The Judicial Committee of the Privy Council Decisions

Schmidt v Rosewood Trust Ltd (Isle of Man) [2003] UKPC 26 (27 March 2003)

ADVANCE COPY

Privy Council Appeal No. 2 of 2002

Vadim Schmidt Appellant

v.

Rosewood Trust Limited Respondent

FROM

THE STAFF OF GOVERNMENT DIVISION

OF THE ISLE OF MAN

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

JUDGMENT OF THE LORDS OF THE JUDICIAL

COMMITTEE OF THE PRIVY COUNCIL,

Delivered the 27th March 2003

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

Present at the hearing:-

Lord Nicholls of Birkenhead

Lord Hope of Craighead

Lord Hutton

Lord Hobhouse of Woodborough

Lord Walker of Gestingthorpe

[Delivered by Lord Walker of Gestingthorpe]

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

Introduction.

1. It has become common for wealthy individuals in many parts of the world (including countries which have no indigenous law of trusts) to place funds at their disposition into trusts (often with a network of underlying companies) regulated by the law of, and managed by trustees resident in, territories with which the settlor (who may be also a beneficiary) has no substantial connection. These territories (sometimes called tax havens) are chosen not for their geographical convenience (indeed face to face meetings between the settlor and his trustees are often very inconvenient) but because they are supposed to offer special advantages in terms of confidentiality and protection from fiscal demands (and, sometimes from problems under the insolvency laws, or laws restricting freedom of testamentary disposition, in the country of the settlor's domicile). The trusts and powers contained in a settlement established in such circumstances may give no reliable indication of who will in the event benefit from the settlement. Typically it will contain very wide discretions exercisable by the trustees (sometimes only with the consent of a so-called protector) in favour of a widely-defined class of beneficiaries. The exercise of those discretions may depend on the settlor's wishes as confidentially imparted to the trustees and the protector. As a further cloak against transparency, the identity of the true settlor or settlors may be concealed behind some corporate figurehead.

2. All these considerations may encourage a settlor to entrust substantial funds to an apparently secure and confidential off-shore shelter. But the very same features may, as this case strikingly illustrates, present problems to the close relatives of a settlor who dies unexpectedly, as did Mr Vitali Schmidt ("Mr Schmidt"), the co-settlor of two Isle of Man settlements called the Angora Trust and the Everest Trust.

3. The main issue on this appeal is the claim of Mr Schmidt's son Vadim, the appellant, to obtain trust accounts and other information from the trustees of the two settlements. The sole trustee of each settlement is Rosewood Trust Ltd ("Rosewood"), an Isle of Man company which is in business as a provider of corporate and trustee services. It is the respondent to the appeal. The appellant's claim for disclosure of accounts and other information has been made in two capacities, that is both personally and as the administrator of Mr Schmidt's estate. Mr Schmidt died intestate in Moscow on 31st August 1997 and letters of administration to his estate in the Isle of Man were granted to the appellant on 17th August 1998.

4. Before any summary of the trusts and powers of the settlements it is appropriate to say something about the background to this litigation, and the evidence which was before the lower courts. When Mr Schmidt died (according to the appellant's evidence, unexpectedly and alone at his Moscow apartment) the appellant was nineteen years of age (he was born on 28 September 1977). He was entitled to share in his father's estate together with his mother Svetlana and his paternal grandmother Alisa. He has deposed that he has devoted his time and resources to tracing the assets of his father's estate. His father was a senior executive director of Lukoil, which is the largest oil company in Russia and one of the largest oil companies in the world. The appellant believes that his efforts to trace his father's assets (in Liechtenstein, Austria, Cyprus and the Isle of Man) have been frustrated by some of his father's co-directors.

5. In June 1998 the appellant commenced proceedings ("the 1998 proceedings") in the Isle of Man against Rosewood, its directors and several other defendants. In the 1998 proceedings the appellant alleged breach of trust and breach of fiduciary duty. He obtained an ex parte order made on 10th July 1998 (and varied by consent on 16th July 1998) prohibiting Rosewood and other defendants from dealing with assets comprised directly or indirectly in the Angora Trust and the Everest Trust. The order also provided for extensive disclosure of information.

6. The appellant commenced the present proceedings by petition on 1st June 1999. His case, as set out in an affidavit sworn by him on 7th June 1999, is that the disclosure made to him pursuant to the ex parte order "raised more questions than it answered", especially as parts of some disclosed documents had been obliterated. It appears that over US$105m was received by the two settlements between their creation and 1998. The appellant deposed that sums totalling about US$14.6m had been paid to him (as his father's administrator) between August and October 1998. These sums were not accepted by him by way of compromise of the claims made in the 1998 proceedings, and he believed them to represent only a fraction of Mr Schmidt's total entitlement under the two settlements. The appellant has obtained a lengthy accountants' report dated 4th June 1999 from Ernst & Young setting out alleged deficiencies and inconsistencies in the material provided.

7. The appellant's present proceedings seek fuller disclosure of trust accounts and information about the trust assets, not by way of discovery in the 1998 proceedings but by virtue of the discretionary interests or expectations which the appellant claims that he himself has, and that his late father had during his life, under the two settlements. The present proceedings were initially attacked by Rosewood as being an abuse of process, but that submission was rejected at first instance and was not pursued on Rosewood's appeal to the Staff of Government Division.

8. That is only a brief summary of the background to the litigation. It has many features which might be thought to prompt further questions. However neither side has suggested that the settlements should be regarded as sham documents, or as documents entered into for illegal purposes. It is unnecessary to try to go further, at this stage, into the origins of the settlements. Indeed it would not be possible to go much further since the judgments below are very short on any findings of fact, and the Record before the Board does not contain all the affidavit evidence sworn in the two sets of proceedings. It appears that Rosewood does not accept that Mr Schmidt was a settlor of either settlement. Certainly he was not named as a settlor. But (in responses filed in the High Court of the Isle of Man) Rosewood has described Pacquerette Limited ("Pacquerette"), the named settlor, as "simply a nominee" and has stated that Mr Schmidt "was involved in the setting up" of each trust. Rosewood also stated in its answer that its involvement was "simply to receive and pay out such funds as [Mr Schmidt] chose to channel through the Isle of Man". In the absence of other evidence that justifies the conclusion that he was one of those who joined in causing the settlements to be made and funded, and was in substance a co-settlor.

   The terms of the settlements.

9. The Angora Trust was executed on 6th April 1992 between (1) Pacquerette and (2) Lorne House Trust Limited ("Lorne House"). The Everest Trust was executed on 20th June 1995 between Pacquerette and Lorne House. Lorne House (an Isle of Man company apparently controlled by the same persons as Rosewood) retired from the trusteeship of both settlements, and Rosewood was appointed as sole trustee in its place, on 3rd May 1997.

10. The Angora Trust and the Everest Trust are in similar but by no means identical form. The convenient course is to summarise the trusts and powers of the Angora Trust and then identify the points of difference in the Everest Trust.

11. Clause 1 of the Angora Trust contains definitions. Some are self-explanatory but the following should be noted:

    (1) "The Trust Period" and "the Accumulation Period" are both defined as the period of 80 years from the execution of the settlement (and that period is by clause 2 the perpetuity period permitted by the law of the Isle of Man) subject to powers for the trustees to terminate either period before the expiration of 80 years.

    (2) "The Beneficiaries" are defined as the Royal National Lifeboat Institution ("the RNLI") and the persons listed in the Second Schedule. That schedule (originally copied in a redacted form, but later without redaction) contains the names of Mr Schmidt and other Lukoil senior executives.

    (3) "The Protector" is defined as Mr Schmidt "or any other person holding the office of Protector hereunder".

    Clause 3 provides for the settlement to be regulated by the law of the Isle of Man, subject to a power for trustees to change the proper law.

12. Clauses 4, 5, 6 and 7 contain the dispositive trusts and powers affecting the trust fund. These present some difficulties of construction but their general effect is as follows:

    (1) Clause 4(1) contains a wide power of appointment exercisable by the trustees, with the prior written consent of the Protector, by deed executed during the Trust Period. The objects of the powers are all or any one or more of the Beneficiaries, subject to certain restrictions set out in the proviso to clause 4(1). Some of these restrictions relate to United States taxation and need not be set out in detail, but it is notable that provisos (ee) and (ff) are as follows:

    "(ee) no distribution of income shall be made to any beneficiary who has not attained the age of twenty-one years except in payment of educational or necessary medical expenses or for the alleviation of hardship.

    (ff) no distribution of capital shall be made to any beneficiary who has not attained the age of twenty-five years except in payment of educational or necessary medical expenses or for the alleviation of hardship."

    (2) Clause 7 contains a comparable power (except that the consent of the Protector is not required) for the trustees to make transfers to other settlements benefiting all or any of the Beneficiaries.

    (3) Clause 5 contains a power for the trustees to accumulate income during the Accumulation Period.

13. It is clauses 4(2) and 6 which present the difficulties of construction. Clause 4(2) is in the following terms:

    "(2) Upon the death of any Beneficiary the Trustee shall hold that portion of the Trust Fund to which the deceased Beneficiary had been entitled during his lifetime UPON TRUST for such person or persons as the deceased Beneficiary had notified the Trustee in writing and in the absence of such notification for that person or those persons whom the Trustee believes to be the closest surviving relative or relatives of the deceased Beneficiary."

    The problem here is that unless and until the trustees have exercised their power of appointment under clause 4(1) in such a way as to give any of the beneficiaries a fixed entitlement (for instance, a life interest in some fraction of the trust fund) the reference to "that portion of the Trust Fund to which the deceased Beneficiary had been entitled during his lifetime" is inapposite.

14. Clause 6 is in the following terms:

    "Subject as aforesaid the Trustee shall hold the Trust Fund and the income thereof so far as not effectively dealt with pursuant to the foregoing trusts and powers UPON TRUST for such purposes as are according to the laws of the Isle of Man or any subsequent forum of administration."

    The problem here is that there is a gap. It is possible to conjecture (especially by reference to the comparable clause of the Everest Trust) that the gap should be filled by "charitable" (or "charitable and public") but it would be a strong thing for any court to fill the gap as part of a process of construction (as opposed to rectification).

15. The remaining clauses of the Angora Trust contain wide administrative and other powers and provisions, mostly in a familiar form. The following call for mention:

    (1) Clause 29 enables a sole corporate trustee to act, and clauses 28 and 29 together provide for the remuneration of trustees.

    (2) Clauses 30, 31 and 32 provide for the devolution (and possible lapse) of the functions of the Protector (since Mr Schmidt's death, Mr Alexander Djparidze has been appointed as Protector).

    (3) Clause 33 confers on the Protector the power of appointing and removing trustees, and (in subclause (c), which also seems to have a gap) "power to require the Trustee or Trustees upon request information relating to the Trust including minutes accounts documents papers records or other information no matter how maintained or any part thereof whether specified or in general".

    (4) Clause 34 gives the trustees power to have the trust accounts audited and ends with the words "and any Beneficiary may require such an audit".

    (5) Clause 38 gives the trustees power to release or restrict their own powers.

16. The principal points of difference in the Everest Trust are as follows:

    (1) Clauses 1, 2 and 4 correspond to clauses 1, 2 and 3 respectively of the Angora Trust except that (i) the definition of "the Beneficiaries" includes any person or charity added under clause 3; (ii) the same definition refers to a Second Schedule which was apparently never included in the deed, but the names of Lukoil senior executives were added in manuscript at the end of the deed (the full list has never been disclosed, but it included Mr Schmidt); and (iii) the Protector was named as Mr Ralif Safin.

    (2) Clause 3.3 confers on the trustees power exercisable by written instrument during the Trust Period to add to the class of Beneficiaries "any person or persons or class or classes of person (including an individual then unborn) or charity" other than a current trustee or (while the trustees are resident in the Isle of Man) any Isle of Man resident. Other provisions in clause 3 permit the trustees to exclude beneficiaries and permit a person (if sui juris) to exclude himself.

    (3) Clause 5(a) contains a wide power of appointment very similar to that in clause 4(1) of the Angora Trust, except that the provisos are different (in particular, they do not refer to United States tax nor do they refer to benefits for persons under 21 or 25 years of age). There is no clause 5(b).

    (4) Clause 6 contains, in default of and subject to any appointment under clause 5, (i) a discretionary trust of income in favour of the Beneficiaries, subject to (ii) a proviso containing a power for the trustees to accumulate income during the Accumulation Period.

    (5) Clauses 7 and 8 correspond to clauses 6 and 7 respectively of the Angora Trust, except that the apparent gap in clause 6 of the Angora Trust is in clause 7 of the Everest Trust filled by the words "both charitable and public as the Trustee shall from time to time determine".

    (6) Clause 35 of the Everest Trust (corresponding to clause 34 of the Angora Trust) does not include the final words "and any Beneficiary may require such an audit".

17. Each settlement contained (in its First Schedule) particulars of the property initially settled. In the Angora Trust this was 50 ordinary shares in Nizam Limited, a company incorporated in the Republic of Ireland. The First Schedule to the Everest Trust has never been fully disclosed, but it included two US$1 shares in a company called Petragonis Limited ("Petragonis"). The Ernst & Young report contains some further information (but, the appellant says, incomplete information) about the web of companies (incorporated in different parts of the world) apparently associated with the two settlements. For present purposes it is sufficient to identify a company called Gingernut Limited ("Gingernut") and Petragonis as to the two companies which were used as the principal vehicles for the distribution to or for the benefit of Mr Schmidt of funds from the Angora Trust and the Everest Trust respectively. These two companies were apparently the source of the sums distributed to the appellant, as Mr Schmidt's personal representative, in 1998.

18. Before leaving the terms of the two settlements there is one further feature of the Angora Trust which calls for mention, although the weight to be attached to it is a matter for debate. An unredacted copy of the Angora Trust found by the appellant with his father's papers shows that the Second Schedule originally contained eight names. Mr Schmidt's name was followed by the words "as to a three-tenths share (30%)". Each of the other seven names was followed by the words "as to a one-tenth share (10%)". All the references to fractional shares were crossed through (but so as to remain legible). In addition four of the names had been crossed through, and one further name had been added in manuscript (without any reference to a share). This feature, together with the other problematic points on the Angora Trust already mentioned, gives much force to the submission of Mr Steinfeld QC (for the appellant) that the settlement was "cobbled together". Mr Brownbill (for Rosewood) agreed that the settlement appeared to contain mistakes.

19. On 23rd June 1992 Mr Schmidt wrote to Lorne House as trustee of the Angora Trust a letter ("the Angora letter") in the following terms:

    "I understand that I am a beneficiary of the Angora Trust. If I should die prior to the termination of the Trust I wish any portion to which I might have been entitled to be held upon trust for [and then he gave the appellant's name and date and place of birth]."

20. On 31st October 1995 Mr Schmidt wrote to Lorne House as trustee of the Everest Trust a letter ("the Everest letter") in the following terms:

    "The Everest Trust. While I recognise the discretionary powers vested in you as Trustees of the above Trust, it would be my wish if I were to die prior to the termination of the Trust that my share of the trust property be given to Vadim Schmidt."

21. There are some indications (in particular in Rosewood's answer, which was later amended in this respect) that Rosewood initially accepted the Angora letter as an effective exercise of a power conferred on Mr Schmidt by clause 4(2) of the Angora Trust. However Rosewood's present position is that neither letter had any legal effect.

    The proceedings below.

22. The appellant's petition first came before Deemster Cain on 5th October 1999. It was supported by the appellant's affidavit of 7th June 1999 already mentioned. Rosewood had an opportunity to put in evidence but did not do so. The time for argument was limited and the Deemster indicated that he was not going to make an order at that hearing. But he gave a short extempore judgment declining to dismiss the petition as an abuse of process. He indicated that in principle Rosewood as trustee must make disclosure to its beneficiaries – "each beneficiary", he said, "is entitled to know what the trustees have done with the money" – but that he was concerned about the confidentiality of third parties. He made the helpful suggestion that disclosure might be made, not to the appellant himself, but to the appellant's legal or accountancy advisers. He adjourned the matter to see whether the parties could agree a form of order.

23. The parties were unable to agree on the form of order and the matter came back to the Deemster on 19th November 1999 when he made an order. The order as finally drawn up is elaborate, running to nine pages of typescript and including (in its second schedule) the names of over 60 companies (or other legal entities) established in different parts of the world. The essential provisions of the order were for Rosewood to make extensive disclosure of unredacted documents and to provide information to three named members of the firm of Ernst & Young ("the Inspectors") on the Inspectors and their lawyers giving undertakings to hold all the disclosed documents and information in confidence, to be used only for purposes authorised by the order. It is not necessary, for present purposes, to go further into the detail of the order.

24. Rosewood appealed to the Staff of Government Division and the appeal was heard by Deemster Kerruish QC and Mr G F Tattersall QC (sitting as a Judge of Appeal) on several dates between January and April 2000. The court heard but rejected an application by Rosewood to adduce further evidence. It gave a reserved judgment on 26th February 2001 allowing the appeal and setting aside the order of Deemster Cain.

25. Rosewood had initially put forward four grounds of appeal. The first, abuse of process, was not proceeded with; the second and third raised points as to the extent and efficacy of the safeguards for confidentiality embodied in the order; and the fourth related to costs. But a week before the appeal hearing Rosewood asked for permission to amend its grounds of appeal and to contend that the appellant was not a beneficiary in any sense of the word under the two settlements, and that his father, Mr Schmidt, was never more than "a mere object of a power who as such had no entitlement to trust documents or information". The amendment was allowed even though it necessitated an adjournment and allowed Rosewood to run a quite different case from that which it had put forward below.

26. Most of the judgment of the Staff of Government Division is concerned with the point of principle raised in the new ground of appeal. Since it allowed the appeal on that ground, it did not go far into the other grounds of appeal. But in para 65 of the judgment the court indicated that if it had concluded that the Deemster had jurisdiction to make the order appealed from, it would have "felt constrained to exercise [its] discretion in favour of, in principle, making an order for disclosure". The court noted in para 66 that it had not found it necessary to consider:

    "… firstly, the privacy and confidentiality of the other beneficiaries of the settlement; secondly, the scope of the order; and thirdly the appointment of the [appellant's, ie Vadim Schmidt's] lawyers [which seems to be a mistake for 'accountants'] as inspectors."

    Her Majesty in Council gave special leave to appeal on 19th September 2001.

    Issues of construction.

27. Before turning to the point of law which is the main issue on this appeal it is convenient to clear the ground by referring briefly to some preliminary issues of construction which were not argued below, but which were to some extent debated before the Board. Three of these relate to the Angora Trust, and two can be dealt with quite briefly.

    (1) The appellant's printed Case contended that clause 4(1) contains not a power of appointment, but a discretionary trust. Mr Steinfeld rightly did not seek to develop this contention in his oral submissions, although he did not formally abandon it. It is a hopeless argument.

    (2) The gap in clause 6 may well have been intended to be filled by a reference to charity, but it would be a strong thing for the court to fill the gap as part of a process of construction (see for instance In re Whitrick, decd [1957] 1 WLR 884). The trustees could at any time fill the gap, if they thought fit, by appointing an ultimate trust to the RNLI. But it is not necessary to pursue the point because it has no practical bearing on the outcome of the appeal.

    (3) The effect (if any) of clause 4(2), in conjunction with the Angora letter, is however potentially of very great importance, and it must be considered more fully.

28. Both sides agree that the Angora Trust is a badly-drafted document. It may be that the Lukoil executives who caused it to be brought into existence had an inadequate understanding of its effect. The deleted reference to fixed shares in the Second Schedule strongly suggests that they may not have seen the settlement as conferring on the trustees the very wide discretions which it appears to confer.

29. Mr Steinfeld submitted that each of the putative settlors should be regarded as having been in some sense entitled to the funds which he caused to be brought into the settlement, and that this enabled clause 4(2) to be given a sensible meaning. The alternative, he said, would be that the settlement would become a sort of tontine for the longest-living settlor, an absurd result which cannot have been intended.

30. Mr Brownbill submitted that clause 4(2) was obviously a mistake of some sort, and that it would do unacceptable violence to the language of the settlement to treat Mr Schmidt as having been entitled to a portion of the trust fund during his lifetime, when he was never more than a mere object of a discretionary power of appointment.

31. Their Lordships take the view that they cannot decide this issue of construction, for two reasons. First, it is a question which is likely to be of acute concern to the other beneficiaries interested under the Angora Trust, but they are not at present parties to the proceedings. The other beneficiaries (or at any rate a representative of them in respect of whom a representation order could be made) would have to be joined and given the opportunity of making submissions on the point of construction. They would then be bound by any order deciding the point. The second reason for the Board to refrain from deciding the point at this stage is that there is insufficient evidence (and no findings below) as to the surrounding circumstances (or "matrix of fact") in which the settlement was made.

32. The Board does not therefore express any view on this issue of construction, beyond stating that it (unlike the point on clause 4(1)) is arguable. Mr Steinfeld's proposed construction does do some violence to the language of clause 4(2), but the modern approach of the court is not to reject any part of a legal document as meaningless without first trying hard to give it a sensible meaning: see the observations of Lord Upjohn in Whishaw v Stephens (on appeal from In re Gulbenkian's Settlement) [1970] AC 508, 522B-D. The appellant may take steps to have the proceedings reconstituted and to have this issue of construction determined in the High Court of the Isle of Man. But unless and until he takes that course and succeeds in his argument, it cannot be assumed that the appellant in his personal capacity is a beneficiary (in any sense) under the Angora Trust.

33. In relation to the Everest Trust the appellant in his personal capacity is no more than a possible object of the very wide power to add beneficiaries conferred by clause 3.3. The Everest letter provides clear evidence of Mr Schmidt's wishes and confirms (as would in any case be fairly evident) that the appellant may have a particularly strong claim on the trustees' discretion. But neither the Everest letter nor any other document put in evidence had any further effect on his status as a possible beneficiary, and ultimately Mr Steinfeld did not contend otherwise.

    Disclosure to discretionary beneficiaries: the background.

34. Counsel have very properly referred the Board to a considerable number of authorities, some of them going back to the early years of the 19th century. It is appropriate to reflect that during the long period covered by these authorities (but especially during the second half of the 20th century) the forms and functions of settlements have changed to a degree which would have astonished Lord Eldon. By the 1930's high rates of personal taxation led some wealthy individuals to make settlements which enabled funds to be accumulated in the hands of overseas trustees or companies (see for instance Vestey's executor v IRC (1949) 31 TC 1). This practice increased enormously with the introduction of capital gains tax in 1965. But increasingly stringent anti-avoidance measures encouraged legal advisers to devise forms of settlement under which the true intended beneficiaries were not clearly identified in the settlement. Indeed their interests or expectations were often barely perceptible. Rarely did a beneficiary take an indefeasibly vested interest with an ascertainable market value. Tax avoidance is therefore one element which has strongly influenced the forms of settlements; and once the offshore tax-avoidance industry has acquired standard forms its inclination is to use them, subject perhaps to some more or less skilful adaptation, even for clients whose aim is not to avoid United Kingdom taxation.

35. There is another element, also linked (though less directly) to taxation, which has encouraged the inclusion in settlements of very widely defined classes of beneficiaries. After the second world war estate duty was charged in the United Kingdom at very high rates, with much less generous reliefs for agricultural and business property than those now available. A wealthy landowner or businessman might be advised that the safest way to preserve his fortune was to give most of it away, while he was still in the prime of life, to trustees of an irrevocable settlement in discretionary form under which the settlor himself was not a beneficiary. It is not surprising that a settlor in such a position should wish to cover as comprehensively as he could all possible current and future claims on his bounty, since he was being asked to make an immediate, irrevocable disposition of much of his wealth, rather than being able to review from time to time the ambulatory dispositions in his will and codicils. But his lawyers might also advise him that the most natural expressions for defining discretionary objects of his bounty (such as "relatives", "old friends", "dependants" or "persons with moral claims") were of doubtful legal efficacy. So there was a tendency to define the class in the widest possible terms. The process can be seen in a long line of cases starting with In re Gestetner Settlement [1953] Ch 672. It led to In re Manisty's Settlement [1974] Ch 17, upholding the validity of an "intermediate" power comparable to that in clause 3.3 of the Everest Trust (that is, a power to add as beneficiaries anyone in the world apart from a very small class of excluded persons).

36. The Board have to consider what rights or claims to disclosure the appellant has, either personally or as his father's personal representative, under two badly-drafted settlements whose terms have been moulded by the sort of influences mentioned above. One possible reaction would be that Mr Schmidt and his colleagues have made their bed and they must lie on it; if they have deliberately entered into a web of camouflage, it is hardly for anyone claiming through them to complain that the position is not transparent. As Lord Greene MR observed, giving the judgment of the court in Howard de Walden v Inland Revenue Commissioners [1942] 1 KB 389, 397 if a taxpayer plays with fire it scarcely lies in his mouth to complain of burnt fingers. However, the Board consider that that inclination must be resisted. As already noted, it has not been suggested that the settlements are shams, or tainted with illegality. It is fundamental to the law of trusts that the court has jurisdiction to supervise and if appropriate intervene in the administration of a trust, including a discretionary trust. As Holland J said in the Australian case of Randall v Lubrano (unreported 31st October 1975) cited by Kirby P in Hartigan Nominees Pty Ltd v Rydge (1992) 29 NSWLR 405, 416:

    "… no matter how wide the trustee's discretion in the administration and application of a discretionary trust fund and even if in all or some respects the discretions are expressed in the deed as equivalent to those of an absolute owner of the trust fund, the trustee is still a trustee."

    Discretionary trusts and powers.

37. Several of the numerous authorities referred to by counsel have been concerned with the general characteristics of interests (or rights) under discretionary trusts (on the one hand) and mere powers (of a dispositive character) conferred on trustees in their fiduciary capacity (on the other hand). It is convenient to refer to these before coming to the authorities directly concerned with disclosure of trust documents or information.

38. In the important case of McPhail v Doulton (on appeal from In re Baden's Deed Trusts) [1971] AC 424 the House of Lords finally settled the vexed question of whether the test for validity, in point of certainty of objects, is the same for trusts and powers, or whether the test for trusts is more demanding. It held the test to be the same. That general question arose in the context of a provision (which the Court of Appeal took to be a power, but the House of Lords held to be a trust) for trustees to distribute income "to or for the benefit of any of the officers and employees or ex-officers or ex-employees of [a named company] or to any relatives or dependants of any such persons" (with a power for the trustees to hold up income which did not, as the House of Lords held, prevent the trustees distributing the retentions as income): see p 428.

39. That was the context in which Lord Wilberforce (at pp 448-9) made some general observations which, although fairly lengthy, need to be set out in full:

    "… some general observations, or reflections, may be permissible. It is striking how narrow and in a sense artificial is the distinction, in cases such as the present, between trusts or as the particular type of trust is called, trust powers, and powers. It is only necessary to read the learned judgments in the Court of Appeal to see that what to one mind may appear as a power of distribution coupled with a trust to dispose of the undistributed surplus, by accumulation or otherwise, may to another appear as a trust for distribution coupled with a power to withhold a portion and accumulate or otherwise dispose of it. A layman and, I suspect, also a logician would find it hard to understand what difference there is.

    It does not seem satisfactory that the entire validity of a disposition should depend on such delicate shading. And if one considers how in practice reasonable and competent trustees would act, and ought to act, in the two cases, surely a matter very relevant to the question of validity, the distinction appears even less significant. To say that there is no obligation to exercise a mere power and that no court will intervene to compel it, whereas a trust is mandatory and its execution may be compelled, may be legally correct enough but the proposition does not contain an exhaustive comparison of the duties of persons who are trustees in the two cases. A trustee of an employees' benefit fund, whether given a power or a trust power, is still a trustee and he would surely consider in either case that he has a fiduciary duty: he is most likely to have been selected as a suitable person to administer it from his knowledge and experience, and would consider he has a responsibility to do so according to its purpose. It would be a complete misdescription of his position to say that, if what he has is a power unaccompanied by an imperative trust to distribute, he cannot be controlled by the court unless he exercised it capriciously, or outside the field permitted by the trust (cf. Farwell on Powers, 3rd ed., p. 524). Any trustee would surely make it his duty to know what is the permissible area of selection and then consider responsibly, in individual cases, whether a contemplated beneficiary was within the power and whether, in relation to other possible claimants, a particular grant was appropriate.

    Correspondingly a trustee with a duty to distribute, particularly among a potentially very large class, would surely never require the preparation of a complete list of names, which anyhow would tell him little that he needs to know. He would examine the field, by class and category; might indeed make diligent and careful inquiries, depending on how much money he had to give away and the means at his disposal, as to the composition and needs of particular categories and of individuals within them; decide upon certain priorities or proportions, and then select individuals according to their needs or qualifications. If he acts in this manner, can it really be said that he is not carrying out the trust?

    Differences there certainly are between trust (trust powers) and powers, but as regards validity, should they be so great as that in one case complete, or practically complete, ascertainment is needed, but not in the other? Such distinction as there is would seem to lie in the extent of the survey which the trustee is required to carry out: if he has to distribute the whole of a fund's income, he must necessarily make a wider and more systematic survey than if his duty is expressed in terms of a power to make grants. But just as, in the case of a power, it is possible to underestimate the fiduciary obligation of the trustee to whom it is given, so, in the case of a trust (trust power), the danger lies in overstating what the trustee requires to know or to inquire into before he can properly execute his trust. The difference may be one of degree rather than of principle: in the well-known words of Sir George Farwell, Farwell on Powers, 3rd ed. (1916), p. 10, trusts and powers are often blended, and the mixture may vary in its ingredients."

40. This passage gives a very clear and eminently realistic account of both the points of difference and the similarities between a discretionary trust and a fiduciary dispositive power. The outstanding point of difference is of course that under a discretionary trust of income distribution of income (within a reasonable time) is mandatory, the trustees' discretion being limited to the choice of the recipients and the shares in which they are to take. If there is a small, closed class of discretionary objects who are all sui juris, their collective entitlement gives them a limited power of disposition over the income subject to the discretionary trust, as is illustrated by In re Smith [1928] Ch 915 and In re Nelson (1918) reported as a note to In re Smith. But the possibility of such a collective disposition will be rare, and on his own the object of a discretionary trust has no more of an assignable or transmissible interest than the object of a mere power.

41. Apart from the test for certainty being the same and the fact that an individual's interest or right is non-assignable, there are other practical similarities between the positions of the two types of object. Either has the negative power to block a family arrangement or similar transaction proposed to be effected under the rule in Saunders v Vautier (1841) 4 Beav 115 (unless in the case of a power the trustees are specially authorised to release, that is to say extinguish, it). Both have a right to have their claims properly considered by the trustees. But if the discretion is exercisable in favour of a very wide class the trustees need not survey mankind from China to Peru (as Harman J, echoing Dr Johnson, said in In re Gestetner Settlement [1953] Ch 672, 688-9) if it is clear who are the prime candidates for the exercise of the trustees' discretion.

42. That thought was developed by Templeman J in In re Manisty's Settlement [1974] Ch 17, although he was mainly concerned to contrast the exercise by trustees of an intermediate power (in the sense mentioned above) with the exercise by trustees of a wide special power. He said (at p 27) that a wide power, whether special or intermediate, does not negative or prohibit a sensible approach by trustees to the consideration and exercise of their powers. After referring to some very well-known observations by Lord Eldon LC in Morice v Bishop of Durham (1805) 10 Ves Jun 522, 539, Templeman J continued (at pp 27-28):

    "Nor does an intermediate power break the principles laid down by Lord Eldon LC in the passage which I have read because, in relation to a power exercisable by the trustees at their absolute discretion, the only 'control' exercisable by the court is the removal of the trustees, and the only 'due administration' which can be 'directed' is an order requiring the trustees to consider the exercise of the power, and in particular a request from a person within the ambit of the power."

    However in Mettoy Pension Trustees Ltd v Evans [1990] 1 WLR 1587, 1617-8, Warner J (after referring to Lord Wilberforce's observations in McPhail v Doulton at pp 456-7 and to some authorities not cited in In re Manisty's Settlement) took a broader view of the court's power to intervene in the case of a fiduciary dispositive power.

    Disclosure to discretionary beneficiaries: a proprietary basis?

43. Much of the debate before the Board addressed the question whether a beneficiary's right or claim to disclosure of trust documents should be regarded as a proprietary right. Mr Brownbill argued that it should be classified in that way, and from that starting point he argued that no object of a mere power could have any right or claim to disclosure, because he had no proprietary interest in the trust property. Mr. Brownbill submitted that this point has been conclusively settled by the decision of the House of Lords in O'Rourke v Darbishire [1920] AC 581. It is therefore useful to go straight to that case to see what it did decide.

44. The facts of the case were unusual. Sir Joseph Whitworth, a man of considerable wealth, had died in 1887. In 1884 he had made a will appointing three executors and leaving his residuary estate to charity. By a codicil made in 1885 he altered his will to leave his ultimate residue to his executors for their own benefit, with a precatory expression of his wishes that it should be used for charitable purposes. Two further codicils executed in 1886 extended the scope of the first codicil's gift to the executors. Sir Joseph's intestate successors would have been Mrs Uniacke (as to realty) and Mrs Uniacke and Mrs McGowan (as to personalty). Mrs McGowan threatened to challenge the will and codicils, but in 1889 there was a compromise between all interested parties. Then in 1916, after Mrs Uniacke, Mrs McGowan and the executors had all died, Mrs Uniacke's administrator (Mr O'Rourke) sought to challenge both the will and codicils and the compromise, alleging fraud by Mr Darbishire (who was one of the executors and had been Sir Joseph's solicitor). Mr O'Rourke sought to obtain disclosure of documents containing legal advice given to Sir Joseph during his lifetime, and to his executors after his death.

45. The House of Lords dismissed Mr O'Rourke's appeal, primarily because he had not made out even a prima facie case that the will and codicils were invalid, or that the communications had been promoting fraud. Viscount Finlay (at p 603) referred to Mr O'Rourke's reliance on a "proprietary right" and Lord Sumner (at p 617) referred to "what has been called the 'proprietary' ground". Lord Parmoor said (at pp 619-20):

    "A cestui que trust, in an action against his trustees, is generally entitled to the production for inspection of all documents relating to the affairs of the trust. It is not material for the present purpose whether this right is to be regarded as a paramount proprietary right in the cestui que trust, or as a right to be enforced under the law of discovery, since in both cases an essential preliminary is either the admission, or the establishment, of the status on which the right is based."

46. It is on what was said by Lord Wrenbury that Mr Brownbill most relied. Lord Wrenbury said at pp 626-7:-

    "If the plaintiff is right in saying that he is a beneficiary and if the documents are documents belonging to the executors as executors, he has a right to access to the documents which he desires to inspect upon what has been called in the judgments in this case a proprietary right. The beneficiary is entitled to see all trust documents because they are trust documents and because he is a beneficiary. They are in this sense his own. Action or no action, he is entitled to access to them. This has nothing to do with discovery. The right to discovery is a right to see someone else's documents. The proprietary right is a right to access to documents which are your own."

    On the facts of the case, what Lord Wrenbury said was very apposite. If Mr O'Rourke was right in his claim, the executors had had no proper legal or equitable title to Sir Joseph's estate. The grant of probate to them should have been revoked and Mr O'Rourke (together with the representatives of Mrs McGowan) would have been entitled to the whole of the estate, including any documents which formed part of it.

47. The same can be said of the earliest authority to which the Board was referred, the decision of Lord Eldon LC in Clarke v Earl of Ormonde (1821) Jacob 108. Under private Acts of Parliament real estate in Ireland was vested in trustees in trust to raise money to pay the debts of Walter Marquis of Ormonde, and subject to that trust for the Marquis in fee simple. In those circumstances it is hardly surprising that Lord Eldon (at pp 119-20) referred to the Marquis' estate in fee simple before observing:

    "But he would have the right to say to the trustees, What estates have you sold? What debts have you paid?"

48. In In re Cowin (1886) 33 Ch D 179 a beneficiary with a vested future interest in one-eighth of a testator's residuary trust fund, subject to his mother's interest during widowhood, wished to mortgage his interest and for that reason sought inspection of the title deeds to the trust property. North J made an order for inspection but rejected the notion that the beneficiary had an absolute right. He quoted (at p 185) from the then current (8th) edition of Lewin, p 975:

    "All documents held by the trustee in that character must be produced by him to the cestuis que trust, who are in equity the true owners."

    But North J clearly considered that the particular interest of an individual beneficiary might in some circumstances run counter to the collective interest of the beneficiaries as a body. He said at p 187:

    "I do not say that he is entitled as of right, but only that he is entitled under the circumstances, because there might be a state of circumstances under which the right to production would not exist."

    Chitty J took a similar approach in In re Tillott [1892] 1 Ch 86, noting (at p 89) that a trustee is not bound to give a beneficiary information about a share in which he has no interest.

49. In In re Londonderry's Settlement [1965] Ch 918 the Court of Appeal had to consider one of the most important limitations on the right to disclosure of trust documents, that is the need to protect confidentiality in communications between trustees as to the exercise of their dispositive discretions, and in communications made to the trustees by other beneficiaries. That issue can alternatively be seen as an inquiry whether such confidential communications are indeed trust documents. The judgments of the three members of the court (Harman, Danckwerts and Salmon LJJ) are not easy to reconcile. All three referred to O'Rourke v Darbishire but Harman and Danckwerts LJJ found that Lord Wrenbury's general observations gave little assistance on the issue which concerned them. Only Salmon LJ (at p 937) expressly adopted the proprietary basis of the principle.

50. Lord Wrenbury's observations in O'Rourke v Darbishire have also been cited in several Australian cases, and they were referred to by Lord Lowry in A T & T Istel Ltd v Tully [1993] AC 45, 65. The Board does not find it surprising that Lord Wrenbury's observations have been so often cited, since they are a vivid expression of the basic distinction between the right of a beneficiary arising under the law of trusts (which most would regard as part of the law of property) and the right of a litigant to disclosure of his opponent's documents (which is part of the law of procedure and evidence). But the Board cannot regard it as a reasoned or binding decision that a beneficiary's right or claim to disclosure of trust documents or information must always have the proprietary basis of a transmissible interest in trust property. That was not an issue in O'Rourke v Darbishire.

51. Their Lordships consider that the more principled and correct approach is to regard the right to seek disclosure of trust documents as one aspect of the court's inherent jurisdiction to supervise, and if necessary to intervene in, the administration of trusts. The right to seek the court's intervention does not depend on entitlement to a fixed and transmissible beneficial interest. The object of a discretion (including a mere power) may also be entitled to protection from a court of equity, although the circumstances in which he may seek protection, and the nature of the protection he may expect to obtain, will depend on the court's discretion: see Lord Wilberforce in Gartside v Inland Revenue Commissioners [1968] AC 553, 617-8 and in McPhail v Doulton [1971] AC 424, 456-7; Templeman J in In re Manisty's Settlement [1974] Ch 17, 27-8; and Warner J in Mettoy Pension Trustees Ltd v Evans [1990] 1 WLR 1587, 1617-8. Mr Brownbill's submission to the contrary effect tends to prove too much, since he would regard the object of a discretionary trust as having a proprietary interest even though it is not transmissible (except in the special case of collective action taken unanimously by all the members of a closed class).

52. Their Lordships are therefore in general agreement with the approach adopted in the judgments of Kirby P and Sheller JA in the Court of Appeal of New South Wales in Hartigan Nominees Pty Ltd v Rydge (1992) 29 NSWLR 405. That was a case concerned with disclosure of a memorandum of wishes addressed to the trustees by Sir Norman Rydge (who was in substance, but not nominally, the settlor). Kirby P said at pp 421-2:

    "I do not consider that it is imperative to determine whether that document is a 'trust document' (as I think it is) or whether the respondent, as a beneficiary, has a proprietary interest in it (as I am also inclined to think he does). Much of the law on the subject of access to documents has conventionally been expressed in terms of the 'proprietary interest' in the document of the party seeking access to it. Thus, it has been held that a cestui que trust has a 'proprietary right' to seek all documents relating to the trust: see O'Rourke v Darbishire (at 601, 603). This approach is unsatisfactory. Access should not be limited to documents in which a proprietary right may be established. Such rights may be sufficient; but they are not necessary to a right of access which the courts will enforce to uphold the cestui que trust's entitlement to a reasonable assurance of the manifest integrity of the administration of the trust by the trustees. I agree with Professor H A J Ford's comment, in his book (with Mr W A Lee) Principles of the Law of Trusts, 2nd ed (1990) Sydney, Law Book Co, at 425, that the equation of rights of inspection of trust documents with the beneficiaries' equitable rights of property in the trust assets 'gives rise to far more problems than it solves' (at 425):

    '... The legal title and rights to possession are in the trustees: all the beneficiary has are equitable rights against the trustees. ... The beneficiary's rights to inspect trust documents are founded therefore not upon any equitable proprietary right which he or she may have in respect of those documents but upon the trustee's fiduciary duty to keep the beneficiary informed and to render accounts. It is the extent of that duty that is in issue. The equation of the right to inspect trust documents with the beneficiary's equitable proprietary rights gives rise to unnecessary and undesirable consequences. It results in the drawing of virtually incomprehensible distinctions between documents which are trust documents and those which are not; it casts doubts upon the rights of beneficiaries who cannot claim to have an equitable proprietary interest in the trust assets, such as the beneficiaries of discretionary trusts; and it may give trustees too great a degree of protection in the case of documents, artificially classified as not being trust documents, and beneficiaries too great a right to inspect the activities of trustees in the case of documents which are, equally artificially, classified as trust documents.'"

53. Mahoney JA (at p 435) favoured the proprietary basis but recognised that it extended to information of a non-documentary kind. Sheller JA (at p 444) considered that inquiry as to an applicant's proprietary interest was "if not a false, an unhelpful trail". All three members of the court expressed reservations about the reasoning and conclusions in In re Londonderry's Settlement.

54. It will be observed that Kirby P said that for an applicant to have a proprietary right might be sufficient, but was not necessary. In the Board's view it is neither sufficient nor necessary. Since In re Cowin well over a century ago the court has made clear that there may be circumstances (especially of confidentiality) in which even a vested and transmissible beneficial interest is not a sufficient basis for requiring disclosure of trust documents; and In re Londonderry's Settlement and more recent cases have begun to work out in some detail the way in which the court should exercise its discretion in such cases. There are three such areas in which the court may have to form a discretionary judgment: whether a discretionary object (or some other beneficiary with only a remote or wholly defeasible interest) should be granted relief at all; what classes of documents should be disclosed, either completely or in a redacted form; and what safeguards should be imposed (whether by undertakings to the court, arrangements for professional inspection, or otherwise) to limit the use which may be made of documents or information disclosed under the order of the court.

55. The proprietary basis of a beneficiary's right to disclosure was fully argued before the Staff of Government Division, which (para 35) accepted the submission (made on behalf of the appellant) that a proprietary interest, although often found, was not necessary. On this part of the case the Board agrees with the conclusion reached by the Staff of Government Division, and does not accept the criticisms of it put forward by Mr Brownbill. It has nevertheless been necessary to look at the authorities in some detail, because they lead on to part of the Staff of Government Division's judgment (paras 42 to 60) in which the court reached the conclusion that Deemster Cain had no jurisdiction to make the order for disclosure which he did make. In reaching that conclusion the court distinguished (or treated as unhelpful) a number of cases, of which the most important are Chaine-Nickson v The Bank of Ireland [1976] IR 393, Spellson v George (1987) 11 NSWLR 300, Hartigan Nominees Pty Ltd v Rydge (1992) 29 NSWLR 405, A-G of Ontario v Stavro (1994) 119 DLR (4th) 750 and Murphy v Murphy [1999] 1 WLR 282.

    Disclosure to discretionary beneficiaries: the recent cases.

56. Chaine-Nickson v Bank of Ireland [1976] IR 393 was a decision of the High Court of Ireland (Kenny J). It was concerned with an application for disclosure of accounts and other information made by the beneficiary of what was described as a discretionary trust. But the report does not give an adequate summary of the trust. The only dispositive provision set out verbatim looks more like a power than a trust, as it begins "upon trust to pay, divide or apply the whole or such part (if any) of the income and capital respectively thereof as [the trustees] shall from time to time in their absolute and uncontrolled discretion think fit …". It appears that initially the settlement infringed the rule against perpetuities and had to be corrected, but again the summary is terse.

57. After stating the general rule that a beneficiary with a vested interest is entitled to disclosure Kenny J continued (at p 396):

    "However, in the case of a discretionary trust, none of the potential beneficiaries have any right to be paid capital or income. All the trust fund is held by the trustees in this case on discretionary trusts and, if the plaintiff is not entitled to the trust accounts and particulars of the investments, it follows that none of the potential beneficiaries have a valid claim to any information from the trustees. The result is that the trustees are not under an obligation to account to anyone in connection with their management of the trust fund. This logical conclusion from the defendants' argument leads to remarkable consequences.

    The amount of remuneration to which the trustees are entitled is specified in the settlement and the potential beneficiaries have an interest in seeing that the amount is not exceeded, for they are the persons who will ultimately benefit by payments of capital and income. The defendants' contention, however, has the result that they do not have to account for or disclose the amount of their remuneration. This seems to me to be contrary to the basic concept of a trustee being accountable for his management of the trust fund."

58. The Staff of Government Division seem to have regarded Chaine-Nickson as a special and unusual case, in that there would normally be some beneficiary who had a fixed interest in default of exercise of the trustees' discretion, and who would be in a position to require disclosure of accounts. But off-shore settlements do very commonly, for reasons already noted, have no ascertained beneficiary with a fixed interest and a real economic stake in the enforcement of the trustees' fiduciary duties. The Staff of Government Division was, with respect, too ready to dismiss Chaine-Nickson as untypical and unhelpful. The contrary view does indeed lead to remarkable results.

59. Spellson v George (1987) 11 NSWLR 300 was a decision of the Equity Division of the Supreme Court of New South Wales (Powell J). The plaintiff (who had recently been divorced) was suing both on his own account and on behalf of his children for information about four separate settlements made by members of his ex-wife's family, three of which had apparently been wound up. Much of the report is concerned with the plaintiff's right to sue on behalf of his children without the concurrence of their joint guardian. However there is at the end of the judgment of Powell J (pp 315-6) a valuable discussion of the authorities and the underlying principle (though the clarity of the discussion is slightly marred by the expressions "power" and "trust" being used almost interchangeably). The whole passage merits study. The conclusion (at p 316) is as follows:

    "The question then is, whether a person whose status is only that of a potential object of the exercise of a discretionary power can properly be regarded as one of the cestuis que trust of the relevant trustee. I do not doubt that he can, and should, properly be so regarded, for although it is true to say that, unless, and until, the trustee exercises his discretion in his favour, he has no right to receive, and enjoy, any part of the capital or income of the trust fund, it does not follow that, until that time arises, he has no rights against the trustee. On the contrary, it is clear that the object of a discretionary trust, even before the exercise of the trustee's discretion in his favour, does have rights against the trustee (see, eg, Gartside v Inland Revenue Commissioners (at 605-606) per Lord Reid, (at 617-618) per Lord Wilberforce) – those rights, so it seems to me, are not restricted to the right to have the trustee bona fide consider whether or not to exercise his (the trustee's) discretion in his (the object's) favour, but extend to the right to have the trust property properly managed and to have the trustee account for his management, a view, I am glad to say, which appears to have been shared by both Holland J in Randall v Lubrano and Kenny J in Chaine-Nickson v Bank of Ireland."

60. The Staff of Government Division quoted the above paragraph (and the preceding paragraph) from Powell J's judgment and criticised him for having given no reasons (apart from referring to Randall v Lubrano and Chaine-Nickson) for his extension of the discretionary object's rights to a right to require the trustees to account for their trusteeship. Their Lordships regard this criticism as surprising and unwarranted. Powell J had avowedly begun with first principles: (at p 315) "the fundamental proposition that one of the essential elements of a private trust, be it a discretionary trust or some other form of trust, is that the trustee is subject to a personal obligation to hold, and to deal with, the trust property for the benefit of some identified, or identifiable, person or group of persons". From there he had proceeded to the trustees' correlative duty to account. The observations of Lord Reid and Lord Wilberforce in Gartside v Inland Revenue Commissioners [1968] AC 553, 605-6, 617-8 were naturally directed to a discretionary object's rights in respect of distribution of income from the trust fund, because (in the context of construing the estate duty legislation) that is what the case was about (cf Sir Robert Megarry V-C in In re Hay's Settlement Trusts [1982] 1 WLR 202, 210: "so far as relevant to the case before me").

61. Hartigan Nominees Pty Ltd v Rydge (1992) 29 NSWLR 405 was, as already noted, a case about a settlor's letter of wishes. There was two main issues: whether the letter of wishes should be disclosed, despite its confidential nature; and how much weight trustees, in exercising their discretions, should give to a letter of wishes. Much of the discussion of well-known authorities was in the context of the second question. For present purposes the case is of most interest for the division of opinion in the Court of Appeal of New South Wales on whether a beneficiary's right or claim to disclosure of documents is proprietary in nature (see paras 52-54 above). In its comments on the case the Staff of Government Division mentioned only the judgment of Mahoney JA, who was in a minority on this point (although in the majority as to the outcome of the appeal). For the reasons already mentioned their Lordships prefer the views, as to the true basis for ordering disclosure, of Kirby P and Sheller JA.

62. The fourth of the cases considered by the Staff of Government Division, in this part of its judgment, was Attorney-General of Ontario v Stavro (1994) 119 DLR (4th) 750. In that case Lederman J had to consider an application for disclosure made by a beneficiary with a contingent interest in an unadministered estate. He referred to the argument for a proprietary basis and (echoing Sheller JA in Hartigan Nominees) considered that it "leads one astray" (see p 756). The case is not directly in point, but it is in line with the prevailing view in Australia as it emerges from Spellson v George and Hartigan Nominees (the point has recently been left open by the Full Court of the Supreme Court of South Australia, in the context of a commercial trust resembling a unit trust, in Rouse v IOOF Australia Trustees Ltd (1999) 73 SASR 484).

63. Finally there is the decision of Neuberger J in Murphy v Murphy [1999] 1WLR 282. That case was concerned with a beneficiary's right to obtain information (as to the identity of the trustees) from the settlor himself but Neuberger J discussed the position of the plaintiff as a discretionary object and stated (at p 290):

    "The facts that in this case the plaintiff is merely within the class of discretionary beneficiaries (as opposed to being someone with a vested beneficial interest in the trust property) and that there is no suggestion of wrong-doing on the part of the trustees appear to me to go to the question of whether to exercise the discretion [to exercise what the judge called the equitable jurisdiction] rather than whether the discretion exists at all."

64. The Staff of Government Division found this of no assistance, stating that it was based on a concession and that the judge did not have the benefit of any detailed analysis or submissions. However in considering whether to accept the concession Neuberger J did (like Powell J in Spellson v George) start from first principles; and his discussion (at p 291) of the principles to be derived from the authorities, from Morice v Bishop of Durham to Chaine-Nickson, appears to the Board to be illuminating and helpful.

65. To this review of the authorities considered by the Staff of Government Division it is appropriate to add three footnotes.

    (1) The Board were also assisted by references to a number of leading textbooks. Of the leading English works on trusts Lewin on Trusts (17th ed, 2000) appears to favour the proprietary approach and Underhill and Hayton's Law Relating to Trusts and Trustees (15th ed, 1995) the alternative approach. But each has a balanced survey of the cases and neither takes an extreme position.

    (2) Neither side made any submissions seeking to distinguish between trust documents and documents relating to the affairs of a company controlled by the trustees (see for instance Butt v Kelson [1952] Ch 197). This may have represented a realistic decision made by the trustees in the light of how the two settlements were in fact administered.

    (3) Mr Steinfeld did not base himself on the final words of clause 34 of the Angora Trust, or on any other express provision of either settlement. Mr Brownbill made some brief submissions based on the Protector's powers to obtain documents and information. These points may conceivably bear on the exercise of the court's discretion but they cannot in the Board's view go to the issue of jurisdiction.

    Conclusion.

66. Their Lordships have already indicated their view that a beneficiary's right to seek disclosure of trust documents, although sometimes not inappropriately described as a proprietary right, is best approached as one aspect of the court's inherent jurisdiction to supervise (and where appropriate intervene in) the administration of trusts. There is therefore in their Lordships' view no reason to draw any bright dividing-line either between transmissible and non-transmissible (that is, discretionary) interests, or between the rights of an object of a discretionary trust and those of the object of a mere power (of a fiduciary character). The differences in this context between trusts and powers are (as Lord Wilberforce demonstrated in McPhail v Doulton) a good deal less significant than the similarities. The tide of Commonwealth authority, although not entirely uniform, appears to be flowing in that direction.

67. However the recent cases also confirm (as had been stated as long ago as In re Cowin in 1886) that no beneficiary (and least of all a discretionary object) has any entitlement as of right to disclosure of anything which can plausibly be described as a trust document. Especially when there are issues as to personal or commercial confidentiality, the court may have to balance the competing interests of different beneficiaries, the trustees themselves, and third parties. Disclosure may have to be limited and safeguards may have to be put in place. Evaluation of the claims of a beneficiary (and especially of a discretionary object) may be an important part of the balancing exercise which the court has to perform on the materials placed before it. In many cases the court may have no difficulty in concluding that an applicant with no more than a theoretical possibility of benefit ought not to be granted any relief.

68. It would be inappropriate for the Board to go much further in attempting to give the High Court of the Isle of Man guidance as to the future conduct of this troublesome matter. But their Lordships can, without trespassing on the High Court's discretion, summarise their views on the different components of the appellant's claims:

    (1) It seems to be common ground that during Mr Schmidt's lifetime substantial distributions were made for his benefit, all or most by allocation of funds to the two companies (Gingernut and Petragonis) which were regarded as being (in some sense) Mr Schmidt's. The appellant as Mr Schmidt's personal representative does not accept that these funds have been fully accounted for. His contention is that in respect of allocated funds Mr Schmidt ceased to be a mere discretionary object, and became absolute owner. On the face of it the appellant (as personal representative) seems to have a powerful case for the fullest disclosure in respect of these funds.

    (2) The appellant as personal representative would also, on the face of it, have a strong claim to disclosure of documents or information relevant to the issue whether, but for breaches of fiduciary duty (such as for instance overcharging) more funds would have been available for distribution to Mr Schmidt, and would or might have been allocated to him in practice. The Board express no view whatever as to whether the appellant has a case for overcharging or any other breach of fiduciary duty. But claims of that sort have been put forward in the 1998 proceedings, and the possibility must be noted in order to make the position clear.

    (3) As regards the appellant's personal claims under the Angora Trust since his father's death, his status as beneficiary of any sort depends on the issue of construction discussed at paras 28-32 above.

    (4) As regards the Everest Trust, the appellant is (see para 33 above) a possible object of the very wide power in clause 3.3, but an object who may be regarded (especially in view of the Everest letter) as having exceptionally strong claims to be considered.

69. Their Lordships will therefore humbly advise Her Majesty that the appeal should be allowed, the order of Deemster Cain restored and the matter remitted to the High Court of the Isle of Man for further consideration in the light of the Board's judgment.

70. Rosewood must pay the appellant's costs before the Staff of Government Division and before the Board. The High Court will determine whether (and if so to what extent and out of which funds) Rosewood should be entitled to reimbursement of those costs out of the trust property. The appellant does not contend that Rosewood acted unreasonably but does contend that in this litigation it has in substance acted for its own benefit (see Order 48A, rules 3(3) and 6(2) of the High Court rules).