11 enquiries which were open at the time in respect of Lansburys’ clients.
In around November 2004, the enquiries into the tax returns of Mr O and the O trust, as well as most settlements advised by Lansburys were taken over by Mr Duncan Cameron of HMRC.
Correspondence relating to DH
As regards DH:
DH submitted a personal tax return for 2000/01. In his return there is no reference to the gains which accrued to the Harris Trust when the relevant shares were disposed of or to the disposal itself. The only reference to gains on assets was to a disposal on 26 January 2001 for a small number of shares in GED Technology Group Limited made by DH. HMRC did not open any enquiry into DH’s personal tax return.
It is clear that a return for that year was also submitted for the Harris Trust but the bundles contained only HMRC’s computer extracts or “screenshots” from that return. Mr Bentley confirmed that the original return would have been a paper return, information from that would be inputted by a member of HMRC staff into the system, there is a box to tick if for example additional information was entered in the white spaces on the return but as he was not responsible for this he did not know otherwise how the inputting worked. Mr Morris confirmed that neither DH nor Mr Gaskell (as director of Allglory) has been asked to give evidence to explain what information was contained on that return. It appears from the extract that it make makes no reference to the disposal of the shares in the company or to any gain on that disposal.
On 19 December 2002, Mr Robertson wrote to Allglory, the then trustees of the Harris Trust, and opened an enquiry into “the Trust’s residence and capital gains” and sought information relevant to the implementation of the RTW planning. The documents requested included the sale agreement for the sale of the Trustees’ holding of shares.
Mr Morris said that he understood this letter to mean that there were no procedural issues between the parties, such as whether claims for treaty relief had been made.
Mr Bentley accepted that (a) this letter shows that the enquiry into the Harris Trust was opened on the basis that the trust had been involved in the RTW planning, (b) some of the questions raised were aimed at least in part at establishing the POEM of the trust, and (c) Mr Robertson would have opened this enquiry following some sort of risk assessment process. He did not know if Mr Robertson would have looked at the trust tax return before opening an enquiry into it. Mr Bentley said he did not know where Mr Robertson obtained the information that the trustees of the Harris Trust had sold the shares held in the trust.
Mr Bentley said that in some instances, the enquiries and therefore the tax records for individual settlors and beneficiaries, including the enquiry file, would be managed and maintained by the regional offices responsible for them. From the records available, there was no clear determining factor which dictated why a particular approach was chosen. In this case matters relating to DH personally were managed by a regional office (Bristol and Somerset) and not by Mr Robertson, the RTW planning lead. In such cases, the RTW planning lead would liaise with colleagues in that office as appropriate. It would have been standard practice for a file of relevant documents to be maintained by the regional office, but he had found no evidence that it was ever sent to him or his predecessors. Due to the passage of time, and HMRC’s move away from paper records, any file that was once held by the regional office would not exist any longer. He had been unable to contact the individuals at that office mentioned in the correspondence referred to below. It appears that they have left the department.
On 21 January 2003, Allglory replied and argued that the information sought was irrelevant. Mr Bentley accepted that this letter shows that Allglory understood that one point in dispute was the way in which the treaty applied and quite probably they derived that understanding from other cases in which Lansburys clients had corresponded with Mr Robertson. Mr Gaskell signed the correspondence as director of Allglory. Mr Morris confirmed that no attempt was made to call him as a witness.
Mr Robertson replied on 6 February 2003 and sent other letters in similar terms on that date stating that his understanding was that a trust could not be a resident of NZ for the purposes of the treaty but that he was awaiting a definitive statement of the position by the NZ Competent Authority. He accepted that Allglory may wish to defer providing the information he had sought until the competent authority had confirmed the trust was resident in NZ.
As Mr Bentley accepted (a) in this letter Mr Robertson confirmed Allglory’s understanding that the construction of the treaty prevented a trust from obtaining relief, (b) Mr Robertson believed that POEM was going to be relevant if a trust was resident in NZ. He said “there was clearly some discussion about the application of” the treaty, and Mr Robertson appeared to believe that if the trust was resident, considering POEM was necessary.
On 8 August 2003, Mr Robertson wrote to Allglory confirming that he accepted the trust was resident in NZ at the time but asserting that it was also UK resident such that he considered it necessary to consider POEM.
As accepted by Mr Bentley, Mr Robertson had not said in any of the letters so far that other issues might arise in this case. When it was put to him that a person reading this letter might reasonably believe that HMRC’s only real issue was with POEM under the treaty, he said “they could, but it would be naive to do so.”
On 11 September 2003, in a letter which is stated to be in response to the letter from Mr Robertson of 8 August 2003 Allglory stated:
“We note that you now accept the Trust was resident in New Zealand for the purposes of the [treaty] at the time of the disposals by the Trustees.
You say “the Trusts were also residents of the UK for the purposes of the Treaty and so their Treaty residence is decided under Article 4(3) by the location of the place of effective management”. That is incorrect. Apart from the fact that we are dealing with only one settlement, for capital gains tax purposes the Trustees were New Zealand resident from the date the New Zealand trustees were appointed until those trustees were removed and replaced by ourselves.
Thereafter the Trustees were resident in the UK. At no time were the Trustees dual resident. The place of effective management is accordingly irrelevant. In view of this will you confirm your enquiry is closed.”
It was put to Mr Bentley that Allglory’s understanding was that, even on Mr Robertson’s case, the application of the POEM test would determine the dispute. He said they confirmed that Mr Robertson thought that POEM was in point although they disagreed; in their view it was irrelevant.
On 14 October 2003, Mr Robertson replied to Allglory’s letter of 11 September 2003 setting out detailed arguments about the application of the treaty to the Harris Trust. He said:
“However certain of the information requested in my letter of 19 December 2002 is I think relevant to considering the Treaty claim irrespective of the effective management point. I believe that…
I hope that you will now feel able to supply all of the information requested but without prejudice to the remainder of the Appendix can I please have the information requested in paragraphs 1, 2 and 7 of the Appendix to the letter of 19 December 2002 as it is necessary to support the trustees’ claim that because of the operation of the Treaty there were no chargeable gains in the year ended 5 April 2001.” (Emphasis added.)
Mr Bentley said that he did not think the use of the emphasised words necessarily meant that Mr Robertson had seen a claim and therefore accepted that a claim has been made. Mr Bentley seemed to mean a formal claim. However, as the appellants submitted, reading this letter in the context of the prior correspondence, it demonstrates that Mr Robertson had accepted that Allglory was seeking to rely on treaty relief and he was here communicating that acceptance to Allglory. Moreover the use of the word “claim”, particularly in the first paragraph as “the Treaty claim”, indicates that Mr Robertson considered that DH/the trustees had done what was necessary in order to be able to seek to rely on treaty relief.
On 20 November 2003, Allglory provided a lengthy response to Mr Robertson’s arguments. As Mr Bentley accepted this letter proceeds on the basis that the sole dispute between the parties is about the interpretation and application of the treaty.
On 23 December 2003, Mr Robertson (a) wrote to DH saying that an estimated CGT assessment for 2000/01 would be issued to him shortly which would charge a capital gain under s 77, (b) sent a memo to Mr Peter Langton, Investigation Manager of the Bristol & North Somerset office, in which he asked for an estimated assessment for the year ended 5 April 2001 to be issued to DH. The memo stated that the amount of the assessment should be based on an estimated £840,000 gain chargeable under s 77 TCGA and that “Mr Harris has used his offshore settlement to try and avoid capital gains tax on the disposal of shares in GED Technology Group Limited”, and (c) wrote to Allglory in reply to their letter of 20 November saying he now had an additional reason for believing tax was due; his view was that the omission of the word “only” from article 14(4) of the treaty meant that NZ did not have exclusive taxing rights over relevant disposals made during periods of residence in NZ. He subsequently dropped this argument. Mr Bentley accepted that this letter refers exclusively to the way in which the treaty applies.
On 6 January 2004, J F Lye of the Bristol & North Somerset office sent a fax to Mr Robertson which stated: “With reference to your memo of 23 December 2003, a Further Assessment has today been issued with respect of estimated Capital Gains of £840,000.”
On 27 January 2004, Allglory wrote to Mr Robertson challenging the additional point he had made in his previous letter and said:
“So that this enquiry is not unduly protracted can we suggest that in your reply you set out all arguments you wish to make in this matter with technical reasons so that they can be dealt with at the same time.” (Emphasis added.)
On 4 February 2004, Rossiter Smith & Co, wrote to Mr Robertson appealing against what they understood to be a notice of assessment. They do not refer to a specific document and neither party was able to locate a notice of assessment sent to DH but it appears it was issued on 6 January 2004 as stated in the memo referred to above. In that letter it was stated that the grounds of appeal are that “we understand there is still continuing correspondence between the trustees and Mr Robertson and that the gains made by the settlement are not assessable on Mr Harris”. It was also stated that DH was applying for a complete postponement of the tax charged in the sum of £333,120 – which seems to be the CGT charge on a gain of £840,000.
On 12 February 2004, Mr Robertson wrote to Allglory saying:
“I agree that the best approach would be if I set out in detail what liabilities I believe fall on your client and the reasons for those liabilities. In order to make that statement as comprehensive as possible I am consulting with colleagues and I will write to you again in due course.”
On 16 February 2004, Mr Robertson sent a memo to J F Lye in which he referred to the fax dated 6 January 2004 and confirmed he had received an appeal against the assessment raised in respect of the tax year ended 5 April 2001 and asked for the tax charge of £333,120 to be stood over.
On 29 July 2004, Mr Robertson wrote to Allglory setting out the arguments on which he was relying. The introductory section of the letter includes the following:
“A charge to capital gains tax arises on the trustees under Section 2 Taxation of Chargeable Gains Act 1992 subject to Section 77 TCGA from the sale of their holding of GED Technology Group shares. The trustees claim that the gain is not a chargeable gain as they were resident in [NZ] at the time of the disposal so that under the terms of the [treaty] only [NZ] can tax the gain.”(Emphasis added.)
Mr Bentley accepted that a reasonable person in Allglory’s position in receipt of the letter of 12 February 2004 would have thought they were going to get a letter, after careful thought and consideration, setting out all the arguments. He said that it was not clear that the reference to a “claim” in the July letter meant necessarily “claim” in a legal sense and that Mr Robertson appears to be putting what the trustees believe. He accepted that the letter then sets out the arguments on which Mr Robertson relied which were concerned with the interpretation and application of the treaty and that probably, in the absence of anything further coming to light, in view of this letter and the correspondence that preceded it, a recipient would understand that these were the only issues between the parties. He added that ultimately there would still need to be evidence provided and that it is not unusual for concepts to be explored in enquiries in the absence of a full statement of facts and evidence.
In response Allglory set out what it understood HMRC’s case to be and said that the POEM issue does not arise and therefore the enquiry should be closed. Mr Bentley accepted that looking at this letter it is apparent that Allglory appear to have understood that the only issue between the parties was the interpretation and application of the treaty.
In our view the July letter could simply be read as meaning that the trustees were maintaining or arguing that no CGT was due on the basis they were entitled to treaty relief but in any event it is again clear from this correspondence that Mr Robertson considered there was no issue with the trustees being entitled to rely on that argument.
Shortly after this HMRC and Lansburys agreed to discuss one informal lead case with all other cases waiting behind and it appears there was then little correspondence specific to DH’s case for the next decade. There is a letter from Lansburys to Mrs S March dated 13 September 2006 in response to a letter of 11 August 2006 from Mrs March to DH’s accountants, Rossiter Smith & Co. This letter was written at a time when Lansburys were interacting with HMRC as agents of both DH and the Harris Trust. It referred Mrs March to Mr Peter Wood, who was the HMRC officer corresponding with Lansburys (a successor to Mr Robertson) and states:
“I think the essence of your letter is why was
1) any gain on the disposal of Mr Harris's No. 1 settlement’s holding of 249 shares in GED for 45,427 shares in Sitec and cash of £840,400 not a chargeable gain; and
2) any gain on the appointment of Mr Harris's No. 1 settlement’s holding of 45,427 shares in Sitec to Mr Harris's 2001 settlement not liable to CGT in the hands of Mr. Harris.
The answer is that at the time of the disposal and appointment above the No 1 settlement was tax resident in [NZ] and under the terms of the [treaty] the gains could only be taxed in [NZ] and could not be imputed through to Mr. Harris.”
As Mr Bentley accepted, up to 2007, all correspondence relating specifically to DH and his trust proceeded on the basis that the interpretation and application of the treaty was the only issue between the parties, and that a reasonable person in the position of either DH or the trustees would have understood that that was HMRC’s only objection.
Evidence on submission of returns by the Murphy Trusts
The bundles did not contain any tax returns submitted by the Murphy Trusts for the 2001/02 tax year in which the disposals took place. It did contain the Murphys’ personal tax returns for that year but they made no reference to the disposals or any gain realised on the disposals. On the evidence given by the witnesses, overall we consider it unlikely that returns were submitted for that tax year by the trustees of the Murphy Trusts
In his statement Mr Morris said that he understood the NZ Trustees submitted tax returns in NZ, Mr Paul submitted forms 50FS to HMRC and informed them of the change of residency of the Murphy Trusts and assisted TM with what needed to be returned on their tax returns. He also said: “When the tax planning was implemented, I did not consider DTA claims to be necessary due to my understanding of how the [treaty] functioned. I believed that the trust and settlors made tax returns in a way which entitled them to benefit from the [treaty]” and in any event he believed claims were made in respect of the Murphy Trusts when their notices of appeal against the assessments were submitted to HMRC (see below).
He said that (1) he did not search for UK tax returns for the Murphy Trusts and did not ask the NZ trustees for copies, (2) he did not know if the trustees of the Murphy Trusts filed UK tax returns for the 2001/02 tax year but “when they did file some tax returns it would…be obvious from what they’ve presented on there that what was happening”, and (3) in his comment in his statement he was not referring to the Murphys, but rather “what I considered at the time”.
It was put to Mr Morris that the trustees of the Murphy Trusts adopted the position that they did not need to put in UK tax returns for the tax year in which the disposals took place as shown by a letter to HMRC from Mr Paul, on behalf of Funcode, in relation to the AM Trust in which he stated that the AM Trust “became resident on 3 April 2002 but there were no income or gains arising in the trust for the period to 5 April 2002 (ie. three days)”. He said that surprised him, he would have thought tax returns were sent, and also that information was supplied on Form 50FS. On 24 January 2003, as trustee of the EM Trust, Mr Paul sent a letter to HMRC in which he wrote: “We refer to the Form 50FS (2002) sent to you on 3 December 2002 and note we have not received a tax return for 2001/02”. The form 50FS referred to was in the bundles. It does not contain any reference to the relevant disposal, it records nil gains arising to the trustees and denies that there were any offshore income gains in the 2001/02 tax year. There is no evidence of any form 50FS being sent to HMRC in respect of the other trusts.
Mr Morris accepted that, as regards the Murphy Trusts, (1) he had not provided any copies of any form 50FS or of any UK tax returns, (2) his view at the time was that there was no need to reference the disposals in any UK returns because he thought there was no need to make a claim for treaty relief, and (3) the position that he and Mr Blower at the time adopted, and the advice he gave to settlors and trustees, was that they did not need to reference such disposals. He added that he was just a bit surprised as he thought tax returns were submitted but obviously not. He said: “I didn’t believe a claim was necessary as long as the information was available…on tax returns.” It was put to him that as his view was that relief was automatic, he must have thought that one did not have to put any information about the disposal in the tax return. He said putting aside these cases, he was pretty sure that “we did file tax returns, with that information on it”.
He was taken to a letter from him about a different client of his in which he stated that it was Lansburys’ and Tax Counsel’s view that a disposal made in the circumstances under consideration in this case is not liable to CGT under the treaty and therefore nothing needed to be put on the client’s personal tax return for the relevant tax year and that HMRC were aware of the situation and were corresponding with Lansburys’ in respect of an earlier case. He accepted that was the advice he gave at the time on that particular client but said that he cannot remember whether in that case a trust tax return was submitted with information on it. He said from memory he was emphatic about all these trusts filing UK tax returns.
Mr Morris accepted that (1) the correspondence is consistent with the trustees of the Murphy Trusts not filing UK tax returns for the relevant tax year but added that Mr Robertson, “picked up very quickly from the tax returns that were sent in that we were looking at the [treaty]. That was admittedly on another client”, (2) the Murphys submitted individual personal tax returns for the 2001/2 tax year but there was no reference at all in them to the disposal of shares by the Trusts and, as a matter of fact, there was no claim made for relief under the treaty in those returns, (3) even if tax returns had been submitted for the Murphy Trusts, they would not have contained a claim for relief under the treaty because his view was that it was not necessary to make a claim on the basis that the relief was obtained automatically – he added provided the relevant information had been supplied, and (4) as regards the Murphy Trusts, he cannot tell the tribunal what information was supplied because he does not know if tax returns were submitted for them.
Mr Morris confirmed that (1) he now accepts that claims have to be made to obtain relief under the treaty, albeit that the form it has to be in is a matter of dispute between the parties, (2) in his view if the relevant information is on the trust tax return or in a document that could show what was happening then that by itself could be classified as a claim, (3) his understanding at the time was that one did not have to make any claim for relief at all provided the information was available. He did not know when his understanding changed.
In his statement Mr Morris said that all the correspondence which Lansburys had with HMRC regarding his clients who had used the RTW planning proceeded on the basis that claims for treaty relief had been accepted by HMRC. He referred to various specific letters, including those of 11 May 2001 and 13 September 2001 and commented that he understood from them that there were no procedural issues between the parties such as whether claims for treaty relief had been made and this was reinforced by HMRC’s actions as they only discussed the substantive issues. At the hearing he accepted that in the correspondence, HMRC never said or accepted that a claim had been made. He said that, however, the whole basis of the correspondence from 2001 onwards until 2018 was that HMRC accepted that a claim had been made in all these cases: they never said they did not believe claims had been made and if they had claims would have been put in; his understanding during all those 20-odd years of correspondence, was that they had accepted that claims had been made. He noted again that if “it [information] was in tax returns” that could be accepted as a claim; that would automatically follow. He suggested that information was provided, as Mr Robertson wrote to the trustees of another client of Lansburys only a few months after the return for that trust was submitted stating that he inferred that trustees believed that the relevant gains were exempt from CGT under the provisions of the relevant treaty. He accepted that it is not known what information was provided to HMRC in that case and that this sheds no light on whether HMRC accepted a claim had been made by the Murphy Trusts.
Mr Paul was questioned about his comment in his statement that he filed tax returns as director of Funcode for the Murphy Trusts:
He said that he filled in every return that he got but he cannot say what happened as regards the Murphy Trusts. It was put to him that the letter relating to the AM Trust referred to above shows that he did not put in a UK tax return in respect of this trust because, in his view, there had been no income or gains in the trust in that period. He was initially insistent that, if he had received a tax return, he would have sent it in with this type of letter even if he would have put there were no gains arising.
He was taken to the letter relating to EM Trust and the form 50FS dated 18 April 2002. It was put to him that this shows that, where he had not received a UK tax return, he simply sent this type of letter. He said he could not recall. He accepted that (a) the EM Trust was different from the others as he was once the trustee of it through Altech and so received this form 50FS (as HMRC presumably did not know that Altech had resigned as trustee) and he completed it and sent it back to HMRC but (b) for the other Murphy Trusts he would not have completed any form 50FS because he/Altech was not the trustee, and (c) given that there were no UK tax returns submitted for the Murphy Trusts, HMRC were not told of the disposals.
He was asked if his understanding at the time was that as trustee he did not need to make a claim for relief under the treaty. He suggested the disposals may have been intimated in another manner as HMRC had identified these cases as involving RTW planning. He was asked if his understanding at the time was that, as the disposals were made by offshore trustees, it was not necessary to tell HMRC about the gains. He said that there were no gains that could be taxed. He then said that (a) he did not think it was necessary to put it on a UK tax return, and (b) whether it was necessary proactively to make a claim for treaty relief is a technical issue which he could not really answer.
It was put to him that he was the trustee through Funcode after the sales took place for some months and he did not contact HMRC to make a claim for relief under the treaty. He said he did not think it was necessary particularly. He said he did not recollect if that understanding was informed by discussions he had had with Mr Morris about his understanding.
Correspondence concerning all Lansbury clients in 2005 to 2007
In 2005 to 2007:
On 3 March 2005, Mr Cameron (who took over from Mr Robertson) met with Mr Morris and Mr Blower. It was agreed that (a) HMRC would engage with Lansburys directly in respect of all individuals and settlements advised by Lansburys, and (b) one case would be taken forward with other cases waiting behind it. Initially, this was intended to be Mr O’s case.
In August 2005, the correspondence relating to Mr O, DH and most or all other Lansburys clients was taken over by Mr Peter Wood.
On 2 November 2005, Mr Wood met with Mr Morris and Mr Blower. The notes of meeting show that Mr Wood understood that at the earlier meeting on 3 March 2005, Mr Cameron had proposed that a test case:
“would be selected through which information would be supplied to test the place of effective management argument under the [treaty]. If, in the light of all relevant evidence, DC could concede that the place of effective management had been in NZ while the trustees were resident there, the point could be conceded on all similar cases.”
Mr Wood proposed an agreement to proceed with a test case covering all “the trusts which had exploited the [treaty]”. Lansburys suggested that the test case be Mr E. Mr Wood’s note of the meeting says:
“The points at issue common to the UK/NZ schemes were: -
[1. A point relating to trusts that had been in the UK before the planning and which is therefore not relevant to the appellants.]
2. Interpretation of Article 4(1) of the DTA – does Article 4(3) apply?
3. If Article 4(3) does apply, where was the place of effective management while the trustees were based in NZ”
After initial hesitation, Mr Bentley accepted that the “trusts which had exploited the [treaty]” would have included the appellants’ trusts. He did not accept that if Mr Wood had been concerned with the claims issue it would have been described as another point at issue common to all the UK/NZ schemes. He said that (a) “they went very quickly into discussion about the application of the [treaty]”, and the technicalities. He could not see that there had been a forensic analysis done of the returns and at some point there would have needed to have been that kind of forensic assessment so they could point to a document that was a claim, (b) that analysis happened when he took over the enquiry, and (c) this is an avoidance arrangement. The discussion appears to have proceeded on the basis of an assumption that it had been implemented correctly, but when it comes to avoidance it is not unreasonable for HMRC to check on implementation and that is where there appears to have been a failing in these cases. He accepted that the impression a reader would have from the notes of the meeting is that there was no procedural issue with relief under the treaty at that time, and Mr Wood did not take any such issue in the following correspondence. He added that until an enquiry is closed points can arise.
It appears that at this point, as HMRC were dealing with the RTW planning implemented by Lansburys’ clients in the round and identifying a lead case, they were not specifically focussing on issues such as whether formal claims had been made in particular cases. However, it is again apparent that they were proceeding on the basis that the only issue, as regards the RTW planning, was the application and interpretation of article 4 of the treaty.
On 7 November 2005, Mr Wood wrote to Lansburys agreeing to proceed with a test case (Mr E). He asked for information to allow him to come to a view on the POEM of the E Trusts. Mr Morris referred specifically to this letter in his witness statement noting that HMRC set out the issues in this letter and none of them referred to procedural steps.
On 17 November 2005, Lansburys confirmed to Mr Wood that they were collating the requested information.
On 13 January 2006, Mr Wood wrote to Lansburys thanking them for the documents he had received and asking for more information relating to POEM.
On 31 January 2006, Mr Blower wrote to HMRC to ask for confirmation of his understanding of the November meeting. He said “you confirmed you wish to raise only three technical points, namely ...” none of which related to procedural issues/claims.
Mr Bentley first said that the claims issue was a “non-technical point”. He later accepted that an HMRC officer receiving this letter should have replied making it clear that there might have been wider points in dispute on which he was reserving his position if he wanted to be able to raise the claims issue later.
Between January 2006 and August 2006 Mr Wood and Lansburys exchanged correspondence about the application of the treaty.
On 13 February 2006, Mr Wood wrote to Lansburys noting that s 788 ICTA allows relief as appropriate from UK taxation in order to avoid double taxation by another territory and “the instrument of relief from double taxation under s 788 is the treaty”.
Mr Bentley said at this stage there was a difference of view as to whether the POEM test was in point at all as Lansburys’ position was that the trust was only resident in NZ so that there was no dual residence and “it is an incidence of dual residence and if you want to claim treaty relief then that claim needs to be met”. He then accepted that Mr Wood did not reserve his position or offer any caveat and, in the subsequent correspondence with him, there was no suggestion that a claim had not been made and he did not reserve his position. He said Mr Wood appeared to be focused on the technicalities and the application of the treaty was the sole focus of the discussion.
Mr Morris thought it was likely that Mr Blower looked at s 788 at the time. He could not recall any conversation with him about s 788(6) which refers to how to make claims under s 788. It was put to him that there is no evidence that he or Mr Blower turned their minds to s 788(6) after receiving this letter. He pointed out that in his witness statement he referred to this letter and noted that HMRC described in it how, in their view, relief from double taxation would be given in cases involving this type of tax planning but “at no point did HMRC express any concerns that such a claim had not been made”.
They arranged another meeting on 5 October 2006, which appears to have taken place because on 6 March 2007 Lansburys wrote to HMRC following up on discussions at the meeting. The bundles did not contain a copy of any notes from that meeting.
On 8 March 2007, Mr Wood replied saying that he had “submitted the papers for consideration by the appropriate HMRC specialist, who feels that the outcome of the litigated cases will be relevant to this case, DTA wording notwithstanding”. The litigated case referred to is likely to be Smallwood, which was heard by the Special Commissioners in December 2007. There appears to have been a substantial pause in the correspondence relating to Mr E and Lansburys’ cases generally while the litigation in Smallwood proceeded.
Correspondence in 2007 relating to the Murphys
There was the following correspondence in 2007 relating specifically to the Murphys:
On 9 August 2007, Lansburys wrote to Mr Wood on behalf of the Murphys referring to letters of 6, 9 and 19 July 2007. The bundles did not contain copies of these earlier letters. In this letter Lansburys said:
“for your information -
1. On 1 March 2002 the non-resident trustees of the settlements of the above individuals retired in favour of trustees tax resident in [NZ].
2. On 20 March 2002 the [NZ] trustees sold their holdings in non-resident companies wholly owned by them for market value to an arms-length purchaser.
3. On 3 April 2002 the [NZ] trustees retired in favour of UK tax resident trustees.
We think you will gauge from the above why in our opinion Section 739 ICTA 1988 does not apply to the above individuals for 2002/03.
As we are already in correspondence regarding what effective management means as regards trustee corporations under the [treaty] can these cases be left in abeyance pending the outcome of a case you are taking to the Special Commissioners later this year under the Mauritius/UK double tax treaty. You are already well versed in our arguments as to why chargeable gains realised by [NZ] resident trustees are not liable to CGT (or more correctly cannot be imputed through to the relevant settlors).”
As Mr Morris accepted, this was the first time that HMRC were made aware specifically of the RTW planning involving the Murphy Trusts. He also accepted that in this letter he/Lansburys were not making a claim for relief under the treaty (in the way HMRC require) and there was no attempt in this letter to quantify how much CGT was at stake in the 2001/2002 tax year in respect of the disposals. We note, however, that as Mr Bentley accepted, it is clear from this letter that the Murphys were specifically seeking to rely on relief under the treaty as regards CGT otherwise due in respect of the disposals.
On 3 October 2007, Mr Wood replied in respect of TM. Letters in the same terms were sent in respect of the other three Murphys. In his letter, Mr Wood said: “Whatever our opposing views of the application of the [treaty], we can I [sic] agree I think that the facts are crucial to determining the matter” and he requested documents clearly intended to address POEM of the trusts. He concluded:
“Please note that the time limit for the making of an assessment for 2001/02 expires 31 January 2008. I propose, therefore, to make shortly a Capital Gains Tax assessment for that year on the basis that S86 TCGA 1992 will apply to the gain realised by the trustees.”
As Mr Bentley accepted, it is clear from Mr Wood’s reply that he understood that the Murphys were relying on relief under the treaty and a reasonable person reading this letter and aware of all the correspondence preceding it, would understand that the debate with Mr Wood was about how the treaty applied and not whether it applied at all. Mr Bentley said that the letter is focused very much on the technicalities, the assessments were raised on the basis that s 77 applies and he thought there was an assumption that the planning was implemented as designed. He accepted that there was no mention of the need to claim treaty relief in the other correspondence with Mr Wood set out below.
It was put to Mr Morris that in these letters Mr Wood said that the relevant disposal by the trustees was previously unknown to him. Mr Morris accepted that that is consistent with the position that no trustee tax returns were submitted which contained information about the disposals.
On 19 October 2007, HMRC issued assessments to each of the Murphys in respect of CGT on the disposals.
On 12 November 2007, AM appealed against his assessment. On 14 November 2007, the other three Murphys appealed against their assessments. In each case the grounds of appeal were stated as follows:
“1. The assessment is invalid as it has been issued out of time.
2. If not invalid is excessive as any chargeable gain would be considerably lower, and
3. In any event the chargeable gain is not liable to capital gains tax under the provisions of [the treaty] for reasons well known to HMRC and which are the subject of communication with Peter Wood...”
Each of the appeal letters referred to a specific sum of tax which the Murphys requested to be postponed.
It was put to Mr Morris that, in these notices of appeal and related correspondence, he/his clients intended to appeal against the relevant assessment and postpone the amount of tax and not to make a claim to relief under s 788 ICTA. He said he had already assumed that by way of correspondence with HMRC since 2001 they had accepted that claims could have been made, and if they thought otherwise they could easily have told him within those five/six years that it was not in the form that they wanted and he/Lansburys would have done what HMRC wanted: “why carry on all these years…if you believe there was no claims? We've wasted 20 years, 25 years of my life”. When pressed he maintained that “we thought that if…there wasn’t a claim…this is a claim” and that there were already claims in place but if not this could be treated as such. He then accepted that he could not have thought that in 2007 as, at that time, he did not think it was necessary to make a formal claim; he said he was getting mixed up.
Following further correspondence, none of which suggested there was any doubt that the treaty was in point in relation to the Murphys, on 3 April 2008, Mr Morris wrote to Mr Wood asking whether the Murphys’ appeals could be left “in abeyance pending what might transpire in relation to other [NZ] treaty matters”. On 8 April 2008, Mr Wood wrote to Mr Morris agreeing that the Murphy cases could be suspended on this basis.
Correspondence concerning all Lansbury clients from 2008 to 2012
There is no record in the bundles of further correspondence relating specifically to Mr E, DH or the Murphys in this period. However, in a letter dated 23 February 2010, Mr Blower wrote to Mr Wood’s successor, Mr Ian Keenan, that:
“In his letter of 2nd September 2008, your predecessor confirmed that if the taxpayers win the “time of disposal” residence argument in the Smallwood case the taxpayers win in the New Zealand cases.”
It is possible that this is the letter quoted by Mr Morris in his letter dated 1 September 2011, in which he said this:
“A predecessor of yours Peter Wood stated -
“the points of contention between us in the [NZ] cases are -
1. Article l4(4) – whether the trustees were resident in [NZ] only at the time of alienation or whether Article l4(4) has no application in a case of serial residence in both contracting states and cannot deny the country of residence taxing rights. Any double taxation arising can be relieved under Article 22.
2. If neither of the above applies, whether the trustees were indeed resident with the meaning of Article 4(3);
3. If so the place of effective management of the trust under Article 4(3).
If either side wins on 1 that will decide the matter for all DTA schemes without recourse to 3.
2 is particular to NZ cases but will only be relevant if neither of the arguments at 1 are accepted.
If you win on 2, 3 will be irrelevant.””
When shown these two documents, Mr Bentley accepted that Mr Blower and Mr Morris both appear to have relied on statements by Mr Wood that the particular interpretation of the treaty would lead to a win for all their NZ clients – including, therefore, the Murphys and DH.
On 3 August 2009, Mr Keenan wrote to Lansburys saying he had taken over Mr Wood’s cases following his retirement and that he had “reviewed the case papers” (for at least the E Trusts). In a later letter, he describes this review as “very detailed”. Mr Keenan said that he had identified “a number of additional issues that need to be clarified regarding this and other related cases before matters can be taken forward”. It is clear that those additional issues did not include determining whether valid claims had been made because on the following page Mr Keenan wrote:
“The trustees of both settlements claim that their respective disposals are exempt from capital gains tax under the [treaty].” (Emphasis added.)
Between August 2009 and February 2010 Mr Keenan and Lansburys exchanged correspondence about the application of the treaty. Mr Bentley accepted that (1) all the correspondence between Lansburys and Mr Keenan proceeded on the basis that the treaty applied to the cases in question and the only question was about how it applied, (2) it is apparent from the correspondence that Mr Blower and Mr Morris had relied on statements by Mr Wood saying that interpretation of the treaty would lead to a win for all their clients
There was the following further correspondence from July 2010 onwards:
On 20 July 2010, Mr Hugh McGillivray wrote to Lansburys saying that he had now taken over from Mr Keenan and that since HMRC’s appeal had been allowed by the Court of Appeal in Smallwood CA he would like to agree the amounts of gains and tax that were in dispute.
On 27 July 2010, Lansburys replied explaining why they believed the different wording in the treaty resulted in a different approach from the UK/Mauritius treaty. Mr McGillivray did not reply to that letter until August 2011.
On 26 April 2011, Mr McGillivray wrote to Lansburys saying that HMRC were “proposing a general meeting with professional representatives on behalf of those persons implementing the scheme to consider any options” for settlement and:
“As you will appreciate from previous correspondence, the Revenue view is that your clients have made a claim and it is for them to show to the Revenue that the point of effective management is NOT IN THE UK.” (Emphasis added.)
Mr Bentley said this letter was sent to Mr Morris and a number of other agents in light of a meeting designed to try and carve a path to settlement when it came to RTW planning; it was a template letter. He accepted that (a) the reference to the claim was not being made in a non-technical sense but said he did not think it was deliberately aimed at Mr Morris and his clients, and (b) it could be read as meaning that Mr McGillivray accepted that the recipients of the letter had made a claim.
In our view, a reasonable recipient of this letter, reading it in the context of the previous correspondence, would take it as meaning that HMRC accepted that Lansburys’ clients who had implemented the RTW planning had done what was necessary from a procedural perspective to make a claim to be entitled to treaty relief – such that the focus was on them demonstrating that the test for relief to apply was met.
On 27 May 2011, Lansburys wrote to Mr McGillivray explaining again why Lansburys believed the different wording in the treaty led to a different result from that under the Mauritius treaty.
On 15 July 2011 or 18 July 2011, Mr Morris met with representatives of HMRC including Mr McGillivray to discuss the RTW cases on which Lansburys had advised. HMRC’s notes of that meeting record Mr McGillivray saying that Mr Keenan’s letter dated 3 August 2009 “outlined the position”. Mr Bentley accepted that Mr McGillivray did not caveat or row back from the position stated earlier by Mr Keenan and he appeared to take exactly the same view as Mr Keenan and that is what one would expect Mr Morris to have taken away from the meeting.
On 19 July 2011, Mr McGillivray wrote to Lansburys following up on the points discussed in the meeting. This letter does not suggest HMRC’s view has changed from that set out in Mr Keenan’s letter dated 3 August 2009.
In the further correspondence Mr McGillivray did not reserve his position as regards the claims issue.
On 3 August 2011, Mr David Lovell wrote to Lansburys setting out HMRC’s analysis of the treaty and why they believed it functioned identically to the Mauritius treaty.
Between August and November 2011 Lansburys and Mr McGillivray exchanged further correspondence about the application of the treaty. A list, which appears to have been attached to a letter from Mr McGillivray dated 14 September 2011, shows that the cases covered by this correspondence include the appellants, Mr O and Mr E. In that letter, Mr McGillivray wrote:
“We are of the opinion that Point [sic] of Effective Management (POEM) has to be considered in the cases for which Lansburys are the accountants…we now need to consider all of your clients’ claims under the DTA that the POEM of their trusts was not the UK.” (Emphasis added.)
We make the same comment on this letter as that made in (3) above.
On 7 November 2011, Mr Paul Callaway wrote to Lansburys saying he had taken over from Mr McGillivray and would be reviewing Lansburys’ cases. Between November 2011 and November 2012 Lansburys and Mr Callaway exchanged correspondence about Mr Callaway taking over the appeals, his plan to conduct a review of the papers and the application of the treaty (as well as some additional issues that do not apply to the appellants). There were delays in this correspondence due to health issues for Mr Morris and Mr Blower’s wife.
Lists attached to two letters from Mr Callaway dated 8 and 9 August 2012 show that the cases covered by this correspondence include the appellants, Mr O and Mr E. This correspondence plainly proceeded on the basis that the treaty applied. For example, Mr Callaway’s letter of 9 August 2012 stated:
“I refer to earlier correspondence with your agent in respect of a claim for relief under the [treaty] in relation to gains that arose in trusts for which you are the settlor.”
As Mr Bentley accepted, the recipient of this letter would have understood that the only issue between the parties was substantive, and that there were no procedural issues.
Mr Morris referred to this letter in his witness statement as further basis for his view that it was common ground that there were no procedural issues. At the hearing he said: “we consciously believed, based on…all the letters that flowed for all that long period that claims had been made. They may not have been made in the form that…HMRC think they should be, but they were definitely accepted as such...by the inspectors dealing with the case”. When it was put to him that his subjective belief at the time was that one did not need to make a claim, he said he may not have believed they needed to make a claim (which we take to mean a formal claim) but he took it that through all those letters that HMRC had accepted that there was a claim, or they would have said they did not accept that and that they wanted the claim in a particular form. He said, in effect, that it may be true that his understanding of the law was that one did not need to make a claim but he “can still have the opinion that claims were made even if in my view was maybe they needn’t be, at an early stage”.
It was put to Mr Morris that given this letter came after the deadlines for claims to be made, this letter cannot have played any part in his thinking about whether to make a claim within the deadline. He said that he can only assume that he believed that what had “passed between us before then was tantamount to a claim”. When pressed he said, he could not recall. He accepted that this follows as a matter of logic and that he knew that identical letters were sent to a number of people who had implemented the planning, and the purpose of the letter was to set out HMRC’s position on a case they were litigating, Smallwood.
We have commented on Mr Morris’ evidence below.
On 13 November 2012, Mr Morris and Mr David Ewart KC met with Mr Callaway and other representatives of HMRC. At this meeting HMRC suggested for the first time in respect of any of Lansburys’ clients that valid claims had not been made. HMRC’s meeting notes state:
“[Mr Callaway] passed over copies of Returns covering the [E] cases and said he could not identify where a claim to DTA had been made. [An HMRC officer] confirmed that a claim was required. [Mr Callaway] suggested that if the claim ought to have been made by the Trusts on departure then it ought to have been repeated on the Settlor’s Return as part of the necessary disclosure (to put HMRC ‘on notice’ as to what was taking place and why no Section 77 gains were being reported). [Mr Morris and Mr Ewart] will review the position.”
On 8 January 2013, Mr Callaway wrote to Lansburys setting out the “Action Points” that he understood had been agreed at the 13 November 2012 meeting which included: “8. You will review & report on the position of DTA claims made by the Trusts and Settlors.” On 21 January 2013, Mr Callaway wrote to Lansburys enclosing HMRC’s notes of the 13 November 2012 meeting and repeated his record of the “Actions Points”.
On 28 May 2013, Lansburys wrote to Mr Callaway with comments on HMRC’s meeting notes including that:
“Based on the returns submitted and the correspondence that has taken place between us over the last ten years we cannot see how you could possibly say there is a “claims position”.
In a letter dated 25 June 2013 Mr Callaway acknowledged receipt of the letter of 28 May 2013 and said he would respond substantively. Mr Bentley accepted that even though Mr Callaway continued to be in charge of these appeals for more than 18 months, until early 2015, he did not appear to have replied to Lansburys on the claims issue.
During 2013 HMRC sought, and Lansburys supplied, documents relating to all of Lansburys’ RTW cases.
Mr Morris was taken to the notes of the meeting of 13 November 2012, at which he was present. He accepted that (1) as recorded, at the date of the meeting, HMRC were still awaiting a reply to a letter of August 2009 and the purpose of this meeting was to “take stock of current position [and] discuss ..technical issues”, (2) from this date onwards he understood HMRC’s position was that a claim needed to be made for treaty relief, and (3) action point 8 identified that Lansburys were to review and report on the position as regards claims as was repeated in a number of pieces of correspondence, and (4) despite these reminders, he/Lansburys did not identify a claim to HMRC in terms of pointing to a document in which the claim is made. He added in effect that various health issues of those involved made dealing with this difficult and that he thought there was a letter in response which reflected advice from Mr Ewart.
He was taken to the correspondence in 2013 in which HMRC served an information notice which included a request for a copy of the claim by the trustees of the relevant Trusts for relief under the treaty. He accepted that (1) from this it was very clear that in 2013 HMRC asked him/Lansburys to provide a copy of any claim, and (2) in Lansburys’ response they did not identify the claim or claims they considered had been made but rather said it is not open to HMRC to take a claims position. When it was put to him he did not identify the claims because he/Lansburys knew at the time that the trustees and/or settlors had never made a claim under s 788 he said: “ Not in that form, no” and if they had made something he understood at the time to be a claim, he would have provided it as the claim.
In his statement Mr Morris noted that the claims issue was not pursued by any of the technical experts at HMRC in the meeting held in November 2012. He also said that between mid-2013 and 2018 HMRC did not raise any significant concerns as regards the claims issue despite a forensic analysis of all the information and documents he supplied in response to the information notices, which he dealt with in full. He said he took HMRC’s silence on the claims issue during this period to mean that they were not raising it and were satisfied with the procedural aspects of these appeals. It was put to him that in the record of the meeting in November 2012 there is no statement from any of the HMRC officers to the effect that everything had been implemented correctly and nothing that would lead him to believe that there could not be any implementation issues. He pointed to the following comment in the note relating to a later meeting in May 2017:
“TM expressed his concern about the way HMRC had handled matters. He pointed out that he had sent all of the documents and details his clients could get and he had also provided CGT computations. TM went on to say that he had not heard back from HMRC on these matters for a very long time. He said as far as he was concerned he did everything HMRC had asked. He added that a test case originally identified in 2005 should have been taken forward to test Lansbury’s [sic] contentions concerning the different DTAs. He pointed out that had this been done we wouldn’t be in the position now. TM stressed that his clients had been treated unfairly by HMRC. CW agreed that he’d come to the same conclusion when completing the reviews and apologised on behalf of HMRC. JB echoed CW’s comments and agreed that the case had not been handled correctly. CW went on to say that HMRC had missed opportunities to progress the cases and this fell short of what the clients might expect. JB confirmed that HM wanted to avoid any more loss of momentum and move forward with this case, and is committed to move these cases towards settlement.
JB said that in HMRC’s view if it’s accepted that there’s no difference in the DTAs ultimately it comes down to POEM. TM commented that after seeking advice if he needs to reconsider POEM then he will do so then see where we differ adding CW commented that from his review the detailed issues were covered but not agreed.”
Mr Morris said, in effect, that he left that meeting in November 2012 thinking that HMRC were saying there was an issue as to whether the treaty applied in terms of residency, and if it did, then an issue on POEM. He accepted there was no discussion at all in this meeting about the Murphy Trusts and the Harris Trust and the focus was on another case which had been identified as a lead case. When it was put to him that one cannot take from the discussion on that case any assurance that HMRC accepted a claim had been made in these cases, he said he still looked at them collectively and when he went to this meeting he was looking at all outstanding cases. He wanted to clear up all of them, if possible. He added that for five years nothing was said in response to the letter of 2013 as a “retort to what we said about the claim issue”. He accepted that during 2013 to 2015, there was no correspondence at all on these cases. When it was put to him that these appeals were not actively pursued by either side during this period, he said he would have thought that “if [HMRC] got something that’s five years outstanding and they’ve not done anything it would be unusual”. He said he genuinely believed at the time they entered into these transactions that they were dealing with just good future tax planning, nothing more, avoidance never came into it, it was just purely good forward-thinking tax affairs. He did not accept that this scheme is a form of tax avoidance. He said that to him it is tax planning: “the consequences may be that you avoid tax, but I see it as tax planning, which never used to be frowned upon”. He then accepted, in effect, that the plan was to avoid CGT but said “but in my view it arose out of good tax planning” and he just had a different view of it.
Mr Morris seemed to accept that there is no duty on HMRC to point out every implementation error made by an advisor. In his view, however, when corresponding HMRC could have said that they cannot accept that as a claim unless it is in a formal way but that was never ever mentioned at all. He thought it was probably correct generally that an implementation error may not be clear at the start of an enquiry by HMRC and may become clear only later on. He said in this case there were regular letters going all the time and not once was it ever said that HMRC were considering that there was no claim, and the correspondence on the substantive issue started very quickly. He considers that HMRC believed the claims had been made.
In his statement Mr Bentley said it cannot be said that all correspondence proceeded on the basis that claims for treaty relief had been made for the relevant taxpayers. At the hearing he accepted, however, that the application of the treaty was a subject of significant discussion and, on the RTW element of the planning, it was the only issue that was the subject of discussion. He added the fact is, for the treaty to apply, there needs to be a claim and the claim has not been made in these cases. He confirmed that he was not aware of any instance of HMRC expressing doubt about whether a claim had been made before November 2012
In his statement Mr Bentley also said that Mr Morris’ statement that claims would have been made if HMRC had been told him this was necessary is contradicted by the fact that Mr Morris did not attempt to submit a claim when Mr Callaway told him on 13 November 2012 that he could not identify where a claim to relief had been made. He clarified that it is not his view that valid claims could have been made in November 2012 or later. He said his comment was a direct response to Mr Morris’ statement - he was told of the need to make a claim in November 2012 and nothing happened. Now he could only speculate as to how HMRC might have responded had a claim been made at that time. He suggested that HMRC may have allowed a late claim. He accepted that all of Lansbury’s clients to whom a closure notice was issued attempted to make consequential claims for treaty relief but HMRC did not accept those claims.
Correspondence concerning all Lansbury clients from 2014 onward
During 2014 and 2015 the parties corresponded about the quantum of CGT that would be due if the RTW planning was unsuccessful. In early 2015, Mr Richard Griffiths took over from Mr Callaway. By June 2015, the claims issue as regards Lansburys’ clients appears to have been forgotten or set aside by HMRC:
On 30 June 2015, Mr Griffiths wrote to Lansburys saying “our position remains as stated in our letter to your clients on 8 August 2012; the tribunal will reach the same conclusion as in the Smallwood judgment”. This is presumably a reference to the letters that were sent to Lansburys’ clients on 9 August 2012 (a draft of which was sent to Lansburys on 8 August 2012). These letters begin “I refer to earlier correspondence with your agent in respect of a claim for relief under the Double Taxation Agreement in relation to gains that arose in trusts for which you are the Settlor”.
As Mr Bentley accepted, Mr Griffiths’ reference to the position set out in the 9 August 2012 letter could fairly lead a reasonable reader to the conclusion that HMRC had dropped the claims issue.
On 30 November 2015, HMRC wrote to many of Lansburys’ clients, including the Murphys, inviting them to settle. Again, these letters stated HMRC’s belief that Smallwood applied and did not refer to the claims issue.
There then appears to have been another substantial pause in correspondence until early 2017:
On 28 February 2017, Mr Bentley called Mr Morris to say he had become the latest Compliance Lead for RTW cases.
On 31 May 2017, Mr Morris met with Mr Bentley and two other HMRC officers. Mr Bentley informed Mr Morris that HMRC had reviewed the available documents in Lansburys’ cases. The meeting notes do not record any discussion of the claims issue, but record Mr Bentley saying on several occasions that the cases were likely to be determined by the application of the treaty:
“[Mr Bentley] said that by agreeing [Mr Morris’] position on the options it leads to considering the NZ DTA”
(b “[Mr Bentley] said that in HMRC's view if its [sic] accepted that there's no difference in the DTAs ultimately it comes to POEM”
“[Mr Bentley] said that in his view the cases will come down to POEM in the end”
“[Mr Bentley] added that regardless of whether the shares have been sold there will be occasions of charge in most cases which will give rise to CGT so the matter boils down to POEM and when these occurred”
Mr Bentley accepted that between June 2013 and September 2017 HMRC did not raise any significant concerns about the claims issue and confirmed that at the time of the May 2017 meeting he was unaware of the claims issue. Mr Morris said in his statement that for him this meeting further confirmed that the implementation of the RTW planning was correct as did the fact that nothing was raised between 2005 and 2012 and, with a couple of exceptions, all the correspondence before 2018 proceeded on the basis that the only issues between the parties were the substantive ones and HMRC did not raise the claims issue at this meeting.
In his statement Mr Bentley said he did not agree that the fact that HMRC did not raise the claims issue between 2013 and 2017 is an indication that HMRC were satisfied and that HMRC never indicated they were so satisfied. At the hearing:
He said he was not aware whether Mr Callaway responded to Mr Morris on the claims point but noted that he was soon replaced as the HMRC lead on these cases. He accepted that Mr Morris then sent a lot of information to HMRC over the summer of 2013 and Mr Callaway did not write to Lansburys stating the point about claims still needed to be addressed. It was put to him that Mr Griffiths appears to have taken over in January 2015 and in his letter he said that the position remains as stated in the letter of 8 August 2012. He was asked if he could see why Mr Morris would have believed that HMRC were satisfied on the claims issue. He said, having had a successful decision in Smallwood, HMRC believed that the POEM of the trusts would demonstrate that chargeability would fall in the UK. He then said he can see why Mr Morris may think that and accepted it was a fair conclusion for him to have reached.
He confirmed that there was no discussion of the claims position at the meeting of May 2017. He said this meeting took place in the context of him having recently taken over responsibility for the enquiries, and its main purpose was to reinvigorate the Lansburys enquiries, in light of a recent decision in a RTW planning case. They had completed reviews of the enquiry correspondence. They had not had any meaningful discussion with Mr Morris for some time, they were aware there were differences between the Mauritius cases and Lansburys’ cases and they wanted to get to grips with how the Lansburys’ iteration of the planning was designed to work.
When asked why there was no discussion of the claims issue at this meeting, he said “because the discussion was setting the scene…in light of recent litigation. It was having reviewed…the files, but we were more concerned with the level of documentation we had received…What we wanted to do was to get ourselves in…a clean deck…of cases to take forward in as an effective way as possible”. He said, in effect, that he did not believe HMRC/the relevant officers were aware of the claims issue when they had this meeting; it emerged subsequently. He accepted that the review undertaken by his colleagues did not identify the claims issue at this stage but said it was not focused on the implementation but on the technicalities. He said that (a) it was possible that to Mr Morris this might have suggested that HMRC were no longer taking the point but (b) if that was the case, HMRC corrected that understanding at the earliest opportunity subsequent to that meeting. He accepted that this meeting could have implied to Mr Morris that HMRC accepted the treaty applied but not that it could be taken from this that HMRC had dropped the claims point because they had not explicitly referenced it. He noted that HMRC did not say they had considered the claims point and dropped it but rather that these were the matters that they had considered. HMRC’s position at the time was and still is that the trusts were dual resident and hence it would come down to POEM. At the time they had not checked and assured themselves that the scheme had been implemented correctly.
He accepted that (a) in his statement he meant that HMRC have never said in terms they had considered the claims issue and, having done so, accepted that claims had been made, (b) they did say lots of times that there were claims, and (c) at one point he thought that a claim had to be made in a return.
On 25 October 2017, Mr Carl Whitehead, a Compliance Caseworker, wrote to Lansburys returning to the claims issue for what appears to be the first time in four and a half years. His comments included that (a) while researching this matter he had attempted to locate “the formal DTA claims under s788 ICTA in respect of certain disposals on the lead case, (b) he could not find the appropriate claims, (c) relief under the treaty cannot be accessed unless claims for relief are made to the Board under s788(6), (d) he had a brief look at a number of other cases on Lansburys’ client list but had been unable to locate claims for these either, (e) before “we can satisfactorily deal with the issues surrounding the effects of the DTA it’s essential to establish that the DTA has been brought into effect”, and (f) “For each of the ongoing cases please would you confirm whether formal claims to the Board have been made and if so by whom and the date these were submitted?”
On 13 December 2017, Lansburys provided a detailed response to Mr Whitehead’s letter concluding:
“Having reviewed correspondence over the last 15 years it is clear HMRC were fully aware of what claims were being made, the treaty issues involved and how they impacted the trusts and settlors.”
From this time, HMRC’s correspondence was focused to a much greater extent on the claims issue.
In his statement Mr Bentley said it is common for different issues to be raised at different points of an enquiry and from his review of the correspondence HMRC were still engaged in technical discussions. At the hearing he said, in effect, that, in his experience, it is not common for issues to be first raised 11.5 years into an enquiry. He was asked if it is usual for HMRC to note that there may be further arguments where they want to keep their options open. He said it is not usual for enquiries to go on for this length of time but it can happen in light of a normal enquiry that issues emerge as facts emerge. He accepted that the best approach is to flag up that there may be further issues and it is not common for an officer not to raise such a point if, several years into an enquiry, an officer agrees to set out all their arguments and takes six months consulting with colleagues with the stated intent to do so. He added that there have been a number of pairs of hands that the enquiries went through and in some instances arguments will be picked up and just continued with; in others, if somebody spots something new, they will raise it. He accepted that where an officer said they are setting out everything it is not common for a new issue to be raised seven years later and any suggestion that this process fitted in with HMRC’s usual or common approach is not right. He added “but I mean the enquiries don’t fit in with HMRC’s normal approach”.
Submissions and decision on discovery issue
HMRC made the following main points:
Neither DH’s personal tax return for the tax year 2000/01 nor that submitted by the trustees of the Harris Trust for that tax year make any reference to the disposal.
It is irrelevant what prompted Mr Robertson to open an enquiry into the trustees’ tax return in 2002. For the purposes of s 29(5) TMA information is only made available to an officer if it falls within one of the exhaustive categories in s 29(6). There is no evidence of any information that falls within this save for DH’s personal tax return which makes no reference to the disposal.
Moreover it is highly unlikely that DH provided any information regarding the RTW planning. The consistent position of Mr Morris was that information about the planning did not need to be provided to HMRC by the settlors. That was the position in respect of the Murphys and the advice that Mr Morris gave to BDO who represented another user of the planning.
Mr Robertson’s letter and memo of 23 December 2003 make clear precisely why he decided to issue the assessment and in doing so they illustrate the discovery he had made, namely that the Harris Trust had made a disposal on which it had made a gain, neither that disposal nor the gain were included in DH’s 2000/01 tax return and DH was chargeable to CGT as a consequence of that gain under s 77 as succinctly set out on the memo of the same day.
Hence, Mr Robertson made the discovery when it newly appeared to him that there was an insufficiency in DH’s self-assessment and once his belief that there had been an insufficiency to tax had come to be more than a suspicion. Nothing turns on the precise point in time at which Mr Robertson became sufficiently confident that there had been an insufficiency to tax so as to satisfy the test in s29(1) TMA. A discovery was plainly made at some point before 6 January 2004 when the assessment was issued.
The appellants submitted that HMRC have not shown that there was a discovery or that the test in s 29(5) is met. In their view, as in Strategic Branding, the fact that a discovery assessment was issued does not of itself suffice to demonstrate that a discovery was made; that is all that HMRC’s contentions amount to. They noted that Mr Bentley has accepted that he cannot provide evidence about the decision to issue the assessment to DH. In their view there is no evidence of the knowledge of the particular officer who issued the assessment nor of any other officer said to have made a relevant discovery: (1) the letter in which Mr Robertson gave notice he would issue an assessment is silent as regards the state of mind of Mr Robertson and his knowledge. It states only that tax may be due. It is not evidence that Mr Robertson had formed the view there was an insufficiency of tax, and (2) similarly the letter he sent to Mr Langton asking for an assessment to be issued does not give sufficient insight into his state of mind and knowledge.
As regards s 29(5) the appellants submitted that:
The enquiry into the Harris Trust’s tax return for 2000/01 was opened on the basis that the trustees undertook RTW planning and the questions set out by Mr Robertson in his letter were clearly focused on POEM and identify all the trustees and relevant disposal of shares. Accordingly, that letter demonstrates that Mr Robertson had knowledge of the circumstances of the RTW planning. However, it is apparent that that knowledge could not have been gleaned from the returns HMRC received and there are no other documents in the bundle produced by HMRC which demonstrate where this knowledge came from.
As regards the returns made: (a) there is no mention of the trustees holding the relevant shares or the disposal in DH’s personal tax return for the tax year 2000/01 and there is no evidence as to what further letters or information was sent by DH or on his behalf in respect of that tax year, (b) HMRC have only been able to produce “screen prints” of the trustees return for that year. It was originally a paper one and these prints are merely the product of HMRC’s own inputting manually into their systems certain information. Mr Bentley accepted that he cannot give any evidence as to the inputting of information, (c) when Mr Robertson opened the enquiry into the Harris Trust’s tax return for 2000/01 he also sent the letter and schedule to the adviser who was on the record as the trust’s adviser and who was also DH’s agent, and (d) Mr Bentley says that a file of documents relating to DH would have been maintained by a regional office but that file no longer exists and it is not clear from the documents available (acknowledging they are not all available) why the enquiry into the Harris Trust’s return was opened and no enquiry was opened into the DH return.
In light of the above, it is therefore impossible to tell where Mr Robertson obtained knowledge that the Harris Trust engaged in RTW planning. HMRC are unaware of what documents Mr Robertson might have considered before opening the enquiry in December 2002. The documents available to the tribunal are incomplete, the enquiry letter shows that other information was provided to HMRC before January 2003 and it is impossible to say whether that information was sent on behalf of DH or the Harris Trust or both, particularly when both had the same agent.
Section 29(5) requires HMRC to keep proper records of the information listed in s 29(6) which was available at the relevant time otherwise the protection offered by those sections is meaningless. It is apparent that there must have been further documents/information which prompted the opening of the enquiry which it was incumbent on HMRC to produce. This is a rare case where the burden of proof comes into play given that HMRC cannot put all the documents and information within s 29(6) before the tribunal, particularly in circumstances where other relevant information must have existed in some format. HMRC have simply failed to meet their burden of proof.
HMRC’s argument is fundamentally inconsistent with the decision in Burgess and misunderstands how the burden of proof works. If HMRC’s evidence was that they had systems in place to ensure they kept everything of relevance when they made the discovery assessments and these are all the documents they had under those systems, the appellants would have to prove there was further information. However, in fact HMRC’s evidence is they probably had further correspondence relating to DH but they have lost it. Hence, the burden of proof is on them. Mr Stone’s submissions on this ground are little more than a concession that HMRC cannot actually meet the burden of proof.
We have concluded that:
For the reasons set out by HMRC, there is sufficient evidence to demonstrate that Mr Robertson made a discovery that, in his view, acting reasonably there was an insufficiency of CGT in respect of the disposal of shares by the Harris Trust. We did not find the decision in Strategic Branding useful given the different factual circumstances in point in that case.
This is a case where HMRC have failed to provide sufficient evidence that the requirement in s 29(5) is met. We note that (a) it is plain that HMRC received information that DH/the Harris Trust had implemented RTW planning in respect of the disposal on the basis that relief was available under the treaty but it is not known where that information came from, (b) HMRC did not retain their file relating to DH’s tax position in respect of the disposal, and (c) HMRC have simply not demonstrated what relevant information/documents they had for the purposes of s 29(5) and (6) TMA by the relevant deadline (31 January 2003). In our view, it would deprive s 29(5) TMA of having any meaningful protection for a taxpayer and render the requirement for HMRC to satisfy the burden of proof on this issue ineffective if, in these circumstances, the tribunal were to seek to plug the gap in the evidence by simply assuming that no relevant documents/information falling within s 29(6) was made available to HMRC by the relevant time, from which a hypothetical officer acting reasonably could have concluded there was an insufficiency of CGT in respect of the disposal made by the Harris Trust. It does not suffice for us to make any such conclusion that there is evidence that generally Lansburys did not advise settlors/trustees to include relevant disposals made under the RTW planning structure in their tax returns. The consistent position of Mr Morris was that information about the RTW planning did need to be provided to HMRC. It is speculation as to what, if anything, of relevance for the purposes of s 29(6) was provided to HMRC but it is for HMRC to keep appropriate records in order to be able to satisfy the burden of proof.
Submissions and decision on the claims issue
As regards DH and the Harris Trust, the appellants submitted that, in light of the statutory requirements and principles set out above, they consider that the documents referred to below show that a claim for relief was made by DH/the trustees of the Harris Trust:
A reasonable officer of HMRC in Mr Robertson’s position would have understood from Allglory’s letter of 21 January 2003 that Allglory was making a claim for treaty relief. Mr Robertson acknowledged that Allglory as trustee of the Harris Trust was relying on the full relief from CGT available under the treaty (most clearly, but not exclusively, in his letter dated 29 July 2004). Mr Robertson did not suggest that a valid claim for relief had not been made and when setting out all the reasons he believed that CGT was due did not refer to the absence of a valid claim.
If the requirement that the claim be quantified means figures must be given, they were given in the letter from Rossiter Smith dated 4 February 2004.
Lansburys, as agent of DH and the Harris Trust, made clear to Mrs March that reliance was being placed on the full relief from CGT available under the treaty and referred Mrs March to Mr Wood.
On the basis of the principles set out above, the appellants submitted that the trustees of the Murphy Trusts made valid claims for relief outside of a return:
The facts set out above show that, the trustees of the Murphy Trusts made clear to Mr Wood that they were relying on the full relief from CGT available under the treaty in Lansburys’ letter of 9 August 2007 and in their notices of appeal on 12 or 14 November 2007. A reasonable HMRC Inspector in Mr Wood’s position would have understood that this was what Lansburys and the Murphys/their trusts were doing.
Mr Wood acknowledged that they were relying on the full relief from CGT available under the treaty in his letter of 3 October 2007 and again when agreeing that the cases could wait behind the outcome of other treaty cases. At no stage did Mr Wood suggest that a valid claim for relief had not been made and instead he made clear that his concern was with the POEM of the Murphy Trusts.
If the requirement that the claim be quantified means figures must be given, they were given in the notices of appeal on 12 or 14 November 2007.
HMRC accepted that the correspondence proceeded on the basis that the application and interpretation of the treaty was in point and the claims issue came to light late in the enquiry process. Mr Stone said it is recognised that it is unfortunate it was not identified sooner but made the following general points: (1) it is a feature of HMRC’s enquiries that they involve an ongoing investigation during the course of which issues will arise and fall away; until an enquiry is complete a taxpayer cannot proceed on the basis that the issues under discussion are the only issues that HMRC will raise, (2) HMRC could tell a particular taxpayer that a particular point will not be taken against them but that did not happen in these cases, (3) Mr Morris accepted that Lansburys were never told by HMRC that claims had been made, (4) the evidence demonstrates that Mr Morris and Mr Paul believed that no claim needed to be made as is demonstrated by their conduct, and (5) after the claims issue was identified at the meeting of 13 November 2012, Mr Morris did not produce anything purporting to be a claim. There was then four years of silence from both parties. When Mr Bentley became involved, he did not immediately identify the issue for the reasons he gave but that only lasted a few months and the claims issue was identified and communicated to the appellants in October 2017.
On the particular correspondence that the appellants rely on as showing they made a claim for treaty relief, HMRC made the following main points:
None of the documents relied on are addressed to the Board or contain a declaration of accuracy.
As regards DH, none of the correspondence relied on contains an active, deliberate and unambiguous claim for relief:
None of the appellants’ witnesses are able to say whether there was any intention to make a claim in any of the letters sent to HMRC pursuant to the enquiry into the Harris Trust or what the effect was of any of the correspondence that was sent by HMRC.
In the letter of 21 January 2003 Allglory simply declines to provide the information sought by HMRC. It is not an attempt to claim relief and cannot reasonably be read as one. It does not purport to quantify any claim. It makes no reference to any specific article of the treaty/form of relief or any specific disposal and does not set out the desired tax treatment. Essentially the same points apply as regards the other letters referred to.
The letter of 13 September 2006 was sent on behalf of DH personally and its purpose was to seek to persuade Mrs March that the information sought from DH did not need to be provided. Mr Morris explained his understanding that the gains could only be taxed in NZ which reflects his belief that claims did not need to be made. This is to be contrasted with the wording that follows in which Mr Morris tells Ms March that in respect of the repatriation of the trust to the UK “hold over relief was claimed under section 165 TCGA 1992”. Given the context of that letter, if Mr Morris wanted to claim relief, he would have said so unambiguously.
The correspondence relied on by the appellants in respect of the Murphy Trusts also does not contain an active, unambiguous and deliberate claim for relief from CGT:
The letter of 9 August 2007 was intended to set out the transactions that took place in 2001/02 with the intention of explaining why no charge arises under s 739 ICTA. It is not, and cannot reasonably be read as, a claim for relief. It was sent on behalf of the settlors and not the trustees of the trusts. It does not refer to any specific transactions and does not purport to quantify either the transactions or the relief required. Mr Morris accepted that he was not making a claim for relief in this letter, there was no attempt to quantify any claim and this was the first time HMRC were made aware of the Murphys’ RTW planning.
None of the appellants’ witnesses have suggested that the notice of appeal was intended to be a claim for relief. On its face, it is no more than an appeal against the notice of assessment and a request for postponement of tax. That is how any reasonable reader would understand it. It was sent on behalf of each appellant personally. The wording of the “reasons of appeal” does not refer to any specific transaction and does not set out the form of relief that is sought and there is no quantification. At the time, Mr Morris did not intend to claim relief from tax. Although a figure is contained in the letter, that is the amount of tax for which complete postponement is sought and there is no quantification by the appellants. The notice of assessment is based on an estimated assessment by HMRC. The appellants did not, by their notice of appeal, accept that estimated figure. There are time limits for opening enquiries into a claim, and if HMRC do not do so, the relief must be given automatically. On the appellants’ view, whenever HMRC receive a notice of appeal, they would have to scour it to determine whether there may potentially be a claim hidden away in it that they have to separately enquire into, otherwise they would have to give effect to it. That is one reason why a claim must be unambiguous because it must be understood by HMRC that that is what is being done; there are serious consequences in terms of time limits and giving effect to claims. As supported by the comments in Post-Prudential if the taxpayer wishes to make a claim he has to say so. A notice of appeal can only be treated as a claim if the taxpayer asks for it to be treated as such.
We have concluded that, having regard to the relevant statutory requirements and the guidance in the caselaw on what is required for the making of claims under those provisions, each of the appellants made a claim for treaty relief at the latest, in the case of the Murphys, when the notices of appeal were submitted and, as regards DH, when the letter of 4 February 2004 was sent. That each of the appellants wished to obtain relief from the CGT charge arising in respect of the disposals under article 14(4) of the treaty is how any reasonable officer of HMRC would interpret the letters and notices of appeals which the appellants referred to, whether viewed in isolation or in combination. In making that assessment, we consider it appropriate to have regard to the background information and knowledge available to a reasonable officer in the course of the preceding correspondence. With that background knowledge, a reasonable officer could not fail to comprehend that Mr Morris/the appellants were seeking to rely on relief under article 14(4) of the treaty to exempt them from any charge on the full amount of CGT which would otherwise be chargeable on the gain in respect of the disposals made by the Trusts. That the actual HMRC officers who dealt with this correspondence were so aware is also readily apparent from the correspondence the appellants referred to.
For the reasons set out in full above, (1) we consider that a claim for treaty relief may be made by the appellants or their agents acting on their behalf, (2) we do not accept that for there to be a valid claim in this context there is any requirement for the relevant documentation relied on to be headed “to the Board” or contain a declaration of accuracy or for the appellants/their agents to be conscious of the need for a “claim” to be submitted, and (3) we consider that it suffices for the quantification requirement to be met that the appellants/their agents made clear that they wished to obtain relief for the full amount of the CGT liability which HMRC otherwise asserted was due in respect of the disposals. Viewing the requirement in its immediate context in the general claims position and in light of the overall self-assessment system, its purpose seems to be to ensure that HMRC know the specific impact the relief would have, if successfully claimed, on the taxpayer’s tax position in the relevant tax year. In these circumstances, it suffices that the appellants/their agents made apparent to HMRC what relief they are relying upon to relieve an identified tax liability in respect of an identified disposal.
On that basis, it is not relevant that Mr Morris did not consider that a formal claim for relief was required in the sense put forward by HMRC and/or that the “claim” was not made in a UK tax return or in a single document stating explicitly it was intended to be a claim for treaty relief. Moreover, to the extent that his intentions are relevant, it is plain that Mr Morris did intend to seek to obtain relief from CGT for the appellants on the disposals under article 14(4) of the treaty. At times under cross-examination Mr Morris was muddled between what he thought at the time as regards the obtaining of treaty relief and what he thought later on once the issue was raised and now. There were lengthy discussions as to his understanding of the legal requirement to make a claim and his intentions in the correspondence. However, he was clear in his evidence that he wished to obtain the benefit of treaty relief for his clients and that throughout the correspondence his understanding was that they were entitled to seek to do so on the basis that the only issues between the parties were the substantive issues
Submissions and decision on estoppel issue
The appellants submitted that, if contrary to the appellants’ case on the claims issue any of them have not made valid claims, HMRC are estopped from contending that the treaty does not apply:
The appellants’ evidence is that their understanding in their dealings with HMRC was that all the correspondence with HMRC proceeded on the basis that the treaty applied and the only issues between the parties were the substantive issues; there were no procedural issues such as whether in fact a claim had been made. This is the only understanding one could have had given the content of the correspondence. Mr Bentley accepted that (a) the correspondence between 2001 to November 2012 proceeded on the basis that the only issues between the parties were regarding the interpretation and application of the treaty, and (b) HMRC had said many times between 2001 and 2017 that there had been claims and accepted that the way these cases were dealt with is not normal and it is uncommon for a point to be raised for the first time 11 years after an enquiry was opened.
In the same way as in Tinkler, HMRC have taken at a late stage what can fairly be described, on the facts of these appeals, as a technical point (that some formality necessary to make a claim was not met) even though that has not caused HMRC any prejudice (see Tinkler at [85]). Tinkler shows that an estoppel by convention will arise to prevent such a technical point succeeding. Indeed, other than HMRC being the estopped party, the appellants cannot see any material distinction between these facts and Tinkler.
The correspondence demonstrates that the parties shared a common understanding that there was no procedural bar to accessing relief under the treaty. It is irrelevant whether this understanding was based on a belief that a valid claim had been made or that a claim did not need to be made. Indeed, it seems impossible to know from the correspondence which Mr Robertson believed. Regardless, there was a shared understanding that the treaty was determinative of the issues between the parties and over the course of a decade there was an enormous amount of correspondence between the parties expressing that view.
HMRC conveyed this understanding to the appellants and expected them to rely upon it, by engaging in extensive discussions about the application of the treaty, opening enquiries into the return for the Harris Trust, requesting substantial information about the POEM of the trusts and not suggesting until many years later that there was any procedural reason why the appellants could not rely on the treaty.
The appellants relied on this common understanding by engaging in years of correspondence with HMRC and, in particular, by not making claims within the time limit for doing so (or perfecting imperfect claims within the time limit for doing so), which they otherwise could have done. That reliance occurred during the ongoing correspondence between the parties relating to the enquiries and discovery assessments.
If no valid claims have been made, the appellants have suffered serious detriment by not making claims they otherwise could have made, by being made liable for tax in respect of which they would otherwise have been able to claim relief and by engaging in almost 20 years of stressful, time consuming and expensive debate with HMRC.
In respect of DH, the following conduct of HMRC indicated that HMRC assented to or shared the understanding that the treaty applied to the disposal made by the Harris Trust, is conduct that HMRC must have expected to be relied upon and is conduct that was relied upon:
HMRC’s letters to Allglory of 8 August 2003 and 14 October 2003 and, in particular, that of 29 July 2004 in which Mr Robertson set out all the reasons for which he considered CGT was due. None of Mr Robertson’s reasons suggest that he doubted the treaty applied.
The meeting between Mr Cameron and Lansburys on 3 March 2005 proceeded on the basis that the application of the treaty was relevant to every person advised by Lansburys who had settled a trust that relocated to NZ.
On 2 November 2005, Mr Wood met with Lansburys and suggested the interpretation of article 4(1) of the treaty was common to every person advised by Lansburys who had settled a trust that relocated to NZ.
Mr Wood’s letter of 13 February 2006 to Lansburys in which he gave no indication there was any issue with claims in respect of Mr E or any other person advised by Lansburys and every indication that he accepted that the interpretation of the treaty determined the CGT liability of Mr E and every other person advised by Lansburys who had settled a trust that relocated to NZ.
In respect of the Murphys, the following conduct of HMRC indicated that HMRC assented to or shared the understanding that the treaty applied to the disposals made by the Murphy Trusts, is conduct that HMRC must have expected to be relied upon and is conduct that was relied upon:
Mr Wood’s statements in his letter of 3 October 2007 and the fact he requested documents clearly intended to address the POEM of the trust and there is no suggestion in that letter that he doubted the applicability of the treaty and the related correspondence in which he agreed that these cases could stand behind the E cases.
Mr Wood’s letter of 8 April 2008 in which he agreed that the Murphys’ appeals could be suspended pending the outcome of other appeals that the parties understood to turn on the application of the treaty. This shows that Mr Blower and Mr Morris relied on statements that a particular interpretation of the treaty would result in a win in the NZ cases including these cases.
Mr Wood’s letter of 2 September 2008 as summarised in Mr Blower’s letter to Mr Keenan of 23 February 2010. This shows Mr Blower and Mr Morris relied on statements that a particular interpretation of the treaty would result in a win in the NZ cases including these cases.
Further, in respect of both Mr Harris and the Murphys, the appellants rely on all the correspondence put to Mr Bentley but, in particular, the following conduct of HMRC indicated that HMRC assented to or shared the understanding that the treaty applied to the disposals is conduct that HMRC must have expected to be relied upon and is conduct that was relied upon:
The letter of 14 September 2011 from Mr McGillivray.
The letter of 9 August 2012 from Mr Callaway to each of the appellants.
The letter of 30 June 2015 in which Mr Griffiths wrote to Lansburys saying that HMRC’s position remained the same as in the letters of 9 August 2012.
The fact that at the meeting on 31 May 2017, Mr Bentley repeatedly stated to Mr Morris that, having conducted a review of the Lansburys cases, in his view the tax position would turn on the interpretation and application of the treaty.
HMRC made the following main submissions:
The tribunal should bear in mind the serious implications, especially in a mass marketed avoidance scheme such as this where multiple issues arise during the course of an enquiry, of a finding that HMRC are estopped from taking a point that no valid claim was made. That would mean that in any case in which HMRC come up with an implementation error later down the line they are in danger of being estopped. That would have severe consequences on HMRC’s handling of enquiries. It is relevant that Mr Morris/Lansburys were marketing tax avoidance schemes, had experience of enquiries and would have known that points arise in the course of an enquiry.
The parties were never operating under a common misunderstanding. If there was any misunderstanding, it was, at most, a mistaken assumption that claims for treaty relief had been made. However, the appellants were not operating under the misunderstanding that claims had been made; they were operating under a misunderstanding that claims for relief did not need to be made. Mr Morris’ understanding at the time was that obtaining the relief was somehow automatic and no formal claim had to be made. The HMRC officers who did not raise an issue about a claim for the Murphy Trusts must either have assumed there was a claim or overlooked the requirement for a claim at that stage. They certainly did not share Mr Morris’ misunderstanding of law as to what was required.
HMRC cannot have strengthened Mr Morris’s belief that valid claims had been made because he never had that belief. Moreover, for an officer of HMRC, in the course of correspondence about the application of the treaty to mention the treaty does not show that he objectively intended or expected that that letter would strengthen Mr Morris’ reliance on a common assumption. There is a difference between an objective intention or expectation and an innocent statement that refers to relief in the course of correspondence. For example a request for information in the course of an enquiry into a return in order to consider the location of POEM to enable HMRC to consider an element of the case is not an example of a document that HMRC intended or expected would strengthen the appellant’s case; it is simply par for the course in progressing with an enquiry. The appellants’ decision not to make a claim for relief was driven by the independent opinion of their advisor that a claim was not necessary. Mr Morris was relying upon his own independent understanding of the treaty and not on any comments in the correspondence with HMRC. HMRC never gave Mr Morris any reason to believe that his own belief that no claim was necessary was correct.
The appellants allege that because of the HMRC’s conduct, they did not make claims which they otherwise would have made (notwithstanding their advisor’s understanding that they did not need to make claims). It follows from this that the appellants cannot rely upon any conduct of HMRC after the deadline for making a claim had passed: 5 April 2009 for the Murphys and 31 January 2007 for DH. Once the appellants could no longer make a claim, it is irrelevant whether they were influenced by HMRC into thinking they did not need to make a claim. They could not have done so even if they had been put on notice of their mistake. On that basis, the Murphys must identify some conduct of HMRC that crossed the line between the start of the correspondence on 9 August 2007 and the deadline for making a consequential claim for relief 6 April 2009. For DH the equivalent period is between 19 December 2002 and 1 February 2007.
Certainly no conduct of HMRC can be relevant after they informed the appellants’ representatives that they could not identify a claim and asked the appellants to identify the claims as they did in a meeting on 13 November 2012 and followed this up in writing. If the appellants failed thereafter to take any action to protect their position, that was their responsibility (see, by analogy, Benchdollar at [67]).
There was no conduct of HMRC which “crossed the line” in such a way as to assume responsibility for the appellants’ understanding. HMRC never said they accepted claims were made. Rather, this is a case in which the parties proceeded on the basis that one thing was in issue, and then it was discovered that there had been an implementation error. The tribunal must consider how other enquiries would be affected if it is correct that HMRC is estopped from raising implementation errors. It cannot be the case that HMRC are required at the start of the enquiry to state every point that is going to be in issue in that enquiry. The appellants suggested it would make a difference if each inspector had written: “These are the issues but we reserve our position.” That cannot be the basis for founding an estoppel. That would lead to HMRC including in every letter a note reserving their position. The only unusual thing about this case is the length of time that it took between the start of the enquiry and the actual identification of the implementation error.
It is striking that the appellants rely so heavily on correspondence which either relates to all of Lansburys’ cases as a whole or correspondence which relates specifically to another taxpayer. Whether or not the RTW planning implemented by the appellants worked in principle was an issue that was common to each of the appellants and other clients of Lansburys. However, whether or not that scheme had been implemented properly, including by making a valid claim for relief, was a discrete question which depended on the specific facts for each appellant. HMRC’s officers cannot have expected or intended that the appellants would be influenced in relation to their individual implementations of the scheme by correspondence about legal principles that applied to the scheme in general terms or about another specific taxpayer.
Mr Bentley was very candid that there were opportunities to have managed the enquiry better at an earlier stage. But there were also opportunities for the taxpayer to bring this matter to a head by coming to the tribunal at an earlier stage, rather than engaging in repeated correspondence, seeking to avoid in particular the requirement to identify a claim. It is wrong to suggest that there was any fault of HMRC, in particular from 2012 when the lack of claim was identified.
HMRC submitted that HMRC cannot have “crossed the line” in any of the correspondence Mr Morris placed reliance on in his witness statement:
The early correspondence in relation to other taxpayers has no relevance given it does not relate to the appellants/their Trusts and pre-date all of the appellants’ tax returns for the relevant tax years. In that correspondence, HMRC cannot have given Mr Morris/the appellants any assurances about different taxpayers in future tax years.
The letter dated 7 November 2005 does not relate specifically to any of the appellants. The three bullet points on which Mr Morris relies cannot reasonably be read as being a comprehensive list of every single issue that was raised by all of Lansburys’ cases. This letter simply does not mention the claims issue. More than silence is required to “cross the line”.
The letter of 13 February 2006 was specifically about a different taxpayer. It has no relevance to these appellants. It says nothing about whether Mr E had made a claim for relief or not. Even if it did cross the line, it cannot have done so in such a way as to assume responsibility and to influence the appellants. HMRC expressly drew Mr Morris’ attention to s 788, which provides that obtaining relief under the treaty requires a claim to be made. That cannot have had the effect of giving the appellants comfort that no claim needed to be made.
The letter of August 2007 simply shows HMRC “working the enquiry” and seeking relevant documents which are relevant to one aspect of the matter. At this stage Mr Wood cannot have known anything about what the issues would be in the Murphy cases because he only learnt for the very first time about the transactions. The letter of 2 September 2008 relates to a different client of Lansburys.
By the time of the letter dated 9 August 2012 any claims for relief would have been out of time such that it is irrelevant for the reasons set out above. Although there are two references in this letter to ‘a claim for relief’, the letter is not about whether the correct procedure for claiming relief had been followed. Rather, it is about the effect of Smallwood, and the appellants were being encouraged to make a payment on account because the decision in that case was favourable to HMRC. HMRC did not intend or expect that the passing reference to a “claim for relief” would influence the appellants.
In his letter of 5 September 2012, Mr Morris does not refer to any of the content of the letters dated 9 August 2012 save to complain that they were sent to his clients whilst he was away. There is no suggestion that the contents of those letters had made any impression on him.
The other documents relied on by the appellants set out above were not mentioned in Mr Morris’ witness statement so he/the appellants cannot be taken to have relied on them in the required way.
Mr Windle replied as follows:
In these cases there was only an enquiry into the Harris Trust and there is no evidence that the RTW planning was a “mass marketed” scheme. In any event, it does not appear to be disputed that there was an assumption of responsibility as regards DH. The public policy concerns raised cannot possibly apply where HMRC agree to set out all their arguments.
There was a common assumption of law, namely, that the treaty applied. It is irrelevant whether the parties had the same reasons for that common assumption.
As regards the Murphys, it is notable that, as accepted by Mr Bentley, there is no other case in which HMRC accepted that Lansburys’ clients made claims. It is clear that some of the correspondence (not all of it) was intended to apply to all of Lansburys’ clients. Although the correspondence which took place before the issue was raised regarding the Murphys in 2007 cannot be relied on as raising an estoppel, it is not irrelevant. The later relevant correspondence has to be understood in context and the earlier correspondence informs the meaning of that later correspondence.
As regards detriment and unconscionability, the main focus of the appellants’ argument is on the correspondence that preceded the deadline for making claims, but the length of the delay in raising and dealing with the claims issue after that is relevant. It is clear from Benchdollar that conduct/correspondence may be of relevance where it takes place in the period (a) after the date on it comes to light that the common assumption is incorrect and/or (b) after the time limit within which corrective action can be taken has expired. One must take all factors into account including the delay in HMRC fully committing to taking a position on the claims issue albeit the weight one attributes to particular factors may vary. It is notable that following HMRC first flagging the claims issue, they did not then pursue it for five years. The only fair reading of that correspondence is that HMRC dropped the point.
In our view, the correspondence/conduct on which the appellants place particular reliance, namely, that in the period from 9 August 2007 onwards, in the case of the Murphys, and from 19 December 2002 in the case of DH, in particular, in the period in which the appellants could have made formal claims for treaty relief, demonstrates that the requirements for HMRC to be estopped from arguing that the appellants have not made valid claims for treaty relief are met:
The correspondence demonstrates that, until HMRC raised the claims issue in November 2012, the parties shared a common understanding that the appellants were seeking to benefit from relief under article 14(4) of the treaty and there was no procedural bar to the appellants’ being able to do so. We cannot see that the fact that the precise reason for each party making that assumption may differ affects matters. Mr Morris/the appellants were of the view that they did not have to submit a formal claim to obtain treaty relief provided that the relevant information was somehow made available to HMRC. It is not clear whether the relevant inspectors thought a valid formal claim had been made or that a formal claim did not need to be made or had some other thought. It is clear, however, that both parties were acting on the basis that there was no action required to be taken by the appellants in order for them to seek to rely on relief under article 14(4) of the treaty.
That this was HMRC’s understanding was plainly conveyed to the appellants by HMRC continually referring only to the substantive issue between the parties (namely, whether treaty relief in fact applied on the correct interpretation of articles 4 and 14(4) of the treaty) and asking the appellants/their agents for information relevant to that issue and stating that the resolution of that issue would determine the tax position of the appellants in relation to the transactions undertaken by the NZ trustees of the Trusts. That this was the appellants’ understanding also was plainly conveyed to HMRC by them/Lansburys engaging with HMRC on the substantive issue, providing information relevant to that issue and submitting an appeal against the assessments issued on the basis that treaty relief applied.
HMRC assumed some element of responsibility in relation to the mutual assumption by requiring the appellants/Lansburys to spend time and effort in providing information and engaging in correspondence on the substantive issues. The nature of the engagement by HMRC can only be taken to have strengthened or influenced the appellants in their belief that they were entitled to seek treaty relief. We do not accept that references in the relevant correspondence to the substantive issue and requests for information in relation to that issue constitute “innocent statements” made in the course of HMRC’s exploration of “an element of the case” which were par for the course. The correspondence demonstrates that, as regards all of the appellants, once HMRC were aware they had implemented the RTW planning (when the enquiry was opened into the Harris Trust in December 2002 and in August 2007 as regards the Murphy Trusts) (a) the relevant HMRC officers were fully cognisant of the nature of the RTW planning and, in particular, that it hinged on each of the appellants being able to benefit from relief under article 14(4), (b) the HMRC officers were not exploring elements of the case but seeking to resolve the only material substantive issue, namely, whether or not article 14(4) applied to exempt the gains arising on the disposals, and (c) they can only be taken to have intended and expected that the appellants would take that to be the case from their requests for information and seeking to debate points with them which were only of relevance to that issue. The appellants in fact relied on the mutual assumption in their continued engagement with HMRC on the substantive issue.
That the common assumption was not correct came to light only in 2012 after the appellants could have made a formal claim of the type HMRC state is required. In relying on the common assumption, therefore, they would suffer serious detriment by not being able to seek to benefit from treaty relief they argue is available to them and by engaging in years of debate with HMRC, involving costs and effort, on an issue which, if HMRC are not estopped from raising the validity of the claims, would have been entirely pointless and redundant. In that context, we consider it is pertinent that HMRC did not respond to the appellants’ assertion that there could be no doubt about their ability to claim treaty relief for several years after the appellants made that statement. A reasonable person would assume that HMRC had simply dropped the point. Whilst the appellants could not at that time have made valid in time formal claims, they were prejudiced by continuing to spend time on engaging with HMRC on an issue which is in point only if they could be regarded as having made such claims.
We do not accept that, as HMRC seemed to suggest, there is an overriding policy, as regards HMRC’s ability to conduct enquiries into taxpayers’ affairs, that prohibits our conclusion in this case. This decision is of course entirely dependent on the particular facts of these cases and what may reasonably be supposed to be an unusual situation of HMRC conducting correspondence with taxpayers/their agents over many years on the only substantive issue between the parties, the application of relief under article 14(4) under the treaty, in effect, to exempt a specified CGT charge, the whole basis of which was dependent on whether the appellants had made valid claims for relief.
Part C – Treaty issues
Evidence and facts relevant to the treaty issues
To recap:
The appellants’ stance is that under article 14(4) of the treaty the gains realised on the disposals are taxable only in NZ as the Trusts were resident only in NZ for the purposes of article 4(1) during the period of relevance, namely, the period when they had NZ Trustees, during which time they made the disposals. During that period as the Trusts were resident in NZ for the purposes of NZ tax and were not resident in the UK for the purposes of UK tax (they became resident in the UK only when UK trustees were appointed shortly after the disposals took place), they were resident only in NZ under article 4(1) and there is no need to apply the tie-breaker in article 4(3) to determine where the Trusts were resident.
In HMRC’s view the gains are taxable in the UK on the basis that (a) the Trusts were resident in NZ for the purposes of NZ tax and in the UK for the purposes of UK tax in the relevant period, which they regard as the UK tax year in which the disposals were made, (b) therefore they were resident in both States under article 4(1) and the tie-breaker in article 4(3) applies to allocate residence to the UK on the basis that the POEM of the trusts was in the UK.
The appellants dispute that, if the tie-breaker is relevant, the POEM of the Trusts was in the UK. The parties disagree on how the POEM test is to be applied.
We have based our findings of fact on the documents in the bundles and the evidence of Mr Morris, Mr Paul and Mr Richards. We found the witnesses to be honest and credible albeit that due to the passage of time the witnesses all had limited specific recall of the events which have given rise to this appeal.
As set out below, the bundles contained documents evidencing the implementation of the various steps in the RTW planning undertaken by each of the appellants and the Trusts, except in relation to DH, but otherwise relatively little record of correspondence between the various persons involved in the implementation of the RTW planning. The witnesses were questioned about the lack of documentation:
Mr Morris accepted, in effect, that the documents before the tribunal in respect of DH’s appeal are only a limited selection of documents that were provided to HMRC on 20 November 2003 in response to their request of 19 December 2002. He said Mr Blower dealt with that and he did not know if there were any discussions with DH about the request. He accepted that from a very early stage HMRC asked for documents relevant to understanding what happened in the case of DH. He was taken to the letter in which HMRC accepted the information did not need to be provided at that time but said that it would be required if HMRC needed to consider the POEM of the trust. He said that he thought this just brought into play letters going backwards and forwards. So far as he was aware no requests were made to gather together the requested documents. He was taken to a letter from HMRC of 3 October 2007 in which they asked for a full set of documents regarding the implementation of the planning and asked why the full documents were not produced and are not in the bundle. He said that they “would just respond on the basis of we were substantiating the [treaty] as to whether there was any gain to be taxed anyway”, and everything he had, he had given over. It was put to him there must have been some correspondence between No Boundary and the trustees but it is not in the bundles and he was asked if he asked Mr Paul or Mr Richards for it. He said he did not but probably, in retrospect, it would have been appropriate to do so.
Mr Paul said in his statement that he does not hold copies of any documents relevant to the present appeals. He was asked why he does not hold documents from his roles either with Altech or Funcode and from when he corresponded with the NZ Trustees. He said it was because his documents in those days were stored in a back office in Manila because of spaces restrictions in Hong Kong, and in 2010 his company was taken over by another corporate service provider/bank who told the staff in the back office to destroy anything over five years old. He confirmed that the take-over took place in 2011, so after HMRC’s request for documents on 3 October 2007. He said he did not think he would have had the requested information in 2007 but that he was getting very confused because his office shipped documents down to Manila quite regularly. He then said they were definitely gone by 2007. When asked if Mr Morris requested him to provide these documents in 2007, he said he had asked him but the documents had already been disposed of by the staff in Manila – he thought the files were that old. He accepted this was a change from his earlier response but said he was thinking about it more deeply now.
Mr Richards said that then when he was appointed a new trustee for a trust, he created a file in NZ with a copy of the material recording the appointment, and any other paperwork that he received and further down the track he may have also set up a file for the NZ tax situation, and/or other matters. He thought that the files were largely destroyed many years after the event. In NZ one is required for tax purposes to keep files for seven years and most prudent operators would keep them for ten years. He thought that they destroyed the files around about 2017 and he was not aware that there was a requirement to hold on to any of this information. He confirmed that Mr Morris did not ask him to preserve the files or provide him with a copy of the files.
It appears from the evidence that little effort was made by Mr Morris, acting on behalf of the appellants, to obtain/retain documents evidencing the transactions which are the subject of these appeals and any correspondence in relation to the transactions. We accept his explanation that the focus was on the technical point that the tie-breaker/POEM test is not engaged and make no adverse inference. However, the burden of proof is on the appellants and the lack of documents, combined with the witnesses’ lack of recall, has rendered it difficult for them to prove, on the balance of probabilities, that some of the events which they assert happened did in fact happen. We have set this out below.
Evidence of Mr Morris on the RTW planning
Mr Morris confirmed the following as regards the setting up of the Trusts and the inception of the RTW planning:
He and Mr Blower set up John Lansbury Associates Limited in the late 1970s/early 1980s and Lansbury International Limited in April 2001. For ease we refer to either or both of these entities as “Lansburys”.
He and Mr Blower devised the RTW planning implemented by other clients they had and by the appellants. Mr Blower drafted template documentation to implement the RTW planning.
In 2000/01 Mr Morris and Mr Blower lived in Ireland for work purposes and considered they were tax resident in Ireland and not in the UK whilst employed by Lansbury Limited, a company owned by Mr Paul with a branch in Ireland. Mr Morris was living and working back in the UK when this RTW planning was implemented in late 2001 and 2002. We do not have sufficient information to make any finding as to Mr Morris and Mr Blower’s tax status in 2000/01, and in any event we cannot see it has any relevance.
Even though they had their own accountants as advisers, the appellants were all direct clients of Lansburys and that was the case for the Murphys since the 1990s. DH was a client of Mr Blower. The Murphys were Mr Morris’ clients but he only usually had direct contact with TM who passed on his advice to the other Murphys. Advice on the RTW planning was not given in writing and there was no engagement letter/terms of engagement as such.
Lansburys advised DH on setting up his trust in 1991 with a view to different planning to avoid CGT that would otherwise have been paid on the disposal of UK assets or companies with underlying UK assets. In the 1990s, Mr Morris first advised the Murphys to settle their shares in Aerohawk on trust also with a view to CGT planning but that planning became impossible due to a change in legislation sometime in the late 1990s. In fact they sold the underlying business to Enron and the Aerohawk companies ended up holding the cash consideration. He thought the BVI companies were put in place on Lansburys’ advice around the same time as the Murphy Trusts were settled.
He and Mr Blower came up with the initial thoughts about the RTW planning and they knew that they would need offshore trustees to do it. They selected NZ to use for this planning on the basis that the treaty had different wording to that in the Mauritius treaty. Mr Morris had known Mr Paul for many years and trusted him. Mr Paul had previously been involved, since the mid-1980s in CGT planning, not RTW planning, but much of which involved using trusts and had been through many (more than 20) sales when Lansburys were doing CGT planning for other clients over many years.
Lansburys provided Mr Paul with a draft brief to NZ tax counsel and they sought the opinion of NZ counsel before they implemented any RTW planning, crucially as regards establishing that the trusts would be resident in NZ for NZ tax purposes and to check there were no other NZ issues raised by the planning. No specific advice was sought specifically as regards the Murphys’ cases. He agreed that, as regards the Murphys, the aim was to extract the cash from the Aerohawk companies without a charge to CGT.
A letter of 8 April 2000 from Mr Blower in which he said he was “as instructed pressing on with drafting the documentation to implement the CGT planning” shows that by then there had been some discussion about what the planning would be and that Mr Blower drafted the required documentation for implementing it. In 2002 Mr Blower was still involved in the implementation of RTW planning.
The RTW planning Lansburys used in this and similar cases essentially involved having a trustee resident outside the UK who would sell the shares followed by the appointment of UK trustees. That was the essence of the planning. He could not specifically recall any conversations with the Murphys but thought there would have been a discussion on fees with TM.
It appears from the correspondence that, at one time, Mr Morris was corresponding with HMRC about the application of the treaty in respect of 39 trusts and 32 different individuals. Mr Morris did not think he had implemented the RTW planning this many times but had no real explanation for the contrary suggestion in the correspondence. It seems highly likely that Lansburys implemented RTW planning on over 30 occasions. There is some substantiation for this also in Mr Richards’ evidence as set out below.
Mr Morris gave the following evidence as regards the appointment and role of the NZ Trustees:
There was no reason to appoint the NZ Trustees other than the planning and to make the planning work he needed NZ Trustees to implement it. He spoke to Mr Paul to find such trustees and his understanding was that NZ had a favourable regime for offshore trusts and there are a number of established providers of trust companies or trust company services in NZ.
Mr Paul did not go to one of those providers but instead spoke to Mr Richards who was an old contact of his, and when he approached him Mr Richards did not have a trustee company to use. He did not know if Mr Richards had any prior experience or track-record in providing trustee services. He confirmed that (a) Lansburys did not carry out any due diligence on Mr Richards; they accepted that Mr Paul knew him and that he would not introduce them to somebody he did not trust himself, (b) he did not speak to Mr Richards but recommended to his clients that they should put Mr Richards and his wife, who was the other trustee, in control of assets worth millions of pounds, and (c) he did not tell his clients that they were appointing NZ Trustees with no track-record of corporate trustee services.
In his statement Mr Morris said that the NZ Trustees took their own decisions as to how to deal with the trust property, they were aware of the planning and the tax benefits but undertook their own review of the benefits and made their own considerations in deciding whether to proceed with the planning. However, from his own evidence and that of the other witnesses, it is apparent that this is not an accurate description.
It was put to Mr Morris that the lack of any track-record in providing trustee services was not a problem as the NZ Trustees would not be required to do anything other than hold the assets for a few weeks and effect the sale. He said they would have to agree to do it and they would have to implement the purchase and sale agreements. He accepted that the decision to sell the assets held by the Trusts had already been taken before the NZ Trustees were appointed. When it was put to him that therefore they did not have an active decision to make about whether or not to sell, he said “they would have to go through the normal mechanics to make sure there was nothing there that was a problem”. He then accepted that the decision to effect the sale was already taken before they were appointed, so they did not have a decision to take as to whether or not to sell. He said he imagined they would check that the amounts involved were correct but accepted that he does not have any knowledge of any checks or due diligence done by the NZ Trustees and that he knew that they would not have to manage the assets prior to the sale or the funds after the sale because they were only intended to be trustees for a few weeks. It was put to him that it is not correct that, as he said in his witness statement, the NZ Trustees “took their own decisions as to how to deal with the trust property” as they did not actively take a decision. He said “…it’s correct in that they made their own decision. It may have been that they knew what was going to happen but they still had to make that decision themselves” and “they would’ve reviewed that decision and decided whether to proceed or not” but he accepted that he does not have any evidence or knowledge of that.
The correspondence shows that Mr Blower sent to Mr Richards and Mr Leighton, the NZ Trustees and the Initial Trustees (except for the EE trust) the Deeds of Appointment and Removal of trustees. Mr Morris accepted that the expectation was that the deeds would be executed, without any amendment or discussion as to the terms but added that the deed was “just changing the trustees” and that he co-ordinated the paperwork.
He confirmed that (a) at the time the NZ Trustees were appointed the Murphy Trusts effectively owned businesses that just held cash and no other asset and there was a need to find a way of extracting the cash, (b) as part of the planning he already had a buyer lined up to buy the BVI companies, No Boundary; it had already been decided as part of the design of the scheme that it would be No Boundary that would buy those companies and there was never any thought that the NZ Trustees might try to find an alternative buyer, and (c) when the NZ Trustees were briefed as to what they had to do, they would have been told that buyer was going to be No Boundary. He thought that No Boundary had been involved in many similar tax planning arrangements but did not think that was on the monetary or buying sense. It had a different role. He accepted that although in his witness statement he said the sale of shares was done by “an agreement for sale to arm’s length purchasers,” he does not actually know if that was the case or not.
Mr Paul sent the NZ Trustees the trust deeds, trust accounts and the companies’ accounts and explained the planning to them and dealt with any queries. Mr Morris initially said that when he was liaising with Mr Richards about the steps the NZ Trustees would have to take, Mr Paul was acting for No Boundary but then said he was acting for Altech. He initially said he thought Mr Paul did not get paid for this role in co-ordinating and briefing the trustees but it later became apparent that he was paid a fee as set out below.
In his statement Mr Morris said that the NZ Trustees agreed the sale price of the underlying companies as the net asset value of the relevant company less a percentage fee. He accepted that in fact Mr Blower prepared information about the trusts and their assets and provided that information to Mr Paul if not to the trustees as well; it was really that information which determined the sale price, and therefore there was no actual need for any negotiation about those figures. He confirmed that (a) the percentage fee he referred to comprised a 6% discount that was applied to the true value of the assets, (b) that discount was fixed before the transaction was entered into and included fees for Lansburys which he thought was whatever was left after all expenses (such as banking fees) had been paid, (c) so if the value of the assets was 100, the purchaser bought the companies for 94 and after the sale had assets worth 100 and a gross profit of 6, and (d) as No Boundary was wholly owned by Mr Paul effectively his reward for this whole transaction was a cut of the 6% assuming the transaction went ahead (as it did); if it had not gone ahead he would not have received anything, as was the same for Lansburys.
(a) Lansburys’ clients were the settlors not the trustees; the settlors agreed to enter into the planning but Lansburys’ fees were paid out of the trust assets and not paid by the settlors. That did not seem at all problematic to him at the time, (b) in effect, No Boundary transferred a sum of money to Lansburys as their fees: TM would have been told about the fee structure before planning was entered into and would have passed that on and, as the settlors knew the value of the assets upfront, they knew that at the end of the planning they would end up with 94, (c) at some point there was a payment out of the trust of the 94 made by the trustees to the settlors as an appointment from the trust: the UK trustees did that, and (d) it was of fundamental importance to the tax planning that the NZ Trustees resigned before the end of the tax year after the sale took place and (5) all these five appellants could ensure that the NZ Trustees were removed because the settlement deeds gave the settlors the power to appoint and remove trustees.
Mr Morris was taken to a deed relating to the Harris Trust which was signed by Mr Gaskell. He confirmed that Mr Gaskell was involved with Republic International Trust Company Limited in Monaco, he was a resident of Monaco and he signed document as a director of Allglory. It was also signed by Mr Blower on behalf of JLA Services Limited a secretarial company which was part of Lansbury. He had not tried to call Mr Gaskell to give evidence.
Mr Morris was questioned about the fact that the NZ Trustees resigned the day after the disposals took place. We note that there are no minutes of the NZ Trustees considering the decision to resign as trustees. Mr Morris accepted that because the settlors had power to appoint and remove trustees under the relevant settlement deeds, in a sense there was no need for the NZ Trustees to make an active decision to resign; they did not have to be involved and if they had refused to resign, the settlor would have been able to effect their removal in any event through this power. It was put to him that he knew that the NZ Trustees would resign at the time they had said they were going to. He said “you couldn’t be 100 per cent sure, but the answer is that they probably would, yes”. He accepted that his confident expectation was that the NZ Trustees would resign because they knew that that was an essential part of the planning.
He confirmed that Funcode had its office at the same place as Lansburys had an office, it was owned by Lansburys and Mr Paul was a director of it. He said in effect, that when he was advising the settlors on the planning, they knew that there would be a trustee appointed in the UK, namely, a UK incorporated company. He accepted that (1) he knew, before the start of the implementation, that entity would be Funcode for the Murphy Trusts, (2) there was a very short window between the appointment of the NZ Trustees, the sale of the shares, the resignation of the NZ Trustees and the appointment of Funcode, so he needed to know who each of the actors was going to be before he kicked off the implementation and Funcode was appointed because he needed UK resident trustees to be appointed before the end of the tax year, (3) after Funcode had been appointed for a couple of months, they were removed and replaced with other UK trustees as set out above.
Mr Morris was taken to correspondence from Mr Blower to Mr Paul of 13 June 2002, as the director of Funcode, to which he attached deeds to change trustees “on five busted trusts to Murphy family” which he asked to be executed and sent to Mr Morris. They were already dated 1 July 2002. Mr Morris accepted that (1) at this stage in the implementation Mr Blower was still the person sending out the documents that he had drafted these particular deeds and was collecting and co-ordinating the paperwork, (2) by this stage Funcode had already appointed the monies in the trust to the beneficiaries of the Murphy Trusts, and (3) before this planning was implemented and before the NZ Trustees were appointed, he had already lined up the purchaser of the shares and the first set of five subsequent UK trustees, and they all knew the roles that they were going to have to perform in the planning and what was required, namely that all the parties executed the documents that they were provided with in the correct order, and all was completed before the end of the tax year.
As regards the Harris Trust:
DH was Mr Blower's client, Mr Morris has no direct knowledge of the facts pertaining to his appeal and all of the information he has in respect of DH derives from documents he had seen as contained in the bundles. He confirmed that DH is still alive and said that he was not giving evidence as he did not want to be involved. He said the same of TM. He accepts that DH received a notice of assessment on 6 January for CGT of £333,120 on 6 January in respect of CGT on the disposal by the trustees of his settlement.
He confirmed that for the Harris Trust the NZ Trustees were appointed in addition to the existing trustees and according to the terms of the appointment, the NZ Trustees and Republic International were required to act together. It was put to him that there is no evidence in the bundle of any contact between the NZ Trustees and Republic International. He said “I don’t know, that may be correct.” We note that there are no documents setting out the decision to appoint NZ Trustees and no minutes of the NZ Trustees deciding to accept the appointment.
Mr Morris accepted that an extract from a sale and purchase agreement dated 30 January, in which the second named party is DH and others, is the only evidence in the bundle of any sale of assets in the case of DH. The extract comprises a long index and only the first page. He accepted that (a) there is no reference in the recitals on the first page to the sale of shares in GED, (b) there is only a list of persons showing the same and address of the shareholder as “Republic International Trust” and the “Number of GED Shares” as 249 “Number of “B” shares” and “Cash”; that is all the information there is about any sale. It was put to him there is no mention in this document of the NZ Trustees at all and the only signature is that of Republic International, there is no signature by the NZ Trustees and there is no documentary evidence of this sale and purchase agreement ever having been sent to or seen by them. After some hesitation and initially saying he thought the signatory signed on behalf of all the trustees he said he could not answer that and he does not know. He confirmed he was not aware of this document ever having been sent to the NZ Trustees. He accepted that on the basis of the documents in the bundle alone, there is no evidence of the NZ Trustees having any involvement in the sale of shares but he thought “there must be a lot more than that”.
In his witness statement Mr Morris said: “Neither I nor Lansburys more generally were controlling the transactions.” He accepted that the settlors had agreed to enter into a scheme which involved a series of preordained transactions that were going to be carried out in the requisite order but commented, in effect, that is what all pre-planning is like. He accepted that there were no commercial decisions to be made by any of the actors, everybody knew that exactly what was going to be done and the order in which it was going to be done and he knew it was going to be done. He added that to him “when one talks about preplanning, that’s what preplanning is”.
Evidence of Mr Paul on the RTW planning
Mr Paul gave the following evidence:
He had no knowledge of DH’s case. He was involved in CGT planning for clients of Lansburys before the RTW planning was under consideration. Mr Morris and Mr Blower came up with the idea for this RTW planning, and then brought it to him to get a view on NZ law. As part of ensuring it worked from a NZ perspective they provided him with a draft brief for NZ counsel which he then gave to his business associate, they went through it and made slight amendments. His associate was an English barrister who resided in Hong Kong who specialised in trust law and use of the treaty, and he had the contacts for the chambers from which they got the opinion. Having concluded that the planning worked from a NZ perspective, he then had to find a NZ trustee to act as trustee of the relevant settlements during the sale process.
Altech was paid an annual retainer out of the trust funds for acting as trustee. He remained the trustee through Altech from when the EM trust was set up until the appointment of the NZ Trustees. Whilst he was trustee, he helped Lansburys to implement the planning by, for example, getting counsel’s opinion on NZ law, by locating the NZ Trustees and by liaising with them about implementation of the planning.
When Mr Richards first took on the trusteeship, Mr Paul’s role included briefing him on what the planning involved and the steps that were required to implement it using information and spreadsheets provided by Mr Blower. He provided Mr Richards with the counsel’s opinion (which had been reviewed by another firm of tax lawyers who agreed with it) and Mr Richards said there were no surprises with the conclusions and concurred there was a withholding tax requirement. When Mr Paul was performing this role (of getting counsel’s opinion, selecting trustees and briefing them on the planning) he was working for himself as sole proprietor. He did not consider he was engaged by Lansburys for that purpose. His fee came from the 6% retained when No Boundary bought the companies. There was only a single fee for all his work whether in setting up the planning and subsequent work for Funcode (except that Altech received a retainer as set out above).
He thought Mr Richards had never previously provided trustee services before he contacted him but was not sure of that. He had probably engaged trustee services because he was working for a pretty big IT software company, he did some tax structuring offshore as well for his colleagues, he was a part-time lecturer and he knew what a trust was, obviously. He thought it was probably correct that Mr Richards’ wife had no professional qualifications in respect of providing trustee services. In his statement he described her as an experienced professional who had worked in the administration of schools and companies. He also said that by the time of these transactions, the NZ Trustees had already implemented RTW planning on at least 10 previous occasions in respect of other clients referred by Lansburys and, after the initial briefing stage, he considered they were perfectly able to undertake this type of planning without any further advice or assistance. He accepted in effect that neither the settlors nor Mr Morris asked him to carry out any due diligence on Mr Richards. He said (a) he had known Mr Richards for 20 years; he knows his integrity and knowledge – he is a very smart guy, and (b) one needed somebody with knowledge of trusts, what they are used for, how they are managed, and the fiduciary duties they had to fulfil.
Once Mr Richards had agreed that NZ Trustees were to be appointed, Mr Paul discussed with him that the need for them to enter into a Deed of Appointment and he signed that document with his Altech hat on. He told Mr Richards that, as part of the planning, the trustees would have to sell the Murphy Trusts’ shares in their respective BVI companies to No Boundary which was wholly owned by Mr Paul.
In his statement Mr Paul said that he sent necessary documentation such as the Deeds of Appointment to the NZ Trustees. He was taken to the correspondence which shows that Mr Blower sent the Deeds to Mr Richards. He said that this was duplication in that he dealt with the documents without knowing Mr Blower was also dealing with it; that happened quite often as Mr Blower worked at home. He accepted that there is no documentary evidence of him sending any documents to the NZ Trustees other than the counsel’s opinion and that Mr Blower drafted these documents. He said there were standard templates. He said the comment in his statement was about what generally happened rather than in these particular cases.
Mr Paul confirmed that (a) he never met the Murphys or had any direct dealings with them, (b) the information that he passed to Mr Richards about the Trusts, such as their asset holdings, was supplied by Lansburys, and (c) he never received from the settlors a letter of wishes as to what they wanted to happen with the trust. He said that was because such letters are not legally binding. He seemed to accept that the real reason was that he received his information about what the settlors or the beneficiaries wanted from Mr Morris and Mr Blower and he took that on trust. He added that he believed Mr Morris had the most interaction with the Murphy family and he passed the information to him.
It was put to him that Mr Richards did not have to use any knowledge of trusts or duties because everyone knew that he was only going to be appointed for a few weeks and the only thing he would have to do was to enter into the agreements for the sale of the shares. He said (a) Mr Richards had already done 11 such transactions before he was involved in the Murphy Trusts. He had the practical experience, the knowledge of what templates are used and where everything was, and (b) most of the 11 cases involved RTW planning of the type used here and was for Lansburys’ clients. He did not know if they were all fully implemented, and (c) in all cases where there had been a sale, Mr Richards resigned as trustee before the end of the tax year because he understood that that was an essential part of the planning. It was put to him that when it came to dealing with the Murphy Trusts, because of Mr Richards’ prior experience, Mr Paul was very confident that Mr Richards would do what was required, namely, to enter into the sale and purchase agreement to sell the assets and then retire. He said: “Basically it was an integral part of the planning but it didn’t always go to the plan”. He accepted that, as in these particular cases there was effectively a sale of cash for cash, there was no commercial risk that the transaction could be derailed subject to exceptional circumstances such as force majeure.
Mr Paul signed the agreements for the sale and purchase of the shares as the authorised representative of No Boundary. He advised the NZ Trustees that they would need to enter into the sale of the assets in the trust, and entered into that very same transaction on the other side, as the purchaser. In his statement he said that (a) as director of No Boundary he discussed the latest financial information on the companies held in the Murphy Trusts for determining the fair value of the shares in them and they came to agreement that the fair value would be based on the net asset value as the entities simply held cash less a 6% percentage to cover the effort required to extract the cash from the companies and to cover costs such as bank fees and Lansburys’ fees, (b) No Boundary was the only buyer interested in this transaction and there were no other offers for the NZ Trustees to consider, and (c) he was involved in ascertaining and discussing the value of the Murphy Trusts’ assets on behalf of No Boundary and did not believe there was a conflict of interest as the briefing stage was separate to the pricing discussion and the NZ Trustees were fully aware of his position as director. This evidence was tested as set out below and it is not entirely accurate. In particular, we do not accept that there was any meaningful negotiation of the price and that agreement on it needed to be reached.
It was put to Mr Paul, in effect that he was in a conflicted position as (a) the interest of the seller was obviously to achieve as high a price as possible whereas (b) the interest of No Boundary as buyer was to achieve as low a price as possible. He said that the assets were just cash, there were no fixed assets, so he does not really see it as a practical thing. He accepted that he did not perceive a conflict of interest as the sale price was determined by the cash value of the relevant company and there was no room for negotiation or uncertainty. He said that they were exchanging cash for cash. He accepted that if the value of the asset had been uncertain, there would have been a conflict and the asset value, which determined the sale price, was information provided by Mr Blower. Mr Blower provided spreadsheets but he cannot remember precisely when.
It was put to Mr Paul that there was no negotiation between him/No Boundary and Mr Richards/the NZ Trustees as to the sale price for the companies. He said that they talked about it, “it’s not so much a negotiation, it was a discussion, to see what was logical, and cash for cash, you had no fixed assets in the company, you had no intellectual property in the company, it was just cash. Therefore, you didn’t have to negotiate anything, that was just the fact that we were throwing these ideas around in our heads…..Cash is cash” but less the 6 % deduction.
He seemed to accept that element, the 6%, in effect was fixed. He said he was not clear on the dates, but he thought that it would have been in the original planning. He confirmed that (a) Lansburys’ fees and other expenses were paid out of the 6% such as bank fees, the facility and management fees and the introducer’s fees, (b) in principle the 6% was a figure that was agreed between Lansburys and the settlors before the planning was implemented, and (c) Mr Richards never searched for an alternative buyer who might have been able to offer a lower, say a 4 or 5%, deduction. He thought that was impractical given the time frame involved with the planning and the completion of it and “the fee obviously had a premium….within the time limits in force in the planning. So I doubt you would find anybody who is going to trade on that basis, and have that cash, that type of money, available.” He accepted that the reality is that the whole planning was set up with only one potential buyer in mind, No Boundary. He said that he did not know where the trustees would have sought such a buyer in such a short time frame. He accepted that it was only if the sale to No Boundary proceeded that he and Lansburys would get their fees. He said that was a normal commercial transaction. He seemed to accept that it is not normal that the settlors’ fees to Lansburys were effectively taken out of the trust funds as part of this 6% deduction. He said “you could look at it that way”.
Mr Paul accepted that (a) he had previously informed Mr Richards that it was essential to the tax planning that the NZ Trustees be removed and replaced with UK trustees within the same tax year in which the disposals took place, (b) in accordance with that instruction, the day after the disposals, the NZ Trustees gave their notice of intention to resign, and (c) as the NZ Trustees were to resign straight after the sale, they did not need to worry about managing the assets or investing the sale proceeds. He commented that their duties finished after resignation.
He confirmed that the NZ Trustees were replaced with Funcode. He did not think that Funcode received separate fees for being a trustee; rather his remuneration or reward for acting through Funcode was wrapped up with the fee that No Boundary took.
Mr Paul did not know why there are no documents in the bundles relating to the appointment of funds to the relevant persons. He accepted that during the period from when the trust was first settled until the Murphys and other were appointed as trustees, he was a trustee of the relevant trusts for all bar the few weeks when the NZ Trustees were in place.
It was put to Mr Paul that, as far as he is aware, the NZ Trustees never sought any independent advice about whether to enter into the scheme or whether to sell the assets. He said: “It was explained to them, always, and they never raised any doubts or the need to take independent advice.” He then accepted that they took on trust from him how the scheme worked and that it worked to avoid UK CGT. He thought it was standard procedure for him to show Mr Richards any opinion of UK counsel or anything from Lansburys explaining how it worked. When it was put to him that the NZ Trustees were aware of the steps that they had to take and the order they had to take them in, he said “yes - inasmuch that there were no hiccups along the way, like…any potential buyer or whatever would back out.” He accepted that there was no chance of that in the Murphys case because he was the potential buyer through his company, No Boundary.
He explained that No Boundary was not used as a buyer of assets in the implementation of other RTW planning structures but it had been used as a buyer in other tax planning for other clients - not necessarily in the UK. He said he could not remember who drafted the agreements for the sale of the companies. He thought that given that Mr Blower drafted other documents for the implementation of this planning it could be more likely that he was the one who drafted them but it depended on an individual’s workloads. He confirmed that as regards the Murphy Trusts, the trustees did not use any of their own documents but entered into documents that were provided to them.
It was put to him that he was the trustee through Funcode after the sales took place for some months and he did not contact HMRC to make a claim for relief under the treaty. He said he did not think it was necessary particularly. He did not recollect if that understanding was informed by discussions he had had with Mr Morris about his understanding.
Evidence of Mr Richards on implementation of the RTW planning
Mr Richards gave the following evidence:
Mr Paul contacted him with a business proposal in early 2000 to assist Mr Paul and Lansburys with the implementation of tax planning arrangements for some of Lansburys’ UK clients; as part of that proposal he was due to act as the corporate trustee for a number of trusts; to that end he incorporated two companies. It was put to him that Mr Paul explained to him at that stage each of the steps involved in the plan. He said, in very general terms, he was aware of the overall picture that a number parties were involved, Lansburys in London, Mr Paul in Hong Kong, and himself in NZ. He was aware of that chain. But in terms of going into detail on it “not particularly”. He certainly understood that that they would be appointed trustees of a number of trusts, and in doing so they would pick up some obligation to those trusts; he said this is a standard situation with trustees.
It was put to him that he understood that this planning would involve him, as trustee, selling the assets of the trust. He said that on day one he was not sure that was put quite that clearly but it was not terribly long before it became apparent that was a part of the activity. He said that at the outset he saw a legal opinion that supported the work that was being envisaged. He understood the NZ tax position but the “tax implication overseas was, I would have to say, much less understood by us”. He accepted that to the extent he had any understanding of the UK tax position, it was based on what he was told by Lansburys and Mr Paul. His main contact by far was Mr Paul in Hong Kong. He was asked if Mr Paul explained to him that the purpose of this planning was to avoid CGT on the sales. He was not sure that Mr Paul would ever have used the word “avoid”. He knew it was part of tax planning in the wider sense and it was a tax planning exercise. Mr Paul did not explain to him that the purpose of the planning was not to pay any CGT on the sale of the companies in so many words; it was not spelled out that clearly.
In his statement Mr Richards said that (a) whilst he knew what the planning involved, he always considered as trustee whether this plan was in the best interests of the beneficiaries and took an independent decision in line with his fiduciary duties to proceed with the tax planning, (b) he would not have taken any steps in breach of his fiduciary duties or without proper consideration, and (c) he was aware from the outset that Mr Paul was a director of No Boundary and he gave a similar description of the process for “agreeing” the price for the sale of the shares by the Murphy Trusts as Mr Paul gave. We do not accept the evidence in (a) and (c) is accurate as set out below.
We note that the minutes recording the NZ Trustees’ consideration of the sales of the share are in similar wording. By way of example, the minutes in respect of TM’s trust provide as follows:
“ “Background It was noted that negotiations had been taking place with No Boundary Limited regarding the sale of the Trust's holding.in Cowbit Holdings Limited.
It was felt that the offer of £423,835 for the Trust's holding of 100 shares ofUS$1 each in Cowbit Holdings Limited should be accepted as it was clearly in the best interests of the beneficiaries of the Trust.
Accordingly, it was agreed that the contract presented to the meeting for the sale of the Trust's holding in Cowbit Holdings Limited be executed.
Resolved It was resolved to accept the offer of the purchase of the Trust's 100 shares of US$1 each in Cowbit Holdings Limited by No Boundary Limited for £423,835 and that Gordon Richards and Helen Richards as directors of Oxford Trustees Limited, the trustee of the Trust, be authorized to sign the Sale and Purchase Agreement on behalf of the Trustee.”
In his statement he also said that the NZ trustees undertook the following work generally for the trusts they were responsible for and, in effect, he confirmed this was the case at the hearing albeit he could not remember a great deal of the specifics relating to these particular trusts:
The NZ Trustees carried out accounting work for the trusts, including the filing of tax returns in NZ in respect of income arising in NZ. This included the maintenance of accounting records in relation to banking transactions and withholding tax interest retained by the bank in order to be able to properly complete the trusts’ tax returns.
They drafted and executed minutes of trustee meetings and resolutions made during those meetings; they kept records; they held individual files for each trust and compiled checklist for trust documents (trust deeds, trust resolutions, deeds of appointment and removal of trustees, correspondence, tax file number, tax returns).
They carried out banking transactions for the payment of fees and sundry expenses from funds remitted to the Bank of New Zealand from Lansburys on behalf of the settlors of the trusts.
Mr Richards was asked why he thought it was beneficial to the beneficiaries of the trust to engage in this planning. He said:
“of course the beneficiaries always rely on professional advice…when I say "always", they frequently would rely on professional advice when it comes to tax planning. Most beneficiaries would not have the expertise…Our overriding consideration, given that we were a trustee, clearly was to do the best job we could for the beneficiaries of the trust to which we were appointed…I think that was the overriding factor rather than the tax implication…I’m not an expert in foreign tax. Our number one priority was always to be seen to do the correct thing by the beneficiaries, I mean that was the nature of our appointment…and we certainly did take it seriously, which I think you have to do.”
He was asked how he decided that selling the assets and entering into that sale and purchase agreement was in the best interests of the beneficiaries. He said:
“We believed that was the case when we approved sale and purchase agreements, yes…that was often after seeking information or additional information…if the initial information wasn’t clear to us…it didn't happen very often but on occasion we certainly did request more detail…Well, the Murphy settlements by and large….involved cashed-up entities and it was about, to a large extent, releasing the cash. So we…were comfortable that the information we were given was appropriate and certainly was in the interests of the beneficiaries, yes…I mean clearly to release cash for their benefit, and in…the Murphy examples…some sort of a sale agreement was put in front of us which created the situation where it released funds…for the beneficiaries. And it seemed to us, in the absence of any other advice or any other deal, it seemed appropriate to be - and it was acting in their best interests, yes…We didn’t seek - I’m very much going on memory here. I think by and large we were satisfied with the majority of them because they involved cashed-up assets. And there wasn’t a lot more information that could be given.”
Mr Richards confirmed that (a) he did not seek any advice from anybody other than Mr Paul, his “go-to person”, and (b) he did not seek the views of either the settlors or the beneficiaries as to what was in their interests: “we relied on information that we had in front of us”. He accepted that it was Mr Paul told him that it was a good idea for the beneficiaries for him to enter into the sale and he acted on what Mr Paul told him. He added:
“at all times we did ask was there….any other deal available. Was – the information put in front of us, was that the only deal?...with the Murphy example we were dealing with…largely cashed-up assets…which is rather different from selling a business.”
In respect of the Murphy trusts, the answer from Mr Paul as regards any other deal was: “No, you have to sell to No Boundary Limited” and he did not try to find another company that might be willing to buy the relevant assets of the trust. He added that it would be very difficult to do that from NZ.
Mr Richards said that he knew that there were fees and charges and that allowance was made using the 6% formula. He did not consider whether there may be another buyer willing to pay a smaller discount than 6%.
The Murphy Trusts were some of the last that he was involved in and he acted for about 20/25 trusts which involved Mr Paul and/or Lansburys. By the time he was involved in the Murphy Trusts, he knew that the planning involved a sale of the trusts’ assets and him retiring soon after the sale. In his statement he said the NZ Trustees retired when they did in order to facilitate the tax planning; it was in the interests of the beneficiaries not to incur CGT on the disposals. They had already undertaken this type of planning and knew very well they needed to retire before the end of the relevant tax year. It was put to him that Mr Paul and Lansburys knew that they could trust him to retire within the same tax year as the sale took place as part of the planning. He said that he believed he got most of the information upfront, but he cannot confirm that he got the retirement instructions at the beginning. When pressed he said that he was unsure. “It's too long ago. It really is. I’m unsure.”
He did not recall that Mr Blower of Lansburys drafted some of the documents such as the deeds. He said that (a) his memory of these individual trusts is pretty faint, certainly the names are very familiar, but the detail, he really would struggle to remember, as it is over 20 years ago, (b) without the documents he would probably have little recollection, and (c) he could not recall where the price paid for the companies came from. He said “I can’t recall the detail and again I’m now influenced by more recent information which has been put in front of me. But I mean from a common sense point of view if they were largely cashed-up…there wouldn’t be a lot of debate about how the price was…calculated”.
The information about the cash value of the companies was provided to him by Mr Paul. He definitely had files for the Harris Trust, the name is very familiar and he had a slight recollection that it was a bit different from the Murphy Trusts, but beyond that his recollection is very hazy. He could not comment on the interactions or involvement he had in the sale in which the Harris settlement was involved.
Conclusions on the evidence
As regards the Murphys and the Murphy Trusts, the evidence establishes that:
The RTW planning was implemented as set out in Part A.
There was an overall single plan for the sale of the shares held by the Murphy Trusts in a tax efficient manner which was devised, decided upon, facilitated and orchestrated in the UK by the settlors and their UK advisors, Lansburys, acting through Mr Paul, as regards direct interaction with the NZ Trustees.
It was integral to the plan that the NZ Trustees would be in place as trustees of the Trusts for a brief period only for the purpose of implementing the plan as was in fact the case. The only reason for their appointment was so that they could play their essential role in implementing the plan.
At or around the time the NZ trustees were appointed, (a) the settlors had decided (i) to implement the plan for the tax efficient sale of the shares, as devised by Lansburys, and (ii) as part of that they had decided that the relevant shares were to be sold by the NZ Trustees, and (b) Lansburys, acting on instructions from the appellants, had lined up the NZ Trustees, the purchaser of the assets, No Boundary, and the UK Trustees that would subsequently be appointed. They were all lined up and knew the parts they would have to play. All that was required, once the planning was implemented, was that the parties executed the documents they were provided with in the correct order and that the planning was completed before the end of the 2001/02 tax year.
Given the nature of the planning and in light of the witnesses evidence we consider it reasonable to conclude that:
Neither Lansburys nor Mr Paul were concerned with whether Mr Richards and/or his wife had the skills and experience necessary to provide trustee services of the type which may be required to act as trustee on an on-going basis for multiple settlements holding valuable assets. That was because the role of the NZ Trustees was to sell the relevant assets and retire straight afterwards.
Lansburys’ and Mr Paul’s main concern was to engage reliable professional persons who could be relied on to understand what was expected of them under the pre-agreed plan and to implement it accordingly.
The NZ Trustees were appointed as trustees of the Trusts and agreed to act as such on the basis of a common intention and understanding between them and Mr Paul, acting in effect on behalf of Lansburys, that they would in fact implement the plan by taking all the actions considered to be necessary for it to succeed, namely, the sale of the shares and their subsequent retirement in favour of UK trustees, subject to an exceptional unexpected occurrence beyond the parties control, as Mr Paul put it, “force majeure”. Mr Richards had acted as trustee (via his companies) in at least 10 other RTW planning structures so he knew very well what was expected of him. Mr Paul trusted him to do what was necessary (namely, sell the shares in the relevant companies and then retire before the end of the tax year) and Lansburys trusted Mr Paul to have selected trustees who would do what was necessary in that sense.
The documents used for the appointment of the NZ Trustees were, like the other documents used to implement the planning, standard templates reused from previous transactions. There was an expectation that the deeds of appointment of the NZ Trustees would be executed by the NZ Trustees without any amendment or discussion.
The NZ Trustees’ role appears to have been confined to (a) reviewing the relevant template documents drafted and sent to them by Mr Blower (whether direct or via Mr Paul) and check they were in order from a NZ perspective and possibly that there were no issues under NZ law (although see the comment in (5)), (b) preparing minutes of meetings of the trustees, and (c) performing some other administrative tasks such as dealing with NZ tax returns, accounting and banking matters and dealing with payment of fees. Whilst some of the minutes of relevant meetings were in the bundles no other documentary evidence has been produced regarding these tasks. However, as Mr Richards acted as trustee in a number of these RTW planning schemes, we accept his recollection that this is what the NZ Trustees generally did and accordingly that it is likely they performed at least some of these tasks in relation to these Trusts.
The nature of the highly limited role the NZ Trustees otherwise had is shown by the facts that (a) they had no contact with the settlors and liaised only with Mr Paul, who was effectively instructed by Lansburys, (b) they had no input into the RTW planning whether from a NZ perspective or otherwise. Mr Richards was shown the note from NZ tax counsel and said it accorded with what he expected, and (c) they did not take any independent advice on whether the RTW planning was likely to achieve the desired saving of CGT and indeed did not fully understand the CGT planning. We accept that Mr Richards considered the NZ Trustees signed the relevant documents, including the share sale and purchase agreement in the belief it was in the best interests of the beneficiaries to do so because he had been told there was a UK tax advantage for them in the NZ Trustees implementing the plan. However, he did not fully understand the CGT planning and did not seek advice on the UK tax position. He did not, therefore, consider for himself (as trustee) whether in fact there was a benefit as would have been necessary for him to make decisions in the beneficiaries’ best interests in the context of planning intended to avoid CGT. He simply accepted that it was in the interests of the beneficiaries of the Trusts for the NZ Trustees to seek to achieve the CGT saving and to sign the template documents presented to them in accordance with the pre-agreed plan. We note also that there is no evidence that the NZ Trustees gave any thought to whether it was appropriate for trust assets to be used to pay the personal debts of the settlors.
As regards the sale of the shares, (a) at or around the time the NZ Trustees were appointed, No Boundary was lined up as the purchaser of the shares and the price and fixed percentage discount was fixed by the settlors/Lansburys. When the NZ Trustees were briefed, they were told that the buyer of the trust assets would be No Boundary. No Boundary was wholly owned by one of the key persons involved in orchestrating the planning, Mr Paul. He had used that company in other tax planning arrangements, (b) the template document for the sale and purchase was provided by Mr Blower, (c) whilst Mr Richards may have reviewed the agreements, there is no evidence of any input, comment or consideration of them, and (d) there was no negotiation with No Boundary. Whilst Mr Richards said there was a discussion it seems only to amount to him being informed of what the price was and he did not seek to question the fixed 6% deduction or consider whether an alternative purchaser could be found who would settle for a lesser deduction. Mr Morris accepted that there was no commercial decision to be made by the NZ Trustees.
The evidence in relation to the retirement of the NZ Trustees confirms that from the outset there was no realistic prospect that the NZ trustees would not retire in favour of UK trustees who had already been identified when they were appointed. We note that the settlors/appellants could have exercised their powers to remove the NZ Trustees if needs be.
We note that minutes of trustee meetings have only been disclosed for the decision to accept appointment as NZ Trustees and to enter into the share and purchase agreement and not for the decision to resign as trustees. Mr Richards had little actual recollection of events but we accept that it is likely that meetings were held and the relevant resolutions and documents were signed by the NZ Trustees.
As regards DH:
None of the witnesses had specific recall of the implementation of the RTW planning for DH/the Harris Trust. DH provided very few documents. There was a tax return submitted by the NZ Trustees in respect of the DH Trust for the NZ tax year from 1 April 2000 to 31 March 2001 and an application for a registration number in respect of the trust. The deed by which the Harris Trust was settled was not produced. However, the Deed of Appointment of trustees appointing the UK Trustees purports to exercise the same power as was exercised in the Murphy transactions. There are no documents setting out the decision to appoint the NZ Trustees and there are no minutes of the NZ Trustees recording a decision to accept the appointment. As set out above, the full Sale and Purchase Agreement for the GED shares is not before the tribunal. The extract that is before the tribunal does not contain any terms that would effect a sale of shares in GED Technology Group Limited. The Sale and Purchase Agreement does not contain any mention at all of the NZ Trustees. There is no evidence that it was ever seen by or sent to the NZ Trustees or that the NZ Trustees had any involvement in the sale of the shares. The agreement was not signed by the NZ Trustees. Mr Richards has no memory of his involvement in the transaction. There is no evidence of any negotiation which preceded the Sale and Purchase Agreement.
The appellants said that whilst the recollection of the witnesses is limited, it is relevant that the Monaco Trustees were legally required to hold the 249 shares in GED Technology Group Ltd “to the order of” the Monaco Trustees, Oxford Trustees and Avon Trustees”. In these circumstances, it would have been a breach of their fiduciary duties for the Monaco Trustees to sell the 249 shares in GED Technology Group Ltd without consulting with the NZ Trustees and obtaining their consent. The appellants submitted that the presumption of regularity (as discussed in in CHF Pip! Plc v HMRC [2021] UKFTT 383 (TC)) applies in these circumstances to treat the Monaco Trustees as acting in conformity with their fiduciary duties and thus as having sold the shares with the consent of the NZ Trustees. We cannot see any basis for making any such presumption in this case. DH or the Monaco Trustees could have been called as witnesses on this point. Overall, it is wholly unclear what involvement the NZ Trustees had in this sale but, in any event, if they did sign the document it is highly likely their involvement was highly limited in much the same way as it was in relation to the Murphy Trusts.
We consider that, as Mr Richards was a conscientious professional, he signed such documents as trustee for the other Trusts and he was involved as trustee in many implementations of the RTW planning, it is likely that the NZ Trustees signed the agreement for the sale of the shares by the Harris Trust and other documents required to implement the planning. Otherwise we can make no specific findings as to the NZ Trustees involvement in or knowledge of the sale or of the implementation of this planning. In any event, to the extent that the NZ Trustees did have any involvement, it is highly likely that it was restricted to the role they had in relation to the Murphy Trusts and that the same points apply as made above in relation to those Trusts.
Approach to interpretation of the treaty
It was common ground that the tribunal should interpret the treaty according to the summary of the principles to be applied in interpreting treaties set out by Mummery J in IRC v Commerzbank 63 TC 218 (“Commerzbank”) (as based on the approach laid down by the House of Lords in Fothergill v Monarch Airlines Ltd [1981] AC 251). His summary was later approved by the Court of Appeal in Memec v IRC [1998] STC 754 at 766g) and in Smallwood CA (see [26] to [29]) and more recently by the Supreme Court in Fowler v HMRC [2020] 1 WLR 2227 at [19]. The full passage from Commerzbank is as follows:
“(1) It is necessary to look first for a clear meaning of the words used in the relevant article of the convention, bearing in mind that 'consideration of the purpose of an enactment is always a legitimate part of the process of interpretation': per Lord Wilberforce (at 272) and Lord Scarman (at 294). A strictly literal approach to interpretation is not appropriate in construing legislation which gives effect to or incorporates an international treaty: per Lord Fraser (at 285) and Lord Scarman (at 290). A literal interpretation may be obviously inconsistent with the purposes of the particular article or of the treaty as a whole. If the provisions of a particular article are ambiguous, it may be possible to resolve that ambiguity by giving a purposive construction to the convention looking at it as a whole by reference to its language as set out in the relevant United Kingdom legislative instrument: per Lord Diplock (at 279)
(2) The process of interpretation should take account of the fact that -
'The language of an international convention has not been chosen by an English parliamentary draftsman. It is neither couched in the conventional English legislative idiom nor designed to be construed exclusively by English judges. It is addressed to a much wider and more varied judicial audience than is an Act of Parliament which deals with purely domestic law. It should be interpreted, as Lord Wilberforce put it in James Buchanan & Co. Ltd v. Babco Forwarding & Shipping (UK) Limited, [1987] AC 141 at 152, “unconstrained by technical rules of English law, or by English legal precedent, but on broad principles of general acceptation”: per Lord Diplock (at 281–282) and Lord Scarman (at 293).".
(3) Among those principles is the general principle of international law, now embodied in article 31(1) of the Vienna Convention on the Law of Treaties, that “a treaty should be interpreted in good faith and in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose”. A similar principle is expressed in slightly different terms in McNair's The Law of Treaties (1961) p 365, where it is stated that the task of applying or construing or interpreting a treaty is “the duty of giving effect to the expressed intention of the parties, that is, their intention as expressed in the words used by them in the light of the surrounding circumstances”. It is also stated in that work (p 366) that references to the primary necessity of giving effect to “the plain terms” of a treaty or construing words according to their “general and ordinary meaning” or their “natural signification” are to be a starting point or prima facie guide and “cannot be allowed to obstruct the essential quest in the application of treaties, namely the search for the real intention of the contracting parties in using the language employed by them”.
(4) If the adoption of this approach to the article leaves the meaning of the relevant provision unclear or ambiguous or leads to a result which is manifestly absurd or unreasonable recourse may be had to “supplementary means of interpretation” including travaux préparatoires: per Lord Diplock (at 282) referring to article 32 of the Vienna Convention, which came into force after the conclusion of this double taxation convention, but codified an already existing principle of public international law. See also Lord Fraser (at 287) and Lord Scarman (at 294).
(5) Subsequent commentaries on a convention or treaty have persuasive value only, depending on the cogency of their reasoning. Similarly, decisions of foreign courts on the interpretation of a convention or treaty text depend for their authority on the reputation and status of the court in question: per Lord Diplock (at 283–284) and per Lord Scarman (at 295).
(6) Aids to the interpretation of a treaty such as travaux préparatoires, international case law and the writings of jurists are not a substitute for study of the terms of the convention. Their use is discretionary, not mandatory, depending, for example, on the relevance of such material and the weight to be attached to it: per Lord Scarman (at 294).”
As set out at [95] of Smallwood SpC, article 31 of the Vienna Convention on the Law of Treaties referred to in this quotation from Commerzbank provides as follows:
“1. A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.
2. The context for the purpose of the interpretation of a treaty shall comprise, in addition to the text, including its preamble and annexes:
Any agreement relating to the treaty which was made between all the parties in connection with the conclusion of the treaty;
Any instrument which was made by one or more parties in connection with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty.
3. There shall be taken into account together with the context:
any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions;
any subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation;
any relevant rules of international law applicable in the relations between the parties.
4. A special meaning shall be given to a term if it is established that the parties so intended.”
Mr Windle referred to the following comments of Lord Reed in giving the leading judgment in Anson v Revenue and Customs Commissioners [2015] STC 1777 (“Anson”) at [54] to [56] where he confirmed the importance of the Vienna Convention and then said that, put shortly:
“the aim of interpretation of a treaty is therefore to establish, by objective and rational means, the common intention which can be ascribed to the parties. That intention is ascertained by considering the ordinary meaning of the terms of the treaty in their context and in the light of the treaty’s object and purpose. Subsequent agreement as to the interpretation of the treaty, and subsequent practice which establishes agreement between the parties, are also to be taken into account, together with any relevant rules of international law which apply in the relations between the parties. Recourse may also be had to a broader range of references in order to confirm the meaning arrived at on that approach, or if that approach leaves the meaning ambiguous or obscure, or leads to a result which is manifestly absurd or unreasonable.”
Mr Windle also referred to Lord Reed’s comments at [58]:
“The contemporary background of a treaty, including the legal position preceding its conclusion, can legitimately be taken into account as part of the context relevant to the interpretation of its terms ...”
He noted that that was a case where the previous conventions were a key part of how the Supreme Court determined the context and the meaning.
He also cited Lord Reed’s comments at [110] to [111]:
“Article 31(1) of the Vienna Convention requires a treaty to be interpreted ‘in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose’. It is accordingly the ordinary (contextual) meaning which is relevant. As Robert Walker J observed at first instance in Memec [1996] STC 1336 at 1349, 71 TC 77 at 93, a treaty should be construed in a manner which is ‘international, not exclusively English’.
That approach reflects the fact that a treaty is a text agreed upon by negotiation between the contracting governments. The terms of the 1975 Convention reflect the intentions of the US as much as those of the UK. They are intended to impose reciprocal obligations, as the background to the UK/US agreements from 1945 onwards makes clear...” (Emphasis added.)
These passages were cited with approval by Lord Briggs in the decision of the Supreme Court in Fowler v HMRC [2020] UKSC 22, [2020] 1 WLR 2227 at [19].
Mr Windle referred to the summary of principles of interpretation set out by Falk LJ in HMRC v GE Financial Investments [2024] EWCA Civ 797, [2024] STC 1310 (“GEFI”) (see [70] to [75]). In that case the taxpayer, a UK incorporated company, had its shares stapled to the shares of a US company which had the consequence that it was treated for US federal income tax purposes as if it was a US corporation and so was subject to income tax on a worldwide basis in both the UK and the US. One of the questions was whether the share stapling had the effect that the taxpayer was a resident of the US for the purposes of the relevant UK/US double tax agreement which contains a similar provision in article 4(1) as that in the Mauritius treaty. It was held that “liability to tax” by reason of a criterion similar to those listed in article 4(1) did not extend to share stapling. In particular, Mr Windle drew attention to Falk LJ’s comment that the Upper Tribunal “using the [Model]...and OECD commentary as its starting point, rather than the words of the Convention itself in the light of its object and purpose, may have increased the risk of error”. He submitted that (1) it is notable that it was held that article 4(1) did not apply by reason of a share stapling criterion; that was not a sufficiently similar criterion to those listed in article 4(1), and (2) that shows the impact of the choices that the UK and US had made in how they drafted article 4(1); it did not encompass all forms of liability to comprehensive taxation.
HMRC drew attention to the reference at [38] of this case to the comment in Irish Bank Resolution Corporation Ltd v HMRC [2020] EWCA Civ 1128, [2020] STC (“Irish Bank”) at [18] to [23] that “…the unilateral opinion or practice of a tax authority is not a relevant aid to interpretation”. Mr Windle drew attention to the full context to that comment in the Irish Bank case at [22] and [23]:
“The UT took the view that the practice of the Inland Revenue in relation to the assessment of the profits attributable to the UK branch of a bank was inadmissible as an aid to the construction of Article 8(2) of the 1976 Convention. At [29]-[31] they said:
“29. There is also a further – and in our judgment altogether more fundamental reason, which we put to the parties in argument - why this material is inadmissible. That is because this material is irrelevant to the question of construction that we have to answer. The unilateral practice of a taxing authority - no matter how well-advised -is not material that can support or contradict a particular interpretation of a treaty.
It is permissible to look to the subsequent conduct of the parties to a treaty to see if there is a subsequent agreement or practice that goes to the meaning of the treaty. Such agreement or practice would have to be evidenced, and would have to demonstrate a bilateral agreement or practice involving both parties to the treaty. No such agreement or practice was alleged here; and we consider the point to be a factual one, that could only properly be raised before the FTT.
We do not consider that the unilateral practice of a contracting party – even if that practice shows a careful attempt by that party to abide by a treaty – can affect the meaning of that treaty or constitute material going to its construction.”
This seems to me to be clearly right. Mr Baker submitted that Article 31 of the Vienna Convention should not be treated as an exhaustive and immutable code and I think that may be correct. As with any other set of legal principles, the norms of international law are capable of development and change. But what the UT said in [30] of its decision (which is derived from Article 31(3)(b) of the Vienna Convention) is an established norm of international law. By contrast with that, the Appellants have been unable to identify any established principle of international law which recognises the unilateral practice of a contracting state as an aid to the construction of a treaty. In that respect, there is no divergence between international law and the English private law system which has never received evidence of what a party to a contract believed that the language of the agreement meant except in relation to a claim for rectification. The legal meaning of the words used is an abstract question of law to be determined on an objective basis.” (Emphasis added.)
Mr Windle also referred to the comment in Irish Bank that new OECD Commentary on the OECD Model Tax Convention (“the Model”) introduced after the relevant Model in question in that case, “ although new, would be admissible as an aid to the construction of the wording” of the earlier Model and: “It would only be inadmissible if the new material made substantive changes which are inconsistent with the commentaries in existence at the time of the 1976 Convention.”
HMRC referred to Lord Briggs’ comments at [19] of Fowler that, at [26] to [29] of Smallwood CA, Patten LJ “provided a useful summary of the correct approach to interpretation, largely based on dicta of Mummery J” in Commerzbank and that the whole passage repays reading and [29] is worth quoting in full. We have already set out [26] above. The other passages are set out in our consideration of the decision in Smallwood CA below.
HMRC referred also to the comments of Arden LJ in Bayfine v HMRC [2011] EWCA Civ 304, [2012] 1 WLR 1630 at [17] where she said:
“…the primary purposes of the Treaty are, on the one hand, to eliminate double taxation and, on the other hand, to prevent the avoidance of taxation. In seeking a purposive interpretation, both these principles have to be borne in mind. Moreover, the latter principle, in my judgment, means that the Treaty should be interpreted to avoid the grant of double relief as well as to confer relief against double taxation.”
Mr Windle noted that the prevention of the avoidance of taxation or fiscal evasion of concern in that case related to a situation where parties sought to claim relief in both States but there is no suggestion of double treaty relief in this case. He submitted that there is no support in this case for the proposition that a treaty of general application should be interpreted to prevent relief in circumstances where relief is only claimed once and HMRC’s objection is simply that the Contracting State with the right to tax (NZ) has chosen not to exercise it.
Since the hearing the Court of Appeal has again set out how to approach the interpretation of double tax agreements in Haworth CA. Newey LJ gave the leading judgment with which the rest of the panel agreed. He set out a review of the authorities, including at [38], the comments of Lord Diplock in Fothergill v Monarch [1981] AC 251, at 281-282 which are cited in Commerzbank (see (2) in the citation from Commerzbank at [152] above).
At [39] Newey LJ said this as regards the practice of a Contracting State:
“While subsequent state practice can be an aid to the interpretation of a treaty, in accordance with Article 31(3)(b) of the Vienna Convention, “since there is more than one party to a treaty, what is required is evidence of practice on the part of both parties to that treaty” and “the unilateral practice of one party cannot alter the meaning of a treaty”: see Irish Bank Resolution Corporation Ltd v Revenue and Customs Commissioners [2020] EWCA Civ 1128, [2020] STC 1946, at paragraph 59, per Singh LJ, and also GE Financial Investments v Revenue and Customs Commissioners [2024] EWCA Civ 797, [2024] STC 1310 ("GE Financial Investments"), at paragraph 38, per Falk LJ.”
At [40] he noted that where a treaty is based on the Model, the OECD Commentaries on that treaty can also be of assistance by reference to Fowler at [16] and set out Lord Briggs’ comments at [18] of that case and commented as follows:
““The OECD Commentaries are updated from time to time, so that they may (and do in the present case) post-date a particular double taxation treaty. Nonetheless they are to be given such persuasive force as aids to interpretation as the cogency of their reasoning deserves: see Revenue and Customs Comrs v Smallwood [2010] STC 2045, para 26(5), per Patten LJ.”
Versions of the OECD Commentaries post-dating the treaty at issue are thus to be “treated in a similar way to academic commentaries”: see GE Financial Investments, at paragraph 43, per Falk LJ. See, too, Commerzbank, at 298 (proposition (5)), per Mummery J, and Royal Bank of Canada v Revenue and Customs Commissioners [2025] UKSC 2, [2025] 1 WLR 939, at paragraph 32, per Lady Rose.”
Treaty Issue - Decision in Smallwood CA
In summary, our view is that, as HMRC submitted (1) having regard to the interpretation of the treaty under the principles set out above and, in light of the relevant OECD Commentary, essentially, the provisions in article 4(1) and 14(4) of the treaty are to be interpreted in accordance with the analysis of the corresponding provisions in the Mauritius treaty set out in Smallwood CA, and (2) on that basis, for the purposes of article 4(1) each of the Trusts was a resident of each of NZ and the UK at the relevant time, and therefore the tribunal must proceed to determine where the POEM the Trusts was in order to determine which State they were a resident of for the purposes of article 14(4). As set out in full below, we do not accept Mr Windle’s submissions as to why a different interpretation of article 4(1) and article 14(4) applies. As regards the decision in Smallwood CA, he said that the analysis in this case hinged on the precise wording in article 4(1) of the Mauritius treaty which, as explained, is different to that in article 4(1) of this treaty. We have first considered that decision.
In Smallwood CA Patten LJ, who gave the leading judgment with which the other judges agreed, explained the background to the appeal as follows:
The trustees contended that, for the purposes of article 13(4), the issue of residence falls to be determined at the date of the disposal giving rise to the chargeable gain, and as the trustees were resident in Mauritius at that time the capital gains are only taxable there (see [11]). The tie-breaker in article 4(3) was used by the Special Commissioners to determine the appeal on the basis of a finding that “the trustees were resident within the meaning of Article 4(1) in both Mauritius and the UK in the period culminating in the time when the shares were sold”. To reach this conclusion they interpreted “resident” in article 4(1) (and therefore article 13(4)) as meaning chargeable to tax rather than simply physically resident (see [13]).
He summarised the reasoning of the Special Commissioners as follows at [14] and [15]. Given his later comments it seems he approved this approach:
“Although under UK domestic law there is a difference between residence and chargeability to tax in that residence for only part of the year can result in chargeability for the whole year, this distinction is too subtle to be applied to the provisions of Article 4(1) and would also require one to examine the domestic basis of taxation which the Model Convention on which the Treaty is based is not concerned with. Article 4(1) equates or elides the two by defining residence for Treaty purposes as liability to taxation on the basis of residence. One is therefore resident in a Contracting State within the meaning of Article 4(1) during a particular fiscal year if under TCGA one is chargeable to tax there in that year of assessment.
Article 13 is intended to deal with any conflicts which may exist between taxation of capital gains on the basis of the source or situs of the income or gain and taxation on the basis of the residence of the taxpayer. It is therefore limited to defining the relevant basis of taxation for each type of gain and is not concerned to determine the purely domestic question of where the taxpayer is resident for purposes of the treaty or when. Articles 13(1)-(3) therefore provide for gains from the disposal of immoveable property and certain types of business assets to be taxed in the Contracting State in which they are situate and for all other gains to be taxed “only in the Contracting States of which the alienator is resident”: see Article 13(4). Any issue as to which Contracting State that is falls to be determined by reference to Article 4.” (Emphasis added.)
At [16] he set out the Commissioners’ conclusions on this issue as summarised at [107] of their decision:
“Article 13 in general deals with a conflict between taxation on the basis of source and on the basis of residence. It states that if the alienator is Treaty Resident in one state, the gains are taxable only in that state. Any dual residence, which we have equated with chargeability, should have been solved by the tie-breaker determining the Treaty Residence before one arrives at Article 13."
The Special Commissioners therefore went on to consider the issue of POEM under article 4(3) and (a) they interpreted this as requiring them to decide in which State “the real top level management (or the realistic, positive management) of the trustee qua trustee is found” (see [17]), and (b) applying this test they found that, although trustee meetings took place in Mauritius where the trust was registered, the top level management of the trust was conducted in the UK (see [18]).
He explained, at [19], that on appeal to the High Court the taxpayer submitted (and Mann J accepted) that the tie-breaker in article 4(3) has no application because the trustees were never resident in more than one of the Contracting States at the same time.
“This argument involves reading “resident” as defined in Article 4(1) as meaning no more than resident or ordinarily resident in the sense contemplated by s 2. Article 13(4) allocates the right to tax the gain by reference to the State in which the alienator is Article 4 resident at the time of the disposal. This was therefore Mauritius. It is impermissible to determine tax residence (as the Special Commissioners did) by using the benefit of hindsight as at the end of the tax year in order to see whether the UK has asserted tax residence on the basis of events subsequent to the disposal. The question of residence has to be decided by asking where the taxpayer was actually resident at the date of the sale. As a consequence, there was no need to resort to the tie-breaker in Article 4(3) which is concerned with concurrent tax residence at the same point in time. It is therefore critical to this argument that, in the case of capital gains, the choice between rival claims by the Contracting States to tax on the basis of residence was intended to be resolved not by Article 4 but by Article 13 itself except in cases where the periods of residence forming the basis of taxation were actually concurrent and that questions of timing are for Article 13 exclusively. One can see this, he said, in Article 13(3) where the POEM of the relevant enterprise is used to allocate the rights to tax between the Contracting States. The POEM of a company, like that of a trust, can change during the course of a particular year of assessment and this points to Article 13(2)-(4) needing to be read so as to determine the position in respect of liability as at the date of the disposal.”
At [23] to [25], he set out HMRC’s contentions:
“…the purpose of the DTA was to grant relief against double taxation. It was specifically not its purpose to facilitate the avoidance of tax in both jurisdictions. It therefore requires to be construed purposively with that primary object in mind. The result contended for by the trustees (and confirmed by the judge) is inconsistent with this.
….the purpose of Article 13 is to resolve any conflicts which may exist between the taxation of gains on the basis of the situs of the assets and taxation of such gains on the basis of residence or some similar qualification on the part of the alienator. Subject to the exceptions contained in Articles 13(1)-(3) (two of which are merely permissive), it therefore preserves residence as the basis of the right of each Contracting State to tax the gain. What, however, it is not intended to do is to determine what constitutes residence for this purpose or to resolve any issues of potential double taxation which may arise from the adoption of that criterion. Conflicting claims to tax based on residence are matters to be dealt with under the Articles which eliminate double taxation: i.e. Article 24.
[HMRC’s] criticism of the judge’s acceptance of the taxpayers’ argument that one has to import into Article 13(4) the date of disposal as a way of resolving potential conflicts between resident based charges to tax is that the time of residence is irrelevant to what Article 13 is intended to achieve and ignores the existence of the other provisions in the DTA which are designed to deal with those problems. If both Contracting States seek to tax the gain because the taxpayer is resident in both countries, but in consecutive periods of time, then Article 13(4) is satisfied. The consequence is that each Contracting State is entitled (in conformity with Article 13(4)) to tax the gain on the basis of residence in accordance with their own domestic legislation and the possibility of double taxation is catered for under Article 24. In the present case that problem does not arise because Mauritius has no capital gains tax regime. He therefore sided with Mr Prosser in front of Mann J in construing "resident" in Article 4(1) as meaning no more than resident in the sense used in the UK tax legislation. On this analysis, Article 4(3) does not come into play because there is not concurrent Article 4 residence in point of time and the issue of the trustees' POEM during the period up to the disposal does not arise. The essential and only real difference between him and Mr Prosser was as to whether Article 13(4) requires the test of residence to be applied at the date of the gain. However, on this appeal, he does accept as an alternative that the construction of Article 4(1) put forward by the Special Commissioners is possible with the result that concurrent residence in the sense of concurrent chargeability existed at the time of the disposal. This therefore brings into operation the tie-breaker under Article 4(3) and Mr Brennan supports the reasons given by the Special Commissioners for deciding that the trustees' POEM was the UK.”
Patten LJ then set out his analysis and conclusions as follows:
At [28] he said it is important to identify what articles 4 and 13 are designed to achieve in the context of the treaty because “this largely colours the interpretation of the provisions themselves”. He explained that the 1977 Model Convention adopted in the Mauritius treaty eliminates the possibility of double taxation in two ways: (a) by the allocation the right to tax by reference to the situs or source of the taxable income or gain or the place where the person in receipt of the profit is treated as resident for tax purposes, or (b) the category of rules (represented by article 24) come into play when the first set of rules leaves both Contracting States as eligible to tax the same gain and operates by allowing tax payable in one Contracting State to be credited against the taxpayers’ liabilities in the other.
We note the importance of assessing the purpose of the relevant provisions in the treaty in interpreting them.
At [29] he noted that these provisions (as given effect under s 788 ICTA) are intended to give a form of relief against what would otherwise be the relevant tax liability under UK law. But:
“the DTA is not concerned to alter the basis of taxation adopted in each of the Contracting States as such or to dictate to each Contracting State how it should tax particular forms of receipts. Its purpose is to set out rules for resolving issues of double taxation which arise from the tax treatment adopted by each country’s domestic legislation by reference to a series of tests agreed by the Contracting States under the DTA. The criteria adopted in these tests are not necessarily related to the test of liability under the relevant national laws and are certainly not intended to resolve these domestic issues.”
This accords with the guidance in the OECD Commentary which is equally applicable here as article 14(4) is in essentially the same terms as article 13(4) of the Mauritius treaty/the Model in place at the time.
At [30] he said that the starting point has to be article 13:
“This, as Mr Brennan has submitted, includes a general rule that capital gains from the alienation of property are to be taxable only in the Contracting State of which the alienator is a resident and four exceptions to that general rule.
The first two (which are simply permissive) enable gains from the disposal of immoveable property to be taxed in the State where the property is situated and for gains from moveable business property (forming part of a permanent establishment) of an enterprise based in one Contracting State to be taxed in the other State in which the permanent establishment is situated. The third exception (in Article 13(3)) requires sales of aircraft and ships to be taxable in the Contracting State where the operators have their POEM.
The appellants submitted that on HMRC’s interpretation of article 14 articles 14(2) and (3) would be otiose. However, we consider that the same description of their function applies as that given by Patten LJ here.
He referred again, at [31], to the definition of residence in article 4(1) and noted it is expressly subject to the provisions of article 4(2) and (3) so that “resident” means someone who is so resident after those tie-breaking provisions have been operated when applicable. This is the definition which is imported into article 13(4).
There is no dispute in these cases that the definition of resident of a Contracting State in article 4 of the treaty is imported into article 14 of the treaty.
He noted, at [32], there is no express reference in article 13(4) to the alienator’s residence in the Contracting State being limited to the time of the disposal and so the issue is whether it is necessary to construe the words in that way in order to give effect to the purpose of article 13 and the Mauritius treaty more generally. The judge took the view that article 13(4), like the other provisions in the Mauritius treaty which allocate the right to tax particular types of receipt by reference to the situs of the property, the source of the receipt, the POEM of the trader or the residence of the recipient of the income or gain, was intended to identify which Contracting State was entitled to tax the gain rather than merely to choose a basis of taxation.
At [33] he continued that:
“This analysis supports the submissions of the taxpayer that Article 13(4) is doing the same thing. It is pointing to a single jurisdiction in which tax can be charged, and that is the state of residence. In order to make that workable one has to find a date at which residence has to be judged. There is no other realistic candidate for that point of time other than the date the gain arose (or the date of the disposition, which in this case is probably the same point of time). If there is competition between both states in relation to that point of time, then the tie-breaker applies to produce a single state in respect of which residence (in the status sense used in Article 4) exists.”
At [34] he said that on this basis Mann J concluded that the trustees were resident in Mauritius at the time of the disposal and that there was no concurrent UK tax residence at the time which required the application of article 4(3). He adopted the reasoning put forward by both counsel that there were in the tax year 2000/01 consecutive periods of tax residence by the trustees first in Jersey and Mauritius and then in the UK and that the construction of article 4(1) by the Special Commissioners based on chargeability had, as he put it, no statutory justification because, under the TCGA, the trustees were only tax resident in the UK from March 2001 even though this was sufficient under s 2 to render them chargeable for the earlier gains.
He commented, (a) at [35], that it is, of course, right that the method of avoiding double taxation which is in operation in provisions such as article 13 does in many cases lead to the identification of one of the Contracting States as being the only country eligible to tax the income or gain in question, and (b) at [36] article 13 provides in terms that gains from the alienation of property “other than that mentioned in paragraphs (1), (2) and (3) … should be taxable only in the Contracting State of which the alienator is a resident”:
“Its focus is therefore on specifying the basis of taxation of gains for this residual category of property and nothing else. It is not therefore concerned with how each of the Contracting States chooses to tax gains on the basis of residence. The issue of whether the taxpayer should be resident in the State at the time of disposal or at some other point in time are matters for the State in question to decide as part of its own taxation regime.” (Emphasis added.)
In effect, Patten LJ identified, at [35] and [36] the clear purpose of article 13(4) as he set out. We note that this is not specifically tied to the precise wording of article 4(1) and that it accords with the guidance given in the OECD Commentary on the Model referred to below in the parties’ submissions.
At [37] he said:
“the DTA must be assumed to have been drafted in a way which comprehends any tax treatment of capital gains based on residence or similar criteria. That is what the definition of "resident" in Article 4(1) says and any narrower construction of it would defeat the obvious purpose of the Model Convention. The DTA therefore covers legislation such as s.2 TCGA under which a gain is made taxable in the UK by virtue of residence in a period after the gain has occurred. Neither side on this appeal contends otherwise. The question which this therefore raises (and critically in the present case) is how the DTA is intended to resolve residence/residence based conflicts when the liability under the relevant domestic legislation does not depend upon concurrent periods of residence.” (Emphasis added.)
In light of Patten LJ’s other comments we take the reference to “the obvious purpose” in this context to mean the purpose of avoiding double taxation on any gain arising on residual property due to a person being liable to tax by reason of residence under domestic laws in both Contracting States. Whilst the wording of article 4(1) evidently assisted that construction in that case, the reasoning is not dependent on that wording as is made explicit in his later comments. The point is that article 13(4) would not operate to achieve that intended purpose if it does not, in combination with article 4, (a) recognise that both States can potentially tax the gain due to the residence of the relevant person under domestic law, regardless of the specific domestic law requirements for the period of residence, and (b) operate to provide a mechanism for resolving which State can tax the gain under article 4(3) even though the two periods of residence which render a person potentially liable to tax are not concurrent but consecutive and/or do not coincide with the time the disposal takes place.
At [38] he explained that the taxpayers’ suggested solution was that article 13(4) has to be read as fixing the date of disposal as the reference point for the determination of residence. They accepted that, for the purposes of the applicable treaty, a person can only be “resident” in one Contracting State. But this is achieved on their argument by looking at the factual position as at the date of disposal and nothing more and if that is right:
“it must follow that the tie-breaker in Article 4(3) only applies to concurrent physical residence at the date of disposal and no Contracting State that is party to a DTA in the form of the Model Convention can tax capital gains under Article 13(4) other than by reference to residence at the time. Provisions such as s.2 TCGA can never trump taxation based on residence at the time of the gain.”
At [39] he noted that (a) the High Court accepted the argument that it is necessary to import into article 13(4) a reference to the date of disposal because otherwise the provision is unworkable, (b) but “the snapshot argument is not sufficient in itself to avoid a charge to capital gains tax under the scheme. It also depends (as the judge accepted) on construing “resident” in Article 13(4) as meaning no more than resident for tax purposes in the s.2 sense at that time rather than chargeable to tax in respect of the gain by virtue of a later period of UK tax residence”.
At [40] to [43] he disagreed with the decision of the High Court and set out his conclusions:
“40. For the reasons I have explained, Article 13(4) must, I think, be construed as effective to deal with any liability to taxation for capital gains which either Contracting State may impose regardless of the basis of that charge under the domestic legislation in question. It seems to me unlikely that the draftsman of the Model Convention intended that capital gains which are to be taxable only on the basis of residence should depend exclusively on residence at the date of disposal and so exclude the rights of a Contracting State to tax gains by reference to residence within the same tax year. The definition of “resident of a Contracting State” in Article 4(1) re-inforces this view by making “liability to taxation” by reason of residence the criterion for the taxation of capital gains under Article 13(4). This, I think, must denote what the Special Commissioners described as chargeability and not simply physical residence. That view is, I think, consistent with the purpose of Article 13(4) and avoids descending into whether the UK or Mauritian requirements for residence are satisfied. The definition assumes that they are and allocates the right to tax on the basis that there is liability.
41. For essentially the same reason, I also reject the submission that Article 13(5) confirms that Article 13(4) is concerned with tax residence at the date of disposal and not with chargeability to tax in respect of the gain. The terms of Article 13(5) are not sufficient in my view to contradict the clear scheme of Articles 13(1)-(4) as read in the light of Article 4(1). All that Article 13(5) does is to provide confirmation that residence in previous fiscal years is not excluded as a basis of taxation by Article 13(4). It does not prevent Article 13(4) from applying to attribute the right to tax the gain to whichever of the Contracting States imposes a liability to taxation on the alienator based on domicile, residence or similar criteria in its relevant fiscal year.
42. The emphasis on liability to taxation also disposes of an essential plank in the taxpayers' argument which is that one considers the question of liability as at the date of the disposal and not with the benefit of hindsight. I can see no justification for this in the terms of the DTA. If the provisions of Article 4 (and therefore Article 13(4)) are to cover every form of capital gains tax based on residence then the issue of liability has to be looked at retrospectively having regard to which of the Contracting States seek to make the taxpayer liable for the gain. It seems to me both artificial and self-defeating to ask that question at the date of disposal without regard to the full tax consequences which flow from the gain. That would lead (as it has in this case) to a situation in which the DTA fails to regulate all the tax consequences of the disposal and leads to tax relief being granted even when no double taxation in fact exists.
43. I therefore accept…that the provisions of Article 13(4) are not to be read as incorporating a reference to the date of disposal but (for the reasons already given) I am not persuaded by his submission that one can construe Article 4(1) as meaning no more than tax resident and so avoid any application of the tie-breaker provisions in Article 4(3). The definition of "resident" in Article 4(1) is critical to the meaning of Article 13(4) and Article 4, once applied by the wording of Article 13(4), has to operate in its entirety. The definition of "resident" in Article 4(1) is expressly subject to Article 4(3) which therefore applies whenever the alienator is liable to taxation in both Contracting States in respect of the gain. Article 4(3), as I have explained, is focused on liability for tax regardless of the period of residence under national law which creates that liability. Looked at in this way it becomes meaningless and impermissible to draw a distinction between consecutive and concurrent periods of "residence". The DTA is concerned only with the possibility of a double tax charge on the same gain and not with the period of residence which gives rise to it. If that situation occurs then Article 4(3) operates to resolve the matter as part of Article 13(4) which incorporates it.” (Emphasis added.)
Patten LJ again referred to the purpose of article 13(4) in his comments in [40]. We note that he then said that reference to “liability to taxation” in article 4(1) of that case reinforces this view of the purpose of the provisions. This plainly indicates, therefore, that his analysis of the purpose of article 13(4) in the preceding passages was not dependent on the particular wording in article 4(1) of the treaty albeit that it added further weight to that being the correct interpretation.
When Patten LJ then said that “the emphasis on liability to taxation” also disposes of the argument that “one considers the question of liability as at the date of the disposal”, we take him to mean the emphasis on liability to taxation as evident from the purpose of article 13(4) (as in that case reinforced by the specific wording in article 4(1)).
His comment that assessing the position at the date of disposal without regard to the full tax consequences which flow from the gain would lead to tax relief being granted even when no double taxation in fact exists applies equally in this case. That would be the result on the appellants’ view of how these provisions work.
Again he referred to the provisions being concerned only with the possibility of a double tax charge on the same gain and not with the period of residence which gives rise to it such that article 4(3) operates to resolve the matter as part of the gains provision which incorporates it.
At [44] and [45] he rejected the argument on article 24 for much the same reasons as the High Court did. He noted that although his alternative approach to article 24 is not the same as that adopted by the judge, it has the same consequence. In his view “article 24 fulfils the purpose of providing an ultimate form of relief against double taxation when the earlier provisions of the DTA have the result of allowing the profit to be taxed in more than one Contracting State” and he gave examples of where this might arise.
At [46] he summarised the position as follows:
“…I reject the trustees’ argument that Article 13(4) requires one to look no further than where the trustees were tax resident at the date of the disposal without regard to subsequent events. I also prefer the view of the Special Commissioners that "resident of a Contracting State" under Article 4(1) means chargeable to tax in that State on account of residence and that, for this purpose, one has to take into account the tax treatment of the gain under the domestic legislation of both Contracting States regardless of the period of residence which gives rise to the liability. It follows that I also reject Mr Brennan's argument that there is no need to apply Article 4(3) because the period of residence which gives rise to the UK tax charge in this case under s.2 was consecutive upon the earlier period of Mauritian tax residence up to and including the date of the disposal. It follows from my construction of Article 4(1) that Article 4(3) applies in every case in which there is a “liability to taxation” in both Contracting States.”
In effect Patten LJ’s view was that article 13(4), in combination with article 4, operates to allocate taxing rights to a single State where the taxpayer would otherwise be within the scope of a charge to tax on a disposal of “residual property” due to being “liable to tax” in both States by reason of residence for domestic tax law purposes, regardless of the precise stipulations under domestic law as to when the person must be resident in order to be so liable. Hence, in his view, it is not relevant that under the UK rules a person is liable to CGT on a gain arising on residual property due to residence in a period which does not coincide with the date of the disposal or with the period when the trustees were resident in Mauritius for Mauritius tax law purposes.
Mr Windle made the following main submissions in support of the appellants’ stance on how article 14(4) and 4(1) operate in the treaty:
In August 1983, when the treaty was agreed, both the UK and NZ were members of the OECD and must have been aware of the 1977 Model then in place. At that time, both countries had double tax agreements with other countries that used a definition of residence that was much closer to that Model such as the 1981 Mauritius treaty considered in Smallwood and the 1979 NZ/France agreement. Negotiators would have known that whether a person was liable to CGT in the UK did not perfectly coincide with their residence in the UK for UK tax purposes (under s 2 Capital Gains Tax Act 1979, the predecessor to s 2 TCGA). This is important context and, as a result, in the absence of clear contemporaneous evidence suggesting otherwise, the different definition of residence in the treaty must be understood as a choice taken by the parties to the treaty and construed accordingly. Applying this construction, until the UK trustees were appointed, the Trusts were not resident in the UK for the purposes of the treaty.
This different definition of residence also carries through into the interpretation of article 14(4). Para (4) provides the general rule that gains from the alienation of property are taxable in the Contracting State of which the alienator is resident. Paras (1) and (2) are permissive exceptions that allow the Contracting State in which the alienator is not resident also to tax the alienation of specified types of property. Para (3) provides an exception to para (2) for the alienation of specified types of property. It therefore follows that gains from the alienation of any property other than that referred to in paras (1), (2) and (3), are taxable only in the Contracting State of which the alienator is a resident. This follows from the fact that article 4 is concerned with identifying a single place of residence at each moment in time and that any alternative interpretation would render paras (1) and (2) otiose. Further, it follows from the fact that the treaty is concerned with factual residence and not liability to tax, that the right to taxation in article 14(4) is allocated to the Contracting State in which the alienator is resident.
This is supported by the following example given in the 2000 OECD Commentary:
“... in one calendar year an individual is resident of State A under that State’s tax laws from 1 January to 31 March, then moves to State B. Because the individual resides in State B for more than 183 days, the individual is treated by the tax laws of State B as a State B resident for the entire year. Applying special rules [the tie breaker] to the period 1 January to 31 March, the individual was a resident of State A. Therefore, both State A and State B should treat the individual as a State A resident for that period, and as a State B resident from 1 April to 31 December.”
The different result in Smallwood CA came about because of the different wording in the Mauritius treaty. To apply the same analysis here would require a link between residence under article 4(1) with “liability to tax” or that residence is determined over a period. There is no basis for reading those words into the treaty. HMRC’s approach raises the difficulty that tax years in the two States may not coincide. The UK tax year was relevant in Smallwood because in that case residence was tied to liability to taxation and liability to CGT is tied to a person being resident in the UK for part of a tax year. HMRC’s approach causes difficulties in a number of scenarios. For example, if a company was resident in NZ for NZ tax purposes until midway through the UK tax year and then only resident in the UK for UK tax purposes for the rest of the year, on HMRC’s analysis the tie breaker would apply and it would be necessary to consider how to apply the POEM test – whether over the whole tax year or by reference to the length of time in each place or where the most important decisions were taken. Assessing POEM over a period where there is no concurrent domestic residence becomes unmanageable.
The words of article 4(1) are clear in themselves but they are supported by the context and purpose as divined from the rest of the treaty. Article 4 is a foundational provision. It is applied in every subsequent provision of the convention except articles 25 (exchange of information), 27 (entering into force), and 28 (termination). There are two instances which require the consideration of the situation over a period (articles 15 and 16) which show that where there is a requirement to assess something over a period, that process and period is identified in terms and in neither case is the period tied to the UK tax year.
The case law makes clear that it is instructive in construing a double tax agreement to look back at the previous agreements between the same States. In the original double tax agreement between the UK and NZ it is clear that the test for residence would not have worked if it was applied to a period of time in which there was consecutive residence, because to be resident in NZ a person had to be resident in NZ for the purpose of NZ tax and not resident in the UK for the purposes of UK tax. The 1966 convention and the applicable one do not suggest any major change was intended and certainly not that residence was to depend on liability to tax or that an approach is required looking at a period of time.
As the UK and NZ chose not to follow the Model on article 4(1); the commentary in it should be approached with caution in this context. The 1977 OECD Commentary (a) draws a distinction between taxation on the basis of a connection between the person and the state, such as residence, and taxation on the basis of a connection between the income and/or gains and the State, on the basis of source or situs. Where there is an appropriate connection between the person and the State, generally the State imposes a comprehensive liability to tax on all income and gains, (b) states that double tax agreements do not normally concern themselves with the State’s domestic laws on when a person is subject to comprehensive liability to tax, (c) sets out that when both States claim the right to tax, a choice needs to be made between them, including when both States claim a right to tax on the basis of “residence” (see paras 1 to 7), (d) explains that the definition aims at covering the various forms of personal attachment to a State which, in the domestic taxation laws, form the basis for comprehensive taxation. The Commentary does not state the same thing about companies and it is notable that States are free to expand or reduce the list of criteria set out in article 4(1) in the Model. In the commentary by Robert Couzin which HMRC refer to (see below) he states it is not uncommon for States to add to the list and, whilst it is rarer for States to remove things from the list, it is done. He notes that where the wording used in this treaty is used, all persons who meet the domestic residence test are article 4(1) resident. Equally it must follow that all persons who do not meet the domestic residence test are not article 4(1) resident. So article 4(1) in a tax treaty defines the class of persons that each State claims a default right to subject to comprehensive liability to taxation, as against the other party to the agreement, subject to the tie-breaker. States who enter into double taxation agreements can and often do vary the scope of the class of persons. Where a person is liable to tax under domestic laws of a State but does not fall within the class stated in the agreement, the State has agreed that it does not have the default right to subject such a person to comprehensive liability to taxation; it has made a choice to that effect.
NZ is a former British colony with a common law system. At the time of all three agreements the UK had common law test for both individual and corporate residence and when the treaty was agreed, the common law test for corporate residence in NZ was materially identical to that in the UK. Where both States apply the same test for residence, it is a perfectly sensible choice to allocate the right to tax to the State in which the person is resident. Such a shared common law heritage is the most likely explanation for the other four States HMRC have identified who use similar or the same wording to that in article 4 of the treaty in their agreements with the UK (the US, Israel, Jamaica and Australia). This interpretation does not give a result that is in any way radical and there is no justification for reading the words “liable to tax by reason of” into article 4(1) of the treaty. The UK chose to render persons liable for CGT on gains arising in a tax year when they were not resident, based on their past or future residency in that tax year. The legislation could easily have said if a person is resident in part of the tax year the person is resident in the whole year but it does not. In Smallwood SpC the Commissioners record that by concession chargeability to CGT was often limited to the period of residence, while noting that no concession applied in the case of trustees (see [79] of Smallwood SpC). So the fact that this was the sort of concession that the UK might be willing to make does not seem so surprising.
Smallwood CA is of limited assistance to determining the correct interpretation of the relevant articles given that the treaty under consideration contained different wording. Patten LJ’s comments, in particular, those in [36] and [37], show that Patten LJ’s analysis was reliant on the particular terms of article 4(1) which reflected the same terms in the Model. The UK and NZ did not make the same choice in the wording they include in article 4(1). They chose to give the default or in principle right to impose liability to tax to the Contracting State in which the person is resident for domestic law purposes and, as such, it is necessarily concerned with how the Contracting States chose to tax on the basis of residence. The reasoning in [37] is a product of the different choices the Contracting States have made in article 4(1) in the Mauritius treaty. This is also evident in [41] to [43] where Patten LJ again referred back to article 4(1).
HMRC submitted that the treaty must be purposively construed to achieve its dual purposes as set out by Patten LJ and in the OECD Commentary (see the Commentary on the Model of 2008 at paras 7 to 10 and on the Model of 1977 at paras 7 to 9). Despite some differences from the Model, the treaty operates in the same way as Patten LJ set out and his observations about the purpose of the Mauritius treaty and what articles 4 and 13 are designed to achieve apply equally to the provisions in the treaty. In support of this HMRC made the following main points:
HMRC’s analysis is supported by the relevant OECD Commentary. Article 14(4) of the treaty mirrors article 13(4) of the Model. Accordingly, the OECD Commentary on article 13 and 4 is an admissible aid to the construction of article 14. The Model allows for States to tax the alienation of goods in whatever ways they see fit, and its intention is to provide a framework for all systems of taxation. The 2000 Commentary on article 4 (the 1977 Commentary is in much the same terms): (a) states that article 4 is intended to solve issues of dual residence under the domestic laws of the Contracting States, (b) recognises that, under domestic law of a State, a person may be liable to tax for a full year even if he is only resident in that State for part of the year, (c) makes the point made by Patten LJ that the purpose of the double taxation convention is not to interfere with the domestic rules on residence or lay down any standards for full liability to tax, and (d) sets out the concept of residence and the different criteria that are applied because there are different connecting factors that might make a person liable to tax. In the treaty this extended definition is simply replaced with the compendious line: resident for tax purposes which in effect wraps up all of those different concepts into one more simple test. The 2000 commentary on article 13 states that, “[a] comparison of the tax laws of the OECD Member countries shows that the taxation of capital gains varies considerably from country to country” (para 1), sets out a number of specific issues raised by different systems of taxation and then states:
“3. The Article does not deal with the above-mentioned questions. It is left to the domestic law of each Contracting State to decide whether capital gains should be taxed and, if they are taxable, how they are to be taxed…It is understood that the Article must apply to all kinds of taxes levied by a Contracting State on capital gains.”
It is clear from this that, as recognised by Patten LJ in Smallwood, the relevant articles are not intended to interfere with UK domestic taxation. They are intended to solve issues of actual double taxation and prevent fiscal evasion. There is no relevant distinction between being resident in a State for the purposes of that State’s tax and being liable to taxation in a State by reason of residence (and similar connecting factors). The test in the treaty is merely a simplification of the wording in the Model. That article 4(1) of the treaty states resident in the UK for the purposes of UK tax brings in the UK’s domestic approach to taxation and chargeability. The fact that the UK chooses to tax a person on gains made during another part of the tax year, before the person was resident, is a matter of UK domestic law (see s 2 and s 69 TCGA as regards trustees). In effect, the UK CGT rules render a person “resident” in the sense of liable to CGT on disposals made in a tax year if the person is resident only for part of the tax year. That is no reason to construe article 14(4) so as to limit the enquiry into residence to the time of alienation and exclude residence at a later part of the same tax year. It is legitimate to read into the treaty that residence is for a UK tax year because the concern is with residence for the purposes of UK tax and that is simply how the UK rules work. There is no split year treatment in these circumstances under UK law
If the appellants’ argument were correct, it would never be possible for the UK to impose a charge to tax under s 2 when a settlement was not resident in the UK at the moment of alienation. That would be a substantial interference with the UK’s freedom to determine how to tax capital gains and inconsistent with the fact that the treaty is not concerned with how each State chooses to tax gains and, in a case such as this, result in fiscal of evasion of the type the treaty is intended to prevent. There is no reason to believe that the negotiators of the treaty intended such a radical result. That is exactly the sort of scenario the Commentary makes clear double tax agreements are designed to prevent as one of their purposes is to prevent fiscal evasion.
It is not correct that an interpretation of the treaty that does not look only at residence at a certain point in time would render articles 14(1) and (2) otiose. The Court of Appeal’s construction in of article 13(2), which is identical to article 14(2) did not render it otiose (see [30]). As set out above, we accept that point.
Whilst it must be the case that residence (as opposed to liability) is determined by reference to a point in time, there is no reason that the particular point in time should be the moment of alienation. As noted, the wording is a simplification of that in the Model, which is not intended to have a fundamentally different meaning:
The phrase used in the Model/Mauritius treaty is relatively obscure and its application may require the interpretation of the phrases “liable to taxation” and “criterion of a similar nature”. In Corporate Residence and International Taxation (Amsterdam: IBFD Publications BV, 2002), Robert Couzin argues that using “a form of residence definition that does not rely upon the “liable to tax” test upon which article 4(1) of the Model is based:
“…circumvents the entire discussion of “liable to tax”. All companies that meet the local residence test are treaty residents. This course may be open where the treaty partners can determine that the universe of “residents” in each contracting state is congruent with the intended universe of treaty residents, barring only the matter of dual residence that is dealt with separately in the tie breaker rule. In any particular convention, the implications for transparent or pass-through entities, tax-exempt persons and other hard cases would have to be considered. Nonetheless, in appropriate circumstances this model has considerable attraction.” (pages 128 to 129)
He also said that where neither Contracting State uses a particular connecting factor in the list set out in article 4(1) of the Model, it might be best to delete it. He commented that:
“The disadvantage of such a deletion could be some loss of pre-emptive flexibility, to accommodate unanticipated changes in the criteria utilized by one of the states. The advantage is to remove a source of possible ambiguity or uncertainty. Neither the United States nor Canada require the reference to “place of management” in Article IV(1) of their convention and, had it not been there, the Crown Forest problem would probably never have arisen.” (see page146)
The most likely explanation for the wording of the treaty is as Couzin states. Similar or the same definitions to that in the treaty appear in the UK/USA double tax agreement of 1980 (at article 4(1)), the UK/Israel double tax agreement of 1963 (at article 2(1)(h)(i)), the UK/Jamaica double tax agreement of 1973 (at article 3(1)), and the UK/Australia double tax agreement of 2003 (at article 4(1)). Although Mr Couzin’s book is some years old, it was referred to and cited by the Court of Appeal in GEFI as one of the academic texts that they relied upon. The use of this definition in the agreements with these countries may relate to the fact the States are broadly common law types of jurisdiction. The fundamental point is that this wording is a simplification of the wording in the Model; it is not a fundamental change of the wording and does not lead one to interpret article 4 and 14 of the treaty any differently to the corresponding articles in the Mauritius treaty/the Model.
One can get no assistance in interpreting the treaty from looking at the wording of the previous double tax agreements between the UK and NZ. In particular, the first agreement Mr Windle relied upon predated the Model and the concept of CGT. The wording was quite strange and did not provide for anything like the current circumstances. Mr Windle’s approach of seeking to trace that provision through the later agreements is an impermissible approach on the facts of this case and given the date of the prior agreements as an aid and interpretation of the current treaty. It is not apparent how the example Mr Windle gave relating to an individual assists his case
In our view, the analysis set out by Patten LJ in Smallwood CA applies in the same way to the interpretation of the treaty as it does to the application of the Mauritius treaty. We do not regard the difference in wording in article 4(1) of the treaty compared with that in article 4(1) of the Mauritius treaty/the Model as leading to a different result. The appellants’ analysis and reasoning was very focused on article 4(1) and its application in general terms and not on how that provision is to be construed in conjunction with article 14(4), specifically in light of the purpose and object of that article. There is no material difference in the wording of article 14(4) of the treaty and article 13(4) of the Mauritius treaty/wording of the Model. There is no reason to suppose that in using the simplified definition of a “resident of a Contacting State” the States intended that the gains provisions in the treaty would operate in a different way than the relevant OECD Commentary indicates the equivalent provisions in the Model should operate, as explained by Patten LJ in Smallwood CA.
As the appellants’ emphasise, Patten LJ’s comments were to some extent tied in to the specific wording of article 4(1) of the Mauritius treaty which differs to that in article 4(1) of the treaty, as that was the wording the Court of Appeal was faced with in that case. However, we cannot see that the use of the simplified wording in article 4(1) of the treaty alters what Patten LJ took to be the plain purpose of the gains provisions (in that case in article 13(4) of the Mauritius treaty), as accords with the OECD Commentary on them. Moreover, as set out above, we do not consider his analysis of that purpose to be dependent on the specific wording of article 4(1) of the Mauritius treaty.
We recognise of course that, as Mr Windle emphasised, the UK and NZ made a choice to use the simplified version of the definition of “resident of a Contracting State” in the treaty rather than to use the wording in the Model. It is not readily apparent why that was. The appellants’ suggestion that it may emanate from the common law background of each the UK and NZ be correct. In our view, however, there is nothing to suggest in any of the materials to which the appellants’ referred that this choice was made with a view to article 14, as an article in which the residence test is critical, operating in a way that which would give a result which is clearly contrary to the overall purpose and object of the Model and of the gains provisions specifically, as those objects and purposes are explained in the OECD Commentary. The purpose of article 14(4), reflecting as it does the terms of the Model, is to avoid double taxation in the manner set out by Patten LJ as regards the corresponding provisions in the Mauritius treaty. As in Smallwood, the appellants’ interpretation would lead to the result described by Patten LJ in which the treaty would fail to regulate all the tax consequences of the disposals and would lead to tax relief being granted even when no double taxation in fact exists (as NZ does not tax the gains arising on the disposals, as was similarly the situation in Mauritius as regards the gains arising to the taxpayers in Smallwood).
Interpreting the provisions with their purpose in mind we do not regard this interpretation of them as leading to an impermissible reading in of words. We cannot discern any difference in the underlying purpose of article 4 in defining a resident of a “Contracting State” as a person who is resident in that State “for the purposes of tax” as opposed to a person who is “liable to tax….by reason of residence”. The use of the words “for the purposes of…tax” plainly indicates that the aim is to identify persons who, under the domestic laws of the relevant State, are liable or chargeable to tax in that State, as Mr Windle put it, on a comprehensive basis, by reason of the stated connection with the State. As in the Mauritius treaty and the Model, in article 4 and the gains provisions there is no reference to any specific time period during which “residence for the purposes of tax” or in that case “liability to tax by reason of residence” is to be assessed. However, as in relation to the Mauritius treaty, it would undermine the clear purpose of article 4, as imported into article 14(4), were it not to capture as a “resident of a Contracting State” a person who is within the scope of a tax charge on the gain in that Contracting State due to being resident in that State regardless of precisely how that charge is imposed/when the residence occurs.
We do not consider that this approach raises a difficulty in that the tax years in the two States may not coincide. The intention is to ensure that a person is captured as a resident of a Contracting State if he is liable/chargeable to tax in that State as a resident of it regardless of the precise basis of taxation in the State. We have not found it useful to consider hypothetical examples of how the residence test may apply in different contexts and cannot see how the example given by Mr Windle above supports the appellants’ case. We accept HMRC’s submissions on the relevance of the OECD Commentary and that, in this context, the very different provisions in the earlier double tax agreement between the UK and NZ gives no material guidance on how these provisions are to be interpreted.
POEM test
The parties disagreed on how the POEM test in article 4(3) of the treaty is to be applied They both referred to various comments in the different versions of OECD Commentary on article 4(3) of the Model. The Court of Appeal in Haworth CA set out a full explanation of the Commentary on the Model as follows:
“43. The OECD Commentary accompanying the 1977 version of the Model Convention included these comments on Article 4(3):
“21. This paragraph concerns companies and other bodies of persons, irrespective of whether they are or not legal persons. It may be rare in practice for a company, etc. to be subject to tax as a resident in more than one State, but it is, of course, possible if, for instance, one State attaches importance to the registration and the other State to the place of effective management. So, in the case of companies, etc., also, special rules as to the preference must be established.
It would not be an adequate solution to attach importance to a purely formal criterion like registration. Therefore paragraph 3 attaches importance to the place where the company, etc. is actually managed.
The formulation of the preference criterion in the case of persons other than individuals was considered in particular in connection with the taxation of income from shipping, inland waterways transport and air transport. A number of conventions for the avoidance of double taxation on such income accord the taxing power to the State in which the “place of management” of the enterprise is situated; other conventions attach importance to its “place of effective management”, others again to the “fiscal domicile of the operator”. Concerning conventions concluded by the United Kingdom which provide that a company shall be regarded as resident in the State in which “its business is managed and controlled”, it has been made clear, on the United Kingdom side, that this expression means the “effective management” of the enterprise.
As a result of these considerations, the “place of effective management” has been adopted as the preference criterion for persons other than individuals.””
In a policy paper published in 1990, “Statement of Practice 1 (1990)”, HMRC said this in paragraph 22 about article 4(3) of the Model Convention:
“The commentary in Article 4 paragraph 3 of the OECD Model records the UK view that, in agreements (such as those with some Commonwealth countries) which treat a company as resident in a state in which 'its business is managed and controlled', this expression means 'the effective management of the enterprise'. More detailed consideration of the question in the light of the approach of continental legal systems and of community law to the question of company residence has led HMRC to revise this view. It is now considered that effective management may, in some cases, be found at a place different from the place of central management and control. This could happen, for example, where a company is run by executives based abroad, but the final directing power rests with non-executive directors who meet in the UK. In such circumstances the company’s place of effective management might well be abroad but, depending on the precise powers of the non-executive directors, it might be centrally managed and controlled (and therefore resident) in the UK.”
When the next version of the Model Convention was published in 1992, Article 4(3) was unchanged except that “the State in which” was substituted for “the Contracting State in which”. However, the Commentary now omitted what had been the last sentence of paragraph 23 so that the paragraph read:
“The formulation of the preference criterion in the case of persons other than individuals was considered in particular in connection with the taxation of income from shipping, inland waterways transport and air transport. A number of conventions for the avoidance of double taxation on such income accord the taxing power to the State in which the “place of management” of the enterprise is situated; other conventions attach importance to its “place of effective management”, others again to the “fiscal domicile of the operator”.”
In the 2000 version of the Model Convention, Article 4(3) was still in the same terms, save that “only” was inserted between “a resident” and “of the State in which its place of effective management is situated”. The Commentary was, however, revised to include this at paragraph 24:
“As a result of these considerations, the “place of effective management” has been adopted as the preference criterion for persons other than individuals. The place of effective management is the place where key management and commercial decisions that are necessary for the conduct of the entity's business are in substance made. The place of effective management will ordinarily be the place where the most senior person or group of persons (for example a board of directors) makes its decisions, the place where the actions to be taken by the entity as a whole are determined; however, no definitive rule can be given and all relevant facts and circumstances must be examined to determine the place of effective management. An entity may have more than one place of management, but it can have only one place of effective management at any one time.”
No changes were made to Article 4(3) in the 2008 version of the Model Convention, but part of the third sentence of paragraph 24 of the Commentary was omitted so that paragraph 24 now read:
“As a result of these considerations, “the place of effective management” has been adopted as the preference criterion for persons other than individuals. The place of effective management is the place where key management and commercial decisions that are necessary for the conduct of the entity’s business as a whole are in substance made. All relevant facts and circumstances must be examined to determine the place of effective management. An entity may have more than one place of management, but it can have only one place of effective management at any one time.”
The appellants submitted that:
It is highly relevant that the OECD Commentary on the 1977 Model includes no observations by the UK and the following observation by NZ:
“25. New Zealand’s interpretation of the term “effective management” is practical day to day management, irrespective of where the overriding control is exercised.”
The Commentary sets out the role of an observation at para 27 as follows:
“27. Observations on the Commentaries have sometimes been inserted at the request of some Member countries who were unable to concur in the interpretation given in the Commentary on the Article concerned. These observations thus do not express any disagreement with the text of the Convention, but furnish a useful indication of the way in which those countries will apply the provisions of the Article in question.”
Following the agreement of the treaty, the NZ Inland Revenue Department published a “Public Information Bulletin” providing commentary on the terms of the treaty which states of article 4(3):
“Paragraph 3 provides that where a company or other legal person is resident in both countries, residence will be where the place of effective management of the enterprise is situated. In this context, New Zealand views the term “effective management” as meaning the practical day to day management, irrespective of where the overriding control is exercised.”
For a convention to be effective both signatories have to apply the provisions in the same way. NZ Inland Revenue Department Public Information Bulletins have been used by the NZ courts to assist with ascertaining legislative intention (see for example Marac Life Insurance Ltd v Commissioner of Inland Revenue [1986] 1 NZLR 694, 713 (CA)). On 11 April 1977 the Council of the OECD, which included an ambassador from the UK, recommended that governments:
“when concluding new bilateral conventions or revising existing bilateral conventions between them, to conform to the Model Convention set out in the Annex hereto (hereinafter referred to as the “Model Convention”), as interpreted by the Commentaries thereto and having regard to the reservations and derogations to the Model Convention which are contained in the Report referred to above”.
This confirms that the UK negotiators agreed with, or at least acquiesced to, NZ’s publicly stated interpretation of the phrase “effective management”. The observation is therefore part of a general approach, agreed by the UK, and at a minimum forms part of the background context against which treaty was agreed.
The decision to include an observation is indicative of a particular strength of feeling about the approach to a particular provision in the Model. The UK had previously agreed in the 1966 convention that a company incorporated in NZ would be treaty resident in NZ if it’s centre of administrative or practical control or practical management was in NZ regardless of where higher level control was exercised. The UK had accepted the possibility of using this as a tie-breaker test. Companies are likely to be the most common non-individual persons considered by the treaty. It would therefore be a smaller concession for the UK to extend administrative or practical management as a tie-breaker to non-company, non-individual persons than it would be for NZ to agree to use a different tie-breaker for companies.
In the Irish Bank case Patten LJ said that later commentaries are admissible unless the substantive changes are inconsistent. NZ removing its observation is a substantive change that is plainly inconsistent with the Commentary in existence at the time of the 1984 treaty. In this case, therefore, later OECD Commentary is not admissible as an aid to interpretation of the POEM test or at least must be treated with great caution.
In these circumstances, the only sensible interpretation of POEM in the treaty is the place of practical day to day management irrespective of where overriding control is exercised. It is clear that such management was in NZ at the relevant time.
The alternative position, if this is found not to be correct, is that the second most sensible construction of the POEM test is the CMC test set out in De Beers Consolidated and further explained in Wood v Holden given that this test was laid down by the House of Lords at a time when litigants in NZ had a right of appeal to the Privy Council.
However, the appellants accept that if it is established that the correct test for POEM in the treaty is that propounded in Smallwood CA as understood in Haworth UT, then in this case POEM of each of the Trusts was in the UK.
We note that, since the hearing, the Court of Appeal has held in Haworth CA that the tribunal took the correct approach in that case in interpreting the decision in Smallwood CA as requiring a broader test than the CMC test to be applied for determining POEM, for the purposes of article 4(3) of the Mauritius treaty. We note that article 4(3) of the treaty is in the same terms as article 4(3) of the Mauritius treaty. We have set out details of the relevant part of the decision and the parties’ submissions on it below.
HMRC argued that:
The unilateral opinion or practice of a tax authority is not an admissible aid to interpretation (see Irish Bank at [23], [59] and [60]):
A double tax agreement is an agreement between two sovereign states and an observation by one of them on the Model cannot bind the other. If the parties have a different understanding of the words they have agreed to use in a treaty that does not, as Mr Windle suggested, lead to the agreement being unenforceable. There is no need, therefore, to try to find any meeting of minds between the UK and NZ on the meaning of POEM. If the parties want to record their agreement as to the meaning of the words used, they can do that by way of a protocol, and they often do. If they want to reach an agreement on a particular form of words they can negotiate that form of words into their treaty. An observation is very limited; it is simply a comment on the commentary not, unlike a reservation, on the text of the treaty itself as shown by the OECD guidance set out above
Moreover, on the basis of the comments in Irish Bank, as quoted with approval in GEFI, it is arguable that the observation is in fact inadmissible when determining the meaning of POEM under English law. Certainly it could be said that the unilateral observation of NZ is its unilateral opinion of how the matter should be interpreted. Even if it is not for that reason formally inadmissible, it is certainly a matter that should be given little weight when determining the correct meaning of POEM.
NZ’s observation was deleted from the 2000 version of the OECD Commentary on article 4, indicating the agreement of the NZ authorities with para 24 of that version of the Commentary. It appears that, as NZ did not feel the need to make any observation on the fuller commentary, NZ was content with this version of the Commentary. Hence the appropriate test for POEM is that set out by the Commissioners in Smallwood SpC, as approved by the Court of Appeal in that case and by the Upper Tribunal in Haworth UT. This version of the OECD Commentary informed the decision in those cases. Therefore, even if the views of the NZ authorities were an admissible aid to construction of the treaty, their apparent agreement with the 2000 version would indicate that POEM should be applied consistently with Smallwood, as explained in Haworth. The latest version of the Commentary that has persuasive value on the determination of POEM (prior to the amendment to the Model Treaty) was that adopted in 2008. As set out above, versions of the OECD Commentary that post-date the particular treaty being construed “are to be given such persuasive force as aids to interpretation as the cogency of their reasoning deserves”. This is also supported by paras 35 and 36 of the Introduction to the 2017 version of the Model. In any event the later commentary on POEM does not contradict the earlier commentary. In 1977 there was very little commentary on what is meant by POEM. Effectively there has been an evolution of the commentary over time.
As the Upper Tribunal found in Haworth UT (see [110] and [116]), the Commissioners in Smallwood SpC specifically rejected the test of CMC as the test for determining POEM and that conclusion was endorsed by the majority in Smallwood CA. This tribunal is bound as a matter of authority to similarly reject the CMC test as the appropriate test for determining POEM and to follow the POEM test set out in those cases. This is now reinforced by the decision in Haworth CA.
Mr Windle replied that (1) the purpose of the treaty is to avoid double taxation and a tie-breaker provision that produced different outcomes in each State would be undesirable. HMRC did not refer to any authority where such an approach has been taken. The tribunal should strain to avoid that result, (2) the observation was in the OECD Commentary that was current at the time, it was clearly part of the context, and therefore substantial weight should be given to it, particularly in light of the history of the convention and the fact that a test akin to day-to-day management was used as one of the tie-breakers when the treaty was agreed, and (3) the removal of the observation subsequently is a matter to which no real regard should be had. The treaty is not ambulatory in the sense that its meaning changes over time. Later Commentaries are relevant only according to the cogency of their reasoning. The most credible interpretation is that NZ changed its mind over time. The fact that it changed its mind later gives no guidance as to what was intended by the parties when the treaty was entered into. It is not an expansion of the reasoning and it is not credible to say that NZ realised that the Commentary in 2000 or subsequent Commentaries properly explained its position as set out in the observation.
In summary, our view is that the observation relied on by the appellants does not provide a basis for the tribunal to apply a different POEM test to that applied by the Court of Appeal in Smallwood CA and in Haworth CA:
On the basis of the approach set out in cases such as Irish Bank and GEFI it seems doubtful that, in interpreting article 4(3), the tribunal can or should take account of a unilateral statement made by NZ of how NZ viewed the POEM test which is made only as an observation. We consider that this has been reinforced by comments made in the decision in Haworth CA as set out below.
The only guidance in the Commentary we were referred to indicates that an observation by a Contracting State, included in effect as a note in the Commentary, is to be regarded simply as an indication of how that particular State will apply the provision to which the section in the commentary/observation relates. We do not take from this that the other Contracting State to a treaty is to be taken to have agreed and concurred in the relevant interpretation/application of the particular provisions. Nor can that be taken from the recommendation to which the appellants referred. The recommendation refers to reservations (and derogations) which, as HMRC submitted, have a different role to the limited role of an observation. We cannot see any basis for the appellants’ assertion that the treaty is “ineffective”, by which they apparently mean unenforceable, unless both signatories interpret the provisions in it in the same way.
In any event, it is reasonable to suppose that the fact that the observation was not included in the 2000 Commentary indicates that NZ were content with the fuller explanation of the POEM test in that Commentary. In our view, that is the natural implication of the observation no longer being included. We do not accept the appellants’ submission that the removal of the observation constitutes a change of the kind which would preclude the tribunal having regard to the later 2000 Commentary. As noted, an observation is not part of the Commentary itself as such and does not function in the same way as a formal reservation or derogation. Moreover, the appellants’ approach of disregarding the later Commentary entirely is out of kilter with the approach to later Commentaries set out in the case law and by the OECD as it involves shutting one’s eyes to advances in international thinking. The Court of Appeal has confirmed in Haworth CA that it is appropriate to have some regard to the 2000 Commentary in setting out their understanding of the POEM test in the Mauritius treaty although they considered that the intended approach is evident from article 4(3) of the Mauritius treaty/Model itself.
Correct approach to the POEM test – Haworth CA
In Haworth CA, the leading judgment was given by Newey LJ with whom the rest of the panel agreed. Having set out the case law on the correct approach to interpretation and details of the OECD commentaries, he explained that the issue was the meaning of POEM as that phrase is used in the Model and, hence, the Mauritius treaty. He noted that the UK was not of course the only party to the Mauritius treaty or of the OECD commentary such that the construction of POEM is not to be approached in the same way as if the words appeared in domestic legislation. Whilst no international tribunal exists to rule on the interpretation of the term, it nevertheless “has an autonomous meaning and falls to be construed in a manner which is “international, not exclusively English” (see [49]).
He explained, at [50], that the taxpayers in that case relied on the 1977 Commentary on the Model as an indication that during the period when the Mauritius treaty was concluded POEM was equated with place of CMC. This was on the basis of the comment in para 23 that, as regards conventions concluded by the UK which provide that a company shall be regarded as resident in the State in which “its business is managed and controlled”, “it has been made clear, on the United Kingdom side, that this expression means the “effective management” of the enterprise”. The taxpayers argued that showed that at the time the Mauritius treaty was concluded “effective management” was understood to be located in the place of the relevant entity’s CMC. Newey LJ rejected this, at [51], and commented that even “if the sentence casts light on the United Kingdom’s understanding, that is not determinative. Other States cannot be taken to have subscribed to the United Kingdom’s perception” (emphasis added).
At [52] and [53] he said that the concept of CMC is not well-suited to perform the function of POEM:
“Both the Model Convention and the Treaty proceed on the basis that there will be only one POEM at a point in time. Thus Article 4(3) of the Treaty provides for a person to be deemed to be a resident of “the” Contracting State in which its POEM is situated. Similarly, Article 13(4) states that capital gains are to be taxable only in “the” Contracting State of which the alienator is a resident, Article 8(1) stipulates that certain profits are to be taxable only in “the” Contracting State in which the POEM of the enterprise is situated, and Article 8(2) also speaks of “the” POEM. By 2000, the Commentary explained that, while “[a]n entity may have more than one place of management,…it can have only one place of effective management at any one time”. Those comments were not in existence when the Treaty was entered into, but they reflect its terms. POEM was clearly intended to serve as a tie-breaker.
CMC, in contrast, does not necessarily produce a single answer. Although an entity normally has only one place of CMC, the authorities mentioned in paragraphs 31 to 34 above show that that will not always be the case. That suggests that CMC is not well-suited to fulfil the role of POEM.”
At [54] he noted that the Commissioners referred to this in Smallwood SpC and set out their comments at [111] and [112] of their decision. At [55], he rejected the appellants’ argument that the approach of the Commissioners in Smallwood SpC was problematic because they had had resort to the 2000 Commentary on the Model, which post-dates the Mauritius treaty. He said that the Commissioners recognised that:
“[t]he relevance of commentaries adopted later than the treaty is more problematic because the parties cannot have intended the new commentary to apply at the time of making the treaty”: see paragraph 99 of their decision. On the basis that ignoring them would mean “shutting one’s eyes to advances in international tax thinking”, the Special Commissioners took the view that “[t]he safer option is to read the later commentary and then decide in the light of its content what weight should be given to it”: see paragraph 99.”
He continued that, in any event, at [56], the passage from the Commissioners’ decision he had quoted above did not depend on what was said in the 2000 Commentary:
“POEM’s role as a tie-breaker is evident from the terms of the Treaty and the 1977 Model Convention themselves. Not only, therefore, is there no good reason to take other States to have accepted that CMC, as developed in our domestic case law, should govern the interpretation of POEM, but CMC is not designed to serve the same purpose as POEM and may not yield the single answer which POEM needs to supply.”
At [57] he said: “The position is instead that it can be seen from the Treaty that there was meant to be just one POEM.”
Newey LJ rejected the appellants’ argument that it is relevant that in Wood v Holden, Chadwick LJ said he did not discern a significant difference between CMC and POEM (see [6] and [44]). He said, at [59], these comments are not of assistance; they were obiter, they were to an extent tied to “the circumstances which the commissioners had to consider” and “the circumstances of this case” and, “more importantly, Chadwick LJ did not explain his reasoning”.
He set out his view of the POEM test at [60] to [65] as follows:
“60. Returning to the language of Article 4(3), that requires the focus to be on the place of “effective management”. Addressing the location of a trust’s POEM in Wensleydale’s Settlement Trustees v Inland Revenue Commissioners [1996] STC (SCD) 241, a Special Commissioner (Mr D A Shirley) stressed the word “effective”. He said at 250-251:
“I confess that I am not persuaded that realistically the place of effective management of the settlement was in the Republic of Ireland. I emphasise the adjective “effective”. In my opinion it is not sufficient that some sort of management was carried on in the Republic of Ireland such as operating a bank account in the name of the trustees. “Effective” implies realistic, positive management.””
61. In Smallwood SpC, the Special Commissioners adopted the reference to “realistic, positive management”: see paragraph 114. It also strikes me as apposite.
62. Approaching matters on that basis, it seems to me that POEM can potentially be in a place other than that in which, applying the Wood v Holden approach, CMC would be located. Wood v Holden shows that, where a company is incorporated in a country and its constitutional organs make their decisions there, its CMC will also be there unless “the functions of the company’s constitutional organs are usurped, in the sense that management and control is exercised independently of, or without regard to, its constitutional organs, or if an outsider dictates decisions (as opposed to merely proposing, advising and influencing decisions)”. I do not think the position will necessarily be the same as regards POEM. Identifying the place of “effective” (or “realistic, positive”) management allows matters to be looked at somewhat more broadly.
63. More specifically, it can, as it appears to me, be legitimate to have regard to the circumstances in which trustees from a particular jurisdiction were appointed. The mere fact that, in making decisions during the period they were in office, the trustees made proper decisions in accordance with their duties without the decisions being dictated by outsiders or their functions being “usurped” will not automatically mean that the POEM was in the relevant jurisdiction. If the trustees were appointed because it was appreciated that fulfilment of their responsibilities would cause them to take the decisions, the POEM of the trust might not be in the jurisdiction from which the trustees come.
64. Mr Rivett expressed concern as to the position of professional advisers were POEM to be understood in that way. However, there can be no question of POEM being situated in a country just because advice is given from it. It is the decision-makers who matter. The fact that trustees from, say, Mauritius take advice from professionals in, say, London cannot of itself result in the POEM of the trust being in the United Kingdom.
65. The present case, like the Smallwood case is different. In each, the role of the trustees in Mauritius was effectively pre-determined. In this case, as the FTT explained, the settlors, albeit with the advice and assistance of advisers, decided to adopt “an overall single plan” and, to that end, exercised their powers to appoint the Mauritius Trustees for a limited period “in the confident expectation that they would implement the plan”. While the Mauritius Trustees genuinely made decisions and, in doing so, complied with their responsibilities, there was every reason to believe that they would decide as they in fact did and so further the “overall plan”. Even, therefore, during the period in which the Mauritius Trustees were in office, “effective” or “realistic, positive” management was elsewhere. The decisions which the Mauritius Trustees made had been pre-ordained and the Mauritius Trustees were doing no more than the settlors had (with good reason) foreseen. The Mauritius Trustees were (without impropriety) playing their parts in a script which had been written by others.” (Emphasis added.)
He said, at [66], that the understanding of POEM he had set out is consistent with, and supported by, the decision in Smallwood CA and, at [67], he set out their conclusion at [140] of their decision:
“The facts surrounding the appointment of PMIL lead us to the view that the real top level management, or the realistic, positive management of the trust, remained in the United Kingdom. We accept that the administration of the trust moved to Mauritius but in our view the “key” decisions were made in the United Kingdom.”
He continued that, expanding on this, the Commissioners said this at [143] to [145] of their decision:
“143. We fully accept that the decision to sell the shares that day was taken by the directors of PMIL at the telephone meeting on 10 January 2001. We also accept that if, for example, the price of the shares had fallen to a level that meant that no gain would be realised on their disposal, the shares would not have been sold but would have been retained and perhaps sold later. Nevertheless, in our view this was a lower level management decision as there was no doubt that the shares would be sold; the real top level management decisions, or the realistic, positive management decisions of the trust, to dispose of all the shares in a tax efficient way, had already been, and continued to be, taken in the United Kingdom. The “key” decisions were made in the United Kingdom.
144. Finally the events after the sale of the shares confirm our view. The tax planning exercise was completed by the appointment of United Kingdom trustees….
145. We conclude that the state in which the real top level management, or the realistic, positive management of the trust, or the place where key management and commercial decisions that were necessary for the conduct of the trust’s business were in substance made, and the place where the actions to be taken by the entity as a whole were, in fact, determined between 19 December 2000 and 2 March 2001 was the United Kingdom.”
At [71] he set out that Hughes LJ, with whom Ward LJ agreed on this issue (Patten LJ dissented on this point), considered that the Commissioners had been entitled to find that the POEM was in the UK and quoted Hughes LJ’s conclusions on this in full (at [66] to [70] of Smallwood CA). At [72] he said he did not find this (relatively compressed) judgment easy to interpret in every respect but he concluded, at [73], that it is clear that Hughes LJ considered that the POEM should not be determined by reference only to the circumstances at the “moment of disposal” or on Wood v Holden principles:
“He evidently took a broader view and considered that the POEM could be found to be in the United Kingdom on the basis that there was “a scheme of management of this trust which went above and beyond the day to day management exercised by the trustees for the time being, and the control of it was located in the United Kingdom”. He also appears to have endorsed paragraphs 140 and 145 of the Special Commissioners’ decision (quoted in paragraph 66 above) in which, relying on the “facts surrounding the appointment of PMIL”, the Special Commissioners concluded that in the period when PMIL was in office “the state in which the real top level management, or the realistic, positive management of the trust, or the place where key management and commercial decisions that were necessary for the conduct of the trust's business were in substance made, and the place where the actions to be taken by the entity as a whole were, in fact, determined” was the United Kingdom.”
He concluded, at [74], that the tribunal was right not to determine the POEM of the Trusts by reference to the principles seen in Wood v Holden.
The appellants submitted that:
Newey LJ made clear that the appeal was concerned only with the meaning of POEM as that phrase was used in the Mauritius treaty (at [1], [18] and [23]) and did not discuss article 4(1) of the Mauritius treaty beyond briefly summarising the approach taken in Smallwood (see [68]).
The approach taken by Newey LJ supports the appellants’ construction of POEM in the treaty in that he interpreted POEM in the Mauritius treaty purposively, with a particular focus on its role as a tie-breaker (at [52] and [56]). By similar reasoning, a residence tie-breaker that produces different results in the two Contracting States would frustrate a core purpose of the treaty, namely, to prevent double taxation. HMRC’s submission that it is reasonable to expect that two parties to a treaty might enter into it expecting to interpret a tie-breaker provision in different ways is plainly wrong. If the States had been unable to agree how to break a residency tie for persons other than individuals, it is much more likely that they would have used the mutual agreement procedure (which was already used as a fallback for individuals in article 4(1)(d)). For the treaty, as a member of the OECD, NZ had included its interpretation of POEM as an observation in the OECD commentary on the Model that was current at the time the treaty was agreed. In the absence of any indication that an alternative interpretation had been agreed, and where the UK had not made its own observation, the conclusion a reasonable reader would reach is that the parties had intended to adopt NZ’s interpretation.
HMRC submitted that:
The Court of Appeal’s rejection of the appellants’ submission that the UK interpretation was to be adopted is directly applicable to the appellants’ reliance upon NZ’s observation. The UK cannot have been taken to have agreed or acquiesced with NZ’s observation when concluding the treaty. Further, the subsequent publication of NZ’s unilateral position cannot alter the autonomous meaning of article 4(3). In those circumstances, the correct interpretation of POEM within the treaty is the same “autonomous” and “international” interpretation of the phrase explained by the Court of Appeal in the Mauritius treaty.
The Court of Appeal’s judgment in Haworth CA effectively endorses the test for POEM set out in Smallwood CA as understood and applied by the tribunal and UT in that case. The Court’s conclusion (at [74]) that POEM of the trusts should not be determined by reference to the principles in Wood v Holden is binding on this tribunal and equally applicable to the treaty.
For all the reasons set out above, our view is that the POEM test in the treaty is to be applied in the same way as the Court of Appeal has decided it should be applied in both Smallwood CA and Haworth CA. We consider that, as HMRC submitted, the approach taken to the interpretation of the POEM test in Haworth CA does not support the appellants’ interpretation; rather it reinforces the view that it is not correct to adopt the interpretation/application of the POEM test in the treaty set out by NZ only in the observation. It is entirely clear from the decision in Haworth CA that for POEM under the Mauritius treaty/Model, and in our view, of POEM under the treaty, is a broader one than the CMC test.
We have concluded that, applying the POEM test as explained in Haworth CA, on the basis of the evidence and findings of fact set out above, POEM of the Trusts was in the UK throughout the period when the RTW planning was implemented. We rely on all the findings set out above, but we note in summary that:
In each case, the role of the NZ Trustees was effectively pre-determined. The settlors, acting through Lansburys and Mr Paul, decided to adopt an overall single plan and, to that end, exercised their powers to appoint the NZ Trustees for a short period in the confident expectation they would implement the plan as required, namely, by selling the shares in the BVI companies to the buyer identified at the outset and resigning shortly after the sale (and in the same tax year) in favour or UK trustees. To the extent that the NZ Trustees made any decisions (and it is not at all clear they even signed the relevant documents for the sale of the shares in the case of the Harris Trust), they were simply playing their part in this pre-ordained plan by carrying out the formal acts required for each step to take place.
Therefore, even during the very short period in which the NZ Trustees were in office, “effective” or “realistic, positive” management was elsewhere. As it was put in Smallwood SpC, the “top-level” management of the Trusts during this period was carried out in the UK and decisions made by the NZ Trustees to effect the individual actions required to implement the overall single plan constituted merely “day to day” management of or administration of this plan or, as the Commissioners put it in Smallwood SpC, lower level decisions by, the NZ Trustees, as the trustee for the time being appointed specifically to effect these actions.
Conclusion and right to apply for permission to appeal
For all the reasons set out above, the appeals made by the Murphys are in principle dismissed and the appeal made by DH is in principle allowed.
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.
Release date: 1st DECEMBER 2025