Hostname: page-component-74656f9884-skzmx Total loading time: 0 Render date: 2026-11-05T04:15:32.802Z Has data issue: false hasContentIssue false
  • English
  • Français

Justifying (or Not) the Office of Trusteeship With Particular Reference to Massively Discretionary Trusts

Published online by Cambridge University Press:  22 July 2021

James Penner
James Penner*
Affiliation:
James Penner, Kwa Geok Choo Professor of Property Law, National University of Singapore, Singapore. lawjep@nus.edu.sg

Extract

In a recent article2 I examined the nature of private law offices, and here I extend that analysis to consider the trustee-beneficiary relationship where the trust is of a kind that Lionel Smith has called a “massively discretionary trust.”3 After considering some of the problematic legal features of such trusts, I shall probe the morally problematic features that these trusts present.

Information

Type
Research Article
Information
Canadian Journal of Law & Jurisprudence , Volume 34 , Issue 2 , August 2021 , pp. 365 - 390
Copyright
© The Author(s), 2021

Let me tell you about the very rich. They are different from you and me.

… F. Scott Fitzgerald Footnote 1

1. Introduction

In a recent articleFootnote 2 I examined the nature of private law offices, and here I extend that analysis to consider the trustee-beneficiary relationship where the trust is of a kind that Lionel Smith has called a “massively discretionary trust.”Footnote 3 After considering some of the problematic legal features of such trusts, I shall probe the morally problematic features that these trusts present.

2. The Private Law Office of Trusteeship

A. Representation

In order to understand how private law offices work, we need first to understand the idea of representation and the way representation can solve problems.

The relationship of representation exists when one person is empowered to act for/for the benefit of, or on behalf of, another.Footnote 4

Trust and fiduciary law provide the classic examples of the representor acting for or for the benefit of another. The typical family trustee is a classic example. The classic example of one person acting on behalf of another is, of course, the agent. We could call this ‘representation by attribution’: the agent’s acts, entering into contracts, performing services that discharge the duties of their principal, and so on, are attributed to their principal. Note the difference with trustees. A trustee acts in their own name, albeit for the benefit of the beneficiaries, but not on behalf of or in the place of the beneficiaries. The trustee’s acts are not attributed to the beneficiaries, though these acts are for them, i.e. for their benefit.

There are two basic cases in which the law provides for relationships of representation: sometimes persons need to be represented, and in other cases it is in their interest that they are represented, irrespective of need. The most obvious case of a need for representation arises when a person lacks legal capacity, as with children or the otherwise incapacitated, e.g. by mental infirmity. There can be different representative solutions to this problem. With respect to a representee’s assets, one solution is the trust. In the case of an adult who loses capacity, a common alternative representational relationship can be provided by the grant of a ‘continuing’ or ‘enduring’ power of attorney over their assets (to which they retain title). We might say that these two alternatives, the trust and the power of attorney, are functional though not legal equivalents.Footnote 5

Besides attending to need, private law permits private individuals to employ relationships of representation to achieve valuable purposes. The most obvious case where there is not, strictly speaking, any need for representation, but it is very much in the interests of individuals that the law provides for such representation, is the case of agency. A principal can, in a sense, extend their own ‘agency’ (in the philosophical sense) by acting through agents. The same sort of justification applies to the law of trusts as well, where it facilitates certain arrangements, again extending the ‘agency’ of settlors.Footnote 6 In the family trust context, these arrangements allow the settlor to ‘structure’ the gift of their property in ways not allowed at common law, as in the simple trust “to Nicola for life and then to Ben and James in equal shares.”

B. The Essential Features of a Private Law OfficeFootnote 7

Private law offices have the following essential features:

  1. 1. An office is a legal position, consisting in a package or constellation of rights, duties, powers, and liabilitiesFootnote 8 which a personFootnote 9 can undertake.

  2. 2. There will be procedures for appointing a first officeholder to the office, and procedures for appointing successive officeholders.

  3. 3. Whilst the originator of the office will typically have powers to shape the office, typically also there will be limitations on these powers set by the general law. To use terminology popular in the US, some of the rules governing offices will be ‘defaults,’ others ‘mandatory.’

  4. 4. Offices fulfil one or more representative functions, as set out above.

  5. 5. Offices have primarily instrumental, not intrinsic, value. The officeholder is not required to establish any sort of intrinsic personal relationship with the representee(s). This is why any person who is sui juris is in principle capable of assuming an office. Of course, different individuals will be better- and worse-qualified to assume the office.

  6. 6. There can be joint officeholders, as in the case of joint trustees.

  7. 7. Negatively, a person is not appointed to an office just because they undertake to discharge a representative function, as we saw in the case of agents.

A trustee is clearly an officeholder on this definition. ‘A trust will not fail for want of a trustee’ is a principle of trusts law.Footnote 10

Let me conclude this section with a normative claim, which will figure heavily in the discussion in Part 5, below. Offices are only morally legitimate in so far as a moral justification for the representation provided by the office is. Where the underlying ‘problem’ the office is supposed to solve is no longer, or never was, a genuine problem requiring representation by an office holder, then the office should come to an end or should never have been created in the first place.

And let me point out how powerful a right it is for a settlor to be able to create an office of this kind. It is not at all obvious why anyone, as a matter of private law—that is, as a matter of purely interpersonal moral relations—should be able to pray in aid the law to protect, preserve, and enhance a structure which involves the necessary participation of others, even accepting the fact that an initial or a replacement trustee is not coerced into the role, and they may accept to do so only on the condition that they are remunerated, etc. To my mind the only justification of this willingness of equity to find a replacement trustee is that the settlor’s creation of a trust is regarded as merely them exercising the powers of ownership to make a structured gift of their own property and they, the settlor, are willing to bear the cost of that gift, not only in terms of the property that is given, but also in terms of the costs of making that gift, that is, bearing themself those costs in the sense that part of the trust assets they settle on trust will be expended in the trust’s administration, trustee’s fees, expenses, and so on. So long as the purpose of this structured gift is more or less commendable and the settlor bears the costs, the law’s allowing them the benefit of creating a voluntarily undertaken office for its administration seems justified. The trouble arises when the structured gift is not commendable, or is positively pernicious.

3. Traditional Limitations on Private (Non-Charitable) Trusts

Whilst I have referred to the settlor’s ‘freedom of trust,’ their power to create any sort of trust they wish, there have always been common law and statutory limits on this freedom, which are familiar to all trust scholars, including limits on the duration of trusts,Footnote 11 the invalidity of ‘illegal’ trusts (such as trusts to defraud creditors), and so on.Footnote 12 I shall just say something about two of these limitations.

A. The Ban on Private Purpose Trusts

In virtue of what is called the ‘Beneficiary Principle,’ a private trust must have personsFootnote 13 as objects, that is, persons for whom the trustee must apply the benefit of the trust assets.Footnote 14 This rule can be justified as a matter of the ‘logic’ of the trust, or on moral grounds. As to the former, because non-charitable trusts are private law arrangements, for them to work they have to instantiate genuine jural relations between individuals. But a ‘purpose’ is not a person, and can have no rights against the trustee to ensure that they devote trust assets to its fulfilment, any more than a purpose could have contractual rights against someone. Of course, one might say that we could appoint someone, an ‘enforcer,’ to give the corresponding rights to. Such a creature has been created by statutory purpose trusts legislation in some jurisdictions. But for various reasons which need not detain us, it seems pretty clear that an enforcer cannot be granted such rights in accord with the principles of equitable trusts law doctrine.Footnote 15 As to the latter, moral ground, Smith raises an objection which I have not come across before but which seems entirely sound.Footnote 16 In fact, on reflection it is almost obvious, and Smith deserves thanks for bringing it to our attention. The thought is this: as an individual with assets, you can devote them to any (legal) purpose that you want; of course, as a sui juris economic actor this is subject to the claims of your creditors in the sense that you have to pay your debts, and if you become insolvent then that will end (temporarily) your pursuit of those purposes, for now you will have no available assets to pursue them with. But if private purpose trusts were allowed you could insulate your assets from your creditors, since placing assets on trust removes them from your beneficial ownership. As Smith asks, “Why should a person be able to affect assets to some private purpose, while at the same time insulating those assets from his creditors?”Footnote 17 There are, I think, no good answers to this question.Footnote 18

B. The Principle in Saunders v Vautier Footnote 19

The principle may be stated as follows: whenever a beneficiary with an absolute interest under a trust is sui juris, i.e. of full age and sound mind, they may call for the trust property that represents that interest, and the trustees are obliged to transfer the title of it to them; if they are a sole beneficiary, this will result in the complete collapse of the trust. For example, if a settlor creates a trust under which their son is to receive the income of property until he is aged 30, at which time he is to receive the capital, that son can ‘call for’ the trust property, i.e. demand the trustees transfer the legal title to the trust property to him, as soon as he reaches the age of 18.

This limitation upon the settlor’s ‘freedom of trust’ can be justified in two ways. First, there might be something of an ‘anti-trust’ justification: while it is fine to empower owners to create structured gifts of property where this is essentially the only means of giving the benefits of property, as for example when money is provided for minor children, this power should not be used to allow an owner to control their beneficiaries when they are fully competent to look after themselves. If you give property to someone, you naturally take the risk that they will use that property in ways that are foolish or which otherwise might defeat your hopes. But that is the price of treating people, including donees of property, as responsible, autonomous individuals. The law of trusts should not, therefore, allow settlors to treat sane adults as children, and so the principle of Saunders v Vautier reflects the law’s desire that all individuals, once sui juris, should be treated as capable of running their own affairs, including their rights over property. Another way of putting this point is that a settlor is not allowed to use a trust to delay a person coming into their majority, extending their infancy as it were.

The second justification is related, and concerns the idea of equitable ownership. In the eyes of equity, the beneficiaries are the owners of the trust property, not the settlor. They have the rights against the trustee, and must enforce the trust themselves. When they reach full age, in essence the trust is in their hands. They can enforce their rights against the trustee or not, may consent to the trustee’s acting outside the terms of the trust, i.e. doing what would otherwise be a breach of trust, and may participate in the variation of the terms of the trust. The settlor has no say in any of this. Thus they are (in theory) in full control of the property via the office of the trustee. But, if that is so, why cannot they do with their property what they like, as can any other full owners and, in particular, take the property out of the trust completely if they so desire?Footnote 20

All of the preceding limitations on the settlor’s freedom of trust aim to prevent moral ‘bads,’ defrauding creditors, controlling your property long after you are dead, treating adults like children, and so on. Not surprisingly, they do not point to any moral ‘goods’ that trusts provide. One such good, as we have seen, is the facility of looking after the property of those lacking capacity, and another such good is allowing ‘structured gifts’ which allow distributions unavailable without the trust.

4. Massively Discretionary Trusts and Their Discontents

A. Discretionary Trusts, Powers of Appointment, and the Rise of Massively Discretionary Trusts

What is now called a ‘discretionary trust’ developed in the 19th century. A very simple example is “to my spouse Jessica for life, and then to my children Tom, Dick, and Mary, in such shares as my trustee shall in their absolute discretion decide.” Although tax considerations played an important role,Footnote 21 the main ‘internal’ trusts law purpose in endowing the trustee with such a discretion was flexibility.Footnote 22 Tom, Dick, and Mary might be in very different financial circumstances at the time of Jessica’s death, and the discretion allowed the trustee to adjust their portions accordingly. But it is to be noticed that this discretion was ‘one-off.’ At the time of Jessica’s death, each of the children would get their share and that would be the end of the matter. They received what they received, and it was theirs to use as their own.Footnote 23

Another development was the provision in well-drafted trust instruments allowing trustees to ‘apply’ the trust assets in any way the trustee, in their discretion, judged would benefit the beneficiary. In my trusts text I provide the following example:

Suppose that the trustee thinks it would make sense for Lionel to have a new car. There are three ways in which this objective might be achieved. The trustee could just give Lionel money from the trust funds to buy a car. Or the trustee could use trust funds to buy the car for Lionel, that is pay the car seller on Lionel’s behalf, so that Lionel acquires title to the car. Or the trustee could use trust funds to buy a car in his own name, ie taking title to the car as a trust asset, and then licensing the car to Lionel to use. This would be the safest way of dealing with those irresponsible or feckless Lionels of this world whose existence or perceived existence preys upon the imaginations of fretful settlors—if Lionel had a bad gambling habit, for instance, keeping the car as trust property would prevent Lionel from selling it and blowing the proceeds playing online poker.Footnote 24

Whilst this is something of a toy example, it illustrates a broader phenomenon under modern wealth management trusts. It is not uncommon for trust assets to include residential properties (often through holding companies—there can be very complicated webs of interlocking trusts and companies under modern wealth management structures) which are licensed to certain of the trust objects for their occupation. Similarly, the trust can hold all the shares in a private company under which company credit cards can be issued to discretionary objects as a way of funding their personal living expenses. In essence, then, we have a structure in which objects can be benefited, without making them personally wealthy in the traditional sense, since they are not strictly speaking entitled to the benefit of the trust assets, which are made available to them, but not as theirs either in law or in equity.

Having got these rudimentary points under our belt, we now come face to face with what Lionel Smith calls the modern ‘massively discretionary trust’ (MDT).Footnote 25 For clarity of discussion, following Smith, I shall adopt the following terminology. Someone entitled under a trust provision to a distribution, even if that entitlement is defeasible, is a ‘beneficiary.’ ‘Defeasible’ indicates that the ‘beneficiary’ may not actually benefit. Going back to our simple example of Jessica, Tom, Dick, and Mary, if the trustee distributes all the capital to Mary, Tom and Dick will get nothing. Yet they are still beneficiaries in the sense that the trustee owes them the obligation to invest the trust assets properly, keep the trust accounts, and so on, and more importantly, to properly consider their interests even if, in the end, the trustee thinks Mary’s interests are greater and in fact prevail. In contrast, someone who might take under a power of appointment, which the trustee has no obligation to exercise, is an ‘object.’

What are the characteristics of an MDT? They have the following structure: first, a period for the duration of the trust is set out, typically the longest period allowed by the law governing the trust. The end of this period will be called something like the ‘Vesting Day.’ The trustee will have a power to declare any earlier day as the vesting day. On this day the trustee will have the obligation to distribute any remaining trust assets to what are often termed ‘default,’ ‘residuary,’ or ‘final’ beneficiaries. This is a genuine trust of the capital. Prior to the vesting day, the trustee will have the power to accumulate trust income and add this to the trust assets. Second, the trustee will also have a range of powers to distribute the trust assets to, apply them for the benefit of, or resettle them on new trusts for, named objects or classes of objects (e.g. the settlor’s children or grandchildren) prior to the vesting day. Third, the trustee will have a power to remove any individual from the class of objects, and a power (a ‘hybrid’ or ‘intermediate’ power) to add to the class any person in the world except a few named individuals (typically the trustees themselves or the settlor).Footnote 26 Finally, and perhaps most importantly, it is intended that the trust fund will be exhausted by the exercise of these powers prior to the vesting day, hence the use of the term ‘default beneficiary’ in some trusts to describe the vesting day beneficiaries. One might as well describe them as ‘dummy beneficiaries.’ As Smith succinctly explains, MDTs are:

trust structures in which the trustees’ dispositive discretions do not merely qualify the beneficial interests, but effectively displace them, one might even say overwhelm them.Footnote 27

B. Problems with Massively Discretionary Trusts

(i) Validity Problems

Regarding the validity of MDTs, Smith argues that where the only true beneficiary of an MDT is a trust for charitable purposes or charitable trusts but there are powers to appoint the trust assets to individuals, such a trust is invalid because not exclusively charitable.Footnote 28 I think Smith is right about this, but let me just offer a few explanatory points. Whether such an MDT is valid or not depends on how the trust is structured. We are assuming that the intention is that the trust assets will all be depleted by the trustee’s exercise of their massively discretionary dispositive powers, but the ‘validity’ point turns on how the trust of the income is structured.

I would suggest a trust with the following structure is perfectly valid:

  1. 1. A trust of the income for individuals, whether discretionary or not;

  2. 2. Notwithstanding 1., a power to accumulate the trust income and add it to the trust funds;

  3. 3. A discretionary trust of the trust funds for distribution on the ‘Vesting Day,’ such as “on trust for such UK charities as my trustees shall at their absolute discretion select.”

This structure is valid because there is a trust of the income for individuals, and a charitable trust of the capital to be distributed on the ‘Vesting Day.’ But consider this structure:

  1. 1. A duty to accumulate the trust income and add it to the trust funds;

  2. 2. Notwithstanding 1., a power to distribute the trust funds to individuals;

  3. 3. A discretionary trust of the trust funds for distribution on the ‘Vesting Day,’ such as “on trust for such UK charities as my trustees shall at their absolute discretion select.”

This structure, I submit, is invalid because the entire trust of the assets is for the capital beneficiaries, the charities—a duty to accumulate is simply a duty to add the income to capital, so such a duty makes the capital beneficiaries also the income beneficiaries. The powers under 2. then give the trustees the power to appoint the trust assets away from the charitable capital beneficiaries. The trust is then not wholly charitable, and it fails.

It might seem that this is a very ‘technical’ distinction to draw. After all, as Lord Wilberforce said in McPhail v Doulton,

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.Footnote 29

Leaving aside the perhaps captious point that laymen and logicians are not lawyers, part of what facilitative areas of private law such as the law of trusts do is to provide techniques or ‘technical’ means to create interpersonal jural relations to achieve purposes, and these interpersonal jural relations have consequences. For example, in this case, it is trite law that where there are more than two trustees and they disagree about exercising a power which departs from what would otherwise be their duty, they must comply with their duty.Footnote 30 Thus, if trustees disagreed about exercising their power of appointment in 2. of this structure, they would be required to accumulate the income. But much more important, as Smith rightly emphasises again and again, the very structure of a valid trust hangs on there being identifiable beneficiaries whose interests exhaust the value of the trust assets. And here they are the charities.

(ii) Enforceability Problems

More than twenty years ago, Hoffmann J (as he then was) discussed the enforceability of MDTs in this way:

It was a typical offshore trust and since a number of these trusts feature in these proceedings, it may be useful to describe their general characteristics. The named settlor is usually a stranger to the transaction who has been requested to lend his name to a document which records (in many cases truthfully) that he has provided a nominal sum as the initial trust fund.… The name of the person providing the assets subsequently added to the fund, whom I may call ‘the real settlor’, does not appear.… The trustees are given a wide discretion to pay income and capital to the beneficiaries, but the persons actually contemplated as beneficiaries are not necessarily named in the document. Instead, a local charity or some other person will be named and the trustee given a discretionary power to add to the list of beneficiaries.… Sometimes the trustee will be given a ‘letter of wishes’ by the real settlor saying whom he wishes to benefit. Sometimes the real beneficiary, who may be the real settlor himself, will remain a matter of oral understanding between him and the trustee. These trusts therefore reveal very little and depend upon trust and confidence reposed in the trustee to give effect to expectations and understandings which may not be legally enforceable. The business of providing offshore financial services requires a reputation for probity, efficiency and discretion in executing such trusts in accordance with the confidential wishes of clients.Footnote 31

In a chapter which seeks to reassure the drafters of MDTs that these confidential expectations are indeed legally enforceable, Richard Nolan argues, citing cases such as Schmidt v Rosewood Trust Limited,Footnote 32 that courts, relying on their inherent jurisdiction to administer trusts, will allow objects to enforce MDTs.Footnote 33 The reader will recall that in Schmidt the Privy Council granted restricted rights to information concerning the trust to an object, indeed relying upon its inherent jurisdiction.Footnote 34 Smith is less sanguine.Footnote 35 To summarise Smith’s position roughly and briefly, since a trustee owes obligations regarding the trust property only to the beneficiaries and not to objects, it is the beneficiaries who have the corresponding rights to enforce the trust, and consequently, the right to information from the trustee concerning the latter’s administration of the trust. This is a matter of the logic of the trust, and Smith persuasively argues that there is authority for this position,Footnote 36 despite Schmidt, and he points out that Schmidt is not binding in any court apart from those of the Isle of Man.Footnote 37 I shall not adjudicate this dispute here except to point out one feature of Nolan’s discussion. Nolan accepts that not any object of an MDT should have standing to enforce the trust:

The court can limit the claims it hears to those made by an object who has a substantial possibility of seeing benefit from the trust, … one of ‘the true intended beneficiaries’ of the settlement.… But the court need not even entertain claims brought by those who have no substantial expectation of benefit. Who are the ‘true intended beneficiaries’, and who are not, simply turns on the evidence, and the court’s evaluation of that evidence. And there will often be a letter of wishes which makes the distinction clear, something that the court can always demand to see.Footnote 38

Somewhat oddly, it seems that Smith and Nolan agree on the basic principle that only true beneficiaries should be entitled to enforce the trust. Where they differ is on whether as a matter of legal principle this should be determined on evidence entirely extrinsic to the trust instrument which sets out the trust’s purported terms, which evidence will almost certainly directly contradict the plain meaning of those terms.Footnote 39 We shall return to the issue of letters of wishes, but one thing about the flexibility of MDTs is that they may not even have true terms if these are determined dehors the trust instrument because there is every likelihood that the ‘true intended beneficiaries’ may change, often radically, over time. To take the most common example from the cases, the spouse of the settlor may well be a ‘true intended beneficiary’ just until they file for divorce. Finally, Schmidt gives almost no practical guidance on the ‘evidence’ upon which such determinations are to be made.

This dispute may reveal a profound difference between Nolan’s and Smith’s particular ‘theories’ of the power to create a trust, especially concerning the ‘certainty’ requirements.Footnote 40 I shall not pursue the matter here, but it raises the fundamental question: what sort of power is it to create a legal structure whose defining legal rights, duties, and powers can vary enormously from time to time, often suddenly, and which can only be ascertained by reference to actions and events entirely extrinsic to its purported terms?Footnote 41 It would, I submit, be perfectly acceptable for future courts to act upon the thought Lord Walker entertained just before resorting to the court’s inherent jurisdiction:

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.Footnote 42

(iii) The Problematic Status of Letters of Wishes

The matter in issue here is obvious. In the case of MDTs, on its face the trust instrument allows the trustees to choose to apply the trust assets to anyone they choose, even themselves sometimes.Footnote 43 Practitioners thus developed the use of letters of wishes to provide some guidance to the trustees. Besides pointing out a number of nefarious consequences of this practice,Footnote 44 Smith raises the most profound objection in these terms: “Paradoxically, the settlor’s wishes, which are not binding and are not terms of the trust, are controlling.”Footnote 45 The paradox is obvious—how can the settlor’s power to determine (‘control’) the distribution of the trust assets not displace the actual terms of the trust, or rather, be part of the terms themselves?

I find Smith’s point here to be essentially on par with his disagreement with Nolan canvassed earlier: on what theory of the trust are the ‘real’ trust terms, which identify the ‘true intended beneficiaries,’ held not to be terms of the trust? But let me throw in a further problem, which Smith touches upon, but does not explore in detail.Footnote 46 It appears that it is commonly understood that letters of wishes can be ‘updated’ by the settlor from time to time, or one letter substituted by a new one at the settlor’s choosing.Footnote 47 It seems to me, however, that this violates a basic principle of trusts law. Let me explain.

A trust is a disposition of property. Once it occurs, unless the terms of the trust grant powers to the settlor, they have no right qua settlor to enforce the trust or dictate to the trustees how they exercise their discretions,Footnote 48 for the simple reason that the property is no longer theirs, but belongs to the beneficiaries.Footnote 49 Another way of putting this point is that you can only be a settlor of a trust once. After you have settled the property, you are not entitled to add or delete terms. And the traditional understanding of the proper purposes doctrine regarding the exercise of a trustee’s powers is that these purposes are determined at the time the settlement is created.Footnote 50 Those purposes cannot be ‘updated’ by the settlor from time to time because like the trusts terms themselves they are, unless provision is made to the contrary, fixed at the time the trust is created.

But, of course, the settlor’s actual intentions are determinative of the terms of the trust whether or not they appear in the written trust instrument. If, because of the advice they have been given, it is the settlor’s true understanding that they can, via a letter of wishes, dictate from time to time what the trustees must do, this is just to say that the trust assets are held to their order. Not only is this a general power to resettle or amend the trust terms, but also, of course, a power of revocation. This is not a fiduciary power. The settlor can exercise it just as they like without being concerned in the least with the interests of others. In consequence if, for example, the settlor becomes insolvent or is subject to marital property proceedings, it is a beneficial asset of theirs available to their creditors or spouse as the case may be.Footnote 51 So be it.

There are other concerns with MDTs besides possible invalidity and unenforceability. But I think we now have sufficient to move on to the next topic of discussion, which is whether or not massively discretionary trusts can be justified as a matter of moral principle.

5. Are Trusts, in Particular Massively Discretionary Trusts, Morally Justified Legal Relationships?Footnote 52

It is a old joke about academic papers that every author says something like “this paper fills a gap in the literature which has been strangely neglected and has remained unexplored despite its undoubted importance.” But in this case, that of the moral-philosophical justification of the trust relationship, I think this paper makes a start, at least, in filling a gap in the literature which really has been strangely neglected and has remained unexplored despite its undoubted importance. I have been a trusts scholar for almost thirty years but it only occurred to me recently that, very much unlike the case of the law of contract, the law of torts, and the law of property, each of which almost drowns in its own moral-philosophical literature seeking to explore their justifications (or lack thereof), there is no similar literature concerning the trust. More recently, equity more generally, the law governing fiduciaries, and the law of restitution/unjust enrichment have all seen a growing wealth of similar scholarship.Footnote 53 But there is almost no moral-philosophical literature on trusts law, and I say ‘almost’ not because I can point out an article here or an article there, but out of an abundance of caution—there may be something out there that I have simply failed to notice. There are, of course, articles which interrogate what might be called the ‘external’ aspects of trusts, such as the way in which they can be used to avoid legitimate taxes,Footnote 54 their distributive effects,Footnote 55 and the way they might be seen to reflect but obscure relations of private power.Footnote 56 But what I am interested in here is the ‘internal’ relationship of settlor-trustee-object, and as I say, I don’t know of any moral philosophical analysis of that save, perhaps, recent work by Ripstein who offers a Kantian perspective upon relationships in which one person is entrusted with another person’s ‘means,’ that is, private rights. I shall attend to that work shortly.

Of course just days after I wrote this paragraph, Steve Hedley’s Private Law Theory website announced the impending publication of Mark Bennett’s and Adam Hofri-Winogradow’s “The Use of Trusts to Subvert the Law: An Analysis and Critique.”Footnote 57 And it is such a contribution, which I welcome, but they tackle a different issue than the one I am addressing here.

A. Trusts as Structured Gifts

I said above that the traditional family trust can be conceived of as a structured gift, so it makes sense first to look at how gifts can be justified from a moral-philosophical perspective. In a recent book I reasserted the claim that the different ways in which an owner of property can deal with their title, to license others to do what would otherwise be a trespass, to give the property away, to leave it by will, and so on, each require their own, perhaps related, justifications.Footnote 58 This followed the analysis I set out in The Idea of Property in Law some 23 years ago, so I shall reprise that here.Footnote 59 First, however, let me set out some aspects of ‘family’ trusts that I shall not be exploring.

First, many family trusts are of course testamentary, but I shall not explore the moral-philosophical literature on the justification of inheritance and the power to make post-mortem gifts by will, which has a long tradition.Footnote 60 This is not because it is irrelevant to the justification of such trusts, of course, but simply because it is orthogonal to the investigation of the ‘internal’ settlor-trustee-object relationship. Second, I shall not be looking at trusts which respond to the need of an individual to be represented in respect of their beneficial assets, such as trusts for minor children or those otherwise incapacitated. The justification for such trusts is obvious, although this is not to say that a trust is the only or even typically the best way of achieving this end. Finally, let me acknowledge that substantial gifts of property, whether inter vivos or testamentary, trouble many philosophers on grounds of distributive justice, and this worry would clearly apply to structured gifts like trusts.Footnote 61 I share some of these worries myself but, again, they are orthogonal to my concerns here.

How, then, might one justify trusts conceived of as structured gifts? In The Idea of Property in Law, I sought to do so as an exploration of what can be called the ‘social thesis,’ the thesis that human beings are social creatures and, in particular, how their individual lives go in terms of success and failure is inextricably linked with how the lives of others go, most obviously in the case of family and friends.Footnote 62 The first step to understanding gifts is the power to share one’s property, to use one’s property with others, thereby improving both the owner’s and others’ lives. Gifts are, on this account, just an extension of sharing. In certain cases, one can only achieve the result of enhancing the lives of others one cares about by transferring title to property to them so as to allow them to make more and/or better choices, thus enhancing their autonomy. The idea is that leading an autonomous life in which one makes authentic choices is one very important aspect of a life’s going well,Footnote 63 and for those I care about, I can enhance their lives in some cases by providing them with the wherewithal to do so by making gifts to them. Can this justification extend to trusts?

It seems to me that in many cases it can. A simple trust for one’s surviving spouse for life and then to one’s children (whether in equal shares or under a discretionary trust) seems easily to fall into the class of gifts whereby the donor, here the settlor, aims to provide resources to those who matter to them, so as to enhance their autonomy. But the minute one says this one realizes that considerations of autonomy place a justificatory limit upon the size of these structured gifts. At a certain point no amount of further resources will do anything to enhance one’s genuine autonomy. Moreover, there is a second consideration which arises in the case of discretionary trusts and powers. Depending upon how trustees exercise their discretions, the distributions they make may have the effect of rendering their objects less autonomous, as perhaps in the case of Lionel in the example above.Footnote 64 We shall return to both points when I discuss MDTs specifically below. First, however, it is time to explore what Ripstein’s Kantian analysis may have to offer.

B. Trusts as Kantian Powers Over Another’s Means

(i) Status Rights

As is well known, Kant explained private right in terms of the innate right to our person, essentially our physical bodies and our reputation (our good name), and to ‘acquired rights’ of three kinds: (1) property rights in the external things of the world, objects and land; (2) contractual rights; and (3) status rights.Footnote 65 As I shall explain in the next section, I think the trust is an interesting hybrid of (1)- and (2)-type acquired rights, a hybrid I am not sure Kant or a Kantian would find licit under the structure of private right. Be that as it may, here I shall pursue the thought that because the trustee is an office holder they have a kind of status with regard to the objects of the trust, akin to a Kantian type-(3) acquired right. Although this has not been explored in detail by Ripstein, as we shall see he seems to endorse some suggestion along these lines.

In the Doctrine of Right, Kant discusses three examples of ‘status’ relations: marriage, the relation of parent to child, and the relation of master to domestic servant. Each is framed as one kind of ‘Right of Domestic Society,’Footnote 66 and as ‘Rights to Persons Akin to Rights to Things.’Footnote 67 I shall ignore marriage simply because it seems very far from the matter—trusteeship—we are discussing. In particular Kant is firm in his view that in so far as the marriage relationship is concerned,Footnote 68 the relationship is reciprocalFootnote 69 and the parties are equals,Footnote 70 which is not the case either for masters and servants or parents and children, and since the trustee-object relationship—at least in so far as it is properly characterised as a status relationship—is not a relationship amongst equals either, it seems sensible to look at these latter two types of status.

Regarding the acquisition of these rights, Kant states, “a couple acquires children; and a family acquires servants.—Whatever is acquired in this way is also inalienable and the right of possessors of these objects is the most personal of all rights.”Footnote 71

‘Parental Right’ arises from the duty of parents to the children they have, by procreation, brought into being, and this duty reflects the child’s innate (not acquired) right “to the care of their parents until they are able to look after themselves.”Footnote 72 Kant also puts the point this way:

From this duty there must necessarily arise the right of parents to manage and develop the child, as long as he has not yet mastered the use of his members or of understanding: the right not only to feed and care for him but to educate him, to develop him both pragmatically, so that in future he can look after himself, and morally, since otherwise the fault for having neglected him would fall on the parents.Footnote 73

In relation to the idea that the right is a right to a person akin to a right to a thing, Kant says the parental right is not ‘just’ a right to a thing; a child is not the property of his parents, so the parental right is not alienable.Footnote 74 But their right is akin to a right to a thing in the sense that they have possession of the child, and he can be brought back against his will if taken ‘possession of’ by another or if he has run away: parents are “justified in taking control of them and impounding them as things (like domestic animals that have gone astray).”Footnote 75

As regards domestic servants, the right is framed as the “Right of a Head of the Household.”Footnote 76 His comments on servants are quite sparse. Domestic society (leaving aside marriage) is a society of unequals; servants must obey the commands of the head of the household.Footnote 77 As with parental right and children, the head has a right akin to possession—“for he can fetch servants back and demand them from anyone in possession of them, as what is externally his, even before the reasons that may have led them to run away and their rights have been investigated.”Footnote 78 Servants, however, are not serfs or slaves. The relation to the head is founded upon a contract between them, and so whilst the head is entitled to “use” his servants, he cannot treat them as if he owned them.Footnote 79 He can use them, but not “use them up.”Footnote 80

It seems to me that neither the parental right nor the head of the household’s right have anything much in common with the office of trustee. In the first place, trustees do not have a right in the bodies of their objects akin to parental or head of household right. What trustees hold is title to property under terms (of the trust), which property must be applied for the benefit of the objects. Secondly, as Kant says, status rights of this kind are of the most personal kind—they cannot be alienated. But the hallmark of a trustee’s position is that it can be alienated just because it is an office. Of course a trustee cannot ‘sell’ their office to a successor trustee, nor can they, whilst they remain trustee, delegate their dispositive duties.Footnote 81 The removal and appointment of a successor trustee must be done only with the best interests of the trust’s objects in mind. That is why the powers of removal and appointment are fiduciary powers.Footnote 82 But the trustee’s relation to the objects, or at least competent adult objects, has nothing like the personal quality of parents to children or the head of a household to their servants. Thirdly, and relatedly, I would submit that whilst the trustee holds an office, a parent or head of household undertakes a role.Footnote 83 On the distinction between an office and a role, Essert says this:

A role is also a position or status that a person can have, and that has a set of associated powers and duties. Common instances of roles are things such as sister, mother, teacher, or perhaps citizen. The difference is that an office is something that persists and has an independent existence, as it were, beyond the person who is filling it.… [T]he idea of filling an office gets at this notion: an office is something that can be vacated by one officeholder and then filled by a new one, and the office itself continues to exist independently even if there is nobody filling it during a transition. A role, on the other hand, is unique for each person who plays it. So sister is a role because only one person can play it, and each new person who plays the role creates it anew for herself.Footnote 84

The idea that a role is ‘unique’ for each person who has it can, I think, be spelled out with the following distinction between offices and roles. An office is dominated by norms—rights, powers, and duties of the office-holder—which are relatively ‘personality-poor,’ in the sense that they are not expected to vary in their possibility of performance in relation to considerations other than the office-holder’s competence in being able to carry out the role. We seek office-holders who are, in this sense, up to the job. And offices are jobs also in the sense that we primarily attribute only instrumental value to their performance. Roles, or at least the roles I am considering here, such as the role of being a friend, involves more than this, or at least our ideal of the role of a good friend does. First, performing the duties of a role and exercising its powers and so on has, to my mind, intrinsic value because friendship has intrinsic value. But, returning to Essert’s ‘uniqueness’ point, this makes the relationships ‘personality-rich’ in the sense that they inevitably concern the very personalities involved in them, if only because such roles involve identification and affective attitudes, and people self-identify and engage affectively in different ways.

This brings us to the concept of a fiduciary. In its most basic legal sense, fiduciary simply means ‘not personal’ in the sense that a right or power held by a fiduciary is held not for their own benefit, but for the benefit of another. In this sense all trustees are fiduciaries since they hold their rights and powers qua trustees for the objects of the trust.Footnote 85 But there is a narrower meaning of the term under which someone is a fiduciary if and only if they have discretionary powers that must be exercised on behalf of or for the benefit of someone else. On this narrower view, the fiduciary relationship is founded upon the fiduciary’s having powers, the exercise of which will involve their making judgments, judgments which require them to take into consideration only their principal’s interests. This is sometimes called the ‘discretionary theory’ of the fiduciary relationship. Footnote 86 It responds to the fact that the voluntary obligation (or ‘contract,’ loosely speaking), which underlies the relationship, is ‘incomplete’ in the following sense: the actual exercises of the trustee’s powers in making investment decisions regarding the trust assets, or in exercising a dispositive discretion under a trust, cannot be specified in advance under the objective terms of the trust.Footnote 87 Following Smith and Miller, we can call these powers ‘fiduciary powers.’Footnote 88 A trustee typically has such powers, and a trustee of an MDT has them in abundance. Does Kant’s discussion of the acquired status rights we have just considered, in particular the right of a parent, illuminate this relationship? Ripstein, for one, seems to think that it can.

I use the word ‘can’ here advisedly. As far as I can tell Ripstein’s discussion of fiduciaries as a kind of status rightholder is more of a suggestion than a sustained analysis.Footnote 89 Be that as it may, in trying to fit the law governing fiduciaries within status rights, Ripstein seems to be motivated by the thought that Kant’s theory of private right is comprehensive:

Those entitlements fall under private right, and cover the traditional categories of Roman private law, relations of property, contract, and status, which govern rights to things, to performances by other persons and, in special cases, rights to other persons. These categories provide a complete specification of independence between interacting persons…Footnote 90

So Ripstein’s move seems to be that we have to fit fiduciary law within this framework. As I shall argue in the next section, to my mind it fits more comfortably within contract, a right to the performance of another—the fiduciary—than it does within the category of status rights. Ripstein’s first gesture to relating a fiduciary relationship to a kind of status relationship is by way of a parallel, using a bailment as an example:

Relations of status are the mirror image of contractual relations, because in relations of status one person has possession of another but not the use of another person. Such relationships are possible when people interact interdependently but nonconsensually. The structure of this relationship parallels the situation when one person is in possession of another’s property: if I am repairing your car, I am allowed to take it for a drive to see if it is working properly, but not to take it to visit friends. To do so would be to use what is yours in pursuit of my purposes rather than your own.Footnote 91

Ripstein relates the position of a principal in a fiduciary relationship to the case of parents and children by focusing on the principal’s and the child’s nonconsensual reliance:

Relations of status enlarge the purposiveness of those in control of them by entitling them to make arrangements for others; because the power is nonconsensual, their purposiveness is restricted, so that they can only exercise that purposiveness on behalf of those in their charge. They also enlarge the purposiveness of those for whom others make arrangements. If I entrust you to manage my affairs while I am away, my purposiveness is secured because you make arrangements for me, in pursuit of my normal purposes. Your purposiveness is enlarged because you get to determine how my goods will be used, subject to the condition that they not be used for your benefit.Footnote 92

And a little further on:

Many other examples fit this structure. The legal relation between a fiduciary and a beneficiary is one such case. Where the beneficiary is not in a position to consent or decline to consent, or the inherent inequality or vulnerability of the relationship makes consent necessarily problematic, the fiduciary must act exclusively for the benefit of the beneficiary. It is easier for the fiduciary to repudiate the entire relationship by resigning than for a parent to repudiate a relationship with a child. But from the point of view of external freedom the structure is exactly the same: one party may not enlist the other, or the other’s assets, in support of ends that the other does not share.Footnote 93

On the inequality inherent in status rights, which presumably applies to the fiduciary relationship, Ripstein tells us:

A status right is a right to a person “akin to a right to a thing,” that is, a right to affect that person without his consent, and a corresponding obligation to avoid using that person. Relations of status are inherently asymmetrical, and so can only be made rightful by restricting the freedom of the right holder to act for the purposes of the other person.Footnote 94

Relating ‘acting for the purposes of the other person’ to the idea of consent, Ripstein says, “[I]n relations of status, in which one person makes arrangements for another, the arrangements are limited to those that the other person could have consented to, had he or she been in a position to.”Footnote 95

Two things to note here: first, I think that status rights are a poor fit with the actual legal fiduciary relationships the law recognises. In particular I would question whether such relations are generally nonconsensual in the relevant sense, and similarly whether they are generally asymmetrical in the relevant sense. Secondly, I do not think the fiduciary relationship should be modelled on status rights, in particular the parent-child relationship, for in many typical fiduciary relationships (think of commercial agency) the principal is sui juris, and the law should not treat sui juris individuals as if they were, or even are like, children. That, I hope to show, is one of the signal problems of MDTs.

(ii) Contractual Rights and the Trustee-Object Relationship

It seems to me that the trustee-object relationship and the fiduciary-principal relationship concern voluntarily undertaken obligations which fit well within Kant’s second type of acquired rights, i.e. contractual rights.

As we have seen, not all trustee powers are fiduciary powers. So in examining whether the trustee-object relationship or the fiduciary-principal relationship is better illuminated in terms of status-right or contract-right, we must treat each separately.

I think it would be fair to say that, given Kant’s obvious familiarity with Roman law, and his equally obvious lack of familiarity with the common law, it didn’t occur to him to try to explain the common law trust.Footnote 96 We are, in this sense, left to our own devices to try to figure out what kind of acquired right an object has under a trust.Footnote 97

An object is entitled to the performance of the trustee, to their carrying out the terms of the trust, no more and no less.Footnote 98 The trustee has two different kinds of duties and powers: administrative duties and powers, and dispositive duties and powers. Their administrative duties concern matters such as keeping the trust accounts and investing the trust property, paying relevant taxes, and so on. There is no difficulty in framing these duties as a kind of Kantian contractual duty. They are objective performances which the trustee owes the objects and the trustee discharges their obligations to them by performing. As to the trustee’s dispositive powers and duties, these involve the trustee laying out the trust funds for the benefit of the objects, and although as the example of Lionel above showed this may not involve direct payments to an object, these outlays are still transactional actions which discharge the trustee’s dispositive obligations. By ‘transactional’ I mean that these actions require the trustee to exercise their power of title over the trust assets to distribute all or some of those assets in ways which benefit an object. There is, in other words, so far as the content of the trustee-object relationship is concerned, only Kantian contract as a possible source of illumination. In particular, as a matter of the general principles of trusts law, there is no need and no room to invoke status-right. Whether it needs to be invoked in the case of MDTs is a point to which we shall have to return.

(iii) Contractual Rights and the Fiduciary-Principal Relationship

Fiduciary powers require the exercise of judgment by the fiduciary in good faith and with only the principal’s interests in mind. Does this feature take them out of what might be called ‘Kantian’ contract right and move them into the realm of status right?

The very fact that a fiduciary has a power which requires judgment regarding the best interests of their principal does not, I think, reflect anything like the discretionary ‘domestic society’ powers that Kant places under status right. So, for example, the power to invest trust assets is a fiduciary power, since it requires the exercise of the trustee’s judgment. But the objects’ interests, unless the trust terms specifies otherwise, are normally the objects’ best financial interests (the pursuit of which is basically what the power to invest is for), and this sort of judgment is clearly not very personality-rich, though it may concern balancing the different financial interests of different classes of objects.Footnote 99

The case is different, perhaps very different, in the case of a trustee’s dispositive discretions, in particular those found in an MDT. Such a trust seems not adequately to reflect the Kantian idea that the main aim of private law is that people, or at least sane adults, should be independent of the choices of others with respect to their means. So to those trusts we now turn.

C. Justifying (or Not) the Massively Discretionary Trust

The main positive feature of MDTs is their immense flexibility. But this must be weighed against the following negative features, which are interrelated.

  1. 1. MDTs do have a structure; there are trust terms, in particular the massively discretionary powers of distribution. But in another sense they are very unstructured because the trustees can in theory distribute the property in whatever amounts they choose to anyone. It is perhaps better to call these managed rather than structured gifts. The trustees, or whoever it is that actually decides on distributions, take on a much more role-like character.Footnote 100 That is, their unique personal judgments about the objects will loom large. Contrast this with the much simpler discretion in the case of a trust for a spouse for life, and then to the children in shares determined by the trustee. One would begin with a presumption of equality, on the basis that a settlor presumably is equally interested in the welfare of each of their children, departures from equality only being justified on the basis of fairly clear differences in the circumstances of the children at the time of the spouse’s death. Any sensible person could undertake this discretion. But in the case of an MDT, the field is wide open, and different distributors might take very different views of the relative interests of the objects. In short, the more a trust becomes massively discretionary, the more the distributor’s powers engage their unique, role-like characteristics. In this respect it may be the case that the actual beneficiaries of the trust may change markedly as the person who really determines where the distributions go changes.

  2. 2. A related concern is that these discretions give the trustee the power to extend a person’s infancy, treating them as less than a full adult. As Arthur Ripstein framed the point, MDTs can be seen as the use of trust form to create a Kantian parent-child status relationship.Footnote 101

  3. 3. Where the law allows for the duration of a trust to be very long, or perpetual, the donor-donee relationship between settlor and the objects becomes attenuated. That is, the intrinsic value in the gift relationship—that of the sharing of interests between donor and donee—cannot with any seriousness be contended to operate across distant generations. At some point the only intrinsic value that can be conjured is that which some people find in family dynastic relations. I am at least sceptical of any such intrinsic value.

  4. 4. As Ripstein has argued, one of the main ideas of private right is that it lets people know where they stand in relation to others, in particular in terms of what ‘means’ they have. Where the rules of law make the trustee’s considerations in exercising their dispositive discretions confidential (as is the case, generally, in commonwealth jurisdictions), or their discretions are much influenced by a confidential letter of wishes, where an object stands in relation to these means becomes uncertain, or hypothetical. For the objects the situation is also, arguably, demeaning. As David Hayton once put it, “No one likes being treated like a mushroom, being kept in the dark and fed you know what.”Footnote 102

  5. 5. As I mentioned above, it is not uncommon for trust assets to include residential properties which are licensed to objects for their occupation, and ways to fund their expenses, such that the objects of such a trust can be made literally legally irresponsible as a matter of property and insolvency law in regard to the means that they factually benefit from. I can see no intrinsic value in such an arrangement, and significant potential intrinsic disvalue.

It is of course a matter of judgment whether these disadvantages outweigh the great flexibility of MDTs, and I shall not try to decide that here. Rather, I have sought to identify these disadvantages and frame them within both a social thesis-autonomy theory of the justification of gifts and a Kantian theory of independence regarding one’s means.

6. Conclusion

Let me end with a diagnosis. Perhaps the reason why there is no substantial moral-philosophical literature on the justification of the settlor-trustee-object relationship is that until the rise of MDTs, that justification was obvious: trusts just allowed donors to fine-tune their power to give, to structure their gifts in a few ways, extending their effects over time (within limits). Now, however, settlors may, MDTs being the most flagrant example, create managed gifts over time, managed in the ways we have explored. I hope to have shown that these managed gifts raise a host of issues that should give us pause in finding these gifts morally acceptable.

Footnotes

Thanks to Larissa Katz, Chris Essert, David Frydrych, Duncan Horne, Arthur Ripstein, Nicole Roughan, and Konstanze von Schuetz for helpful feedback on a first draft; especial thanks to Jessica Hudson, Jeremiah Lau, and Lionel Smith who provided extensive comments on the penultimate draft, and to helpful comments from an anonymous reviewer. I am happy to acknowledge research funds provided to me by the National University of Singapore as Kwa Geok Choo Professor of Property Law. The standard ‘claimer’ applies.

References

1. F Scott Fitzgerald, “The Rich Boy” in All The Sad Young Men (Cambridge University Press, 2006) at 5. See also Janny Scott, The Beneficiary: Fortune, Misfortune, and the Story of My Father (Riverhead Books, 2019).

2. JE Penner, “Private Law Offices” (2020) 70:Supplement 2 UTLJ 299.

3. Lionel Smith, “Massively Discretionary Trusts” (2017) 70:1 Current Legal Probs 17.

4. Representation can also address cases of collective decision-making. See Penner, supra note 2. I shall not address those cases here.

5. In certain respects the powers of a trustee in relation to trust assets are a functional equivalent of a company director’s powers in relation to company assets, though as Jessica Hudson points out, this in no way makes them legal equivalents. See Jessica Hudson, “One Thicket in Fraud on a Power” (2019) 39:3 Oxford J Leg Stud 577.

6. As in the case of pension trusts, where one purpose is to take advantage of expert fund managers.

7. Not all relationships of representation occur within the confines of an office. This is so even when the problem of representation endures. One can, in many circumstances, provide for the function of representation without creating an office. Agency provides a good example. It is a basic principle of agency law that no one who is sui juris is ever required to be represented by another. A principal can peremptorily dismiss their agent, even if this is in breach of a contractual agency relationship. Agents are not office-holders, because one can always terminate the agency relationship and act for oneself instead. There is never a requirement to replace an agent at common law, but when a principal does, it can be on entirely different terms. It is a function of the identity of an office that the constellation of rights, duties, powers, and so on remain the same whoever is appointed to fill it.

8. Whether changes in some features of the package, e.g. a change in the trustee’s charging clause, are such as to change the office into a different one is a question of judgment. See the discussion of the individuation of offices, Penner, supra note 2.

9. Including a ‘legal’ person, as in the case of a trust company.

10. The principle applies somewhat differently in the case of testamentary and inter vivos trusts. In the case of a valid testamentary trust, even if the testator’s chosen trustee declines to serve, the court (or the executor if they have the power to do so) will ensure that a trustee will be appointed. In the UK the trustee ‘of last resort’ is the Public Trustee. Once a trustee is found who is willing to be appointed, the trust will be constituted—that is, the titles to the intended trust assets will be vested in the trustee. In the case of an inter vivos trust, it is up to the living settlor to find someone willing to undertake the trust and accept the titles to the trust assets, thereby constituting the trust. A living person cannot, in this way, just declare a trust ‘in the air.’ Once, however, the trust is constituted, the slogan applies. A replacement trustee will be found if the current one is unable, unwilling, or unfit to act.

11. In the form of the ‘modern’ (i.e. post-17th century) rule against perpetuities. There has been extensive reform of the rule in many jurisdictions. The most recent reform in England and Wales is found in the Perpetuities and Accumulations Act 2009 (UK). The rule has been reformed in British Columbia (Perpetuity Act, RSBC 1996, c 358), Alberta (Perpetuities Act, RSA 2000, c P-5) and Ontario (Perpetuities Act, RSO 1990, c P-9). The rule has been abolished in its entirety in Manitoba, Nova Scotia and Saskatchewan: The Perpetuities and Accumulations Act, CCSM c P33; Perpetuities Act, SNS 2011, c 42; Part XI, The Trustee Act, 2009, being c T-23.01 of The Statutes of Saskatchewan, 2009.

12. See, in general, JE Penner, The Law of Trusts, 11th ed (Oxford University Press, 2019) at 3.44-3.55.

13. Which may be legal persons, e.g. companies.

14. In Alberta, British Columbia and Ontario trusts “for a specific noncharitable purpose that creates no enforceable equitable interest in a specific person must be construed as a power to appoint the income or the capital, as the case may be.” See Perpetuities Act, RSA 2000, c P-5, s 20(1); Perpetuity Act, RSBC c 358, s 24(1); Perpetuities Act, RSO 1990, c P-9, s 16(1).

15. Penner, supra note 12 at 9.26-9.30.

16. See Smith, supra note 3 at 20, n 8. See also Lionel Smith, “Give the People What They Want? The Onshoring of the Offshore” (2018) 103:5 Iowa Law Rev 2155.

17. Smith, supra note 3 at 20, n 8. Similar thinking underlies the ban on self-settled protective trusts. See Penner, supra note 12 at 3.129-3.133.

18. I think this is certainly correct in the case of inter vivos trusts; testamentary trusts raise other considerations, where a personal representative must discharge all the deceased’s debts before any will trusts can be constituted. It is worth noting that the anomalous exceptions to the rule are all testamentary trusts. See Penner, supra note 12 at 9.16-9.20.

19. (1841) 4 Beav 115 (Rolls Ct). See also Penner, supra note 12 at 3.30-3.32.

20. On both justifications see also Paul Matthews, “The Comparative Importance of the Rule in Saunders v Vautier” (2006) 122:2 Law Q Rev 266 at 270-76. It is worth noting that these justifications have not been found persuasive in the United States, where the wishes of the settlor have often been regarded as paramount. Following Claflin v Claflin, 20 NE 454 (Mass Sup Jud Ct 1889), the principle will not apply so long as there remains a ‘material purpose’ of the settlor in the trust continuing. Some other ‘settlor-friendly’ jurisdictions have adopted the ‘material purpose’ limitation on the principle by legislation, e.g., Bahamas Trustee Act, c 176, s 87. In Alberta and Manitoba the rule has been modified to require the court to find that “in all the circumstances at the time of the application to the court, the arrangement appears otherwise to be of a justifiable character.” See Trustee Act, RSA 2000, c T-8, s 42(7); The Trustee Act, CCSM c T160, s 59(7)(b). It is not entirely clear what the extent of the court’s discretion is, but the mere fact that the termination would conflict with the wishes of the settlor does not require the court to disapply the rule; see Knox United Church v Royal Trust Corp of Canada (1996), 110 Man R (2d) 81 (CA).

21. See Richard Nolan, “Invoking the Administrative Jurisdiction: The Enforcement of Modern Trust Structures” in Paul S Davies & James Penner, eds, Equity, Trusts, and Commerce (Hart, 2017) 151 at 152-54.

22. The distinction between ‘internal’ and ‘external’ here is roughly between essential aspects of the legal relationship between the parties and collateral areas of law that shape how parties use that relationship. See Mark Bennett, “Competing Views on Illusory Trusts: The Clayton v Clayton Litigation in its Wider Context” (2017) 11:1 Journal of Equity 48 at 57-58, citing at 57, n 96 Tobias Barkley, Discretionary Interests and Rights to Replace Trustees: Can They Be Property? (LLM Thesis, University of Otago, 2012) at 97-111 [unpublished].

23. A different development occurred in the case of trusts for minor children, which also involved a trustee’s dispositive discretion. In the case of minor children, the trustee was generally given under the terms of the trust and today, unless excluded, is given by statute, a power of maintenance, which allows the trustee to use certain income to support the child during their minority, paying for school fees, food, clothing, and so on. A different power again was typically conferred on trustees, again now on a statutory footing, which allowed the trustee to ‘advance’ capital funds from the trust to those adult beneficiaries whose capital interests would not vest until later (typically following the death of a tenant for life). See Penner, supra note 12 at 10.36-10.52.

24. Ibid at 2.11.

25. Smith, supra note 3.

26. Penner, supra note 12 at 3.12-3.13.

27. Smith, supra note 3 at 27-28 [footnote omitted].

28. Ibid at 29-33.

29. (1970), [1971] 1 AC 424 at 448 (HL (Eng)) [McPhail].

30. Penner, supra note 12 at 3.22.

31. Re T R Technology Investment Trust Plc, [1988] BCLC 256 at 263-64 (Ch D) [emphasis added].

32. [2003] 2 AC 709 (PC) [Schmidt].

33. Nolan, supra note 21.

34. Schmidt, supra note 32 at para 36.

35. Smith, supra note 3 at 22-24, 35-39.

36. For example, Re Manisty’s Settlement (1973), [1974] Ch 17 at 25 (ChD).

37. For a recent review of the rights of objects in Australia, see Jessica Hudson, “Mere Objects in Australia” in Ying Kai Liew & Matthew Harding, eds, Asia Pacific Trusts Law: Theory and Practice in Context (Hart, 2021).

38. Nolan, supra note 21 at 167-68 [footnote omitted] [emphasis added].

39. Most obviously, to turn someone who is the object of a mere power of appointment into a beneficiary of a substantial share of the trust fund.

40. See generally, Penner, supra note 12 at 3.67-3.82, ch 7. Smith (in personal conversation) has said that he thinks the “rot set in” with McPhail, supra note 29, in which the ‘is or is not’ test for certainty of objects was applied to discretionary trusts (see also Smith, supra note 3 at 24, n 19), but I take a more benign view of the decision. McPhail concerned a class of beneficiaries, most of whom could be determined with complete certainty, with a sub-class whose membership (on the reasoning of Sachs J in Re Baden’s Trust Deed No 2 (1972), [1973] Ch 9 (CA)) could be established very simply on extrinsic evidence that they were related to a member of the other sub-classes. Those for whom no such evidence was available were excluded. Nothing in this reasoning is authority for the proposition that the ‘true’ terms of a trust can be more or less at large unless and until the trustees choose to apply the trust assets or the court is called upon, relying upon extrinsic evidence which is not even minimally specified by the terms of the trust instrument itself. In any case, the massive discretions in MDTs concern mere powers, not discretionary trusts and, as I read him, Smith has no objection to the ‘is or is not’ test in relation to mere powers.

41. Nolan, supra note 21, is concerned with the due administration of the trust, but he doesn’t address the question whether the identification of the ‘true intended beneficiaries’ is such as to allow them to invoke the Saunders v Vautier principle and collapse the trust. My strong suspicion is that many MDT proponents would want to have it both ways, that is, being able to identify the ‘true intended beneficiaries’ would satisfy the certainty requirements, but would not then allow them to collapse the trust, pointing to the other trust provisions rendering their interests defeasible and so on, the very provisions which would, of course, have to be ignored or discounted when identifying them as the ‘true intended beneficiaries’ in the first place. For related discussion see Smith, supra note 3 at 29-42.

42. Schmidt, supra note 32 at para 36.

43. Typically where the settlor is also the trustee or protector: see, e.g., Clayton v Clayton, [2016] NZSC 29 [Clayton]; JSC Mezhdunarodniy Promyshlenniy Bank v Pugachev, [2017] EWHC 2426 (Ch) [Pugachev]; Webb v Webb [2020] UKPC 22 [Webb].

44. Briefly, the possibility that in some circumstances they could contribute to a finding that the trust is a sham, Smith, supra note 3 at 45-47; or an invalid non-charitable purpose trust, at 47; their being confidential to the trustees and the disputable grounds for this, at 48-51; and the ways in which they might encourage litigation, at 49-50. A further problem is that depending upon how ‘imperative’ the letter of wishes is, a court may find it is really part of the trust terms: see Chen v Ling, [2000] HKCFI 1356. (See also Taulbut v Davey, [2018] EWHC 730 (Ch) where a letter of wishes was incorporated by reference into a will.) Thus whether a ‘letter of wishes’ forms part of the terms of the trust is a matter of substance not form, that is, it doesn’t matter what the settlor calls the document. For similar judicial reasoning, see Street v Mountford, [1985] AC 809 (HL(Eng)).

45. Smith, supra note 3 at 45 [emphasis added].

46. Ibid at 46.

47. I am not certain how common this practice is, but a perfunctory search of the web found this: “If you have incorporated a trust into your will, or have created a lifetime trust, then it is important to let your trustees know what your wishes are with regards to how you would like the assets of your trust to be managed and distributed. The best way to do this is to set out your wishes in writing or, more specifically, in a letter of wishes to your trustees. Letters of wishes are not legally binding but they do hold strong moral weight and your trustees should have good reason to depart from their guidance. You can include guidance such as ages at which you envisage your beneficiaries receiving capital from the trust fund, what you envisage the trust fund being used for—i.e to fund educational costs or to help with the purchase of a property etc. A great benefit of letters of wishes is that they can be updated as often as you like—if your circumstances change or you simply change your mind—without the need to update your will itself. Letters of wishes should be signed and dated by you and then stored with your will or trust document.” See “Letters of Wishes” (last visited 3 June 2020), online at Birketts www.birketts.co.uk/insights/legal-updates/letters-of-wishes [emphasis added]. Although the italicised passage only mentions wills, I presume the same advice would apply to a trust, given the introductory sentence. See also Breakspear v Ackland, [2008] EWHC 220 (Ch) at paras 3-4, where Briggs J appears to treat written letters of wishes and oral expressions of wishes as standing on the same footing, and at para 6, where he refers to the “flexibility” of “[t]he combination of a broad discretionary trust accompanied by a wish letter.”

48. It seems to me that the idea that updating a letter of wishes is unproblematic rests on a confusion between the trustees ‘consulting’ the settlor or beneficiaries in order to gain information about their circumstances as a preliminary to exercising their discretions in an informed way, which is perfectly alright and commendable, and basically letting the settlor or beneficiaries tell them what to do. The contrast is well brought out by Christopher McCall: “[Y]ou can have trustees bound to consult with their beneficiaries and a protector and then required to form their own views how to act in the light of that consultation.… But not many trusts work like that. Most trustees are if not muppets then at least puppets who charge a high fee for being told what to do and meanwhile safeguarding the funds. That is not what trusteeship was meant to be.” See Christopher McCall, “The Trust as an Enemy of the People” (2013) 19:3-4 Trusts & Trustees 338 at 339.

49. Though they may of course be a beneficiary under the trust.

50. For discussion see JE Penner, “Trustees and Agents Behaving Badly: When and How is ‘Bad Faith’ Relevant?” 128 at 134-36 and Lionel D Smith, “Conflict, Profit, Bias, Misuse of Power: Dimensions of Governance” 149 at 150-54, both in Paul B Miller & Matthew Harding, eds, Fiduciaries and Trust: Ethics, Politics, Economics, and Law (Cambridge University Press, 2020).

51. Re creditors see Pugachev, supra note 43 and Tasarruf Mevduati Sigorta Fonu v Merrill Lynch Bank and Trust Co (Cayman) Ltd, [2011] UKPC 17. Re spouses see Clayton and Webb, supra note 43. See also Bennett, supra note 22.

52. For ease of exposition, in this section objects will be used as a compendious term for both ‘beneficiary’ and ‘object’ as defined in the text accompanying note 24.

53. It would take pages to provide even a sampling of this work. Perhaps, however, it is enough to say that Oxford University Press has published a volume in its Philosophical Foundations of Law series on contract, torts, unjust enrichment, property, fiduciary law, and most recently, the law of equity.

54. McCall, supra note 48.

55. Adam S Hofri-Winogradow, “The Stripping of the Trust: From Evolutionary Scripts to Distributive Results” (2014) 75:3 Ohio State LJ 529 and Adam S Hofri-Winogradow, “The Stripping of the Trust: A Study in Legal Evolution” (2015) 65:1 UTLJ 1. In the latter paper Hofri-Winogradow considers the way in which recent trusts law reforms alter the ‘corrective justice’ balance between the trustee and the beneficiary in favour of the former. There has also been an analytic philosophical analysis of some of the features of trusts, e.g. Lionel Smith,“The Foundations of Proprietary Remedies” 281 & James Penner, “Value, Property and Unjust Enrichment: Trusts of Traceable Proceeds” 306 both in Robert Chambers, Charles Mitchell and James Penner, eds, Philosophical Foundations of the Law of Unjust Enrichment (Oxford University Press, 2009) and Avihay Dorfman, “On Trust and Transubstantiation: Mitigating the Excesses of Ownership” in Andrew S Gold & Paul B Miller, eds, Philosophical Foundations of Fiduciary Law (Oxford University Press, 2014) 339. Dorfman focuses on cases of representation in the case of those such as children unable to own or manage property which I explored here earlier (see the text accompanying note 6) but which is not my focus here.

56. Roger Cotterell, “Power, Property and the Law of Trusts: A Partial Agenda for Critical Legal Scholarship” (1987) 14:1 JL & Soc’y 77.

57. Mark Bennett & Adam Hofri-Winogradow, “The Use of Trusts to Subvert the Law: An Analysis and Critique” Oxford J Leg Stud [forthcoming in 2021]. DOI:10.1093/ojes/gqaa008

58. JE Penner, Property Rights: A Re-Examination (Oxford University Press, 2020) at 27-28.

59. JE Penner, The Idea of Property in Law (Clarendon Press, 1997).

60. For my own (minor) contribution, see JE Penner, “Intergenerational Justice and the ‘Heredity Principle’” (2014) 8:2 Law & Ethics of Human Rights 195.

61. Ibid at 198-200.

62. Penner, supra note 59 at 74-75, 87-90, 207-10.

63. The claim for this is of course most famously made by Joseph Raz. See Joseph Raz, The Morality of Freedom (Clarendon Press, 1986) at Part V.

64. See the text accompanying note 24.

65. As elaborated by Immanuel Kant, “Doctrine of Right” in The Metaphysics of Morals, translated by Mary Gregor (Cambridge University Press, 1996). All citations are to the Akademie pagination.

66. Ibid at 6:277 et seq.

67. Ibid at 6:276.

68. As regards wives and husbands relationships with the outside world, i.e. beyond their internal marriage relationship, women are not equal to men. For example, they are regarded as dependent upon their husbands, and so can only ever be ‘passive citizens.’ Ibid at 6:314-6:315.

69. Ibid at 6:277.

70. Ibid at 6:278-6:279. For this reason prostitution and concubinage are ruled out as rightful sexual relations, for in these cases the relationship is not one of equal possession of the other’s physical person.

71. Ibid at 6:277 [emphasis in original].

72. Ibid at 6:280.

73. Ibid at 6:281 [emphasis in original].

74. Ibid at 6:282.

75. Ibid at 6:282.

76. Ibid at 6:282.

77. Ibid at 6:283.

78. Ibid at 6:284. Kant makes the same point at 6:283.

79. Ibid at 6:283.

80. Ibid at 6:283.

81. At common law they may delegate their ‘ministerial’ duties, those not involving the exercise of judgment, for example employing a collecting agent to collect rents from those leasing trust property.

82. As a default. In theory a personal power to remove or appoint trustees could be conferred by the terms of the trust, though for obvious reasons this would normally have to be set out in the clearest of terms, though see Pugachev, supra note 43 for a case in which a protector’s powers were held to be personal considering surrounding evidence.

83. Michael O Hardimon, “Role Obligations” (1994) 91:7 Journal of Philosophy 333, provides an interesting discussion of the philosophical issues surrounding the obligations that attach to roles, but he provides no clear indication of what amounts to a role, other than it is related to institutions, in particular the family and the state. For Hardimon, because friendships are not ‘institutional’ in the way that the family and the state are institutions, the obligations of friendship are not role obligations, but that sounds counterintuitive to me.

84. Christopher Essert, “The Office of Ownership” (2013) 63:3 UTLJ 418 at 433 [footnotes omitted].

85. A trustee may also be an object.

86. See, e.g., Paul B Miller, “A Theory of Fiduciary Liability” (2011) 56:2 McGill LJ 235; JE Penner, “Is Loyalty a Virtue, and Even If It is, Does It Really Help Explain Fiduciary Liability?” in Andrew S Gold & Paul B Miller, eds, Philosophical Foundations of Fiduciary Law (Oxford University Press, 2014) 159; Lionel D Smith, “Fiduciary Relationships: Ensuring the Loyal Exercise of Judgement on Behalf of Another” (2014) 130:4 Law Q Rev 608; Ernest J Weinrib, “The Fiduciary Obligation” (1975) 25:1 UTLJ 1.

87. Although ‘discretion’ and ‘discretionary’ are the standard terms used to describe the ‘leeway’ the fiduciary has in exercising their powers, the better word for describing this ‘leeway’ is ‘judgment,’ not discretion, at least in so far as the word ‘discretion’ suggests that the fiduciary has any sort of personal freedom in these cases. The fiduciary is required to choose that course of action which, in their judgment, is in the best interests of their principal; in making such judgments, the fiduciary is to consider only the interests of their principal(s); to put this another way, the fiduciary owes their principal what I have called ‘deliberative exclusivity.’ See Penner, supra note 86 at 166-67 and JE Penner,“Fiduciary Law and Moral Norms” in Evan J Criddle, Paul B Miller & Robert H Sitkoff, eds, The Oxford Handbook of Fiduciary Law (Oxford University Press, 2019) 781 at 789, 791-92.

88. Paul B Miller, “Fiduciary Representation” in Evan J Criddle, ed, Fiduciary Government (Cambridge University Press, 2018) 21 at 24-25 and Smith, supra note 50.

89. For example, in Arthur Ripstein, Force and Freedom: Kant’s Legal and Political Philosophy (Harvard University Press, 2009). Ripstein devotes a full chapter to each of the acquired rights of property and contract, but not to acquired status rights.

90. Ibid at 17 [first emphasis in original; second emphasis added].

91. Ibid at 21 [emphasis in original].

92. Ibid at 72 [emphasis in original].

93. Ibid at 73 [footnotes omitted]. See also ibid at 76-77, where Ripstein uses the example of a person investing another’s retirement savings on their behalf.

94. Ibid at 79 [footnotes omitted]. See also ibid at 192, 365, for similar points.

95. Ibid at 110.

96. Kant does cite as one kind of gratuitous contract one for the “Keeping goods on trust (depositum)” (Kant, supra note 65 at 6:285), but in English law this would be described merely as the bailment of goods for the benefit of the bailor. It is not a common law trust.

97. Ibid at 6:271-6:272 and Ripstein, supra note 89 at ch 5. What Kant was obviously most concerned about in explaining the nature of contract is how the united will or choice of the parties explains how one person can acquire the right to the performance of another. Any person aiming to explain the common law trust as a Kantian contract therefore faces an immediate problem, which might be called the ‘bilaterality in formation’ problem. Typically trusts are settled by one party, the settlor, who transfers assets to the trustee, who undertakes the obligation to hold the trust assets, but not as a performance for the settlor, but for the objects of the trust. As is well known, civilian jurisdictions find it very difficult to fit trusts within their private law, and even when they adopt versions of the trust, their trusts do not have the same legal incidents that common law trusts do; see Penner, supra note 12 at 2.94-2.104 (Quebec and Scottish trusts). Whilst civil law does recognise contracts for the benefit of a third party in a way that the traditional common law does not, trusts are not contracts and cannot be enforced by the settlor qua settlor. Trusts can only be enforced by their objects. Another problem, which is something that as far as I can see Kant never addresses, probably because of the very restricted numerus clausus of the civil law, is on what theoretical basis two parties can enter into an agreement in respect of property so as to bind third party successors in title, which a trust ‘agreement’ clearly does in cases of breach of trust: a third party recipient of trust property transferred in breach of trust will be bound by the beneficiary’s interest unless they are a bona fide purchaser. It is not clear to me that Kant provides the intellectual resources to explain why private law should grant powers of this kind, powers to attach obligations to particular titles, which ‘run with the title’ so to speak. But I think we can put all of this aside in examining whether the trustee-beneficiary relationship is a kind of Kantian contractual relationship, in so far as we are focused on the trustee’s obligations to the object.

98. See JE Penner, “The Beneficiary’s Performance Interest in a Trust: AIB v Redler and the March of the Compensatory Principle” in Richard C Nolan, Kelvin FK Low & Tang Hang Wu, eds, Trusts and Modern Wealth Management (Cambridge University Press, 2018) 277.

99. When investing the trust assets, the trustee must hold ‘an even hand’ between different classes of beneficiaries; see Penner, supra note 12 at 10.3-10.5.

100. As a matter of reality, this might be the settlor whilst they remain alive, or a ‘protector’ (see Penner, supra note 12 at 3.28-3.31) or only one of the trustees. For a disastrous example of a trust company leaving the running of the trust to a family member trustee, see Mennen v Wilmington Tr Co, No CV 8432, 2015 WL 1914599 (Del Ch Apr 24, 2015), adopted, (Del Ch Aug. 18, 2015), aff’d sub nom Mennen v Fiduciary Tr Int’l of Delaware, 166 A.3d 102 (Del 2017)).

101. In private conversation.

102. In private conversation.