Neutral Citation Number: [2023] CA (Bda) 4 Civ Case No: Civ/2021/20

IN THE COURT OF APPEAL (CIVIL DIVISION)

ON APPEAL FROM THE SUPREME COURT OF BERMUDA SITTING IN ITS

ORIGINAL CIVIL JURISDICTION

THE HON. ASSISTANT JUSTICE KAWALEY

CASE NUMBER 2018: No. 40 Dame Lois Browne-Evans Building Hamilton, Bermuda HM 12 Date: 17/02/2023 Revised 23/02/2023 Before:

THE PRESIDENT, SIR CHRISTOPHER CLARKE

JUSTICE OF APPEAL SIR MAURICE KAY

and

JUSTICE OF APPEAL DAME ELIZABETH GLOSTER, DBE

- - - - - - - - - - - - - - - - - - - - -

IN THE MATTER OF THE X TRUSTS

Mr Simon Taube KC of counsel and Mr Thomas Fletcher of counsel and Ms Lilla Zuill, Zuill & Co., for D1, D5, D8, D9-D12 (“the B Branch”) Mr Brian Green KC of counsel and Ms Anna Littler of counsel and Mr Matthew Watson, Cox Hallett Wilkinson Limited, for D3 (“the A Branch”) Mrs Elspeth Talbot Rice KC of counsel and Ms Judith Roche, Conyers Dill & Pearman Limited, for the Trustees Mr Keith Robinson and Mr MacKay, Carey Olsen Bermuda Ltd., for the Protectors Hearing date(s): 14-16 June 2022 - - - - - - - - - - - - - - - - - - - - -

REVISED APPROVED JUDGMENT

GLOSTER, J.A:

Introduction

1. 2. This is an appeal from the judgment of the Supreme Court of Bermuda handed down by Kawaley AJ (“the judge”) and dated 7 September 2021 (“the judgment”). The appeal concerns the construction of certain protector provisions (“the Protector Provisions”) contained in 73 settlements which I will refer to as “the X Trusts”1. Of the X Trusts, 51 are governed by English law, 21 by the law of Bermuda, and one by the law of Jersey2 . The primary beneficiaries are descendants of Mr X, a businessman, an original settlor of certain of the X Trusts. I shall refer to all the beneficiaries where appropriate as the X Family. Two of Mr X’s sisters were also settlors of certain trusts. All the relevant Protector Provisions are in substantially identical form. Each provides that the trustees’ powers to appoint capital, and to vote and to deal with “Specified Securities,” as defined in the settlements, may not be exercised “without the prior written consent of the Protectors”. The “Specified Securities” include shares in a substantial quoted public company, with a large market capitalisation, which I shall refer to as “OpCo”. OpCo was established by Mr X. At all material times the Trustees of the X Trusts held, and still hold, in aggregate (i.e. across the various X Trusts) a significant aggregate interest in OpCo. Recent evidence from the Trustees’ expert corporate financier shows that this block of shares (not surprisingly) had, and has, greater value and influence over OpCo if the Trustees of the various X Trusts vote and act in a unified manner.

3. The principal question raised in this appeal is: ‘what role does a fiduciary protector have when it is asked for its consent to the exercise by trustees of a substantive power of appointment and/or the exercise of an administrative power of dealing with, or disposing of, particular assets, where such powers have been specifically entrusted to the trustees specified in the trust deed?’ 1 For the purposes of anonymisation, all names have been changed to letters. My letters do not correspond to the letters in the original names. 2 This trust held assets of relatively small value. Page 2 of 83

4. Factual and procedural background The parties The appellants are certain members of the family of the daughter and younger child of Mr X (and the niece of Mr X’s sisters) (“the Appellants”). In the judgment, the judge referred to the Appellants (for anonymisation purposes) as the “B Branch” or the “B Family”. I shall do the same. Before this Court, they were represented by Mr Simon Taube KC, Mr Thomas Fletcher and Miss Lilla Zuill.

5. The principal respondents are certain members of the family of the son and elder child of Mr X (and the nephew of Mr X’s sisters). In the judgment, the judge referred to these respondents (for anonymisation purposes) as the “A Branch” or the “A Family” (together “the Respondents”). I shall do the same. Where appropriate, I shall refer to them as “the A Family Respondents”. Before this court they were represented by Mr Brian Green QC, Ms Anna Littler and Mr Matthew Watson.

6. 7. 8. The other respondents to the appeal include the current trustees (“the Trustees”) of the 73 X Trusts which are the subject of the appeal before us. They are Bermuda resident corporate trustees. (There are a further 6 X Family trusts which have Jersey resident trustees, and which are not before the Court in these proceedings or, if they are technically respondents, they are not participating in the appeal). The Trustees also appeared below and made submissions to this Court on the appeal. They have stated that they are neutral on this appeal and that their role was limited to assisting the Court. The Trustees were represented before this Court by Mrs Elspeth Talbot Rice KC and Ms Judith Roche. Further respondents to the appeal are the current protectors of the 73 X Trusts (“the Protectors3”). They are companies incorporated and resident in Jersey. The Protectors stated that they have remained neutral at all times. However, they were represented before this Court by Mr Keith Robinson and Mr MacKay and presented submissions effectively in support of the Appellants’ position. I use the term “Respondents” to refer collectively to all the respondents to the appeal. We are grateful to counsel for their detailed and extensive arguments. The Protector Provisions

9. Some of the settlements had included Protector Provisions since their inception; however, in the case of other settlements, Protector Provisions were only added by way of 3 I refer to “Protectors” in this judgment with an uppercase P, when referring to protectors under the X Trusts, whether current, past or future. The reference is not limited to the current protectors.Page 3 of 83

10. 11. 12. 13. amendment, following their inception. It was the Trustees (not a settlor) who introduced the Protector Provisions in 55 of the 73 cases, with the 18 subsequent trusts thereafter adopting a similar approach. In 1989 and 1990, the English resident trustees of many of the X Trusts were replaced with Bermudian resident corporate trustees. The new trustees were both relatively unknown to, and physically distant from, the English-resident beneficiaries. It was in this context that in or about early 1991 the X Family and its advisors considered the introduction of Protector Provisions. According to the evidence (the relevance and admissibility of which is disputed by the A Family), certain family member settlors and beneficiaries of the X Trusts wanted the family’s trusted advisors to have some control over the offshore Trustees. The contemporaneous correspondence, memoranda and notes documenting the settlors’, trustees’, and advisors’ (both legal and lay) discussions in relation to the inclusion of Protector Provisions placed an emphasis on the desire to provide “stability, continuity and coherence in the long-term administration of the X Trusts”: see para 2.4 of the “Operation Protector Bible”. The desire was expressed in particular in relation to the significant shareholding in OpCo and other diversified instruments. In addition to these contemporaneous documents, three of the Appellants also produced affidavits of their alleged personal understanding – having “refreshed their memories” by reference to documents drawn to their attention – that the Protectorate as introduced in 1994/95 and in the (relatively less significant in terms of value) settlements that were made in 1997, had the Wide Discretionary Role. Further, the Protectors adduced evidence from one of the original Jersey Protectors as to what he thought the “flexi-role of the Protectors” involved. I set out in Appendix A a summary of various extracts from the contemporaneous documents which were in evidence before us and upon which Mr Taube sought to rely as aids to construction of the Protector Provisions. The summary is taken, with certain adaptions, from the Appellants’ skeleton argument on the appeal. (Appendix A does not include a summary of the affidavit evidence referred to above). I refer to such documents as the “Contemporaneous Materials.” The result of these discussions was a plan referred to as “Operation Protector”, which was implemented in three stages: (i) In phase 1, the trustees of 49 X Trusts exercised powers of amendment in each of the relevant settlements to introduce the present Protector Provisions. (ii) In phase 2, the trustees of a further 6 X Trusts removed then-existing Protector Provisions in order to replace them with new Protector Provisions in substantially the same form as those introduced into the phase 1 trusts. Page 4 of 83

(iii) In phase 3, the settlement instruments of subsequent X Trusts were drafted to include Protector Provisions in much the same form as those implemented in phases 2 and 3.

14. 15. 16. Both the Appellants and the A Family Respondents agreed that, for the purposes of this appeal, the Protector Provisions in the relevant settlements are substantially identical. In 55 of the 73 X Trusts subject to these proceedings, the Protector Provisions are to be found in self-contained schedules inserted into the X Trust by appointments made by their Trustees in 1994 or /95 (“the Protector Schedule)”. Each provides that the Trustee’s powers to appoint capital, and to vote and to deal with “Specified Securities”, may not be exercised “without the prior written consent of the Protector”. The “Specified Securities” include shares in OpCo. Because the Protector Provisions, the construction of which is in dispute, are in substantially identical terms, the parties used the following sample provision as the template in relation to which this Court’s determination is sought: “2. Restriction on power to appoint capital The Trustees shall not exercise their power to appoint, distribute or pay any part of the Trust Fund to or for the benefit of any member of the Appointed Class or any Beneficiary without obtaining the prior written consent of the Protectorate, nor, if the Trustees’ consent is required for any appointment of capital, shall they give their consent without the prior written consent of the Protectorate.

3. Restriction in relation to Specified Securities Notwithstanding anything to the contrary contained herein or in the Settlement, the Trustees shall not, without in each case obtaining the prior written consent of the Protectorate: sell, charge, exchange, transfer or otherwise deal with any Specified Securities or any interest therein, whether legal or equitable; give any consents that may be required of them in relation to any sale, charge, exchange, transfer or other dealing with any Specified Securities or any interest therein, whether legal or equitable; nor exercise, or take or omit to take any action in relation to the exercise of, voting rights attaching to any Specified Securities ...

13. Joint Protectors Page 5 of 83

17. 18. 19. If the Protectorate comprises more than one Protector, any decision must be taken unanimously. If any power vested in the Trustees requires the prior written consent of the Protectorate and the members of the Protectorate cannot agree as to whether it should give or withhold its consent to a proposed exercise of such power in relation to a particular matter, the Trustees shall then be free to exercise the power (in relation to the matter in question but not further or otherwise) without having the written consent of the Protectorate. In such a case the Trustees shall nevertheless consult with each Protector and shall take into account the views expressed before making a final decision.” The present dispute The issue in the present case is essentially what role do the Protectors have when he is asked for his consent to a decision which the relevant trust deed has entrusted to the Trustee? Obviously, the Protectors have to make a decision as to whether to say “yes, I consent”, or “no, I do not consent”. But the critical question is what role is the Protector meant to play as a fiduciary when he considers whether to say yes or no? The ambiguity arises in the present case because the trust deed, in common with very many other similar trust deeds which incorporate protector consent provisions, does not spell out what function the Protectors are to perform when he is called upon by the trustee to consent to a decision which the trustee has the power to make and has made. Of course, a settlor would be able to impose on a Protector an express obligation as to what considerations he is obliged to take into account, and what he is not to take into account, in deciding whether to give consent. But there are no such express provisions in the present case. The present claim arises against the backdrop of a proposed division of the assets of the X Trusts between the A Family and the B Family. The Trustees hold assets worth several billion. The assets include a significant aggregate interest in OpCo and other diversified investments. The Trustees decided, in principle, to use their powers to allocate the assets between the A Family and the B family in the proportions 2/3: 1/3, subject to limited adjustments. The Trustees drew up the preliminary proposals for this course of action which they put before the court for approval under the jurisdiction identified in Public Trustee v Cooper [2001] WTLR 901. The judge gave his approval to these preliminary proposals on 23 October 2020. The Trustees are now working on detailed plans to implement their proposals. They intend to ask the Bermuda Supreme Court to approve their plans under the jurisdiction identified in Public Trustee v Cooper. However, on 24 August 2020, the Protectors informed the other parties that the Protectors would be unlikely to consent to some of the steps necessary to implement the Trustees’ proposals for the future administration of the X Trusts, because they did not view them as being in the best interests of the beneficiaries. In particular, the Protectors indicated that Page 6 of 83

20. 21. they would not consent to a division of the Trust assets, nor, even if they were minded to consent to such a proposal, would they consent to the proposition that any division should proceed from the Trustees’ proposed starting point of a 2:1 split between the A Family and the B Family respectively. The Protectors therefore indicated that they were unlikely to provide their consent to the Trustees’ proposed course of action. Without the consent of the Protectors, the Trustees were unable to move forward with their plans. By a Summons dated 20 January 2021 (“the Protector Summons”), the Trustees brought the issue of the nature and extent of the Protectors’ role before the Supreme Court for determination in in the following terms: “(1) whether, on the proper interpretation of the relevant trust instruments, the role of the 16th and 17th Defendants as protectors of the trusts identified in Appendix A to the Originating Summons dated 21 February 2018 (the X Trusts) (or any of them) (save for the settlement known as [X Trust] numbered 365 in Appendix A to the Originating Summons) in exercising their powers to consent to the exercise of powers vested in the Plaintiffs (or any of them) is: (a) to exercise an independent discretion as to whether or not to give consent to a proposed exercise of power by the Plaintiffs (as trustees of the X Trusts) (or any of them) which requires the protectors’ consent, taking into account relevant considerations and disregarding irrelevant considerations so that the protectors might withhold their consent to a proposed exercise of power by the Plaintiffs even if the proposed exercise of power was an exercise of power which a reasonable body of properly informed trustees was entitled to decide upon (the latter being a relevant factor, but not the only relevant factor, for the protectors to take into account) [which the parties generally described, and I shall refer to, as the “Wide Review Role”]; or (b) to satisfy themselves that the proposed exercise of a power by the Plaintiffs (as trustees of the X Trusts) (or any of them) is an exercise which a reasonable body of properly informed trustees is entitled to undertake and, if so satisfied, to consent to the same [which the parties generally described, and I shall refer to, as the “Narrow Review Role”].” The A Family (viz. the A Family Respondents) contended, as they did before this Court, that the Narrow Review Role was the correct construction of the relevant provisions; whereas the B Family (viz. the Appellants) contended that the Wide Review Role was the correct construction. Page 7 of 83

22. 23. 24. 25. It can be seen from the terms of the Summons that the Wide Review Role includes the Narrow Review Role. No other rule was promoted before the judge or was promoted before this Court. The judgment In his judgment, dated 7 September 2021, the judge concluded that the role of the Protectors was limited to the Narrow Review Role. He therefore concluded that: “126. The Trustees are entitled to declarations pursuant to paragraph 1 (b) of the January 21, 2021 Summons that: (1) on the proper interpretation of the relevant trust instruments, the role of the protectors in exercising their powers to consent to the exercise of powers vested in the Trustees is to satisfy themselves that the proposed exercise of a power by the Trustees is an exercise which a reasonable body of properly informed trustees is entitled to undertake and, if so satisfied, to consent to the same”. In fact, after hearing from certain of the parties, the relevant declaration which the judge made in his subsequent order dated 22 November 2021 was in the following terms: “On the proper interpretation of the trust instruments that confer powers on the Protectors in relation to each of the X Trusts other than the trust identified as [X Trust], the role of the Protectors in exercising their powers to consent to the exercise of powers vested in the Trustees is to satisfy themselves that the proposed exercise of a power by the Trustees is a proper exercise of the power and one which a reasonable body of properly informed trustees was entitled to undertake and, if so satisfied, to consent to the same.” So far as relates to the issue under appeal, the judge identified two key questions: (i) On a proper construction of the Protector Provisions, did the role of the Protectors reflect the Narrow Review Role or the Wide Review Role? (ii) Were there grounds for implying a term into the relevant settlement instruments limiting the role of the Protectors to the Narrow Review Role? Page 8 of 83

26. The judge concluded that, on the proper construction of the Protector Provisions in the settlement instruments, the Protectors’ role was the Narrow Review Role for the following reasons: (i) (ii) The judge held that the extrinsic evidence as to the purpose of the Protector Provisions was a neutral factor in deciding between the Wider and Narrow Review Roles. If applied consistently, both interpretations of the Protector Provisions were capable of adding to the stability and coherence of the X Trusts’ administration. The judge considered the context of the settlement instruments to be the most important criterion when construing the Protector Provisions. At [78] the judge summarised the contextual factors which he concluded indicated that the Narrow Review Role was the correct construction: “(a) the consent powers themselves are expressed in terms which suggest that the substantive decision-making powers are vested in the Trustees; (b) the Protectors can waive their consent and, where there are more than one Protector, in the absence of unanimous consent the requirement for consent falls away; (c) the Trustees are appointed on terms which include the benefit of indemnities while the Protectors are not. Although no indemnity protections are conferred in relation to even those powers fully vested in the Protectors (in particular the power to appoint and remove Trustees), those powers do not relate to the day-to-day operations of the administration of the X Trusts.” (iii) (iv) At [79]-[95], having reviewed the existing literature and obiter dicta on the wider subject of Protectors’ powers in general, the judge concluded that the predominant view in relation to Protectors generally is that their role is subsidiary to that of trustees, and that the former’s power of veto should not be used to ‘refuse consent to a trustee decision which was consistent with the settlor’s intentions by virtue of the fact that it is both a lawful and rational decision on an issue requiring protector consent’. The judge considered the Bermudian Court of Appeal’s decision in Re Information About a Trust [2014] Bda LR 5 to be of particular significance in this regard. At [96]-[100], the judge considered the practical implications of the Narrow and Wide Review Roles. The judge concluded that the Narrow Review Role still gave Page 9 of 83

27. (v) the Protectors a sufficiently robust role in the decision-making process by allowing the Protectors to ‘stress-test the Trustees’ initial proposals by positing alternative proposals of their own’. On the other hand, the judge concluded that the Wide Review Role turned the Protectors into de facto co-trustees, and therefore was sufficiently atypical as ‘to give rise to a reasonable expectation that the draftsman of the relevant power would use clearer language than was actually deployed’. Subsequently, at [101]-[112] the judge considered the English High Court case of PTNZ v AS [2020] WTLR 1423, in which Master Shuman had held that similar powers of consent by protectors constituted an independent discretion jointly held with the trustee. The judge held that PTNZ did not assist in determining the present dispute, and distinguished PTNZ from the present case for various reasons which he set out. He summarised his conclusions on the construction issue at [113 to 119] as follows: “113. The Narrower View reflects the true construction of the consent powers conferred on the Protectors of the X Trusts primarily because it is clear from the terms of the relevant instruments that their dominant purpose is to ensure the due exercise of the powers vested in the Trustees. The preponderant view of the text writers whose learning on this topic was placed before this Court supports the following critical conclusion. Unless a contrary meaning can legitimately be discerned in the instrument conferring the relevant consent powers, the usual role of a protector is not to exercise a power jointly with the trustee in relation to the matter requiring protector consent. The protector’s role is to be a “watchdog” to ensure due execution by the trustee of the powers vested in the trustee. I arrive at this conclusion based on an analysis of the terms of the instruments without implying any additional terms following both (a) the iterative approach to construction commended by Lord Hodge in Barnardo’s-v-Buckinghamshire [2018] UKSC 55, [2019] ICR 495 at [13] to [17] and (b) the contextual analysis commended by Sir Christopher Clarke in Grand View Private Trust Company-v-Wong et al [2020] CA (Bda) 6 Civ at paragraphs 178-179.

114. In the present case the relevant instruments are substantially expressed in the same terms with the Protector Provisions created by the Trustees reflecting the Phase 1 Trusts created by the original settlors themselves. The drafting approach clearly distinguishes between powers expressly vested in the Trustees, powers expressly vested in the Protectors and powers expressly vested in the Trustees Page 10 of 83

subject to Protector consent. It is true that on a literal reading of the wording of the consent powers, ignoring the wider context of the instruments of which they form part, a power of veto is imposed. However, particular clauses in trust instruments, like most legal documents, cannot properly be construed in isolation from other pertinent parts of the instrument and ignoring altogether the practical and legal dimensions of the competing constructions.

115. A contextual reading of the Protector Provisions suggests that the consent powers were not intended to be powers exercised jointly with, or entirely independently from, the powers conferred on the Trustees subject to Protector consent. There is no explicit wording used to signify an absolute discretion. But more importantly still, the powers requiring protector consent are expressed to be powers vested in the Trustees. This view is not only reinforced by the fact that the ‘normal’ function of ‘standard’ protector consent clauses appears to be understood by most legal writers as an ancillary power rather than a power exercised jointly with the trustee. This understanding has also received the imprimatur of the Bermudian Court of Appeal (Evans JA) in Re Information About a Trust [2014] Bda LR 5. In these circumstances, clear language would be required to signify the intention of achieving an atypical result in terms of the scope of consent power conferred.

116. In PTNZ v AS [2020] WTLR 1423, Master Shuman admittedly held that consent powers conferred on a protector embodied an independent discretion jointly exercised with the trustee; it was not limited to ensuring the due administration by the trustee of the trust. This sole identified judicial authority directly considering the point raised by the present construction dispute was potentially the most powerful support for the Wider View. On closer consideration, however, its persuasive value was very weak for the following main reasons: (a) the protector’s powers were seemingly drafted in wider terms than in the present case; (b) the authorities on protectors’ powers placed before me were not considered; and (c) the point did not receive the benefit of full adversarial argument.

117. In rejecting the Wider View construction argument, it is important to reiterate that I have also rejected the thesis that the Narrower View results in defining the Protectors’ role as being a fundamentally limited one. Ensuring the Trustees properly exercise their important powers is in and of itself an important and substantial role. Depending on the content of the proposed action for which Protector consent is required, the Protectors will bePage 11 of 83

28. 29. 30. entitled to undertake greater or lesser degrees of independent analysis before deciding whether to grant or withhold consent. In many cases the Protectors’ decision, affirmative or negative, will obviate the need for the Trustees to seek Court approval; in other cases the Protectors’ consent may mean that “blessing” applications can be dealt with in a more economical manner.

118. It is likely to be the exception rather than the rule that the Protectors’ deployment of their undoubted veto powers will result in the legality of the Trustees’ proposed course of action being adjudicated on a contentious basis in the context of a Category 2 Public Trustee-v-Cooper application. That this is the efficient way the Narrower View operates in practice may well in large part explain why there is a dearth of judicial authority on the scope of the powers of consent conferred on protectors under so-called standard form trust instruments.

119. For the avoidance of doubt, I find that the expert evidence on the UK tax implications of the competing constructions was ultimately inconclusive and shed no material light on which construction should be preferred. As can be seen from [113], in the light of his findings in relation to the construction issue, the judge concluded that he did not need to reach a conclusion on the implied term issue. Grounds of appeal By a Notice of Appeal filed on 15 December 2021, the B Family gave notice of their intention to appeal the judge’s decision that the Protectors’ role under the X Trust instruments was that described by the Narrow Review Role. The B Family contended that the judge ought to have held that the Protectors’ role under the X Trust instruments was the Wide Review Role. In summary, the B Family relied upon the following principal grounds of appeal: (i) The judge wrongly concluded that the role of the Protectors was the Narrow Review Role “on an analysis of the terms of the instruments without implying any additional terms”. The express terms of the instrument relating to the role of the Protectors, whether read in isolation, or in the context of each relevant trust as a whole, were not capable of bearing the meaning which the judge ascribed to them. Contrary to the judge’s holding, his decision necessarily involved the implication of a term that the role of the Protectors, in deciding to give or withhold consent, Page 12 of 83

(ii) (iii) (iv) (v) (vi) (vii) (viii) was limited to the Narrow Review Role. However, there was no basis for implying such a term. The effect of the Narrow Review Role construction was that, if the Protectors were satisfied that the Trustees’ proposed exercise of a power requiring their consent is rational, the former were obliged to consent to it. On that construction, the Protectors had no choice or discretion. On the natural and ordinary meaning of the language in the relevant Protector Provisions in their context, the words used showed that the Protectors indeed have a choice or discretion. At [65] and [114] the learned judge correctly stated those words “do indeed suggest a power of veto when the relevant words are literally read”. But the judgment wrongly adopted an interpretation of those words which they could not bear and which gave the Protectors no discretion. The judge failed to conclude that the wording of the Protector Provisions in the X Trust instruments unambiguously indicated that the Protectors’ role was the Wide Review Role. Further, the judge read in words that were not in the X Trust instruments and ignored the evidence that the Protectors’ role was the Wide Review Role. The judge was wrong in law to hold that the duty of the Protectors obliged them to consent to the Trustees’ exercise of their power if the Trustees were acting rationally, even when the Protectors rationally and bone fide, and otherwise in accordance with their fiduciary duties, considered such exercise not to be in the best interests of the beneficiaries of the X Trusts. He should have concluded that that Protectors indeed had an independent discretion in choosing whether to consent to the Trustees’ proposed course of action. The judge failed to give appropriate weight to the fact that the rules governing the Protectors’ power of consent were those generally applicable to fiduciaries. The judge gave undue weight to irrelevant factors, such as the wording of irrelevant provisions and the absence of irrelevant provisions, in construing the Protector Provisions. The judge was wrong in law to find that the Wide Review Role described a joint power shared by the Trustees and Protectors. The judge was wrong to hold that the preponderant view of the relevant textbooks was that a protector’s role was that of a ‘watchdog’. The judge was wrong in law to distinguish the decision of the English High Court in PTNZ, so as to hold that it was not persuasive. Page 13 of 83

31. Respondent’s Notice By a Respondent’s Notice filed on 31 March 2022 (“the Respondent’s Notice”), the A Family set out additional grounds for upholding the judgment. Those additional grounds were, in summary, as follows: (i) The constitutionally distinct roles of trustees and protectors required that the Protector Provisions be construed such as to give effect to the Narrow Review Role. (ii) The respective practical implications of the Narrow and Wide Review Roles militated against construing the Protector Provisions such as to give effect to the Wide Review Role. (iii) (iv) The argument that the Narrow Review Role did not confer a sufficiently substantial role upon the Protectors was ill-founded. The extrinsic evidence as to the background of the Protector Provisions should be excluded as a matter of law.

32. Matters of common ground Before this Court there was common ground in relation to a number of matters: (i) The parties agreed that the Protectors were fiduciaries and were therefore subject to the fiduciary duties to act in good faith and in the best interests of the beneficiaries. (ii) They agreed that the Protectors were not co-trustees in the administration of the X Trusts, and the Trustees’ powers were not intended to be exercised jointly with the Protectors. (iii) They agreed that, regardless of the role that the Protectors had under the consent provisions, the court had jurisdiction to hear applications under the principles summarised in Public Trustee v Cooper [2001] WTLR 901. In particular, the parties agreed that the court may hear a Public Trustee 2 application, which allows the court to bless a proposed course of conduct by trustees such as to immunise them from a later breach of duty claim. The approach to the construction of trust instruments Page 14 of 83

33. The legal principles applicable to the questions of construction and implied terms were not substantially in dispute between the parties, although there was disagreement as to the extent to which, if at all, extrinsic evidence leading up to the introduction of the Protector Provisions was relevant.

34. The principles applicable to the construction of written instruments are well established. The English House of Lords and Supreme Court have articulated these principles on numerous occasions, initially in relation to the construction of contracts: see Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896; Rainy Sky SA v Kookmin Bank [2011] 1 WLR 2900, Arnold v Britton [2015] AC 1619; Wood v Capita Insurance Services Ltd [2017] AC 1173. As Lord Hodge JSC (with whom the other members of the court agreed) stated in Wood v Capita: “10. The court's task is to ascertain the objective meaning of the language which the parties have chosen to express their agreement. It has long been accepted that this is not a literalist exercise focused solely on a parsing of the wording of the particular clause but that the court must consider the contract as a whole and, depending on the nature, formality and quality of drafting of the contract, give more or less weight to elements of the wider context in reaching its view as to that objective meaning. In Prenn v Simmonds [1971] 1 WLR 1381, 1383H—1385D and in Reardon Smith Line Ltd v Yngvar Hansen-Tangen (trading as HE Hansen-Tangen) [1976] 1 WLR 989, 997, Lord Wilberforce affirmed the potential relevance to the task of interpreting the parties' contract of the factual background known to the parties at or before the date of the contract, excluding evidence of the prior negotiations. When in his celebrated judgment in Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896, 912—913 Lord Hofmann reformulated the principles of contractual interpretation, some saw his second principle, which allowed consideration of the whole relevant factual background available to the parties at the time of the contract, as signalling a break with the past. But Lord Bingham of Cornhill in an extrajudicial writing, ""A New Thing Under the Sun? The Interpretation of Contracts and the ICS decision'' (2008) 12 Edin LR 374, persuasively demonstrated that the idea of the court putting itself in the shoes of the contracting parties had a long pedigree.

11. Lord Clarke of Stone-cum-Ebony JSC elegantly summarised the approach to construction in the Rainy Sky case [2011] 1 WLR 2900, para 21f. In the Arnold case [2015] AC 1619 all of the judgments confirmed the approach in the Rainy Sky case:Page 15 of 83

Lord Neuberger of Abbotsbury PSC, paras 13—14; Lord Hodge JSC, para 76 and Lord Carnwath JSC, para 108. Interpretation is, as Lord Clarke JSC stated in the Rainy Sky case (para 21), a unitary exercise; where there are rival meanings, the court can give weight to the implications of rival constructions by reaching a view as to which construction is more consistent with business common sense. But, in striking a balance between the indications given by the language and the implications of the competing constructions the court must consider the quality of drafting of the clause (the Rainy Sky case, para 26, citing Mance LJ in Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (No 2) [2001] 2 All ER (Comm) 299, paras 13, 16); and it must also be alive to the possibility that one side may have agreed to something which with hindsight did not serve his interest: the Arnold case, paras 20, 77. Similarly, the court must not lose sight of the possibility that a provision may be a negotiated compromise or that the negotiators were not able to agree more precise terms.

12. This unitary exercise involves an iterative process by which each suggested interpretation is checked against the provisions of the contract and its commercial consequences are investigated: the Arnold case, para 77 citing In re Sigma Finance Corpn [2010] 1 All ER 571, para 12, per Lord Mance JSC. To my mind once one has read the language in dispute and the relevant parts of the contract that provide its context, it does not matter whether the more detailed analysis commences with the factual background and the implications of rival constructions or a close examination of the relevant language in the contract, so long as the court balances the indications given by each.

13. Textualism and contextualism are not conflicting paradigms in a battle for exclusive occupation of the field of contractual interpretation. Rather, the lawyer and the judge, when interpreting any contract, can use them as tools to ascertain the objective meaning of the language which the parties have chosen to express their agreement. The extent to which each tool will assist the court in its task will vary according to the circumstances of the particular agreement or agreements. Some agreements may be successfully interpreted principally by textual analysis, for example because of their sophistication and complexity and because they have been negotiated and prepared with the assistance of skilled professionals. The correct interpretation of other contracts may be achieved by a greater emphasis on the factual matrix, for example because of their informality, brevityPage 16 of 83

35. or the absence of skilled professional assistance. But negotiators of complex formal contracts may often not achieve a logical and coherent text because of, for example, the conflicting aims of the parties, failures of communication, differing drafting practices, or deadlines which require the parties to compromise in order to reach agreement. There may often therefore be provisions in a detailed professionally drawn contract which lack clarity and the lawyer or judge in interpreting such provisions may be particularly helped by considering the factual matrix and the purpose of similar provisions in contracts of the same type. The iterative process, of which Lord Mance JSC spoke in Sigma Finance Corpn [2010] 1 All ER 571, para 12, assists the lawyer or judge to ascertain the objective meaning of disputed provisions.

14. On the approach to contractual interpretation, the Rainy Sky and Arnold cases were saying the same thing.

15. The recent history of the common law of contractual interpretation is one of continuity rather than change. One of the attractions of English law as a legal system of choice in commercial matters is its stability and continuity, particularly in contractual interpretation.” These principles were usefully summarised by Popplewell J (as he then was) in Lukoil Asia Pacific Pte Ltd v Ocean Tankers (Pte) Ltd (Ocean Neptune) [2018] EWHC 163 at [8] as follows: “8. There is an abundance of recent high authority on the principles applicable to the construction of commercial documents, including Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896; Chartbrook Ltd v Persimmon Homes Ltd [2009] 1 AC 1101; Re Sigma Finance Corp [2010] 1 All ER 571; Rainy Sky SA v Kookmin Bank [2011] 1 WLR 2900; Arnold v Britton [2015] AC 1619; and Wood v Capita Insurance Services Ltd [2017] AC 1173. The court's task is to ascertain the objective meaning of the language which the parties have chosen in which to express their agreement. The court must consider the language used and ascertain what a reasonable person, that is a person who has all the background knowledge which would reasonably have been available to the parties in the situation in which they were at the time of the contract, would have understood the parties to have meant. The court must consider the contract as a whole and, depending on the nature, formality and quality of drafting of the Page 17 of 83

36. 37. 38. contract, give more or less weight to elements of the wider context in reaching its view as to the objective meaning of the language used. If there are two possible constructions, the court is entitled to prefer the construction which is consistent with business common sense and to reject the other. Interpretation is a unitary exercise; in striking a balance between the indications given by the language and the implications of the competing constructions, the court must consider the quality of drafting of the clause and it must also be alive to the possibility that one side may have agreed to something which with hindsight did not serve his interest; similarly, the court must not lose sight of the possibility that a provision may be a negotiated compromise or that the negotiators were not able to agree more precise terms. This unitary exercise involves an iterative process by which each suggested interpretation is checked against the provisions of the contract and its commercial consequences are investigated. It does not matter whether the more detailed analysis commences with the factual background and the implications of rival constructions or a close examination of the relevant language in the contract, so long as the court balances the indications given by each.” It was common ground that these principles of construction applied to other instruments including wills and trust documents: see Marley v Rawlings [2015] AC 129; Barnardo’s v Buckinghamshire [2018] UKSC 55, [2019] ICR 495 and, in Bermuda, Wong (i.e. Grand View Private Trust Co Ltd v Wong [2020] CA (Bda) 6 Civ – cited above) at [93]. Lord Neuberger articulated the principles in Marley v Rawlings at [19] as follows: “When interpreting a contract, the court is concerned to find the intention of the party or parties, and it does this by identifying the meaning of the relevant words, (a) in the light of (i) the natural and ordinary meaning of those words, (ii) the overall purpose of the document, (iii) any other provisions of the document, (iv) the facts known or assumed by the parties at the time that the document was executed, and (v) common sense, but (b) ignoring subjective evidence of any party’s intentions.” Lord Neuberger said that the approach equally applied where the document was unilateral such as a will, as in Marley v Rawlings; and he continued at [20] that in all cases – “the aim is to identify the intention of the party or parties to the document by interpreting the words used in their documentary, factual and commercial context”.Page 18 of 83

39. 40. In this context Mr Taube, for the Appellants, also drew attention to the comment of Lord Clarke in Rainy Sky v Kookmin Bank at [23] which highlighted an important point (which, Mr Taube submitted, was overlooked in the judgment): “Where the parties have used unambiguous language, the court must apply it.” The most recent guidance in relation to the approach to construction of a trust is the UKSC’s decision in Barnardo’s. Although Barnardo’s concerned a pension scheme trust deed, it is apparent from what Lord Hodge JSC said at [14] that what was said by him is apposite to all trust instruments – reading references to “beneficiaries” and “potential [future] beneficiaries” in place of references to “members” and “potential [future] members” where they occur: “13. In the trilogy of cases, Rainy Sky SA v Kookmin Bank [2011] UKSC 50; [2011] 1 W.L.R. 2900, Arnold v Britton [2015] UKSC 36; [2015] A.C. 1619 and Wood v Capita Insurance Services Ltd [2017] UKSC 24; [2017] A.C. 1173, this court has given guidance on the general approach to the construction of contracts and other instruments, drawing on modern case law of the House of Lords since Prenn v Simmonds [1971] 1 W.L.R. 1381. That guidance, which the parties did not contest in this appeal, does not need to be repeated. In deciding which interpretative tools will best assist in ascertaining the meaning of an instrument, and the weight to be given to each of the relevant interpretative tools, the court must have regard to the nature and circumstances of the particular instrument.

14. A pension scheme, such as the one in issue on this appeal, has several distinctive characteristics which are relevant to the court’s selection of the appropriate interpretative tools. First, it is a formal legal document which has been prepared by skilled and specialist legal draftsmen. Secondly, unlike many commercial contracts, it is not the product of commercial negotiation between parties who may have conflicting interests and who may conclude their agreement under considerable pressure of time, leaving loose ends to be sorted out in future. Thirdly, it is an instrument which is designed to operate in the long term, defining people’s rights long after the economic and other circumstances, which existed at the time when it was signed, may have ceased to exist. Fourthly, the scheme confers important rights on parties, the members of the pension scheme, who were not parties to the instrument and who may have joined the scheme many years after it was initiated. Page 19 of 83

Fifthly, members of a pension scheme may not have easy access to expert legal advice or be able readily to ascertain the circumstances which existed when the scheme was established.

15. Judges have recognised that these characteristics make it appropriate for the court to give weight to textual analysis, by concentrating on the words which the draftsman has chosen to use and by attaching less weight to the background factual matrix than might be appropriate in certain commercial contracts: Spooner v British Telecommunications Plc [2000] Pens. L.R. 65, Jonathan Parker J at [75]–[76]; BES Trustees v Stuart [2001] Pens. L.R. 283, Neuberger J at [33]; Safeway Ltd v Newton [2017] EWCA Civ 1482; [2018] Pens. L.R. 2, Lord Briggs, giving the judgment of the Court of Appeal, at [21]–[23]. In Safeway, Lord Briggs stated, at para 22: “the Deed exists primarily for the benefit of non-parties, that is the employees upon whom pension rights are conferred whether as members or potential members of the Scheme, and upon members of their families (for example in the event of their death). It is therefore a context which is inherently antipathetic to the recognition, by way of departure from plain language, of some common understanding between the principal employer and the trustee, or common dictionary which they may have employed, or even some widespread practice within the pension industry which might illuminate, or give some strained meaning to, the words used.” I agree with that approach. In this context I do not think that the court is assisted by assertions as to whether or not the pensions industry in 1991could have foreseen or did foresee the criticisms of the suitability of the RPI, which later emerged in the public domain, or then thought that it was or was not likely that the RPI would be superseded.

16. The emphasis on textual analysis as an interpretative tool does not derogate from the need both to avoid undue technicality and to have regard to the practical consequences of any construction. Such an analysis does not involve literalism but includes a purposive construction when that is appropriate. As Millett J stated in Re Courage Group’s Pension Schemes [1987] 1 W.L.R. 495, 505 there are no special rules of construction applicable to a Page 20 of 83

41. 42. 43. pension scheme but “its provisions should wherever possible be construed to give reasonable and practical effect to the scheme”. Instead, the focus on textual analysis operates as a constraint on the contribution which background factual circumstances, which existed at the time when the scheme was entered into but which would not readily be accessible to its members as time passed, can make to the construction of the scheme.

17. It is nevertheless relevant to the construction of pension schemes that they are drafted to comply with tax rules so as to preserve the considerable benefits which the United Kingdom’s tax regime confers on such schemes. They must be construed “against their fiscal backgrounds”: National Grid Co Plc v Mayes [2001] UKHL 20; [2001] ICR 544, para 18 per Lord Hoffmann; Stevens v Bell [2002] EWCA Civ 672; [2002] Pens. L.R. 247, para 30 per Arden LJ.” [My emphasis] Also worthy of reference is this Court’s decision in Wong where Sir Christopher Clarke, President, said at [93]: “As to the scope of the power, the principles of construction which apply to a document such as a declaration of trust are the same as those which apply to a contract: Marley v Rawlings [2015] A.C. 129, [17]- [23]; Richards v Wood & Wood [2014] EWCA Civ 327. The most important aspect of the process of construction is to consider the meaning of the words used; per Lord Neuberger at [17] and [18] of Arnold v Britton [2015] A.C. 1619.” It is with the above principles in mind that I approach the construction of the Protector Provisions. Relevance and/or admissibility of the Contemporaneous Materials The principal area of disagreement as to the correct approach to the construction of the Protector Provisions as between Mr Taube KC for the B Family and Mr Green KC for the A Family, was as to the relevance and admissibility of the Contemporaneous Materials to assist in the construction of the Protector Provisions. The parties’ submissions as to the relevance and/or admissibility of the Contemporaneous Materials Page 21 of 83

44. In brief summary, Mr Green KC for the A Family submitted that the Contemporaneous Materials (together with the evidence in the Appellants’ affidavits in relation to subjective views and opinions et cetera) were neither relevant nor admissible for the following reasons: (i) The issue in the appeal concerned the meaning to be accorded as a matter of substantive trust law to provisions for fiduciary protector consent. There was nothing special about the consent wording used in the Protector Provisions. It was ordinary wording providing that the Trustees’ appointment of capital and dealings with specified securities should not occur “without obtaining the prior written consent” of the Protectors. The expert evidence adduced below confirmed what was common ground, namely that the wording in the present case was not out of the ordinary – in effect standard form wording used in a multitude of trusts and settlements. (ii) (iii) There was no getting away from the fact, therefore, that the issue of construction raised a question of general application in relation to the inclusion of provisions for protector consent in relation to trustees’ powers – namely did they connote the Wide Review Role or the Narrow Review Role? The words used were (as per the experts’ agreement) “no more and no less than an expression of a requirement of consent”. Whilst different, and specific, express terms of a trust deed might create a fiduciary protector consent role that equated to the Wide Review Role, which was different from that created by the ordinary wording in the present case, that was not the case here. Similarly, different legislative regimes supplying the governing law of any given trust could create a fiduciary protector consent role different from that created by this ordinary wording if there was included provision to that effect. But that was not the case here. Neither English legislation, nor Bermuda legislation, nor Jersey legislation – as the governing laws of any of the X Trusts – contained any such provision. Fiduciary protectors were in each of England, Bermuda and Jersey exclusively the product of (what may be described as) the common law relating to trusts in each such jurisdiction. It followed that background documents in relation to a particular trust or trusts were irrelevant. Subjectively expressed views of advisers or others in connection with the insertion of the consent provision into the particular trust were likewise irrelevant. In this context, Mr Green relied upon the statement in the House of Lords’ decision in AIB Group v Martin [2001] UKHL 63, [2002] 1 WLR 94 to the effect that it was well-established that, unless a standard form provision had been used in an inapposite context, individual factual background had no part to play in its interpretation. Mr Green referred to the following statement of principle given by Lord Millett at [7]: Page 22 of 83

45. (iv) “…A standard form is designed for use in a variety of different circumstances. It is not context- specific. Its value would be much diminished if it could not be relied upon as having the same meaning on all occasions. Accordingly the relevance of the factual background of a particular case to its interpretation is necessarily limited. The danger, of course, is that a standard form may be employed in circumstances for which it was not designed. Unless the context in a particular case shows that this has happened, however, the interpretation of the form ought not to be affected by the factual background.” Without prejudice to his submissions on irrelevance, Mr Green also submitted that the Contemporaneous Materials were inadmissible as subjective intention materials as an aid to construction. In support of this submission: (a) He contended that a clear application of the parole evidence rule rendered such documents inadmissible. In this context he relied upon: Halsbury’s Laws of England (2019) vol 32, para 385: Lewin on Trusts (20th edn) “7- 005 to 7-006. (b) He referred to the recent authorities in the UKSC commenting on principles of construction over the last decade, which I have already referred to above, which he said, on a proper analysis, supported the conclusion for which he contended. (c) He additionally referred to the decisions of the English Court of Appeal in: Rabin v Gerson Berger Association Ltd [1986] 1 WLR 526, a case where counsel attempted (unsuccessfully) to rely on counsel’s prior opinions in the construction of a charitable trust deed; and IRC v Botnar [1999] STC 711 at [15], where Morritt LJ endorsed (and both sides acknowledged the correctness of) the trial judge’s decision that a memorandum prepared by the settlement’s protector (who was also the settlor’s lawyer) setting out the origin and underlying purpose of the settlement was not admissible in relation to the construction of the settlement to ascertain the scope of the powers it conferred. Mr Taube KC, on behalf of the B Family, rejected the argument that the extrinsic evidence which the B Family had adduced was inadmissible or irrelevant. He emphasised Lord Neuberger PSC’s formulation in Marley v Rawlings [2015] AC 129 at [19], of the factors to be taken into account when construing a document: Page 23 of 83

46. 47. 48. 49. “[T]he court is concerned to find the intention of the party or parties, and it does this by identifying the meaning of the relevant words, (a) in the light of (i) the natural and ordinary meaning of those words, (ii) the overall purpose of the document, (iii) any other provisions of the document, (iv) the facts known or assumed by the parties at the time that the document was executed, and (v) common sense, but (b) ignoring subjective evidence of any party's intentions.” [Emphasis added] He further submitted that the Supreme Court in Marley was not saying that background material was inadmissible, but rather that the weight which should be given to a textual analysis on the one hand, and to the background or factual matrix on the other, should be different in the case of pension trusts. And that for the reasons given in Barnardo’s, in the case of a pension deed weight should primarily be given to a textual analysis. But the case was not a decision that evidence of the factual matrix was inadmissible in relation to the construction of a private family trust. Mr Taube contended that the extrinsic evidence constituted facts known to the Trustees at the time that the protector consent provisions were executed. He submitted that one can discern the purpose of the protector consent provisions, not only from their terms, but also from such evidence constituting the background to the inclusion of such terms in the trust deeds. He emphasised that the Appellants were not relying on subjective intention evidence. He said that the factual matrix demonstrated that the settlors and the Trustees who introduced the Protectors must have intended the Protectors to have an independent discretion as a matter of construction of their role. The relevant evidence did not describe the subjective intent of the Trustees, but rather described the facts known to the Trustees that inspired their decision to create the protector consent provisions. As such, it was neither inadmissible as evidence of the executing party’s subjective intent nor was it irrelevant – as the wishes of Mr X and his children would have been significant features of the factual background. Somewhat confusingly, Mr Taube only sought to rely on the Contemporaneous Materials as a guide to construction of the Protector Provisions in the event that the Court concluded: “that there is an excludable substantive rule of trust law in the form that the [Respondents] contend for that the role of the Protectors where the Trustees have a power and the consent of the protectors is required is the Narrow Review Role.” Determination of the relevance and/or admissibility of the Contemporaneous Materials I do not consider that even if, as Mr Green submitted, the Court in this case has to decide what the role of a protector comprises as “a matter of substantive trust law”, that isPage 24 of 83

50. 51. determinative of the admissibility of the Contemporaneous Materials. The reality is that the Court is construing these particular trust deeds, against the relevant factual matrix, one of the aspects of which is that standard terms, or, perhaps more properly, terms in common usage, are used in the consent provisions. In my judgment neither the Contemporaneous Materials, nor the affidavits of certain family members and a former director of the Protectors, are admissible as a guide to the construction of the relevant wording. My reasons may be summarised as follows. It is a cardinal principle of construction of documents (often referred to as the parol evidence rule) that, as set out in Halsbury’s Laws of England (2019) vol 32, para 385: “Where the intention of the parties has been reduced to writing it is, in general, not permissible to adduce extrinsic evidence, whether oral or contained in writings such as instructions, drafts, articles, conditions of sale or preliminary agreements or memoranda provided for the 'protector' of a settlement,4 either to show that intention5 or to contradict, vary, or add to the terms of the document. ….

52. Extrinsic evidence cannot be received in order to prove the object with which a document was executed; or that the intention of the parties was other than that appearing on the face of the document.” Lewin on Trusts (20th edn) endorses the application of this – the parol evidence rule – as regards the interpretation of trust instruments (emphasis added): “7-005 The intention that the court seeks is the intention as expressed, that is, the way in which the document is to be understood, not the purpose or motive, desire or other subjective state of mind of the settlor. The reason for the rule is that otherwise no lawyer would be safe in advising on the construction of a written instrument, nor any party in taking under it. … 7-006 Although the rule is called the parol evidence rule, it does not exclude oral evidence only, but all extrinsic evidence. For instance, drafts of a deed cannot be referred to in order to interpret 4 Citing IRC v Botnar [1998] STC 38 at 87 to 88 per Evans Lombe J. 5 Citing Re Atkinson’s Will Trusts [1978] 1 WLR 586 (evidence that testator used words in a special sense inadmissible), and Rabin v Gerson Berger Association Ltd [1986] 1 WLR 526, CA (evidence of counsel’s opinion given before the trust deed executed inadmissible). Page 25 of 83

53. 54. it, nor can preliminary negotiations, nor can the written opinions of counsel who drafted a settlement be looked at to ascertain the subjective intention of the settlor. An unambiguous declaration of trust will not be altered even by a recital of the intention of the settlor, let alone by any extrinsic expressions of intent on the settlor’s part.” As Mr Green correctly submitted, and I accept, references to the settlor in the above text obviously apply equally to trustees making a unilateral instrument in exercise of a power of appointment under a trust such as the Trustees making the 55 Phase 1 and Phase 2 Appointments introducing Protector Schedules in this case. The authorities referred to in the passages cited from Halsbury’s Laws and Lewin above reflect long established authoritative case law, prior to the chain of United Kingdom Supreme Court on construction issues referred to above. For example, in Rabin v Gerson Berger Association Ltd, a case where reliance was attempted to be made on counsel’s prior opinions: (a) At pp.530D-F Fox LJ adopted the statement from Halsbury’s Laws quoted above, in its then 4th edition iteration. (b) Then at pp.531G to 532A he stated: “In the notice of appeal the first two grounds which are stated for basing the appeal are: The said opinions were admissible as part of the factual matrix surrounding the making of the said charitable trust deeds; consideration of the said opinions would enable the court to place itself in thought in the same factual matrix as the settlors. It seems to me that to look generally at the opinions can only be for the purpose of asking the court to conclude that, because counsel thought that the words he used had a particular effect, the court should give them that effect, even though that is not their meaning in law according to ordinary English usage. That, it seems to me, is contrary to the rule against the admission of parol evidence; it is no different from tendering documents in which the settlorPage 26 of 83

or grantor himself tells his solicitors the effect which he wishes to be achieved, and is, indeed, no different from the settlor or grantor giving evidence of the general aim which he wishes to be perfected. Such evidence, I think, is simply parol evidence of the intention of the grantor, either personally, or formulated by counsel and imputed to (or accepted by) the settlor personally. The result, in my view, is that the opinions cannot be referred to generally for the assistance that their contents may give.” Then at 533H to 534F: “In my view, what is being attempted in the present case under this head is to impute to the settlor certain aspects of the legal knowledge of the draftsman at the time when he advised upon and settled the documents. To my mind, in practical terms, that is only a disguise for evidence of the draftsman's actual intention because it has no purpose except to throw direct light upon, and lead directly to, the draftsman's intention in relation to the deed…. What is being sought, in my view, goes far beyond anything contemplated by the “surrounding circumstances” principles, as stated by the House of Lords in Prenn v. Simmonds [1971] 1 W.L.R. 1381 and in the Reardon Smith Line case [1976] 1 W.L.R. 989. They were concerned with objective external facts, and what is sought here is the admission of evidence of the personal legal knowledge of the draftsman. It seems to me that that has no purpose, except as an indication of his intention in relation to the deed which he drew, and it would be quite artificial to regard reference to the opinions as having any other effect. The opinions (or any part of their contents) are not “surrounding circumstances” in the terms in which that has been understood on the authorities; they go directly to intention. …” Fox LJ was clear that this is not acceptable: “…direct evidence of the intention of the draftsman to any degree is inadmissible. The fact that the evidence may not go the whole way does not alter the fact that where you have circumstances, such as these, where you arePage 27 of 83

55. 56. referring directly to the actual knowledge of the draftsman of the deed in relation to legal concepts which directly affect the construction, you are directing your attention solely to the intention of the draftsman, and that, in my view, is inadmissible.” It is clear that the statements by Fox LJ in Rabin remain good law today, even after ICS v West Bromwich. For example, Halsbury’s Laws (2019), as quoted above, cites the case as authority, as did the English Court of Appeal in IRC v Botnar [1999] STC 711. At [15], Morritt LJ endorsed (and both sides acknowledged the correctness of) the trial judge’s decision that a memorandum prepared by the settlement’s protector (who was also the settlor’s lawyer) setting out the origin and underlying purpose of the settlement was not admissible in relation to the construction of the settlement to ascertain the scope of the powers it conferred: “… whatever the width of the principles established in Mannai Investment Co. Ltd v Eagle Star Life Assurance Co Ltd … and Investors Compensation Scheme Ltd v West Bromwich Building Society … they did not warrant the introduction of evidence of the subjective intention of the settlor such as was excluded in Rabin v Gerson Berger Association Ltd …” Mr Taube sought to rely on Lord Hoffmann’s much quoted propositions (1) to (3) in the contract case of Investors Compensation Scheme Ltd v West Bromwich Building Society at 912-913 that: “(1) Interpretation is the ascertainment of the meaning which the document would convey to a reasonable person having all the background knowledge which would reasonably have been available to the parties in the situation in which they were at the time of the contract. (2) The background was famously referred to by Lord Wilberforce as the “matrix of fact,” but this phrase is, if anything, an understated description of what the background may include. Subject to the requirement that it should have been reasonably available to the parties and to the exception to be mentioned next, it includes absolutely anything which would have affected the way in which the language of the document would have been understood by a reasonable man. Page 28 of 83

57. 58. (3) The law excludes from the admissible background the previous negotiations of the parties and their declarations of subjective intent. They are admissible only in an action for rectification. …” and to treat, in particular, proposition (2), as a justification for the introduction of the Contemporaneous Materials. He also sought to rely in this context on Lord Neuberger PSC in the UKSC in Marley v Rawlings [at [19] where he said: “When interpreting a contract, the court is concerned to find the intention of the party or parties, and it does this by identifying the meaning of the relevant words, (a) in the light of (i) the natural and ordinary meaning of those words, (ii) the overall purpose of the document, (iii) any other provisions of the document, (iv) the facts known or assumed by the parties at the time that the document was executed, and (v) common sense, but (b) ignoring subjective evidence of any party's intentions. In this connection, see Prenn at 1384–1386 and Reardon Smith Line Ltd v Yngvar Hansen-Tangen [1976] 1 WLR 989 , per Lord Wilberforce, Bank of Credit and Commerce International SA v Ali [2002] 1 AC 251 , para 8, per Lord Bingham, and the survey of more recent authorities in Rainy Sky, per Lord Clarke at paras 21-30.” But again, it is clear that propositions (ii) and (iv), on which Mr Taube sought to rely, cannot provide carte blanche for introducing anything and everything in the background to the making of a trust instrument. The Contemporaneous Materials and the affidavits are subjective evidence of various parties’ and their advisors’ intentions, promoting or supporting the introduction of the relevant provisions with the Trustees, and thus undoubtedly come within the prohibited category referred to in (3) by Lord Hoffmann and (b) by Lord Neuberger. They cannot properly be characterised as “objective factual background” to the introduction of such provisions. To allow the introduction of such materials would, in my judgment, be a recipe for uncertainty and confusion. Indeed, the evidence in this case demonstrated that a variety of views had been held at different times by advisors (legal, accounting and lay) and by family members. In my judgment, it is wholly unrealistic to suppose that such documentation and evidence would be of any utility in answering the question that arises in this case - namely whether the Wide Review Role or the Narrow Review Role is the correct one. The objective factual matrix may easily be distilled from the objectively known - and therefore admissible – facts relating to the X Trusts. These can be articulated, as including for example: Page 29 of 83

(i) (ii) Many of the settlements had included Protector Provisions since their inception; however, in the case of other settlements, Protector Provisions were only added by way of amendment, subsequent to their inception. In 1989 and 1990, the English resident trustees of many of the X Trusts were replaced with Bermudan resident corporate trustees. The new trustees were both relatively unknown to, and physically distant from, the English-resident beneficiaries.

(iii) (iv) (v) (vi) It was in this context that in or about early 1991 the X Family and its advisors considered the introduction of Protector Provisions. As can be inferred from the subsequent inclusion of such provisions, and, in particular, the consent requirement, some measure of protection was considered desirable by members of the X Family, their advisors, and the Trustees. The “Specified Securities” held across all the X Trust include shares in a substantial quoted public company, with a large market capitalisation - i.e., OpCo. which had been established by Mr X. At the time of the introduction of the Protector Provisions, the Trustees of the X Trusts a significant aggregate interest in OpCo. Even in the absence of any evidence, common sense would suggest that this block of shares had, and would continue to have, greater value and influence over OpCo if the Trustees of the various X Trusts voted to act in a unified or tactical manner.

59. It follows that I would exclude the Contemporaneous Materials and the affidavits as evidence in relation to the determination of the issues in this case, on the grounds that they are irrelevant and inadmissible.

60. But even if I were wrong in that conclusion, and such materials were admissible, I conclude that there is nothing in such documentation which assists the Court in the determination of the issue of construction which arises in relation to the extent of the Protectors’ powers. There is certainly nothing which provides support for the Wide Review Role as formulated at paragraph 1(a) of the Protector Summons - or indeed for the Narrow Review Role. The construction of the relevant Protector Provisions- the Appellants’ submissions

61. I turn now to address the construction of the relevant Protector Provisions having excluded the Contemporaneous Materials as a guide to their interpretation. I set out the Appellants’ submissions at some length. I do not adopt the same course in relation to the A Family’s Page 30 of 83

62. 63. submissions because, as will be seen, to a large extent my analysis and determination reflects those of Mr Green (and Mrs Talbot Rice KC for the Trustees) on this issue. The Appellants’ submissions (i) The construction of the relevant Protector Provisions Mr Taube contended that, as a matter of construction, the ordinary and natural meaning of the express language used in the Protector Provisions, was that the Protectors had the discretion whether to give or withhold consent to the exercise of the Trustees’ specified powers. And, as a matter of construction, that discretion was uncontrolled; there were no words of limitation. Accordingly, the correct construction was that the Protectors had the wider powers contemplated under the Wide Review Role. In order for the Narrow Review Role to be the correct construction, it would be necessary to imply words such as “such consent not to be withheld unless the trustee is acting irrationally”. And that is what, in reality, the judge did. But there was no basis, either as a matter of law, or in fact, for implying such a term. Mr Taube supported his argument by the following submissions: (i) The relevant Protector Provisions state the Trustees’ specified powers may not be exercised “without obtaining the prior written consent of the Protectorate”. (ii) In this context the natural and ordinary meaning of the word “consent” was agreement or permission: see the Oxford English Dictionary. (iii) The reference in the Protector Provisions to the Protector’s “consent” – to its agreement or permission – indicates the Protector has a choice whether to consent to the Trustees’ proposed exercise of the specified powers. (iv) It followed, as a matter of ordinary language, that the Protector had a discretion in the matter whether to choose to consent. (v) This conclusion was reinforced by the description of the role of the Protectorate where there were “Joint Protectors”: “If any power vested in the Trustees requires the prior written consent of the Protectorate and the members of the Protectorate cannot agree as to whether it should give or withhold its consent to a proposed exercise of such power in relation to a particular matter”, then the Trustees do not require its consent. The underlined words reinforced the conclusion that the members of the Protectorate had a choice. Page 31 of 83

(vi) (vii) (b) By contrast, the reference to the Protector’s “consent” was not appropriate to describe the Narrow Review Role. The Narrow Review Role, for which the Respondents contended, involved not a discretion but rather an adjudication whether particular circumstances existed. If the Trustees’ decision was rational, the Protector was obliged to consent. The Narrow Review Role as described in para 1(b) of the Protector’s Role Summons echoed the test applied by the court in a Public Trustee v Cooper Category 2 case: see:

(a) The decision of the Jersey Court of Appeal in Kan v HSBC International Trustee Limited [2015] JCA 109, where Bompas JA gave a convenient summary of the law in a Public Trustee v Cooper Category 2 case: “14. Where a trustee has made a momentous decision, that is a decision of real importance for the trust, and seeks the court’s approval for the decision, the legal test to be applied by the court is well- established in this jurisdiction. As explained in Re S Settlement [2001] JLR N 37, the court must satisfy itself (i) first, that the trustee’s decision has been formed in good faith, (ii) second, that the decision is one which a reasonable trustee properly instructed could have reached, and (iii) third, that the decision has not been vitiated by any actual or potential conflict of interest …

18. When the court is to give approval for a momentous decision the court needs to be satisfied as to the rationality of the decision; the lengths to which the court must go in examining the process by which the trustee arrived at the decision must depend upon the particular decision.” Once the court adjudicated and answered the three questions affirmatively, and concluded the trustees’ decision was rational, there was no longer a choice or discretion for the court. Martin J.A. made this point at [11] in the Guernsey Court of Appeal in Re F [2013] Judgment 32/2103:- “In the second type of application [i.e. a Public Trustee v Cooper Category 2 case], however, the court is not exercising a discretion. What it is doing is in effectPage 32 of 83

(viii) (ix) (x) making a declaration that the trustees' proposed exercise of the power is lawful; in other words, that the proposed exercise is within the proper ambit of the power, that the trustees are acting honestly, and that in reaching their decision the trustees have taken into account all relevant matters, have taken into account no irrelevant matters, and have not reached a decision that no reasonable body of trustees could have reached.” By contrast, as a matter of ordinary language, in the X Trusts’ Protector Provisions the requirement for the Protector’s “consent”, as well as the agreement of the members of the Protectorate “as to whether it should give or withhold its consent”, showed that the Protector has a choice or discretion. This discretion was inconsistent with the adjudication function of the Narrow Review Role. Further, there was nothing in the remainder of each deed of appointment which introduced the Protector Provisions into the Phase 1 and 2 Trusts, or in the Modern Trusts, to suggest the role of the Protector was limited to (i) the adjudication of the rationality of the Trustees’ decision and (ii) an obligation to consent where the decision is rational. Indeed, the remaining Protector Provisions strongly supported the view that the Protector was intended to have an independent discretion. By reference to a specimen deed of revocation and appointment dated 13 May 1994 it could be seen that: (a) Paragraphs 1(b) and 7 conferred on the Protectorate an independent discretion to designate “the Specified Securities” for the purpose of the powers of consent in relation to the trustees’ powers to deal with and vote “Specified Securities”, so as to extend the scope of the Protectorate’s control over investments. (b) (c) Paragraph 8 authorises the Protectorate to waive its “powers”. If the Protector’s role were the Narrow Review Role – to adjudicate the rationality of the trustees’ decision to exercise their power – it would have been appropriate to use a different term to describe the abolition of the Protector’s function in connection with the adjudication of the rationality of the Trustee’s proposed exercise of powers. Moreover, paragraph 13 (relating to “Joint Protectors”) provides that, where there is more than one Protector, the members’ decisions must be taken unanimously. If the members of the Protectorate cannot agree, thenPage 33 of 83

64. 65. 66. (xi) the Trustees do not require the Protectorate’s written consent. But in such a case “the Trustees shall nevertheless consult with each Protector and shall take into account the views expressed before making a final decision”. The role of the Protector is thus wider than the narrow role of adjudicating the rationality of the trustees’ decision-making. If the Trustees must ascertain the differing views of the Protectorate’s members on the merits of their proposal where the members disagree, it is implicit that any agreed view of the Protectors must have involved the Protectors’ discretionary consideration of the merits of the Trustees’ proposal, not merely an evaluation of its rationality. In conclusion, Mr Taube submitted that the natural and ordinary meaning of the powers to give or withhold consent showed that (i) the trustees of the Phase 1 and 2 Trusts and (ii) the settlors of the Modern Trusts intended to confer on the Protectors an independent discretion. There was no ambiguity or inconsistency in the language used in the Protector Provisions and there was nothing in the words to suggest that the draftsman intended the role of the Protector to be limited to an expert adjudication role like the court’s role in a Public Trustee v Cooper Category 2 case. Consequently, the Court had to give those words their natural meaning, namely that the Protector has an independent discretion: see Lord Clarke at [23] in the Rainy Sky case above. (ii) Background factual context favours Protectors having independent discretion Mr Taube next submitted that the conclusion for which the Appellants contended, namely that the Protectors had an independent discretion, was reinforced by looking at the surrounding circumstances in 1994 and 1995 at the time of Operation Protector and in 1997 to 2003 when the Modern Trusts were settled. I should interpose to say at this juncture that, having rejected the Appellants’ attempted reliance upon the Contemporaneous Materials and the Affidavits, I consider Mr Taube’s attempted articulation of the objective factual matrix set out below includes what I regard as inadmissible subjective materials. However, as I have summarised in paragraph 58 above, the Protector Provisions can, and indeed should, be construed against the objectively ascertainable background facts. If one strips out his reliance on the inadmissible settlor’s and others’ wishes and intentions, his “background features”, to the extent that they can be inferred objectively, are not that different from those which I have summarised above. Mr Taube’s submissions in support of the contention that the background factual matrix supported the Appellants’ contention may be summarised as follows. (i) There were three important features of the background to the introduction of the Protector Provisions which identified their purpose.Page 34 of 83

(ii) (a) (b) (c) By the 1990s, the residence of the Trustees had moved across the Atlantic to Bermuda. Three settlors and beneficiaries of the X Trusts wished the family’s trusted advisers to have some control over the offshore trustees. They requested the Trustees to introduce the Protector Provisions in Operation Protector. Five of the directors of the two corporate Protectors were based in London or the Channel Islands and were trusted advisers of the settlors and the family. The Operation Protector Bible, which the family’s advisers prepared at the request of the settlors and the Trustees with a view to the introduction of the Protector Provisions by the Trustees, stated at para 2.4: “The veto powers over capital and the control of specified securities would be designed to provide stability, continuity and coherence in long term planning for the benefit of the X Family as a whole in relation to primary assets (i.e. shareholdings in [OpCo]).” The first two factors were and remained common reasons for settlements with non- resident trustees to incorporate provisions that required the trustees to obtain the prior written consent of a powerholder, whether named a protector or given some other name: see [91] and [117] in Re Piedmont and Riviera Trusts [2021] JRC 249 (which I shall refer to as “Piedmont”). As Sir Michael Birt noted, in this sort of case – “The settlor has decided that a protector (often himself or a longstanding friend or adviser whose judgment he trusts) should be appointed pursuant to the trust deed and has specified those matters where the protector’s consent is required. The settlor must be taken in those circumstances to have intended that the protector should exercise his own judgment in exercising those powers; otherwise why bother to go to the trouble of appointed [sic] a trusted friend or adviser (or himself) as protector rather than someone with a legal qualification to judge issues of rationality.” As Sir Michael said at [117], it is “inherently unlikely” that the settlors would request the Trustees “to go to the trouble of appointing … trusted friends or advisors as protectors if they intended the role of protector to be limited to that of assessing rationality”. Page 35 of 83

(iii) In the case of the X Trusts, there was a further important factor. The Trustees of the X Trusts held, and still hold, a significant aggregate interest in OpCo, a valuable quoted company. This bloc of shares had greater value and influence over X plc if the Trustees voted and acted in a unified manner.

67. (iv) One purpose of introducing the Protectors’ powers to give or withhold consent to the Trustees’ exercise of their powers was “to provide stability, continuity and coherence in long term planning for the benefit of the X Family as a whole in relation to primary assets (i.e. shareholdings in [OpCo]).” The judge accepted that the evidence established this fact; see [31 - 38], [45 – 47] and [67] in the judgment. (v) Had the Protectors’ role simply been to consider the rationality of the proposed exercise of the Trustees’ powers, this would not have enabled the Protectors to use their powers “to provide stability, continuity and coherence in long term planning for the benefit of the X Family as a whole in relation to primary assets (i.e. shareholdings in [OpCo]).” The Trustees would have been able to get over the low bar of rationality in their decisions about distributions or dispositions of the shares in OpCo; and the Protectors could not have withheld consent to decisions of the Trustees which indeed jeopardised the “stability, continuity and coherence” of long-term planning affecting this bloc of shares. (iii) Relevance of practical implications if Protectors have an independent discretion Mr Taube then submitted that the judge was wrong to come to his conclusion on the construction of the relevant Protector Provisions not by reference to the ordinary and natural meaning of the words used), but instead on the basis of their legal and practical implications, as he understood them. In summary, in this context6 Mr Taube submitted as follows: (i) The judge went wrong in this regard in two respects. (a) First, he misunderstood the legal and practical implications of the relevant Protector Provisions. (b) Secondly, in the face of the unambiguous ordinary and natural meaning of the Protector Provisions which conferred an independent discretion on the 6 Mr Taube dealt with the legal and practical implications of the Protector Provisions in this context without prejudice to the Appellants' primary submission that the judge was, in effect, wrongly implying terms into the Protector Provisions. Page 36 of 83

(ii) (iii) (iv) (v) Protectors, it was not permissible for him to depart from that meaning by reference to the legal and practical implications which he identified. He quoted [113] of the judgment, where the judge said that the Narrow Review Role: “reflects the true construction of the consent powers conferred on the Protectors of the X Trusts primarily because it is clear from the terms of the relevant instruments that their dominant purpose is to ensure the due exercise of the powers vested in the Trustees ... The protector’s role is to be a “watchdog” to ensure the due execution by the trustee of the powers vested in the trustees”. However, submitted Mr Taube, there were no words in the relevant Protector Provisions that referred to the Protectors’ role being to “ensure the due exercise” or “due execution” of the Trustees’ powers or to act as “watchdog”, let alone words which are capable of bearing this meaning. On the contrary, the Protector Provisions showed that the Protectors had their own choice to make. It was inherently unlikely the draftsman intended the settlor’s trusted advisers in the Protector companies to act solely as experts on rationality. The judge based his conclusion on the propositions:

(a) that absent a contrary intention discerned in the trust instrument, “the usual role of a protector is not to exercise a power jointly with the trustee in relation to the matter requiring protector consent” at [113];

(b) that the trust instruments distinguished between powers vested in the Trustees alone, powers vested in the Protector alone, and powers vested in the Trustees subject to the Protectors’ consent at [114] and [115]; and

(c) that the Protector Provisions contained no express statement that the Protectors had an “absolute discretion” at [115]. Mr Taube submitted that the judge appeared to have thought that, if the Protectors had an independent discretion to give or withhold consent to the Trustees’ relevant powers, then the Trustees’ relevant powers would be “joint powers”. Thus, he misunderstood the legal effect of the Protectors’ powers to give or withhold consent to the exercise of the Trustees’ relevant powers, if the Protectors had an independent discretion. Such an independent discretion would not have had the Page 37 of 83

result that the Protectors had a “joint power” with the Trustees or an “absolute discretion”. (vi) It was for the Trustee to make the decision in the first place, and not the role of the Protector to reach the initial decision: see Piedmont at [92])7; at [116(i)]: Sir Michael Birt continued: “we accept that the role of a protector is not to exercise a power ‘jointly’ with the Trustee. On the contrary, as we endeavoured to clarify at paragraph 92 above, the discretionary power to make a distribution lies with the trustee. The protector’s only function is to decide whether or not to consent to that decision by a trustee. It is a separate decision on the part of the protector, not a joint exercise of a power with the trustee. We do not see that a conclusion that a protector does not exercise a power jointly with the trustee points towards the Narrower View rather than the Wider View.” (vii) (viii) In the X Trusts, even if the Protectors had an independent discretion, it was not an “absolute discretion” when deciding whether to give or withhold consent to the exercise of the Trustees’ powers - the Protectors and the Appellants had always accepted the Protectors’ discretion was not “absolute”. Thus, in reaching its decision, a Protector was bound to act in good faith in the best interests of the beneficiaries, not for the Protector’s own benefit, and to take into account only relevant considerations and to ignore irrelevant considerations. (See section D below.) The A Family argued, and the judge accepted below, that the Trustees’ relevant powers had primacy; the Protectors’ powers of consent were ancillary to the Trustees’ powers8; and so, the A Family argues, the Protectors’ role had to be confined to satisfying themselves that the Trustees were acting rationally, even though the Protector Provisions did not say this. If that process of reasoning were correct, it would apply equally to the common case where a trustee’s power to appoint capital or to dispose of investments was subject to the consent of the settlor or a life tenant. Yet, in such cases, it had never been suggested the person with the 7 Mr Taube also referred to the Bermuda Court of Appeal in In the Matter of an Application for Information About a Trust [2014] Bda LR 5 at [67]: “The words in parenthesis “with the prior written consent of the Protector” can only mean, in our judgment, that the Trustees must obtain the Protector’s written consent before any release takes place; they do not have the effect of transferring the exercise of the Trustees’ discretion to the Protector.” 8 At [72 – 73] in the Judgment the learned judge also attached weight to the fact that the Protectors have powers to waive the requirement for their consent. Since the Trustees also have powers to waive their relevant powers, the existence of the Protectors’ powers of waiver are neutral on the issues about their role.Page 38 of 83

power of consent had no independent discretion or was bound to consent if the trustee’s decision is rational. In this context Mr Taube relied on a number of cases as set out in schedule 1 to his written submissions.

68. Mr Taube also submitted that the additional practical considerations relied upon by the A Family (as set out in their “Respondents’ Notice of Appeal”) provided no reason to conclude that the Narrow Review Role was the correct construction. In this context he submitted: (i) (ii) (iii) The A Family’s submission under this head was that if the Protectors had an independent discretion, that was a recipe for “deadlock”; but, by contrast, if the Protectors were obliged to consent to the Trustees’ exercise of their relevant powers because the trustees were acting within the bounds of rationality, this avoided “deadlock”. But that, submitted Mr Taube, was an inappropriately pejorative use of the term “deadlock”. Another way of looking at the same situation – and a more accurate way – was that if the Protector withheld consent to a proposal from the Trustee, that simply gave effect to the settlor’s intention that the Trustee’s power should not be exercised without the Protector’s consent. That was not “deadlock”: see Lord Briggs JSC at [221] in Children’s Investment Fund Foundation (UK) v A- G (sub nom. Lehtimäki v Cooper) [2020] UKSC 33, [2020] 3 WLR 461 (“CIFF”). If the Settlor had given the Protector a power of consent and the latter reached its own rational independent decision to withhold consent to the Trustee’s proposed exercise of a power, the court should respect the Protector’s decision. That approach was consistent with the “non-intervention principle”. It was only in truly exceptional cases that the court might override a fiduciary’s proper decision to withhold consent to the trustee’s exercise of a power: see CIFF (above). In Piedmont (above) Sir Michael Birt specifically dealt with the point about deadlock at [118] as follows: “We acknowledge that the approach we favour carries with it a greater risk of deadlock between trustee and protector if a protector refuses consent. Clearly, if a trustee considers that a protector’s refusal to consent is irrational or otherwise legally flawed, he may have recourse to the Court to overturn the protector’s veto. However, there is the potential for deadlock where the trustee and the protector both reach rational but opposing decisions. In our judgment, this is a natural consequence of the settlor’s decision to introduce the office of protector into the trust deed. A settlor must be taken to have intended (by imposing a requirement for consent) that a trustee should not be ablePage 39 of 83

(iv) (v) (vi) to make certain decisions unless the protector consents. If consent is refused, the trustee’s decision cannot be put into effect. In most cases this is likely to lead to further discussion between trustee and protector in the hope of finding a sensible outcome. In the event of complete deadlock where such deadlock is causing real damage to the interests of the beneficiaries, we leave open the possibility of recourse to the Court. The Court has power to break a deadlock where this is caused by lack of agreement among trustees where they have to act unanimously (see Garnham v PC [2012] (1) JLR 204, approved by the Supreme Court in Children’s Investment Fund Foundation (UK) v Attorney General [2020] 3 WLR 461 at [219]. It may be arguable that the Court has a similar jurisdiction in the event of damaging deadlock between a trustee and a protector. However, we say no more about that. We have not heard any argument on the point and it does not arise in this case.” In practice, it was unlikely that there would be permanent deadlock, save in the most exceptional case. Where the Protectors had rationally refused consent to a proposal from the Trustees on the grounds that the proposal was not in the best interests of the beneficiaries, it was to be expected they would discuss the position and try to find an alternative solution. In the X Trusts which are governed by English law, there was another solution in the case of damaging “deadlock” relating to the management or administration of the trust property. Mr Taube gave the example of a decision by the Trustees on rational grounds that a proposed transaction with the X shares was in the best interest of the beneficiaries, but the Protectors withheld their consent. In that case, the Trustee were able to apply to the court and obtain authority to carry out the transaction if they could show it was “expedient” for the trust and beneficiaries as a whole. (See section 57 of the English Trustee Act 1925 and Re Beale’s Settlement Trusts [1932] 2 Ch 15, where consent could not be obtained to exercise a power of sale.)9 The position was identical in the X Trusts which were governed by the law of Bermuda: see section 47 of the Bermuda Trustee Act 1975. Additionally, section 47 permitted the court in Bermuda to authorise beneficial dispositions as well as administrative transactions which were “expedient”, where the Trustees have no power to implement them because the Protectors refuse their consent: see the 9 Mr Taube informed the court that, as regards the single X Trust governed by Jersey law, article 47(3) of the Trusts {Jersey) Law 1984 (as amended) had similar effect. Page 40 of 83

summary of the law by Hargun C.J. at [9] in Re GA Settlement [2019] SC (Bda) 38 Civ.

69. 70. 71. (vii) Thus, he submitted, the judge’s concerns around impasse between the Trustees and Protectors were misplaced. Likewise, Mr Taube rejected the Trustees’ argument (relied upon by the A Family in their Respondent’s Notice) that there would be “waste (in terms of both time and money) and duplication” if the Protectors had an independent discretion to give or withhold consent to the Trustees’ power all, which was referred to at [63] by the judge. Mr Taube submitted that it was inevitable, that both the Trustees and the Protectors would have to consider carefully any proposed exercise of the Trustees’ relevant powers which required “the prior written consent of the Protector”. Time and money would have to be expended whether the role of the Protectors was the Narrow Review Role or the Wide Discretionary Role. The settlors or trustees who introduced the Protector Provisions must be taken to have intended that such additional time and money would have to be expended by the Protectors in considering their decisions and regarded it as beneficial. Accordingly, submitted Mr Taube, this factor could not sensibly affect the construction of the Protector Provisions. Respondents’ implied term argument I turn next to set out the A Family’s alternative argument in relation to implication of a term into the Protector Provisions to the effect that the Protectors’ role was the Narrow Review Role; and also the submissions of the Appellants in response. Ground 1.2(f) of the Respondents’ Notice set out the alternative position that the Narrow Review Role arises as a matter of implication of law. As is clear from the transcript10, this was raised by Mr Green before the judge as an alternative way of looking at the case. Mr Green emphasised that it was not necessary for the A Family to put their case in this way, but suggested that it might be helpful to do so for two reasons: (i) first, because it might be said to be another way of putting the A Family’s point that the identification of the fiduciary protector’s role was a matter of substantive trust law; (ii) second, if the matter was so looked at, it was clear that one was not looking at what were described as cases of terms implied in fact, but rather with what are described as cases of terms implied by law – the relevant analysis and test being different in each case. 10 Transcript Day 4 page 74 lines 15-20 to page 77 line 21; in particular at page 76 line 3 to page 77 line 21.Page 41 of 83

72. 73. 74. Mr Green submitted that the implication for which he contended was reflective of the respective roles of: (i) the trustees as officeholders under the trust having a substantive power vested in them; (ii) the protector as an entirely separate officeholder under the trust, having a qualitatively different role from that of the trustee, pursuant to which a consent power only is vested in it; and (iii) the beneficiaries who have an interest in the proper exercise of the trustees’ powers in accordance with their terms. Whilst he accepted that, as set out by Lord Neuberger in Marks & Spencer plc v BNP Paribas Securities Services [2016] AC 742 at [25 to 31], the processes of construction and implication were separate and sequential, they nonetheless led to the same conclusion. Inasmuch as the differential roles between the trustees and the protector gave one a clear steer as a matter of construction as to what the answer to the issues raised in the present case was, so too, the different roles of the trustees and the protector may be said to give one a direction or steer as to where the implication of law might be found. The parties were in agreement as to the principles applying to the implication of terms. The position is best summarised by Lord Collins in the Privy Council case of Vizcaya Partners Limited v Picard [2016] UKPC 5, [2016] 3 All ER 181 at [57]: “In English domestic law, there are, broadly, two classes of implied term. The first class, sometimes called terms implied as a matter of fact, consists of terms implied from the circumstances in order to give effect to the intention of the parties to the contract. The authorities on this class of implied term have been reviewed comprehensively by Lord Neuberger of Abbotsbury PSC in [Marks & Spencer], and it is not necessary to repeat what he said there. The policy of the common law is not to imply such terms lightly, and that is why the principles have been formulated in terms of necessity or business efficacy or “it goes without saying”. The second class consists of terms implied by law, which are implied into classes of contractual relationship as a necessary incident of the relationship concerned. An example is the obligation of confidentiality in banking contracts or in arbitration agreements: Emmott v Michael Wilson & Partners Ltd [2008] Bus LR 1361 (“really a rule of substantive law masquerading as an implied term”: at para 84).” Mr Green also referred to the distinction between the two types of implied term as acknowledged by Lord Neuberger in Marks & Spencer plc at [15] (emphasis added): “As Baroness Hale of Richmond JSC pointed out in Geys v Société Générale, London Branch [2013] 1 AC 523, para 55, there are two types of contractual implied term. The first, with which this case is concerned, is a term which is implied into a particular contract, in the light of the express terms, commercial common sense and the Page 42 of 83

75. facts known to both parties at the time the contract was made. The second type of implied term arises because, unless such a term is expressly excluded, the law (sometimes by statute, sometimes through the common law) effectively imposes certain terms into certain classes of relationship.” In this context Mr Green also referred the Court to the following textbooks: (i) Chitty on Contracts, 34th ed., 2021, at §16-005 which refers to the categories as terms implied in fact and terms implied in law, states that there are important differences between the two, and describes them as follows: “Terms implied in fact are implied into a particular contract in the light of the express terms of the contract, commercial common sense and the facts known to both parties at the time of entry into the contract. Terms implied by law, by contrast, are implied into ‘…a class of contractual relationship, such as that between landlord and tenant or between employer and employee, where the parties may have left a good deal unsaid, but the courts have implied the term as a necessary incident of the relationship concerns, unless the parties have expressly excluded it.’” (citing Société Genéralé v Geys) and then continues (ii) “…when deciding whether or not to imply a term as a matter of law into a contract of a particular type, the courts do not confine themselves to a narrow test of necessity but instead can draw upon a wider range of factors such as reasonableness, fairness, policy when deciding whether the proposed term is a necessary incident of the type of contractual relationship in question.” Lewison, The Interpretation of Contracts, 7th ed., 2020, at §6.13 states: “Where the court is asked to imply a term as a legal incident of a particular legal relationship, the strict test of necessity need not be satisfied. The court is concerned with broader questions of policy”, and goes on in §6.14 Page 43 of 83

76. (iii) “… the court has recently isolated a special category of implied terms, namely those where the court is asked to imply a term as a legal incident of a particular relationship, as a default rule which will apply unless specifically excluded. These kinds of implied terms are not based upon the intention of the parties, actual or presumed, in a given instance, although the provenance of a particular term may well have been the commonplace use of such a term in earlier times of contracts of that type, so establishing what later would become the default rule.” And then at §6.15, Lewison makes the point that referring to “implied terms” at law may be a misnomer: “Where the court decides that a term should be implied as an incident of the legal relationship it is really deciding a question of substantive law. In Emmott v Michael Wilson & Partners Ltd [2008] Bus. L.R. 1361, Lawrence Collins LJ, speaking of the implied obligation of confidentiality in an arbitration agreement said: ‘The implied agreement is really a rule of substantive law masquerading as an implied term.’” (at para [84]).11 Mr Green submitted that such was indeed the correct analysis. Mr Green further submitted that, whilst it was sometimes said that the term “protector” was not a defined term of art, or, in other words, the fact that somebody was described as such did not conclusively determine the nature of his function, once it was apparent that the nature of the protector’s function as such was fiduciary, and that he was someone who held his powers for the benefit of the beneficiaries, in a separate capacity from the trustees, as officeholders, then it followed that the common law would imply the incidents of the Narrow Review Role into the office.

77. Thus, Mr Green submitted that whether one approached the question on the basis of the second type of implied term, or as a question of substantive law defining the role of a protector, the answer was the same. 11 See also Lord Collins in Vizcaya Partners Ltd v Picard [2016] UKPC 5, [2016] 3 All E R 181 as quoted by the Appellants at para 101 of their Skeleton – in the passage following that emphasised by them.Page 44 of 83

78. 79. 80. 81. Mr Taube, on the contrary, submitted that there was no basis for the implication of the second type of implied term, since there was no recognised statutory or common law rule creating such a term. Nobody had identified a statutory or common law rule to the effect that there was an implied term that the role of a powerholder is the Narrow Review Role whenever a trustee has a power which requires the powerholder’s written consent. Mr Taube further submitted that in England and Jersey the courts had specifically held that there was no implied term to this effect: see PTNZ (above) and Piedmont (above). 51 of the X Trusts were governed by English law, 21 by the law of Bermuda, and 1 by the law of Jersey (plus an additional 6 which were not subject to the current proceedings although one of the respondents was the trustee thereof). There was no statute or case law in any of these jurisdictions to support the suggested rule. The second type of implied term could only be established on the basis of its being a necessary incident of the relationship between the trustee, in whom the power is vested, and the protector, whose prior written consent is required. But the suggested implied term was not a necessary incident of the relationship between a trustee and a holder of a power of consent under the second test for the implication of terms and there was no sound basis for the court to invent one now. In this context Mr Taube relied upon a review of the cases, textbooks and commentary as set out in schedule 1 to his skeleton argument. He submitted that his review illustrated the following points: (i) (ii) (iii) (iv) First, the term “protector” was not a defined term of art. The powers and duties of any “protector” depended on the terms of the particular trust instrument appointing him. The forms of trust instrument were manifold and diverse, as were the types of powerholder and the divisions of responsibility between different fiduciaries in a single trust. Second, prior to the case of PTNZ in 2020, there was no hint of a suggestion in the case law or literature that there might exist an implied term of the type for which the A Family now contends. Third, reported cases on powers of consent showed that, absent a special statutory jurisdiction or exceptional circumstances, the court may not dispense with a requirement in the trust instrument for X’s prior written consent to a power vested in a trust fiduciary. Fourth, an analogy existed under the Settled Land Act 1925 and its predecessors. The tenant for life had fiduciary powers to dispose of the settled land, and some powers required the consent of the settlement trustees. The court held there were Page 45 of 83

82. 83. 84. (v) (vi) limited duties imposed on the trustees in exercising their power to consent, but the duties did not include a duty to consent if satisfied the tenant for life’s proposed disposal was proper and rational. Fifth, in England, Bermuda and Jersey there was no legislation creating a statutory implied term or rule that the protector has the Narrow Review Role. Furthermore, in the legislation concerning “protectors”, which has appeared in recent decades in “offshore” common law jurisdictions, there was no suggestion that a rule or implied term exists of the kind for which the Respondents’ Notice contends. Sixth, prior to 2020 and the decision in PTNZ, none of the textbooks or articles on trusts and powers, whether those cited by the judge at [79 – 89] of the judgment or elsewhere, stated or even adverted to the argument that the role of the person with a power of consent was confined to the Narrow Review Role. In these circumstances Mr Taube submitted that there was no sound basis for implying a term on the second basis. However, Mr Taube accepted that, if under the relevant Protector Provisions in the X Trusts, the Protectors indeed had an independent discretion, then there necessarily had to be implied terms that the Protectors must exercise the discretion (i) in good faith, (ii) in the manner that, subjectively, they considered to be in the best interest of the beneficiaries, not for their own benefit, and (iii) rationally. Mr Taube submitted that, in contrast to the implied term for which the A Family contended, such implied terms were obvious and necessary, because under the X Trusts each Protector was an office holder, with no beneficial interest, whose function was to serve the beneficiaries; and that these implied terms had been recognised by the courts. (See IRC v Vestey [1949] 1 All ER 1108, 1115; Pitt v Holt [2013] 2 AC 108 at [10]; CIFF at [42] – [51]; Piedmont at [89] above); and the summary of the duties of fiduciaries in chapter 29 section 3 of Lewin on Trusts 20th edn., especially paras 29-031 to 29-033. Such implied terms were consistent (i) with the express terms in the Protector Provisions, which conferred an independent discretion on the Protectors to consent or not, and (ii) the duty of a fiduciary to act in the manner that subjectively he rationally considers to be in the best interests of his beneficiary: see IRC v Pilkington [1964] AC 612, 629 and 632; Re Hampden’s Settlement Trusts (1977) 23 March, [2001] WTLR 195; CIFF at [41], [100], [218] and [232]: and Regentcrest plc (in liq) v Cohen [2001] BCC 494 at [120]. Discussion and determination of the construction issue I agree with the conclusion of the judge, that, on the proper construction of the Protector Provisions in the relevant X Trust deeds, the role of the Protector is limited to the Narrow Review Role. The reasons for reaching my conclusion are somewhat different from those of the judge, and, accordingly, I set them out below.Page 46 of 83

85. Approach to construction of the Protector Provisions The starting point is that the relevant Protector Provisions must, in accordance with the various House of Lords and Supreme Court decisions on construction which I have referred to above, be construed in the context of the whole trust deed and against the admissible factual matrix, excluding inadmissible evidence of the subjective intentions of settlors, beneficiaries, trustees and advisors. And that admissible context, as submitted by Mr Green on behalf of the A Family, includes “the context of general principles which apply to the office and role of the protector under the settlements”. Support for this approach is to be found in the case of Davidson v Seelig [2016] WTLR 627 where Henderson J (as he then was, and who was and is an extremely experienced trusts lawyer) stated at [54] to [56] in relation to protectors’ powers materially similar to those in issue in the present case12: “[54] Counsel for the claimants and the trustees submit that the amendments need to be considered in the context of general principles which apply to the office and role of the protector under the settlements. There are currently two protectors, Promenade and Mr Haringman. The terms of their joint office are defined by the protectorship regime set out in the 2003 deeds. As I have already pointed out, where there are two protectors they have to act jointly: neither has power to act alone, subject to limited exceptions (such as lack of capacity) which are for present purposes irrelevant. [55] Assuming that the protectorship regime was validly introduced, the protectors have four principal functions to perform in relation to the administration of the trusts. First, they have power to give or withhold consent to any exercise by the trustees of their beneficial powers of appointment, or revocation of earlier appointments, from time to time. Secondly, they have power to remove any trustee from office, with or without cause, provided that there will still remain a minimum number of trustees. Thirdly, they have a contingent power to appoint new trustees which will be exercisable only after the death or incapacity of both settlers. Finally, the protectors may together appoint new protectors. These powers are fiduciary, and they must be exercised in the interests of the beneficiaries. The protectors do not, however, have a general power or duty to supervise the administration of the settlements, and they may only apply to the court for relief which relates to the proper exercise of their own powers. 12 These provided for (i) the exercise of overriding powers of appointment to have to be “with the consent of the protector” and (ii) the protector having powers to appoint and remove trustees in defined circumstances.Page 47 of 83

86. 87. 88. [56] I would provisionally accept these submissions, which appear to me firmly based on general principles of trust law and to reflect the limited nature of the powers conferred on the protector by the 2003 deeds. In the light of these principles, I can now examine the main forms of relief sought by Mr Haringman.” [My emphasis.] Thus, one starts from the position that, when one comes to consider the ambit of the consent powers of the Protectors under the relevant Protector Provisions, one is considering a specific power of a separate fiduciary office holder in relation to the exercise by the trustees of a settlement of certain substantive powers conferred under the trust instrument on the latter. And, as Henderson J made clear at [56] above, powers of the type conferred on protectors under the X Trusts, and under the similar provisions in Davidson v Seelig, and indeed in many other trust instruments where similar wording is used, are necessarily limited in nature. It was common ground below, based on the various expert reports from English leading counsel, put forward by the Appellants, the A Family, the Trustees and the Protector, that the consent wordings used in the relevant settlements are (and were in 1994/5) of general application, standard in concept, and commonly used, and contained nothing that was out of the ordinary or exceptional in form. But, as Mr Green pointed out, whilst the law relating to the trustee’s role as regards the exercise of a substantive power vested in it is well developed, having been declared and affirmed by courts over many years, the law relating to the fiduciary protector’s role where a consent power is conferred on it in relation to a trustee’s substantive power is relatively undeveloped. Determination of the ambit of that latter role is the subject of this appeal. The wording of the relevant Protector consent provisions must take into account the separate functions of the Trustees and the Protectors I turn first to the wording of the consent provisions. As I have already set out above, Mr Taube emphasised the natural meaning of the express language used, which he said conferred on the Protectors an absolute and unfettered discretion which was only consistent with the Wide Discretionary Role. There was no ambiguity or inconsistency in the language used in the Protector Provisions and there was nothing in the drafting to suggest a limited role. Consent, he submitted, meant just that – the giving or withholding of consent; it did not mean that the Protectors had to engage in some adjudication exercise every time they were called upon to give their consent to the exercise of the relevant power by the Trustees, so as to determine whether the Trustees had acted rationally and in accordance with their fiduciary duties; even if the Trustees had reached what might be characterised as a rational decision from their perspective, it was open to the Protectors to disagree and refuse their consent, provided as he accepted, that their conclusion was reached on rational grounds and consistently with the Protectors’ own fiduciary duties. In the absence of words limiting the Protectors’ discretion, Mr Taube argued that the only Page 48 of 83

89. 90. 91. restrictions upon the Protectors’ role were those applicable to any fiduciary, namely the obligations to act in good faith, in the manner they consider to be in the best interests of the beneficiaries, and rationally (IRC v Vestey [1949] 1 All ER 1108, 111517; Pitt v Holt [2013] 2 AC 108 at [10]; Children’s Investment Fund Foundation (UK) v A-G (sub nom. Lehtimäki v Cooper) [2020] UKSC 33, [2020] 3 WLR 461 (“CIFF”) at [42] – [51]). I cannot accept this submission. It disregards, and is inconsistent with, the fact that under the express language of the X Trusts, it is the Trustees who have the paramount substantive role of administering the trusts and exercising the powers conferred upon them. If the role of the Protectors was indeed the Wide Review Role, the Protectors, by threatening the exercise of their (on this hypothesis) absolute right of veto, would in effect themselves be taking the decision (for example) whether, to whom and in what amount to appoint capital or whether to sell Specified Securities and at what price. That would not be a joint decision with the Trustees, but rather an entirely separate decision by the Protectors, trumping that of the Trustees. Envisage the situation where the Trustees, acting rationally and properly, wish to make an appointment of capital to beneficiary M; the Protectors take the view, again rationally, that beneficiary N should also receive an appointment of capital – a view with which the Trustees, again rationally, disagree; the Protectors accordingly refuse their consent to the appointment of capital to beneficiary M. The result is that the decision (effectively taken by the Protectors) is that there should be no appointment at all. Mr Taube’s argument that the consent to be given by the Protectors is an unfettered consent, affording, as a matter of simple language, an uncontrolled choice on the Protectors, is also inconsistent with the acceptance by Mr Taube of the fact that, by the very nature of their office, the Protectors have, whether by way of an implied term or otherwise as an incident of their office as a fiduciary, fiduciary duties towards the beneficiaries which necessarily constrain the manner in which they exercise their discretion. The reality is that, whatever the language says, the discretion vested in the Protectors to refuse consent is not absolute. Let us take another example. On Mr Taube’s approach, the Protectors could refuse their consent, for example, to a properly reached, rational decision by the trustees to advance capital for the purposes of paying for the education of a particular female beneficiary, on the grounds that the vehemently expressed views of the settlor were that university education for that particular child would be a waste of time, and that no appointment of capital should be made to her until the occasion of her marriage. Let us assume that such refusal of consent by the Protectors on such grounds could, objectively, be regarded as rational and proper, because of the past irresponsible conduct of the girl. I find it very surprising that in such circumstances the Protectors could indeed withhold consent to the Trustees’ exercise of their powers, in circumstances where the Trustees had properly considered and taken into account the views of the settlor, and the girl’s conduct, but nonetheless decided to make the appointment because they had sufficient confidence in her current stated intention to obtain a degree. I reach this conclusion as a matter of construction based on the wording used and a consideration of the respective constitutional roles of trustees and fiduciary protectorsPage 49 of 83

92. 93. 94. under the relevant provisions of trusts such as the present. As Mr Green submitted, it is the Trustees who operate the trusts and the Protectors are there to act as watchdog or enforcer13 of the trusts imposed on the Trustees, not to operate such trusts themselves. Moreover, as all parties agreed, it is not a joint power of appointment. In my judgment, a provision for protector consent is providing for protector supervision; thus, the role of the protector is to sign off in relation to what the trustee is proposing to do in exercise of the discretion conferred on the trustee and the trustee alone. Protectors on terms such as the present cannot, in my judgment, be the ultimate decision makers. Obviously, to the extent that there are special, express, contrary provisions contained in the trust instrument which alter the balance of power between trustees and protector, conferring more extensive or different roles on the protectors and (correspondingly) less extensive or different roles on the trustees, the situation might be different. But there are no such provisions here; the Protector Provisions are the standard provisions which are found in numerous trusts. There is thus nothing in the language to displace the supervisory role of the Protectors when it comes to the performance of their consent functions. I do not find the logic of Mr Taube’s reasoning in relation to the Wide Review Role compelling. It appears to be based on the proposition that, since the Protectors are fiduciaries – and so, like the Trustees, obliged to do what they do in performance of their office in the interests of the beneficiaries not themselves – that predicates that, therefore, the Protectors are obliged to take their own independent view of whether and, if so, how the Trustees’ power should be exercised, when the substantive powers in question (in this case, powers of appointment over capital and dealing with Specified Securities)are that of the Trustees, and the Trustees alone, and there is nothing conferring such power on the Protectors. But recognition that a consent provision involves a fiduciary role, and so is not exercisable beneficially by an office-holder, does not have the effect, in my view, of conferring a parallel discretion on protectors in relation to the exercise by trustees of the substantive power vested in them (not the protectors), any more than it has the effect of transferring the trustees’ discretion to the protectors. I base my conclusion on my construction of the Trust Deeds against the background of substantive law relating to fiduciaries, trustees and protectors and their respective roles and obligations. I agree with Mr Green that it is difficult to discern a real difference between approaching the question on the basis of the second type of implied term or as a matter of substantive law. Indeed, if it were necessary, I would be prepared to imply the second type of implied term. I find support for my conclusion by a consideration of: 13 Re an Application for Information about a Trust [2013] CA (Bda) 8 Civ at [11].Page 50 of 83

(i) (ii) (iii) other relevant Protector Provisions; these do not in my judgment support the Wide Review Role and to a certain extent support the Narrow Review Role; some of the authorities to which we were referred, refer to, or reflect upon, the role of a protector appointed on similar terms to the present; these authorities demonstrate that, as a matter of substantive law, the protector’s role is not to usurp the trustee’s functions; none of them can be regarded as determinative of the issues which we have to decide in this case, save, arguably, for one, the decision at first instance of Sir Michael Birt in the Royal Court of Jersey in Piedmont with which I disagree; the respective practical implications of the Narrow Review Role and the Wide Review Role; including, in particular, the available methods of resolution provided by the court for any impasse between a protector and trustees in relation to the absence of consent and in relation to the approval of the exercise of trustees’ powers generally, or the surrender of such powers; these, in my view, support the Narrow Review Role.

95. I turn now to address these factors.

96. 97. 98. Other relevant Protector Provisions do not support the Wide Review Role and arguably support the Narrow Review Role Mr Taube, on behalf of the Appellants, pointed to a number of provisions found in the X Trust instruments which he submitted militated against the conclusion that the Narrow Review Role should be adopted. I deal with some of these arguments below. His primary submission14 was that paragraph 13 of the Protector Schedule was inconsistent with the Narrow Review Role in that it provides that, in a case where the Protectors (if more than one) cannot agree, the Trustees can proceed without Protector consent but “shall nevertheless consult with each Protector and shall take into account the views expressed before making a final decision”. In my judgment, there is nothing inconsistent with the Narrow Review Role in this provision, let alone supportive of the Wide Review Role. The Trustees in exercising their powers are bound to take into account relevant considerations, which necessarily must include the views of the Protectors. Because the latter are likely to be closer personally to a settlor or the principal beneficiaries, they are likely to be in a position to identify relevant considerations to the Trustees. If the Protectors disagree among themselves, the Trustees may be assisted by hearing both sides of the relevant argument. But the fact that the 14 See paragraph 69.2 of the Appellants' written submissions. Page 51 of 83

99. Trustees, where the Protectors are not unanimous, are obliged to consult with each Protector to ascertain his/her views, does nothing to support the construction of the Wide Review Role. They have to do that anyway. Moreover, in my judgment, the fact that the Trustees are entitled to go ahead and exercise their relevant powers, even though the Protectors are not unanimous, supports the construction that the Protectors only have the Narrow Review Role. It emphasises that the Protectors’ consent power is subsidiary to the primary power given to the Trustees as to whether to exercise their discretion to appoint capital or to deal with the Specified Securities. The second point taken by Mr Taube in this context relied on the fact that the Protectors have a unilateral power to designate further “Specified Securities” under paragraph 7 of the Protector Schedule15. He sought to argue from that provision that, in relation to the Trustees’ powers of sale, disposition, voting etc. of OpCo shares as specified securities, this showed that the Protectors should be taken to have the Wide Review Role so far as giving their consent to any exercise of the Trustees’ powers in this respect. I disagree. The fact that the Protectors may, in response to the circumstances prevailing in a given OpCo (e.g. OpCo shares being held for reasons of tax expediency by an offshore investment holding company), designate further specified securities, does no more than inform the reader that the intention of the Trust Deed is that, in appropriate circumstances, the Protectors may take the view that they should also determine that they should be keeping a like eye on the Trustees’ exercise of their powers in relation to additional securities as they are empowered to do in relation to the existing Specified Securities. But that does not say anything about the nature of such supervision by the Protectors or as to whether their function is the Wide or Narrow Review Role. In other words, as Mr Green submitted, the width of the Protectors’ powers where powers are uniquely conferred on them, is no basis for determining the breadth of a Protectors’ role under a provision providing for their consent only to the exercise of a discretion which is, and which remains exclusively, vested in the Trustees.

100. Mr Taube’s third point was that, because the Protectors were authorised to waive their powers under paragraph 8 of the Protector Schedule, that was inconsistent with the Protectors’ consent powers being limited to the Narrow Review Role. He said: “If the Protector’s role were the Narrow Review Role – to adjudicate the rationality of the trustees’ decision to exercise their power – it would be appropriate to use a different term to describe the abolition of the Protector’s function in connection with the adjudication of the rationality of the Trustee’s proposed exercise of powers. “ 15 See Paragraph 69.1 ibid. Page 52 of 83

101. 102. 103. 104. 105. I do not see why, logically, this follows at all. The power of waiver is a wide one which entitles the Protectors to waive or restrict their consent powers either on one, or more, occasions or in their entirety – i.e. for all time. I see no reason why the wording should be different, depending on the width of the consent power. The existing waiver wording is perfectly clear. Moreover, it could properly be said that the power of waiver is supportive of the Narrow Review Role in that the fact that the consent power can be waived at the discretion of the Protectors, suggests that their role is less extensive and less important than the Wide Review Role might suggest. Finally, both parties addressed the significance, or otherwise, of the absence of exoneration provisions in the Protector Provisions. At [76 – 78] of the judgment the judge expressed the view that a factor supporting the Narrow Review Role was that the Protectors, unlike the Trustees, have no rights to exoneration under indemnity provisions in the X Trusts in respect of their decision-making. Thus, he said at [77]: “professionals carrying out substantive decision-making, whether as directors in the company arena or trustees in the trust sphere, will generally not accept such onerous responsibilities without receiving generous indemnities”; and he thought that: “the absence of commensurate indemnities in favour of the Protectors was a further pointer to the conclusion that their powers of consent are merely ancillary to, rather than equal in status to, the Trustees’ relevant power.” The Appellants complained that these paragraphs of the judgment appear to have been based on submissions made on behalf of the Trustees in reply at the end of the hearing, after all other parties had made their submissions, so they had no chance to reply. Mr Taube pointed out that, as a matter of fact, in all the 18 Modern Trusts the Protectorate had the benefit of an exoneration provision preventing the Protectors from being liable for anything “except wilful fraud or wrongdoing” and complained that the judge appeared to have overlooked this point. Mr Taube submitted that, since some of the X Trusts contained provisions exonerating the Protectors and others did not, it was not easy to see how this factor could affect the construction of the relevant identical Protector Provisions in all the X Trusts (i.e. the requirement for the prior written consent of the Protectors to the Trustees’ exercise of a power). Page 53 of 83

106. 107. 108. 109. 110. He also pointed out that, as a matter of law, the individual “professionals” who acted as directors of the corporate Protectors owed their duties to the corporate Protector, not to the beneficiaries: Bath v Standard Land Company [1911] 1 Ch 618 and Gregson v HAE Trustees Limited [2009] 1 All ER (Comm) 457. Accordingly, Mr Taube submitted that the judge’s statement at [77] about the position of “professionals carrying out substantial decision-making” appeared to be based on a mistaken legal assumption. Finally, he submitted that, in any event, the corporate Protectors would have owed duties to the beneficiaries of the X Trusts, whether their role was the Narrow Review Role or the Wide Review Role. I do not find these arguments compelling. As Mr Green pointed out, the original absence of exoneration clauses in favour of the Protectors in the majority of the X Trust instruments might be said to be illustrative of the fact that protectorships were, and are, not seen as carrying the same range and depth of responsibility as trusteeships. The fact that exoneration provisions were included in the Protector Provisions in the Modern Trusts does not impact upon the construction of the earlier trust instruments. I do not consider that the inclusion or omission of such provisions is determinative either way. At best, it is perhaps slightly supportive of the Narrow Review Role as demonstrating the view at the time of the unlikelihood of the Protectors incurring liability in respect of their role. No submission was made by Mr Taube that the Protectors’ role under the X Trusts had changed in the more modern trust instruments. I derived no assistance from Mr Taube’s submission that the distinction between the corporate status of the actual Protectors and the individual directors had significance in this context. The authorities, which refer to, or reflect upon, the role of a protector appointed on similar terms to the present, although not determinative, support the Narrow Review Role I turn next to consider some of the authorities, to which we were referred, which refer to, or reflect upon, the role of a protector appointed on similar terms to the present. In my judgment, although not determinative, these support the Narrow Review Role and, in particular, emphasise that the role of a protector is not to usurp that of a trustee. Rawcliffe v Steele [1993-95] MLR (SGD) 426 In the leading Isle of Man Court of Appeal case, Rawcliffe v Steele [1993-95] MLR (SGD) 426, the Court had to decide whether just as a trust would not fail for want of a trustee, so it would not fail for want of a protector. The protector in that case had a series of consent roles in relation to trustee powers whose exercisability was essential to the proper functioning of the trust (including the power to populate an otherwise rudimentary beneficiary class). The Court decided that it could indeed appoint a protector to fill the role,Page 54 of 83

111. 112. and indeed in an extreme case (i.e., where appointment of a protector was not the solution), the Court could step in and exercise the protector’s consent role itself: see Acting Deemster Hegarty at [1993-95] MLR (SGD) p. 503 lines 30-33; and Acting Deemster Smith at p. 529 lines 29-30. Of particular importance in relation to the present case is the statement of Acting Deemster Hegarty at 510: “It seems to me that it would be wrong entirely to neglect the terminology involved. The word 'protector' seems to me to connote a role for the person holding that position even before one considers the detailed provisions relating to it. A 'protector' is, presumably, someone who 'protects'. But what is he to protect? He is not a protector of any specific individual or interest. At recital (b), he is referred to as 'protector of the trusts created by this Declaration of Trust'; and at cl 1(b) he is referred to as 'protector of the settlement'. It is, therefore, the settlement that he is obliged to protect. I interpret this as indicating that his essential role is to ensure that both the letter and the spirit of the settlement are complied with. For example, it might be difficult in practice for the Court to ensure that the trustees of a settlement would carry out a proper enquiry of the kind contemplated by Lord Wilberforce16 before exercising their powers. A protector with a power to give or withhold his consent to the exercise of such powers could much more readily ensure that they had done so.17” (My emphasis). It appears that Acting Deemster Hegarty contemplated that a trustee might fail to do his duty as assiduously as Lord Wilberforce had envisaged – and that it would be the protector’s role to ensure that he did, by refusing consent to the exercise of a power of appointment where a “proper enquiry” had not been carried out. As Mr Green submitted, the protector’s role would be to ensure that the trustee exercised its powers lawfully and in accordance with its duty. But the Acting Deemster does not envisage or imply, for example, that the protector could simply choose to disagree with the trustee as to who are the objects of the power to benefit from an appropriately assiduous exercise by the trustees of the power of appointment. At 511, the Acting Deemster returns to the point where he considers the protector’s power to give or refuse consent to the nomination of a proposed beneficiary by the trustees, stating that the role of the protector is to “give proper consideration to the trustees' proposals and decide whether to exercise his power of consent having due regard to the extent of the trustees' enquiry, the basis of their selection and the overall purposes of the trust all. He is 16 17 The reference to Lord Wilberforce is to his well-known statement in Re Baden's Deed Trusts (No 1) [1971] AC 424 of the duty of a trustee of a large discretionary trust to “examine the field, by class and category” and so make a sufficient survey to enable him to distribute trust funds This case is also known as Steele v Paz Ltd. Page 55 of 83

113. 114. 115. making the same point that it is not for the protector to exercise a wholly independent discretion as to whether the individual proposed should be a beneficiary, such that the trustees’ nomination will succeed only if the protector happens, independently, to have formed the same view. Rather, the protector’s role is to supervise the trustees’ exercise of their power in accordance with its duty and with the law. Furthermore, the ability of the Court to take over the exercise of the protector’s powers itself, without anything in the nature of a surrender by a protector, in my judgment provides support for the Narrow Role. In the matter of the A and B Trusts [2012] JRC 169A In In the matter of the A and B Trusts [2012] JRC 169A the Jersey Royal Court had to consider applications by trust beneficiaries to remove from office the protector of two trusts. The protector in that case had a “misconceived view of himself as the living guardian and enforcer of the settlors’ wishes”, bound “to ensure that the wishes of [the settlors] are adhered to” [3]. (My emphasis). This understanding of the protector role was rejected by the Royal Court, with Commissioner HWB Page QC holding at [4]: “It can be no part of the function of a protector with limited powers of the kind conferred on S by the trust instruments18 to ensure that a settlor’s wishes are carried out any more than it is open to a settlor himself to insist on them being carried out. A trustee’s duty as regards a letter of wishes is no more than to have due regard to such matters without any obligation to follow them. And a protector’s duty can, correspondingly, be no higher than to do his best to see that trustees have due regard to the settlor’s wishes (in whatever form they may have been imparted): from the moment of his acceptance of the office of protector his paramount duty is to the beneficiaries of the trust.” (Original emphasis) The words “a protector’s duty can, correspondingly, be no higher than to do his best to see that trustees have due regard to the settlor’s wishes” make clear that the Royal Court’s view of the role of a protector was not to take decisions for himself about whether the trustees should be following the settlor’s wishes or doing something else, but was rather to ensure that the trustees “have due regard to the settlor’s wishes”. If the trustees do so then they will not on that account be guilty of “inadequate deliberation”,19 and the protector will therefore have discharged his proper function of ensuring lawful and rational trustee decision-making. That this is what the Commissioner meant was also reflected in his subsequent finding at [11 (ii)] that the protector had cast himself in a role which “went well beyond what was 19 See Pitt v Holt [2013] UKSC 26, [2013] 2 AC 108 at [60]. Page 56 of 83

116. 117. proper for someone in his position and led him, not just to insist on playing an overactive part in the management of the trusts, but also to take up indefensible positions as regards his successor…”. (My emphasis). The clear thrust of the Commissioner’s conclusion was that it was not for a protector officiously to intrude on the business of trust administration, because his proper role was supervisory and did not entail second-guessing the trustees’ own decision-making. Re an Application for Information about a Trust CA (Bda) [2013] Civ App No 8, [2014] Bda LR 5 page 1 The decision of the Isle of Man Court of Appeal in Rawcliffe v Steele has repeatedly been cited and applied in Bermuda,20 including in Re an Application for Information about a Trust where Evans JA, at [43] quoted Acting Deemster Smith at page 529 as follows: “[counsel] described the Protector as being a vital part of the machinery of the trust. I agree with that analysis … his role is clearly vital. Nevertheless, his role in my opinion, is that of assisting in the administration of the trust … The protector must bona fide consider the exercise of the powers from the point of view of the beneficiaries under the trust.”(p.529). Further, at [67 – 69] Evans JA continued as follows: “67. In our view, clause 9.2 does not go so far as to release the Trustees from their duty to make their own decision, nor does it entitle them simply to pass on the request so that the Protector can decide. The clause reads “no person or persons shall be provided with” Trust accounts or information “except to the extent that the Trustees … In their discretion otherwise determine”. The discretion is clearly, and understandably, given to the Trustees. The words in parenthesis “with the prior written consent of the Protector” can only mean, in our judgment, that the Trustees must obtain the Protector’s written consent before any release takes place; they do not have the effect of transferring the exercise of the Trustees’ discretion to the Protector.

68. If that is correct, the Trustees are required to make their own decision, in the interests of the Trust and in accordance with the intentions of the Settlor as set out in the Trust Deed. If they are 20 The decision was recently applied In the Matter of H Trust [2019] SC (Bda) 27 Com at [11]-[15], where Hargun CJ rightly treated it as authority for the existence of an inherent jurisdiction to replace a fiduciary protector.Page 57 of 83

118. 119. 120. minded to release the information, they must seek the consent of the Protector before doing so. The question then arises, as it has done in the present case, on what grounds the Protector’s consent can properly be withheld, in a case where the Trustees are of the view that there should be a release.

69. It is not contended that the Protector’s refusal may be “capricious”, and it is recognised by the Appellant that it may not be “unlawful or irrational”. In our judgment, the Protector is bound by the same constraints as are the Trustees. The clause encompasses the release of information to beneficiaries as well as to strangers to the Trust. There is no indication that the Settlor intended that they should be deprived of information to which they are entitled as of right under the general law. Just as the Trustees were expected to exercise their discretion accordingly, so also in our judgment is the Protector in deciding whether to refuse consent to a proposed release. The Protector cannot lawfully refuse consent in a case where the Settlor is taken to have approved the release, any more than the Protector can vary the terms of the Trust”. (My emphasis.) It is correct, as the Protectors’ counsel pointed out in an earlier skeleton, that Re an Application for Information about a Trust – unlike the present case – concerned a non- fiduciary protector who was seeking rights as a beneficiary to information about the trust. But nonetheless the case is instructive to demonstrate that a protector’s consent powers are not absolute – they are constrained by obligations to respect the terms of the trust, fiduciary obligations and, indeed, subject to the jurisdiction of the court. I do not accept the Protectors’ arguments that this case demonstrated the existence of two separate discretions, one vested in the trustees and another in the protector. That is simply not a correct analysis of the case. The Appellants themselves did not seek to argue to this effect. I do however accept Mr Green’s submission that neither do the relevant Protector Provisions for fiduciary protector consent in relation to an exercise of the Trustees’ power of appointment in the present case confer a joint power of appointment – i.e. a power of appointment exercisable by the Trustees and the Protector together. What the Protector Provisions provide for, in my judgment, is the different function of consenting, or withholding consent, to the exercise of a discretion conferred on the Trustees and the Trustees alone. Accordingly, although the assistance is limited, I do nonetheless derive some support for the Narrow Review Role from the statements of Evans JA in Re an Application for Information about a Trust. Page 58 of 83

121. 122. 123. PTNZ v AS & others [2020] EWHC 3114 (Ch), [2020] WTLR 1423 PTNZ21 was the first case upon which the Appellants (and the Protectors) sought to place reliance to support their contention that the Wide Discretionary Role was the correct construction of the Protector Provisions. However, before considering this case, it is important to appreciate that the Appellants do not contend that the fiduciary consent provisions in the present case make the Trustees’ powers to which they attach joint powers. PTNZ is a recent English judgment of Master Shuman22. She held that, as a matter of construction of the trust deeds, against their factual context, the powers of the protector were in effect joint powers and not limited to a review of decisions taken by the trustee; accordingly, the protector in the exercise of his independent discretion, if he disagreed with the trustees, would therefore in principle be entitled to withhold his consent, even if the trustees were neither acting unreasonably nor for improper purposes; and thus there was no basis for imposing any limit on the role of the protector at the final hearing of the trustee's “blessing” application. The relevant paragraphs of her judgment are as follows: “[97] A protector's power of veto is as the name suggests exactly that and not a power of review. Under the trusts the trustees have a wide range of powers and discretions which require the written consent of the protector; I have summarised the key ones in para [75](b)(i) to (vii) above. In passing I note that this is consistent with how the judge approached the parties' respective positions in Bathurst. [98] Mr Wilson contends that if the settlor required a joint exercise of the dispositive powers by the trustee and the protector the trust deeds could easily have said so. Instead they provide for the trustee to exercise the power with the protector's consent. Mr Wilson suggests that as a matter of construction 21 PTNZ was before Kawaley AJ, who accepted the A Family's submissions that he should not apply it; see [111] to [112] of the judgment. 22 In England and Wales, a Master’s decision, albeit a decision of the High Court, is inferior in terms of binding authority to that of a High Court Judge. In Coral Reef Ltd v Silverbond Enterprises Ltd ([2016] EWHC 874 (Ch) at first instance), on appeal [2016] EWHC 3844 (Ch)) David Foxton QC, sitting as a Deputy High Court Judge, concluded that “the fact that a High Court judge and a master sit in the same court, namely the High Court, is not determinative of the question of whether the doctrine of precedent applies as between them” (para 61) and that “the decision of a High Court judge in terms of its clear ratio is binding on a master, absent either conflicting decisions of another judge at the same level of the High Court judge, or obviously of superior courts” (para 67). That is not surprising, since, as David Foxton QC pointed out (Coral Reef para 61), the fact that appeals from Masters’ decisions are heard by High Court Judges demonstrates their relative status.Page 59 of 83

124. there is a distinction to be drawn between the powers that each has. I do not accept that that follows from the wording of the trust deeds and the mechanism by which the officeholders were to exercise their powers. As SW observed in his 1st witness statement the purpose of the protector holding the power of consent is to control the trustees' exercise of their broad discretionary powers. I have not been referred to anything in the trusts that is consistent with a restrictive interpretation of the protector's role. In contrast the genesis of the trusts (as referred to in para [61] above), the language used in the trusts and the wide expansion of the powers of the protector set out in the deed of variation are consistent with the 1st defendant's intentions when the trusts were established that the protector would hold joint power with the trustee. [99] This position is also consistent with an offshore trust which typically appoints a protector. The trustee may very well be a corporate entity located in a different jurisdiction. The settlor and trustee may not know each other and there may be limited trust between them. In that context the imposition of a power of consent in the sense of being a joint power rather than a restrictive review power provides a solution to control the power exercised by the trustees. [100] I am satisfied that properly analysed the power of the protector is a joint power with the claimant and not a review power. [101] The third point raised by the claimant is what role the 10th defendant should play in the blessing hearing. In light of my decision there is no reason to limit the role of the protector at the blessing hearing.” (My emphasis). I do not regard this authority of any assistance. In my view it is wrong. It is not possible correctly to analyse the protector’s power in that case as a joint power with the trustees. It was an entirely separate power conferred on the protector. It was the scope of that power that was the issue. Moreover, it appears that the relevant point, apparently the last point considered in the case, was (unlike in the present case) the subject of only very limited argument before the Master, with the only opposition to the “joint power” interpretation provided by counsel for the neutral trustees [3], who was only seeking to assist the court and “expressly not adopting [the] position” he was arguing for: see [86]. Moreover, the Master appears to have taken account of inadmissible material at [98] in order to reach her conclusion, in circumstances where counsel do not seem to have argued to the contrary or Page 60 of 83

125. 126. to have cited Barnardo’s v Buckinghamshire [2018] UKSC 55, [2019] ICR 495 (above). She appeared to be uncertain as to whether the power in that case was fiduciary or not (see [80] and [85]), and in coming to her conclusion as to the scope of the role of protector, relied on only one case, namely Re Forster’s Settlement [1942] 1 Ch. 199. But Re Forster’s Settlement is not a case that was useful in the context of deciding the scope of the role of a fiduciary protector. In that case a first (divorced) wife of a dead settlor held a beneficial consent power in relation to the trustees’ power to override her interest as tenant for life. She was an enemy alien and it was uncertain whether she was dead or alive. The issue in Re Forster’s Settlement was whether, in such circumstances, such a beneficial consent power could be disregarded so that the trustees could exercise their powers of advancement to the children of the settlor’s second marriage, with the result that the powerholder (the first wife) would be deprived of her life interest in the property. It was perhaps not surprising that, in the absence of evidence of her death, the judge, Morton J, declined to override her consent. It is not useful authority in relation to the question which this Court has to decide. In the matter of the B Trust [2020] JRC 011. Of more relevance is the recent decision of the Royal Court of Jersey in In the matter of the B Trust. In that case the protectors were carrying out a fiduciary role in relation to the consent provisions for the appointment of capital. It was held by the Court (at [12] to [13] of the judgment) that, in circumstances where there was concern regarding disclosure of certain confidential information to the protectors, with which they needed to be provided in order to “meet their fiduciary functions”, the Court was prepared to dispense with the requirement of the consent of the protectors to the proposed appointment of capital. That was on the basis that: “13. The Trustee provided information to us in connection with the proposed appointment of £50,000 to the Eighth Respondent which we found convincing. Had it been made available to the protectors, and it was not made available for the reasons we have indicated, we have no doubt that the protectors, acting in accordance with their fiduciary functions, would have given their approval to the relevant appointment of capital. In the circumstances, the Court is prepared on this occasion to dispense with the requirement under the Trust that the consent of the protectors be obtained simultaneously with or before the appointment of capital is made and to approve the Trustee’s proposed appointment as requested.” That seems to me to be consistent with the protectors in that case being regarded by the Court as having a defined fiduciary consent power, which they were bound to exercise objectively in the interests of the beneficiaries, without, as the Court made clear, havingPage 61 of 83

regard to the views of the First Respondent who appointed them. That view appears to me to be consistent with the Narrow Review Role. It is certainly inconsistent with the notion that protectors have a wider discretion not to consent to the proper exercise by trustees of their power of appointment, provided that the former can justify their refusal of consent on the basis that it was a decision that was reached on rational grounds and consistent with the Protectors’ own fiduciary duties, albeit in disagreement with the decision reached by the Trustees. Piedmont

127. 128. The principal authority upon which counsel for the Appellants (and indeed the Protectors) relied to support their construction of the Wider Discretionary Role was the judgment of the Royal Court of Jersey23 Piedmont, which appeared shortly after Kawaley AJ’s decision in the present case. Piedmont was a case which raised the same issue as that before this Court. As Mr Green pointed out, it is apparent from [87] to [95] of the judgment that only limited argument was received in relation to the issue - indeed no arguments in favour of the Narrower Role are actually recorded in the main body of the judgment, or addressed; it is further apparent from the postscript at [112] to [120] that no further argument was received by the Royal Court in relation to the critical point, after the circulation of the Royal Court’s decision in draft (which contained the latter’s ruling on the point) to the parties and the subsequent provision to the Court of Kawaley AJ’s judgment ([115]). I quote the relevant paragraphs of Sir Michael’s judgment as follows: “(iii) The role of protector

87. On behalf of the Adult Grandchildren, Advocate Renouf submitted that the duty of a protector with a power of veto is to ask whether the decision of the trustees to which he is being asked to consent is one which a trustee could reasonably arrive at, whether or not it is a decision the protector himself would have made; and that accordingly the role of a protector is the same as that of the Court in a blessing application and is only concerned with the rationality of the trustees’ decision. If the trustees have reached a decision which a reasonable body of trustees could have arrived at, have taken account of relevant considerations and ignored irrelevant considerations, that is the end of the matter; the protector must consent. This submission was opposed by Advocate Christie who submitted that the Protector must reach its own decision in good faith in the interests of the beneficiaries. It was not confined to 23 The Court comprised Sir Michael Birt, Commissioner (and former Bailiff of Jersey), and Jurats Ramsden and Olsen.Page 62 of 83

assessing the rationality or lawfulness of a proposed decision on the part of the Trustees.

88. No assistance is to be derived from any provision in the trust deed of either Trust and there is scant judicial authority on the nature of a protector’s duties, particularly in the context of a requirement for a protector’s consent. This is probably because widespread use of protectors in trusts is a comparatively recent development and also because the role of a protector varies so much, depending on the nature and extent of the powers conferred by the trust deed, with the consequence that it is difficult to develop general principles which are applicable to all protectors or to all decisions of a protector.

89. However, Advocate Renouf was unable to point to any authority which supported his submission and we have no hesitation in rejecting it. In our judgment, as Page, Commissioner said in the passage from A & B Trust quoted at para 63(iii) above, the paramount duty of a protector is to act in good faith in the best interests of the beneficiaries. In pursuance of this duty, as in the case of trustees, he must have regard to relevant considerations, ignore irrelevant considerations and make a decision which a reasonable protector could arrive at; but he must reach his own decision. To like effect is the observation of Acting Deemster Smith on appeal in the Isle of Man in Rawcliffe v Steele 1993-95 MLR 426 at 529; [and he then cited the passage which I have already quoted above]….

90. One of the reasons that the Court exercises a limited review function on a blessing application is that, as described in S v L E & Bedell Cristin Trustees Limited [2005] JRC 109 at [22], a settlor does not choose the Court as a trustee; he chooses his appointed trustee. It is that trustee upon whom the various discretions conferred by the trust deed have been conferred. If the Court were to exercise a wide ranging role on such applications and decide the matter entirely for itself, the effect would be to constitute the Court as a trustee. That is not the Court’s role. The Court’s role is a supervisory one and it is simply to ensure that decisions taken by trustees are reasonable and lawful. Accordingly the Court does not simply substitute its own discretion for that of the trustee.

91. These considerations do not apply to a protector. The settlor has decided that a protector (often himself or a longstanding friend or adviser whose judgment he trusts) should be appointed pursuant to Page 63 of 83

the trust deed and has specified those matters where the protector’s consent is required. The settlor must be taken in those circumstances to have intended that the protector should exercise his own judgment in exercising those powers; otherwise why bother to go to the trouble of appointing a trusted friend or adviser (or himself) as protector rather than someone with a legal qualification to judge issues of rationality. Furthermore, if the role of a protector was simply to review the trustee’s decision in the same way that the Court would do, his role would be almost redundant; he would bring nothing to the table that the Court itself would not bring on a blessing application. It follows that, depending on the circumstances, a protector may well be entitled to veto a decision of a trustee which is rational, in the sense that the Court would bless it.

92. However, in the context of a power to consent, as in this case, a protector’s discretion lies within a narrower compass than that of a trustee. He is not the trustee. It is for the trustee to make a decision in the first place as to distributions or in relation to the exercise of any other discretionary power conferred on the trustee. It is emphatically not the duty of the protector to take that decision himself or to force the trustee into making the decision which the protector would make if he were the trustee by stating that he will only consent to a particular decision. That would be to exceed his proper role and to use the power given to him otherwise than for its intended purpose. Such conduct would also almost certainly not be in the interests of the beneficiaries and would be likely to lead to deadlock requiring the intervention of the Court. A protector may often find that he should consent to a discretionary decision of a trustee on the basis that it is for the benefit of one or more of the beneficiaries even though, if he had been the trustee, he might have made a different decision which he thought to be even more beneficial.

93. In this connection, it is to be expected and indeed encouraged for there to be full and open discussion between trustee and protector, with a view to finding something upon which they can both agree. We see nothing wrong with the sort of discussions which took place between the Protector and the Trustees in this case. A protector is not confined to a simple yes or no to a request for consent. A protector and a trustee should work together in the interests of the beneficiaries. It is therefore perfectly reasonable for a protector to explain his concerns about a particular proposal by a trustee and the trustee may often be willing to modify his proposalPage 64 of 83

to take account of these concerns or the protector may be satisfied after the trustee has explained his thinking.

94. Advocate Renouf submits that the email from the Protector of 15th September 2020 exceeded the proper role for a protector in that it indicated a specific proposal which the Protector would consent to. If the email had indicated expressly or by necessary implication that the Protector would only consent to the suggested distributions which it was putting forward in the email, we agree that that would have exceeded the Protector’s powers for the reasons we have just given. It is not for a protector to dictate to a trustee how the trustee must exercise his powers. However, we do not consider that the Protector was saying that in this case. The passage we have quoted at para 75 above makes it clear that this was merely a suggestion and that the Protector realised that it was for the Trustees to decide what distributions should be made. We see the email as part of a perfectly normal exchange of views between a protector and a trustee with a view to the good administration of the relevant trust.

95. In summary, we reject the criticisms of the conduct of the Protector in this case and we also reject the criticism of the Trustees for deciding to reconsider the November 2019 Proposal and to modify it. It was clear that the Protector was not willing to consent to the November 2019 Proposal and it was therefore perfectly proper for the Trustees to revisit that proposal and decide whether to maintain it or whether to modify it to take account of the Protector’s concerns. The fact that the Protector consented to the January 2021 Proposal even though it was very different from the suggestion which the Protector had put forward in its email of 15 September 2020, shows that the Protector was not seeking to dictate the only form of distribution to which it would consent. Consideration of the blessing application

96. Because of the adjournment of the proceedings as described at paras 34 – 36 above, the decision making process by the Trustees and the parties’ submissions to this Court fall into two parts. First, there is the decision on 8th February 2021 to make the Proposed Distributions and the parties’ submissions in relation to that decision; secondly, there is the decision of the Trustees at the June meeting not to alter their original decision following receipt of the Page 65 of 83

UK tax advice and the parties’ supplemental submissions to this Court on that aspect.

97. For convenience, we shall follow the same pattern in this judgment, but it is of course the case that the ultimate question is whether, at the end of the day, having regard to all the circumstances including the UK tax position, the decision of the Trustees to terminate the Trusts by making the Proposed Distributions is a decision which a reasonable trustee properly instructed could arrive at. …… Postscript

112. At the time of the hearing before us, counsel were not aware of any authority on the nature of a protector’s role when deciding whether or not to consent to a decision by a trustee. A draft of this judgment was circulated to the parties on 28th September for comment in the usual way. In response, Advocate Christie alerted the Court to the existence of a judgment of Kawaley J in the Supreme Court of Bermuda dated 7th September 2021 in the case of Re The X Trusts [2021] SC (Bva) 72 Civ. The issue before Kawaley J appears to have been the same as was raised in these proceedings. The question therefore was whether, in exercising their powers to consent to the exercise of powers vested in the trustees, the protectors were to exercise an independent discretion as to whether or not to give consent, taking into account relevant considerations and disregarding irrelevant considerations so that the protectors might withhold their consent to a proposed exercise of power by the trustees even if the proposed exercise of power was one which a reasonable body of properly informed trustees was entitled to decide upon (“the Wider View”), or whether the correct approach for the protectors was simply to satisfy themselves that the proposed exercise of power by the trustees was an exercise which a reasonable body of properly informed trustees was entitled to undertake and, if so satisfied, to consent (“the Narrower View”). In summary, did the consent provisions in the trust deed confer an independent decision-making discretion on the protectors (the Wider View) or merely a discretion to ensure that the trustees’ substantive decision was a valid and rational one (the Narrower View)? Page 66 of 83

113. At the conclusion of a thorough and detailed review of the position, Kawaley J came to the conclusion that the Narrower View was to be preferred. It is not possible to do justice to his full reasoning and, in any event, it is not necessary for the purposes of this postscript. Suffice to say that he helpfully summarised his reasons at [113] – [117] which we in turn would summarise as follows: (i) The Narrower View reflected the true construction of the consent powers conferred on the protectors because it was clear that the dominant purpose of those terms was to ensure the due exercise of the powers vested in the trustees. Unless there is something to contrary effect in the trust deed, the usual role of a protector is not to exercise a power jointly with the trustee; the protector’s role is to be a ‘watchdog’ to ensure due execution by the trustee of the powers vested in the trustee. (ii) The drafting of the trust deed clearly distinguished between powers expressly vested in the trustees, powers expressly vested in the protectors and powers expressly vested in the trustees subject to protector consent. Whilst on a literal reading of the wording of the consent powers, a power of veto was imposed, the provisions had to be construed in the wider context of the trust deeds. (iii) A contextual reading of the consent provisions suggested that the consent powers were not intended to be exercised jointly with, or entirely independently from, the powers conferred on the trustees subject to protector consent. There was no explicit wording used to signify an absolute discretion but, more importantly, the powers were vested in the trustees, albeit subject to protector consent. As Evans JA had said in Re Information About a Trust [2014] Bda LR 5, the normal function of the protector consent clause was that it was to be regarded as an ancillary power rather than a power exercised jointly with the trustee. (iv) Although the only case in which the issue had been specifically considered, namely PTNZ v AS [2020] WTLR 1423, had preferred the Wider View, this was a decision of a Master and Kawaley J was not persuaded by it, not least because the point had not been fully argued before the Master and the authorities on protector powers placed before Bermudan court had not been considered by the Master. Page 67 of 83

(v) Kawaley J did not accept that the Narrower View resulted in the protector’s role being a fundamentally limited one. Ensuring that trustees properly and rationally exercise their powers was an important and substantial role.

114. Kawaley J also placed some weight at [77] upon the fact that there was no provision in the trust deeds for the X Trusts for an indemnity in favour of the protectors. He felt that this was a further pointer to the conclusion that their powers of consent were merely ancillary to, rather than of equal status to, the trustees’ relevant powers. In the case of the P Trust and the R Trust, there is in each case a provision granting an indemnity to the Protector, but apart from this, there is no material distinction that we can see between the protector consent provisions in the Y Trusts and those in the Trusts. Accordingly, we do not think we can properly say that the decision in Re Y Trusts can be distinguished because of a difference in the wording of the trust deeds.

115. None of the parties in this case sought to argue that the matter should be restored for further argument in the light of the Bermuda decision. Nevertheless, we have considered whether we should amend our view at paras 87–95 above in the light of the decision in Re The X Trusts, which is clearly inconsistent with that view.

116. We acknowledge that Kawaley J had the advantage of much more detailed argument on the point than occurred before us, but we nevertheless respectfully differ from Kawaley J and remain of the views expressed in the above paragraphs. Such views are, we believe, essentially consistent with the Wider View as formulated before Kawaley J. Our reasons remain those summarised earlier in this judgment, but we would take this opportunity of commenting briefly on the matters relied upon by Kawaley J as summarised at para 113 above: (i) As to (i)-(iii), we accept that the role of a protector is not to exercise a power ‘jointly’ with the Trustee. On the contrary, as we endeavoured to clarify at paragraph 92 above, the discretionary power to make a distribution lies with the trustee. The protector’s only function is to decide whether or not to consent to that decision by a trustee. It is a separate decision on the part of the protector, not a joint exercise of a power with the trustee. We do not see that a conclusion that a protector does not exercise a power jointly with Page 68 of 83

the trustee points towards the Narrower View rather than the Wider View. (ii) Nor are we convinced by the reasoning at (ii) and (iii). We do not think that the observation of Evans JA in Re Information About A Trust can bear the weight which Kawaley J placed upon it. We found more convincing the submission of Mr Taube QC in the X Trust case as to the correct interpretation of a protector consent clause (in the terms in which it appeared in that case and in the present case), which submissions were recorded at [42] of the judgment of Kawaley J in the following terms: “93. The Powers of Veto in the X Trusts state the Trustees’ specified powers may not be exercised ‘without obtaining the prior written consent of the Protectorate’.

94. In this context the natural and ordinary meaning of the word ‘consent’ is agreement or permission: see the Oxford English Dictionary.

95. The reference in the Powers of Veto to the Protector’s ‘consent’ – to its agreement or permission – indicates the Protector has a choice whether to consent to the Trustees’ proposed exercise of the specified powers.

96. It follows, as a matter of ordinary language, that the Protector has a discretion in the matter whether to choose to consent.

97. By contrast, the reference to the Protector’s ‘consent’ is not appropriate to describe the [Narrower View]. The [Narrower View] involves not a discretion but an adjudication whether circumstances exist.

98. The [Narrower View] is described in para 1(b) of the Protector Summons, where the court is asked whether the role of the Protectors is ‘to satisfy themselves that the proposed exercise of a power by the Trustees of the X Trusts (or any of them) is an exercise which a reasonable body of properly informed trustees is entitled to take and, if so satisfied, to consent to the same’.

99. This description of the [Narrower View] echoes the test applied by the court in a Public Trustee v Cooper Category 2 case. In suchPage 69 of 83

a case the court determines whether the proposed exercise of the trustee’s power is rational: is the trustee’s decision one which a reasonable body of properly informed trustees is entitled to take? The court adjudicates the point. It is not a matter of the court’s discretion.

100. As a matter of ordinary language, these factors show the requirement for the ‘consent’ of the Protectors in the Powers of Veto is not intended to allocate to the Protectors the [Narrower View]. (iii) As to (iv), we have not seen the judgment in PTNZ and are certainly willing to accept that the Master did not have the advantage of the detailed arguments put before Kawaley J. We accept that no great weight can be placed upon the decision, but it remains of interest that the only known decision (prior to the X Trust and the present case) dealing with this issue has adopted the Wider View. (iv) As to (v), differing from Kawaley J, we would categorise the role of a protector, if the Narrower View is adopted, as being a fundamentally limited one. The protector will simply be fulfilling the same role as the Court. Accordingly, provided the trustee’s decision is a rational one and has not relied on irrelevant considerations or ignored relevant considerations, the protector is helpless, regardless of how wrong he thinks the trustee’s decision to be in terms of the interests of the beneficiaries.

117. The last point has particular force in the context of offshore trusts where the use of a protector is most common. As mentioned in Re X Trust, it is frequently the case that a settlor is recommended to a particular trustee company by his advisers but has no personal knowledge of the trustee company or its officers. Not unnaturally therefore, he will often wish to impose some check on the exercise of the trustee’s powers and to do this by appointing himself or a trusted friend or adviser as protector. To take a common example, he may well have views about how much money should be given to comparatively young children or grandchildren and does not wish to give them too much too early. A decision by trustees to appoint a comparatively large sum (perhaps at the request of a beneficiary) is unlikely to be categorised as irrational but this is just the sort of situation where a settlor would no doubt intend that a protector should be able to see that the trust is administered in accordance with his (the settlor’s) wishes by refusing consent. One can think of Page 70 of 83

many other examples. It seems inherently unlikely that settlors would go to the trouble of appointing themselves or trusted friends or advisors as protectors if they intended the role of protector to be limited to that of assessing rationality. If that were the case, the key requirement for a protector would be a legal qualification rather than knowledge of the settlor’s wishes and sound judgment as to what is in the best interests of particular beneficiaries.

118. We acknowledge that the approach we favour carries with it a greater risk of deadlock between trustee and protector if a protector refuses consent. Clearly, if a trustee considers that a protector’s refusal to consent is irrational or otherwise legally flawed, he may have recourse to the Court to overturn the protector’s veto. However, there is the potential for deadlock where the trustee and the protector both reach rational but opposing decisions. In our judgment, this is a natural consequence of the settlor’s decision to introduce the office of protector into the trust deed. A settlor must be taken to have intended (by imposing a requirement for consent) that a trustee should not be able to make certain decisions unless the protector consents. If consent is refused, the trustee’s decision cannot be put into effect. In most cases this is likely to lead to further discussion between trustee and protector in the hope of finding a sensible outcome. In the event of complete deadlock where such deadlock is causing real damage to the interests of the beneficiaries, we leave open the possibility of recourse to the Court. The Court has power to break a deadlock where this is caused by lack of agreement among trustees where they have to act unanimously (see Garnham v PC [2012] (1) JLR 204, approved by the Supreme Court in Children’s Investment Fund Foundation (UK) v Attorney General [2020] 3 WLR 461 at [219]. It may be arguable that the Court has a similar jurisdiction in the event of damaging deadlock between a trustee and a protector. However, we say no more about that. We have not heard any argument on the point and it does not arise in this case.

119. For these reasons, we think there is no reason to cut down the ordinary and natural meaning of a protector consent provision in the form in which it appears in this case and in the X Trust so as to read the word ‘consent’ as being limited to an assessment of rationality.

120. In summary, for the reasons set out earlier in this judgment and the supplemental reasons, we would respectively differ from the Page 71 of 83

decision in the X Trusts in so far as the role of a protector is concerned.”

129. 130. The decision in Piedmont is not binding on this Court, although, as a decision of a judge who is extremely experienced in dealing with trust matters, it is obviously worthy of appropriate consideration. In my judgment, however, the decision is not correct. My reasons may be stated as follows: (i) (ii) (iii) (iv) It appears that the Royal Court received only limited argument in relation to the issue; see [87] to [95] of the judgment. Indeed, no arguments in favour of the Narrow Review Role are recorded or addressed in the main body of the judgment. It is also clear from the postscript at [112] to [120] that no further argument was received by the Court – which had already indicated its decision on the point in its draft judgment which had been circulated - after Kawaley AJ’s judgment was sent through to it for its attention ([115]). Accordingly, the Royal Court did not have the benefit of the full argument which we have received in this case. The Jersey Court’s first point at [88] was that “No assistance is to be derived from any provision in the trust deed of either Trust”. As I have explained above, in my judgment assistance can be derived from the scheme and machinery of the Protector Provisions and certain of the relevant provisions in the present case. Nor do I agree with the assertion in the same paragraph that the “widespread use of protectors in trusts is a comparatively recent development”; it was common ground in the present case that such provisions have been a feature in offshore trusts for 40 to 50 years. It is correct, as Sir Michael states in paragraph [88], that, depending on the nature and extent of the powers conferred by the trust deed, the role of a protector may be very different under different settlements. Nonetheless, it is possible in my view to identify standard positive powers such as the power to appoint and remove trustees, and consent provisions in relation to the exercise of important powers conferred on the trustees under a trust (e.g. the power to appoint capital or a power to deal with key investments). And from this it is possible for the court to discern, or develop, general principles which are likely to be applicable to protectors, or to decisions of a protector, operating under the terms of such standard powers. As Mr Green submitted, whilst the literal wording of the powers conferred on protectors may vary (just as it may vary in the case of trustees), there is continuity in the core nature of what is being provided for (just as in the case of trustees). Sir Michael’s reasons for preferring the Wide Review Role are set out in paragraph [91] of his judgment. His first assumption is that a protector will often be the settlor himself or a long-standing friend. That is not necessarily the case, as indeed the present casePage 72 of 83

131. 132. 133. demonstrates, and cannot of itself predicate the answer to the issue which we have to determine. What can be assumed in relation to the role of a fiduciary protector, whether a long-term friend of the family or not, is that protectors are usually in a position to communicate information about the particular circumstances of a beneficiary or the settlor’s wishes to offshore trustees. The first reason given by Sir Michael at [91] for concluding that the Wide Review Role is the correct one is that, if the fiduciary protector’s role were indeed the Narrow Review Role, then there would be little point in having the settlor or a trusted friend as protector, rather than a person with a legal qualification to judge the rationality of the trustees. I do not find that argument logically compelling. The answer is that there is a need to have someone as protector who is in a position to bring to the administration of the trust the communication of relevant wishes and circumstances, in order to ensure that the trustees properly take account of such matters in reaching their decision. I agree with Mr Green that to argue that such a person might as well just be a lawyer misses the point. Similarly, trustees have to take account of relevant considerations and conduct their deliberations with due process, but that does not mean that they have to be lawyers. Inevitably, if they need legal advice, they (and indeed frequently protectors) will be advised by lawyers, as was the position here. Moreover, on the Appellants’ case, since the Wide Review Role includes the Narrow Review Role, the Protectors would in any event, even on the hypothesis that the Wide Review Role were the correct construction, be obliged to evaluate whether any proposal put forward by the Trustees was a proposal which “a reasonable body of properly informed trustees was entitled to decide upon”. So, in my view, the point made by the Royal Court is of no assistance in choosing between the Wide Review Role and the Narrow Review Role. Moreover, nor can I agree with the second reason given by Sir Michael at [91] to support his construction that the Wide Review Role is the correct one. He said: “if the role of a protector was simply to review the trustee’s decision in the same way that the court would do, his role would be almost redundant; he would bring nothing to the table that the court itself would not bring on a blessing application” In my judgment that is an inaccurate description of the role of the Protectors; it fails to include any reference to (i) the special role of the Protectors in bringing relevant information relating to (for example) the circumstances of individual beneficiaries or the wishes of the settlor, to the notice of the Trustees; and (ii) and what Mr Green descriptively referred to as the “control mechanism for the real-time assurance of proper administration of a trust” which “reflects the safeguarding function that the investment of a consent power – rather than a joint power – in the separately constituted fiduciary office-holder entails”. Even within the confines of the Narrow Review Role, the Protectors exercise undoubted practical control over the Trustees’ discretion, through the leverage of the requirement for Page 73 of 83

134. the former’s consent. That is a significant benefit to the administration of the trust as it might well obviate the need for the Trustees to obtain the assistance of the Court through a Public Trustee v Cooper24 category 2 confirmation that it has reached a lawful and reasonable conclusion in the exercise of its discretion, as to which I refer further below. Nor do I consider that the reasoning by the Royal Court in the later paragraphs of its judgment support the conclusion that the Wider Role is to be preferred. For example, at [92], the Court made the point that the protector’s discretion operates within a “narrower compass” to that of the trustee. That is certainly so, but it is hard to understand why that can be said to be consistent with the Wide Review Role; on the contrary, the recognition of the limited role afforded to the protectors might be said to be more in keeping with the Narrow Review Role. Nor am I persuaded by the Royal Court’s apparent concerns about deadlock, or why they suggest a construction that favours the Wide Review Role. (I deal with the issue of deadlock in the next section of this judgment where I deal with practical considerations of the alternate constructions.) Finally, I comment that the final sentence of [92]25 is hardly a description of the exercise of an independent discretion in exercise of the Wide Review Role; indeed, it appears to me to be wholly inconsistent with the conclusion that the Wide Review Role is the correct construction. Under the Wide Review Role, the protector should have withheld his consent to (on this hypothesis) the rationally made decision of the trustees in circumstances where the protector thought his decision would have been even more beneficial.

135. Nor was I persuaded by the Royal Court’s reasoning in paragraph [118] about reaching a “sensible outcome”. Of course, in some cases, discussion may achieve just that. But in cases where there is real disagreement, there is a requirement to know with certainty what is the criterion which justifies the refusal of the protectors to provide their consent. If it is, as the Royal Court appeared to accept, for the trustees to make a decision in the first place, and, if there is no question of the exercise of a joint power, then there has to be a clear red line which demarcates in respect of which decisions the protectors can legitimately withhold consent - and, on the other hand, those refusals of consent by the protector which the trustees can legitimately challenge in the courts. I deal with this issue of resolving deadlock further below, but the simple point is that it is highly unsatisfactory if the decision is to be left at large, effectively to be decided by the Court on the basis of evidence. Mr Green summarised the point well at paragraph 128 of his written submissions in his critique of the Royal Court’s decision in Piedmont: “the attempt, to soften the ambit of the Wider Role in [92] (without any explanation as to when a protector will be constrained to consent to trustee proposals as to the exercise of the power) 24 Public Trustee v Cooper [2001] WTLR 901. See further below. 25 “A protector may often find that he should consent to a discretionary decision of a trustee on the basis that it is for the benefit of one or more of the beneficiaries, even though, if he had been the trustee, he might have made a different decision which he thought to be even more beneficial.”Page 74 of 83

136. 137. 138. 139. completely falls away if what is then being said is that the true ambit of the Wider Role extends to the taking up of a negotiating position to bring the trustee closer to the protector’s preferred option – the protector holding the whip hand under the Wider Role of simply withholding consent. And all of this notwithstanding that – it is said – the protector does not hold the power jointly with the trustee, and that the protector’s role is ancillary to that of the trustee (per Evans JA in Re Information About a Trust [2014] Bda LR 5 as quoted at [113(iii)] as apparently accepted by the Jersey Court).” Finally (although there are other points that can be made), I find the articulation by the Royal Court of what the Wider Role is meant to comprise, as somewhat puzzling. On the one hand the Court attempts to narrow the role, but on the other hand it seems to envisage a degree of subordination of the Protector’s consent powers to the discretion of the trustees. For the above reasons, I am not persuaded that this court should follow the Royal Court’s decision in Piedmont or its reasoning. Other consent power cases We were also referred to other consent power cases26 where, for example, beneficiaries had consent powers conferred on them in relation to powers whose exercise (such as trustees’ powers of advancement or to override a life interest) would override their beneficial interests. I found no useful analogy in the line of cases which addressed the position of this category, where the person with the consent power held his powers beneficially and not in a fiduciary capacity. The consent power in such cases is held by the beneficiary so that he can decide whether to agree to his interest in the estate being impaired, diminished or taken away altogether. It is obvious as a matter of common sense that he can exercise his power to withhold consent exclusively in his own interest. Second, Mr Taube also referred to cases where a power of appointment is conferred on A but subject to the consent of trustees. He submitted that, in such cases, the trustees were not obliged to consent to A’s exercise of his power simply because such exercise was proper and rational. Instead, Mr Taube submitted that the court had held that the trustees have a wide independent discretion to give or withhold consent. In this context he relied principally on Re Dilke [1921] Ch. 34 and Commissioner of Estate and Succession Duties (Barbados) v Bowring [1962] AC 171. He submitted that that those two cases provided guidance in relation to consent provisions, namely that where one has a provision saying that a trustee, or anyone else, has a power to appoint not to be exercised without obtaining 26 For example, In re Cooper, Cooper v Slight (1884) 27 Ch D 565; In re Forster’s Settlement [1942] Ch 199; C v C (Ancillary relief: trust fund) [2009] EWHC 1491 (Fam). Page 75 of 83

the prior written consent of X, or a fiduciary, or whatever, if nothing else was said, then as a matter of construction it was an unfettered discretion.27

140. 141. Again, I do not find these cases of assistance in determining the issue in this case. Re Dilke was a case where a tenant for life, Sir Charles Dilke, was given the power to appoint capital under a settlement with the consent of the trustees, and the question in issue was whether the tenant for life’s power to appoint was a general or a special power. As articulated by Warrington LJ at page 42 the issue was: “Now on the one hand it is said that this is a power which can be exercised by the testator in favour of such person or persons as he pleases although for the validity of its exercise the consent of the trustees and their concurrence in the deed are required. On the other hand it is said that it is a special power exercisable only in favour of such persons or classes of persons as shall be approved by the trustees and ascertained or ascertainable by name or description at the date of the exercise of the power.” It was held by Peterson J, and confirmed by the Court of Appeal, that it was a general power which simply required trustee consent. It is interesting to note that the argument28 for the unsuccessful party that there was a “positive duty imposed on the trustees to exercise a discretion in seeing that the testator appoints to such persons as they approve of” was rejected. So that case hardly suggests that the trustees had the Wider Role; on the contrary, the case seems to me to support, albeit weakly, the Narrower Role. The Court of Appeal expressly approved the statement by Peterson J as follows29: “On the whole, I am of opinion that on the true construction of this clause the trustees were not required to approve of the persons who are to benefit under the exercise of the power of appointment or of the extent to which they are to benefit, but that the exercise of the general power is conditional upon their consent.” [Emphasis supplied.] But there was very little, if any, discussion of the circumstances in which the trustees could refuse their consent on the assumption that the power was a general one. As Peterson J pointed out at first instance30, and as the subsequent deed of appointment itself showed, Sir Charles, as at the date of the settlement, was not of sound mind, and so it was argued that the real intention of the trustee consent provision was that the question whether Sir Charles 27 See Day 3, page 221. 28 See per Hughes KC at 38. 29 See per Sterndale MR at page 41. 30 See page 36. Page 76 of 83

142. was competent to exercise the power of appointment should be considered by the trustees, and that their consent to the execution of the deed of appointment, confirmed by their concurrence in the deed, should be obtained before it could be contended that the power had been exercised. But, in fact, three of the four relevant trustees had already consented before the case started, so no issue arose as to the circumstances in which they could refuse consent. As such, the case seems to me to afford little assistance. In so far as it does so, it suggests that the mere addition of a third-party consent provision does not impose a wide discretion on the party required to give consent. Likewise, Commissioner of Estate and Succession Duties (Barbados) v Bowring furnishes little assistance for the purposes of the issue we have to decide. That was a Privy Council case involving the issue as to whether, as a matter of Barbados estate duty law, a beneficiary was “competent to dispose” of property comprised in a deed of settlement governed by Massachusetts law, in circumstances where the effect of a provision reserving power to revoke a settlement with the consent of the trustees had to be determined under Massachusetts law. As is clear, for example, from pages 180 – 181 of the judgment, and the citation of section 330 (1) of the American Restatement of the Law of Trusts, under Massachusetts law the extent to, or circumstances in, which the trustees were entitled to withhold consent depended on the terms of the trust deed. It was also common ground between the experts that, as a matter of Massachusetts law, under the terms of the relevant trust the trustees had a complete discretion to give or withhold their consent provided they acted honestly and from a proper motive. So I do not consider that the conclusion of the Privy Council, that no estate duty was payable because: “upon the true construction of section 3 (a) of the Barbados Estate and Succession Duties Act, 1941, a person cannot be said to have a general power making him competent to dispose of property within the meaning of that paragraph if the consent of the trustees is required to the exercise of that power, and that provision is so framed that the court will not control the trustees in the exercise of the power if they act honestly and do not act from an improper motive…..”

143. assists this court in determining what the circumstances are in which the Protectors in the present case (as opposed to trustees) can refuse to provide their consent. We were also referred by the Appellants to cases31 under the English Settled Land Act 1925, where the tenant for life was a trustee of powers exercisable by him over the estate. Again, these cases were of no assistance, since they involved a very different factual 31 Gilbey v Rush [1906] 1 Ch.11 which was concerned simply with the binary question of whether consent had been given or not; and Re Marshall’s Will Trusts [1945] Ch. 217 involving Settled Land Act trustees who had a fiduciary rather than a beneficial role. Page 77 of 83

144. 145. 146. 147. situation and trust regime under the Settled Land Act 1925 and did not address the scope of a protector’s role under an offshore trust involving very different species of trust property. The respective practical implications of the Narrow Review Role and the Wide Review Role; including the fact that the available methods of resolution provided by the court for any impasse between a protector and trustees in relation to the absence of consent and in relation to the approval of the exercise of trustees’ powers generally, or the surrender of such powers, support the Narrow Review Role. Both Mr Green for A Family and Mrs Talbot Rice for the Trustees made cogent submissions in support of the argument that practical considerations relating to the process of decision-making and resolution of disagreements between the Trustees and Protectors supported the Narrow Review Role. In paragraphs 99 and 100 of his judgment the judge concluded that the practical implications of the competing constructions favoured the narrow rather than the wider view of the Protectors’ consent powers. The A Family Respondents’ Notice of Appeal put forward further detailed grounds in paragraphs 1.2 and 2 to support the judge’s conclusion in this respect. Mr Taube, on behalf of the B Family, on the other hand, submitted that, contrary to the findings of the judge, the risk of wasted time and money, and of the duplication of work could not be a reason to prefer Narrow Review Role. Both the Narrow and Wide Review Roles required the Protectors to take sufficient steps to satisfy themselves that the Trustees’ proposal was one to which they could consent. Expense and duplication were inherent to both, and the draftsmen of the consent provisions must have intended as much. As Mrs Talbot Rice submitted in her oral argument, so far as process is concerned, if the Protector’s role is indeed the Wide Review Role, then the process by which the Protector comes to its conclusion is a lengthy and sequential decision-making process. First, the trustee makes its decision, including taking such expert or professional advice as it thinks it needs, consulting with its beneficiaries and deliberating accordingly. Subsequently, if the Protector’s consent is required to be given to the particular decision, the Protector then has to go through a similar process of making its own enquiries, consulting with the settlor or the beneficiaries, taking its own independent advice, whether expert or professional, before coming to its own independent view on the matter which is the subject of the Trustee’s decision. All that process may have to take place against a time critical scenario, for example in relation to the voting of shares at a shareholders’ meeting. Now of course, as Mrs Talbot Rice accepted, in some circumstances, even in relation to the Narrow Review Role, it may be necessary for the Protectors to take their own professional advice before being satisfied that the Trustees had indeed reached a reasonable, proper decision taking into account all the relevant circumstances. But that, I accept, is a lesser task in the sense that, under the Narrow Review Role, the respective rolesPage 78 of 83

of the Protector and the Trustees are clearly defined. There is no blurring of roles. As the Family A Respondents put it in their Notice of Appeal32: “2.2 On the Narrower Role: (a) the operation of the two fiduciary roles is simple, clear, efficient and complementary. The trustees in deciding whether, and if so in what manner, to exercise their power and the protectors in determining whether to provide or withhold its consent are carrying out their respective functions within boundaries - each knowing where they stand as regards the common criterion to apply, being the criterion well established under trust law for the proper exercise by trustees of a power vested in them; …… 2.3 On the Wider Role:

148. (a) with two separate fiduciary office-holders independently and sequentially exercising separate fully discretionary decision- making as to whether an exercise of a substantive power vested in the trustees is in their respective views in the best interests of the beneficiaries, the roles of the trustees and the protectors are neither straightforward, efficient nor complementary- rather than facilitating the smooth administration of a trust it hard wires in potential duplication, delay, cost and uncertainty as to the trustees' ability to exercise substantive powers vested in them for the purposes of benefitting the beneficiaries, which would otherwise be in compliance with their trust law defined duties;” . So, in my judgment, the need to recognise the complementary roles of the Trustees and the Protectors in the decision-making process strongly supports the application of the Narrow Review Role.

149. 150. The second reason as to why the practical consequences of the two roles, in my judgment, support the Narrow Review Role relates to the resolution of disagreements between the Protectors on the one hand, and the Trustees on the other. Mr Green, on behalf of the Family A Respondents, accepted that jurisdiction would exist, and would be available, to break a deadlock between the Trustees and the Protectors, by 32 Paragraph 2.2 and 2.3. Page 79 of 83

151. 152. means of an application to the court by the Trustees under a category 3 type application pursuant to the categorisation by Hart J (following the decision of Robert Walker J in Chambers in 1995) in Public Trustee v Cooper [2001] WTLR 901 at pages 922 – 994. This would involve the Trustees surrendering their discretion to the court in relation to the decision, whether to appoint or otherwise dispose of shares in OpCo, or in relation to any administrative decision as to how, for example, to vote the shares. Mr Green pointed out that this was a jurisdiction much less frequently resorted to in practice and that the courts were not enthusiastic about being asked to exercise such power where trustees were able to form a view as to how power should be exercised themselves. He submitted that, in accordance with the decision in CIFF, the court would, if it thought it appropriate, direct the Protectors to abide by the court’s decision that the Trustees’ decision should be upheld, notwithstanding the original refusal of consent by the Protectors to the Trustees’ decision - in other words, the court would be prepared to override refusal of consent on the part of the Protectors. But, Mr Green submitted, such a process was cumbersome and not welcomed by the court. On the other hand, Mr Taube, on behalf of the Appellants, argued that the Protectors’ role was to create an impasse where the Trustees have proposed a course of action that the Protectors do not believe conforms to the settlors’ wishes. Once one accepted that the Protectors indeed had an independent discretion as to whether (for example) an appointment should be made and were in position precisely to block such proposals by the Trustees, then the concept of impasse no longer presented a problem for the operation of the X Trusts. The draftsmen of the consent provisions had given the Protectors a veto power to force the Trustees to listen to the Protectors and devise a mutually acceptable course of action. Accordingly, as Mr Taube submitted33 , even if the Trustees surrendered their discretion to the court in a deadlock case, the Court should not be capable of overriding a refusal of consent on the part of the Protectors. In other words, the Protectors could override the Trustees’ otherwise perfectly proper exercise of discretion, but, absent an abuse of power by the Protectors, not even the court could generally override the Protectors, if the Protectors choose to override the Trustees. The Protectors also made submissions on the practicability of the Wide Review Role. Mr Robinson submitted that, on the Wide Review Role, the Protectors could restrict and expand the scope of their inquiries in accordance with the significance of the Trustees’ proposed course of action. For less significant proposals, the Protectors could limit their inquiries to whether the proposal was reasonable, and the reasoning was unobjectionable; with more significant decisions, the Protectors might take into consideration not only what they would have done in the Trustees’ position, but, inter alia, the fact of the Trustees’ own decision, the nature and extent of the Trustees’ consideration of the proposal and their reasoning. He relied upon Edge v Pensions Ombudsman [2000] Ch 602 267D-268B and Pitt v Holt [2013] UKSC 26, [2013] 2 AC 108 for the proposition that the Protectors had 33 See paragraphs 116 – 118 of the Appellants' skeleton argument.Page 80 of 83

153. 154. 155. 34 35 sufficient autonomy in executing their duties as long as they did not act perversely, fail to take into account relevant considerations, or take into account irrelevant ones. He submitted that such a flexible approach would allow the Protectors to avoid duplication and expense when considering less significant proposals. Mr Green criticised the Appellants’ and the Protectors’ reliance on the alleged ability of the Protectors to take a flexible approach under the Wide Review Role. He submitted that no principled basis had been given for adopting a flexible approach under the Wide Review Role. It would be difficult to envisage a situation where the Protectors, charged with the Wide Review Role, might justifiably reduce their responsibilities to the beneficiaries by restricting the scope of their inquiries into a proposal by the Trustees. Rather, Mr Green submitted that the flexible approach was a means of avoiding the impracticalities of the Wide Review Role by leaving open the possibility of adopting a de facto Narrow Review Role on an ad hoc basis where the simpler and more principled conclusion would be to adopt the Narrow Review Role itself. There was some debate between counsel about the ability of the court to avoid any impasse by means of an application by the Trustees: (i) in England under section 57 of the Trustee Act 1925; (but in England it is established that the jurisdiction under this section is exercisable only as regards the “management or administration” of the trust property and not so as to enable the variation of beneficial interests)34; (ii) in Jersey under Article 47 (3) of the Trusts (Jersey) Law 1984; and (iii) in Bermuda under section 47 of the Trustee Act 1975; (in Bermuda it has been held that the section is sufficient to enable its use to vary beneficial interests35). However, it does not seem to me that it is necessary, or indeed appropriate, for this court on this appeal to decide whether or not an application by means of any of the above routes would enable the court, on the application of the Trustees, to override a refusal of consent by the Protectors to a decision by the former. What is clear to me, is that the obstacles and Re Chapman’s Settled Estates [1953] Ch. 218 in holding 3 – the point was conceded on further appeal Chapman v Chapman [1954] AC 429 at p.465, which decision led to the enactment of the Variation of Trusts Act 1958 in England, which in its turn is effectively reproduced in Art 47(1) and (2) Trusts (Jersey) Law 1984 and s.48 Trustee Act 1975 (Bermuda). GH v KL (2011) SC (BDA) 23 Civ per Ground CJ (since applied on several occasions) stating that s47 is “an amalgam of two English provisions, being s57 of the Trustee Act 1925 and section 64 of the Settled Land Act 1925, one effect of which is to remove the limitation to administrative matters contained in the former. I have no doubt, and in any event it must be presumed, that this was deliberate on the part of the [Bermuda] legislature”. Page 81 of 83

156. 157. 158. uncertainties which would lie in the way of the Trustees attempting to set aside, or overcome, a refusal of the Protectors to consent to a decision which the Trustees considered manifestly in the best interests of the beneficiaries, strongly support the conclusion that the Narrow Review Role is the correct one. Problems about impasse do not arise when the criteria entitling the Protectors to refuse their consent are clear – namely a decision by the Trustees which the Protectors regard as not reasonable, tainted by improper process or which has failed to take into account relevant considerations. On the assumption that the Narrow Review Role applies, and accordingly the criteria are clear, any disagreement between the Trustees and the Protectors as to whether the latter were entitled to refuse consent, if not resolved by agreement, could be resolved by a simple application to the court under the Public Trustee v Cooper category 2 jurisdiction. And, contrary to Mr Taube’s submission that the Protectors’ right to refuse consent is absolute, the Supreme Court’s decision in CIFF certainly does not support the notion that a fiduciary, in the position of a protector and holding a consent power, has an absolute right of veto, even in circumstances where the court considers that the primary decision taker, such as a trustee, had reached a rational and proper decision to exercise a power of appointment. Other points In their extensive submissions, counsel raised many other arguments. Although I have carefully considered them, I do not find it necessary to address them in this judgment. The Court was presented with a raft of published commentaries in textbooks and elsewhere, and unpublished views expressed in counsel’s opinions and notes, in relation to the powers of protectors generally and, in particular, the ambit of their role in circumstances similar to the present. The views expressed in some cases supported the Narrow Review Role and, in others, the Wide Review Role. None was authoritative. Although interesting, they reflected the respective arguments on both sides in the present case. For that reason, I did not consider that there was any utility in referring to them. Disposition For the above reasons I would dismiss the appeal.

KAY, J.A:

159. I agree.

CLARKE, P:

160. I, also, agree with my Lady’s comprehensive judgment, and add but a footnote of mine own. Page 82 of 83

161. 162. 163. 164. 165. There are, in my view, four matters which are of critical importance in any determination as to whether the Wide or the Narrow View is to be adopted. The first is the use of the word “Protectorate” itself. A protector is someone who provides a safeguard to others; here the protection afforded is to protect the beneficiaries from a breach of trust by the Trustees. Whilst the adoption of the Wide View could be said to afford the beneficiaries an enhanced form of protection by requiring a separate decision by the Protectorate on whether it would (or would not) exercise the power of the Trustees in the same way, the concept of protection, in this context, seems to me, as a matter of ordinary language, more naturally applicable to some body which is to protect the beneficiaries by ensuring that the decisions of the Trustees are within the scope and ambit of their powers, rather than a body which is to make a freestanding decision of its own, without which an otherwise legitimate decision of the Trustees cannot take effect. The second, which is linked to the first, is that the role of the Protectorate is an important one, but one which is ancillary to that of the Trustees. It is for the Trustee to decide what appointments, distributions or payment from the Trust Funds to make. The Protectorate cannot make such dispositions itself. In those circumstances it seems to me inconsistent with the respective functions of the Trustees and the Protectorate that the latter should be required to make a separate decision of its own as to any appointment, disposition or payment; and thereby be enabled, in practice, to prevent the Trustees from implementing their own decision even though it is not irrational, does not take account of irrelevant and does takes account of all relevant considerations. The third is that although the words “without obtaining the prior written consent of the Protectorate” are not further qualified, the phraseology used begs (but does not answer) the question as to the basis upon which the Protectorate is entitled to withhold its consent. That falls to be answered in the light of (i) the description of its function by the use of the word “Protectorate”; and (ii) the true nature of that function, having regard to the functions of the Trustees and the scheme of the Trusts and the Protector Provisions within them. The function of the Protectorate is not that it shall be a body which is, in effect, albeit not in name, another trustee whose word is determinative even when the Trustees have reached what is otherwise a valid decision. The fourth is that I would accept, as was submitted, that the adoption of the Narrow View produces a result which means that the operation of the two fiduciary roles (of Trustees and Protectorate) is simple, clear, efficient and complementary. By contrast the Wide View involves the separate exercise of a fully discretionary function and is calculated to produce in some cases duplication, delay, cost and conflict which would not arise if the Narrow View was adopted; or, at the lowest, to prevent the Trustees from exercising the discretion given to them in the manner that they wish, even though in making their decision they had complied fully with their legal and equitable obligations. Page 83 of 83

1. 2. 3. 4. 5. 6. A P P E N D I X A

BACKGROUND TO THE PROTECTOR PROVISIONS

The following account is taken from the Appellants’ skeleton argument before this court. It is redacted in certain respects in order to preserve the anonymity of the family. Text which might be said to be presentation of the Appellants’ arguments, as opposed to an historical account of the background, has also been removed. The same definitions are used as those in the judgment. Since the late 1960s, from inception, many of the X Trusts had included protector provisions similar to those introduced by Operation Protector in 1994 and 1995; 9 trusts came into this category. Each of these 9 settlements contained similar requirements for the trustees to obtain “the prior written consent” of the “Protector” or the “Committee of Protectors” before exercising specified powers, in much the same way as in the present protector provisions, i.e. the Trustees’ powers to deal in or vote the X shares or to pay, appoint or distribute capital. In 1989 and 1990, the English resident trustees of many X Trusts were replaced by corporate trustees resident in Bermuda, so as to secure UK tax advantages for the beneficiaries. In consequence, their professional advisers considered the idea of introducing the protector provisions. Their concern arose from the potential risks to the UK resident beneficiaries of having trustees resident on the other side of the Atlantic who were not well known to them. This concern, it was said, was the motive for the introduction of the present protector provisions. The reorganisation known as “Operation Protector” between 1991 and 1995 led to the introduction of the present protector provisions in 1994 – 95. The contemporary documents show that, during Operation Protector, the beneficiaries and trustees of the X Trusts took advice from law firms in London, Jersey, Guernsey and Bermuda and from an eminent QC in the field. At this time protector provisions were commonly adopted in settlements with offshore trustees. The family’s and trustees’ advisers advised that protector provisions were “best practice”.1 They enabled the actions of the offshore trustees on important matters to be subject to the control of trusted advisers of the settlors and their families.2 There were three phases to this introduction of the protector provisions into the X Trusts. a. In May 1994, in Operation Protector, the trustees of 49 X Trusts exercised powers of appointment in each of the relevant settlements to introduce the present protector provisions. (These 49 X Trusts are “the Phase 1 Trusts”.) b. In October 1995, as a further part of Operation Protector, the trustees of 6 more X Trusts (“the Phase 2 Trusts”) (i) removed existing provisions in those settlements conferring powers on a protector or committee of protectors to give or withhold consent to the exercise of specified trustees’ powers and (ii) replaced them with new protector provisions in substantially the same form as those introduced into the Phase 1 Trusts. 1 See paragraph 1.3 of the “Operation Protector Bible”. 2 In the 1990s powers of veto were commonly conferred on protectors of offshore trusts: see e.g. Soares Non-Resident Trusts 4th edn (1993) pp. 7 – 8, Clarke’s Offshore Tax Planning 3rd edn (1994) pp. 17 - 18, and Underhill & Hayton’s Law of Trusts and Trustees 15th edn (1995) at pp. 23 – 25.

7. 8. 9. 10. In the matter of The X Trust

APPENDIX A

Page 2 of 5 c. In each of the 18 X Trusts settled between 1997 and 2003 (“the Modern Trusts”), the settlor included protector provisions in much the same form as those introduced by Operation Protector into the Phase 1 and 2 Trusts. In February 1991 the English firm of solicitors, who acted for the family and the Trustees (“Firm C”), produced a Discussion Paper about the idea of introducing protector provisions into the X Trusts; this document was before the court. In the context of a consideration of various tax issues, the document contained the following statement: “It may be proper to transfer [OpCo shares] to new settlements which require Protector consent to dealings in the voting shares in [OpCo] on the grounds that this accords with the family’s views about control of [OpCo].” In a Memorandum dated July 1991 Firm C again addressed this idea. Changes in 1991 to the UK’s CGT regime meant it was no longer feasible to transfer shares in OpCo to other settlements, so they considered instead introducing new protector provisions into each X Trust; and at paragraph 5 they stated: “[t]he Protector can play a part both in safeguarding the assets held by the Trustees and also the interests of the beneficiaries” . An employee of the family produced a Background Paper, dated 18 March 1993 and primarily addressed to the Trustees, where he said: “… bearing in mind the present capitalised market value of [OpCo] is now in excess of £[X], and having regard to the best interests of the potential beneficiaries of the various settlements, the consensus view of the family’s advisers (redacted…..) is that consideration be given to steps being taken, as soon as possible, in order to provide, within the settlement structures, an effective control mechanism (possibly protectorship structures)” (underlining supplied). He requested the Trustees: “to consider whether they feel they can accept the advice and recommendation of the above advisers, which has been endorsed in full by senior family members, namely [redacted]. If the Trustees accepted that request, then it was proposed that the Trustees should instruct “appropriate professional advisers to proceed with necessary detailed investigations”. On 25 March 1993 there was a board meeting of the Trustees. The Trustees resolved that: “Messrs. [redacted] be and are hereby requested to identify and analyze … the technicalities of introducing an effective additional layer of protection over the trust assets”. The Trustees later instructed Firm C.

11. 12. 13. 14. 15. 16. In the matter of The X Trust

APPENDIX A

Page 3 of 5 Firm C subsequently produced a document pack dated 31 August 1993 setting out the proposal (“the Operation Protector Bible”). At paragraph 1.3 it stated:- “The non-resident trustees have been considering for some time with the family whether the existing arrangements provide the beneficiaries of the settlements and important family assets with the appropriate protection, having regard to current “best practice” where trustees are resident outside the [UK].” The Operation Protector Bible focussed on two aspects of protector provisions: first, the power of the proposed protectors to remove and appoint trustees, with a view to achieving an “exit” from Bermuda if necessary; and secondly, the provisions conferring power on protectors to give or withhold consent to the trustees’ exercise of key powers. At paragraph 2.4 it explained: “The veto powers over capital and the control of specified securities would be designed to provide stability, continuity and coherence in long term planning for the benefit of the family as a whole in relation to primary assets (i.e. shareholdings in [OpCo]” . Appendix 3 to the Operation Protector Bible contained a “Memorandum of Legal Issues”. Paragraph 2 noted the Trustees acknowledged that the “best practice” for non-resident trusts was to have protector provisions whereby – “(b) a coherent overall approach can be taken to the management and control of shares in [OpCo] by requiring the consent of a protector before any disposition of shares can take place; (c) the consent of a protector is required before any distribution of trust capital occurs (or substantive capital appointments on new trusts are made) – again as part of the process of ensuring that a global view can be taken of the interests of the class of beneficiaries as a group, bearing in mind that many of them are beneficiaries (or potential beneficiaries) of a large number of settlements in different jurisdictions who have (or may have) private assets and family expectations – as well as seeking to secure stability and continuity in the total holdings of [OpCo] in the wider interests of all beneficiaries.” On 24 September 1993 Firm C instructed a very distinguished and experienced QC (Mr W) to advise on the proposals. The first consultation with Mr W QC took place on 30 September 1993 with Firm C and the employee of the family in attendance. The Note of the Consultation recorded that: “[t]he purpose of the consultation was to consider whether the trustees could effectively introduce a Protector mechanism into various family trusts having non- resident trustees” (paragraph 1); and it was explained that:

17. 18. 19. 20. 21. In the matter of The X Trust

APPENDIX A

Page 4 of 5 “[t]he trustees and the family regarded the Protector mechanism as being extremely important in order to provide stability and protection not only for the senior generation but also for succeeding generations. ”(paragraph 2). The second consultation with Mr W QC took place on 20 October 1993. Mr X’s son attended this consultation with Firm C, and others. A combined Note of the two consultations was produced which Mr W QC approved on 26 November 1993. It was recorded at paragraph 15 that: “It was emphasised that not only were the [X] family anxious to ensure that the Settlements still maintained the maximum degree of flexibility but that the introduction of the Protector mechanism should not in any way interfere with the family’s desire to achieve overall cohesiveness.” In a letter dated 21 October 1993 the outcome of the consultation was reported to another of the family’s advisers and to the Trustees. The letter noted at paragraph 7: “[Mr X’s son] also found the word which has somewhat eluded me to sum up the perceived benefits from having a Protector namely ‘cohesiveness’. Counsel could quite understand that Operation Protector would introduce a facility whereby family and asset cohesiveness in terms of policy and security could more readily be achieved.” A letter dated 7 January 1994 sent to the Trustees setting out progress on Operation Protector stated that both Mr X’s son and Mr X’s daughter had: “given their verbal confirmation that they would be very happy if the trustees in Bermuda and Jersey were to arrange to implement the introduction of protectorship provisions within the various overseas settlements” and that letters from each of Mr X’s son, daughter and son-in-law would be produced confirming their support for the introduction of the protector provisions. These letters were signed on 10 and 18 January 1994. Each letter was addressed to the Trustees and identical in all material respects:- “In my capacity as settlor of certain of [the X family settlements], I confirm that I have carefully considered the question of introducing protectorship arrangements into the management of those settlements and confirm that I am wholly in agreement with the introduction of such arrangements which will, I am sure, be for the benefit of the beneficiaries of the settlements and also add to the cohesiveness of the protection of the assets held in the settlements.” The Protector companies were incorporated in Jersey on 3 May 1994. The first meetings of the Protectors took place on 6 May 1994 The original directors of the Protectors included a solicitor (who had formerly been a partner at Firm C), a former director of a merchant bank, a senior Bermudan lawyer, and 3 accountants from a leading firm of chartered accountants. The evidence showed that these people were five trusted advisors of Mr X’s son, Mr X’s daughter and her husband, all of whom were settlors of certain of the trusts. It was noted in the Instructions to Mr W QC on 22 March 1994 that the role of the directors of the Protectors was:

22. 23. 24. In the matter of The X Trust

APPENDIX A

Page 5 of 5 “considered to be of great importance and the family are concerned to have as directors only persons whom they know well and in whom they have complete trust” . Similar views to the effect that the original directors were selected because they were “individuals whom my family and I knew and trusted” and were “the principal advisers to the [X]Family” were expressed by various beneficiaries, and Mr X’s daughter in their respective affidavits. The evidence also showed that those individuals were in a position to exercise a discretion based on their long experience of the beneficiaries, OpCo and the X Trusts. The reasons for introducing the protector provisions in relation to the Phase 2 Trusts, were the same as those in relation to the Phase 1 Trusts. The reasons were set out in a letter dated 25 October 1994 from the family’s and Trustees’ adviser, to one of the Trustees. The initial idea had been to supplement the existing protector provisions in the Phase 2 Trusts by adding extra provisions which had been introduced into the Phase 1 Trusts. At paragraph 5, the adviser pointed out the potential for confusion if there were two separate sets of protector provisions, and he noted that “one of the objectives of Operation Protector was to introduce coherence and consistency between the Settlements”. He therefore proposed the replacement of the existing protector provisions with the protector provisions introduced into the Phase 1 Trusts. The Modern Trusts were the 18 further X Trusts which were settled after Operation Protector was completed in 1995. They were settled between 4 April 1997 and 17 January 2003. Unlike the Phase 1 Trusts, the Modern Trusts included the protector provisions from the start. The drafting of the protector provisions in the Modern Trusts was either identical or very similar to those in Phase 1 Trusts, albeit the requirements for the consent of the protectors was more extensive. Certain of the beneficiaries and Mr X’s daughter stated in evidence that they could not recall any specific discussion of the protector provisions in relation to the settlement of the Modern Trusts, and that this was probably because the protector provisions followed the terms adopted in the remaining X Trusts.