Philip Laidlow – Halliwell Landau
(taken from Isssue No 12 –  July 2000)


This article looks at so called sham trusts. I have read elsewhere the view that the percentage of sham trusts is high. In the Trusts & Estates 1999 Legalease Special Report, at p.26, Catriona Syed, wrote,

`It is impossible to estimate how many sham trusts are currently in existence. Estimates vary from 50% to 80% of all trusts.`

Personal experience suggests that the figure is even higher than that. However, that depends on the trusts one is looking at and taking a broad view of what is a sham trust. The latter is one of the issues this article explores; that there may be fewer true sham trusts than we think; `Sham` is perhaps not the best term. The trust deed may be a sham but only to the extent that there is a trust at all and that it is not representative of its terms, as in a blind trust. More often a sham trust is a trust which simply does not exist. That is one of the themes of this article; that usually there either is a trust or there is not a trust, and that many trusts we might be tempted to suggest are shams, are in fact validly created and existing trusts. They may be run and administered in an appalling manner in the way in which the Rahman trust was. However, the manner in which a trust is run as opposed to set up has separate consequences and ramifications and does not lead to the inescapable conclusion that the trustee is purely the nominee of the trust assets for the purported settlor. Therefore while the picture is no rosier than the statistic above suggests, some of the 50-80% may be something other than shams; though still trusts where there is slavish adherence (nominee style) to the instructions of the settlor treated as a client.

The article focusses on offshore settlor-interested trusts, invariably discretionary. Domestic trusts, particularly those used for tax planning, and even more so those where the settlor is excluded, will be much less likely to be shams.

The article then goes on to explore the extent, if any, to which some of the problems which otherwise arise out of shams, can be legislatively cured.

  1. The Cases

Recent cases on the subject have been few, although a vastly increased number can be expected. Both Rahman and Wyatt have been written about extensively elsewhere. Nevertheless, it is still worth summarising some of the salient aspects of each, Rahman in particular, as a few misconceptions flow from the cases themselves.

1.1    Rahman

The seminal case was Abdel Rahman v. Chase Bank (CI) Trust Company Limited and others, a decision of the Jersey Royal Court reported at [1991] JLR 103. Mr Rahman was a Lebanese domiciliary who established (in a loose use of the word) a Jersey trust in 1977. It is not totally clear from the judgment but this seems to have been as a settlement, ie. as a bipartite document. The trust contained a number of provisions whereby trustee actions required Mr Rahman’s consent. These were not particularly unusual or unknown to an offshore trust. However, there were a number of core clauses in the trust which were less usual.

  • Clause 4(1) gave Mr Rahman a general power of appointment over the trust fund and its income during his life (though exercisable only with trustee consent). There was a proviso to the clause which permitted Mr Rahman in any period of 12 calendar months to appoint one third of the capital of the trust fund without that trustee consent. In default of appointment he was the income beneficiary of the trust; clause 5(1).
  • Clause 5(2) directed the trustees after Mr Rahman’s death to hold the trust fund and its income as Mr Rahman should appoint. The power was exercisable in favour of a specified class. The description of this specified class seems to have given an element of designation to Mr Rahman such that he could effectively add to the specified class. There were trusts in default of exercise.
  • Clause 10 gave the trustee an overriding power to pay or apply the capital or income of the trust fund to or for the benefit of Mr Rahman and to have regard exclusively to the interests of Mr Rahman in determining whether or not to exercise such power.
  • Curiously there was also a protector appointed although it is not clear from the judgment what the scope of the protector’s powers were.

Mr Rahman purported to exercise the power of appointment in September 1980. Following his death, his widow and the estate of his mother attacked the validity of the settlement. The two main grounds were, first, that the provisions of the trust deed were contrary to an old French element of Jersey law encapsulated in the maxim donner et retenir ne vaut, and, secondly that the trust was void as a sham.

The Court held that the trust deed did breach the maxim donner et retenir ne vaut. It is clear, though, that had the Court not made its decision on that ground it in any event would have found the trust to be invalid as a sham. The matters which allowed the Court to find the trust to be a sham also went to support the Court’s judgment on the first ground.

Donner et retenir ne vaut is no longer part of Jersey law. It has been removed by statute. Once one gets through the first 45 pages on that maxim, the judgment becomes a fascinating one. At p.146 Tomes, D.B. said,


We have to make a review of the evidence. We were unanimously satisfied that the oral evidence, together with the documentation placed before us, established clearly that from the date on which [Mr Rahman] purported to constitute the settlement he exercised dominion and control over the trustee in the management and administration of the settlement, including distributions of capital to himself, to others as gifts or loans, and the making and disposal of investments. He treated the assets comprised in the trust fund as his own and the trustee as though it was his mere agent or nominee. There was rétention by [Mr Rahman]. We are unanimously of the opinion that the settlement was a sham on the facts, in the sense that it was made to appear to be a genuine gift when it was not. Thus, even if the Court were wrong on the construction of the settlement and the impact of the maxim donner et retenir the settlement would still be invalid.`

The Court set out a series of examples of the dominion and control exercised by Mr Rahman over his assets and over the trustee. These run to approximately 20, although the Court also suggests that this was not an exhaustive list. They provide an object lesson in how not to `administer` a trust.

`The Court was impressed by the evidence of the Plaintiff. Despite the difficulties of language, we believe that the Plaintiff tried to give her evidence truthfully and without exaggeration. In general, the documentation supported and corroborated her evidence. By contrast, Mr Johnson, who was successively trust officer, trust manager and general manager of the trustee, struggled unsuccessfully to defend the indefensible.`

One can almost picture Mr Johnson squirming in the witness box in an attempt to persuade the Court that there was something other than complete acquiescence in the action taken by, or obedience to any instruction given by, Mr Rahman. The examples given by the Court will be recognised by many as a tale of everyday life in some offshore trust companies. Absolutely no instance of saying no to Mr Rahman could be identified. Two or three of the more damning examples are:Mr Rahman selected the investment adviser, and instructed the trustee to appoint that investment adviser. The trustee had no role in the selection, although under the trust deed it should have selected the adviser (subject to Mr Rahman’s consent). On matters of investment policy, the investment adviser met and corresponded directly with Mr Rahman, who is described as returning his approval directly to the investment adviser. The trustee seems to have been copied into the correspondence and nothing more, ie. played no role in approving the investment policy.

A scenario is described of approximately $12m being contributed to the trust yet withdrawals totalling $1m being made the same day at the hand of Mr Rahman without the knowledge or concurrence of the trustee. There seems to have been no complaint or enquiry about the $1m but acquiescence rather.

Funds were paid out of the trust without any real thought as to the basis on which they were paid, ie. under which clauses of the trust deed, whether out of capital or income etc. No reason seems to have been given or case made for payment. Generally, they seem to have simply been payments on instruction from Mr Rahman without any consideration by the trustee.

Mr Rahman obtained payments direct from the trust banker so the trustee did not know about the payments until afterwards. There was no request from Mr Rahman to the trustee, and no instruction from the trustee to the banker. An example is noted of an executive of the trustee making a file note in which he recorded that a representative of Chase (Switzerland) rang to say that Mr Rahman had contacted him from the South of France requesting an advance of $100,000 which had been credited to him. The telephone call seems to have been more to establish whether a withdrawal fee applied than to apprize the trustee. No action seems to have been taken to prevent a repeat.

Midland Bank v. Wyatt

Midland Bank plc v. Wyatt [1995] 1 FLR 696 at first glance was a very different case in several regards, mainly that it turned on concealment of a trust rather than an attempt to exercise dominion and control over it. Mr & Mrs Wyatt, on legal advice albeit seemingly given over Sunday lunch, executed a trust over their home as to 50% for Mrs Wyatt and 25% for each of their minor children. The document was professionally drawn by their solicitor. The trust was put away in their safe. Mr Wyatt later borrowed from the bank on the security of `his` house. There was no disclosure of the trust to the bank. When, in due course, the bank started proceedings to recover, the trust was produced. The Court took the view that the trust was a sham. It had been signed, put away, ignored, not disclosed and generally not acted on. As in Rahman the Court took account of the subsequent acts and behaviour of Mr Wyatt in determining the trust to be a sham.

  1. What is a sham?

A sham trust is a trust document which is not a trust. A trust may ostensibly have been constituted in the sense that the trustee may be in possession of the intended trust assets and there may be a signed declaration or settlement of the terms of the trust. Nevertheless, from inception there is no trust. Either there was no valid and effective delivery to a trustee rather than a nominee, as in Rahman,or simply there was no intention to create a trust as in Wyatt. Some views of a sham trust presuppose that the trust in question somehow exists but is nevertheless ineffective as a sham. This cannot be so. Once a trust is validly created and exists, that existence cannot be denied. Sham-like trust behaviour, such as a total kowtowing to the wishes and directions of the settlor, or simply behaviour which might not be in accordance with the terms of the trust or in accordance with the duties imposed upon the trustee, will potentially have its own consequences. It is true that in Rahman the trustee’s behaviour (as did Mr Rahman’s behaviour and attitude) subsequent to the purported creation of the trust led to finding that the trust was in any event a sham. However, this was not because of the situation itself but because it was evidence that there was never intended to be a trust. Similarly in Midland Bank -v- Wyatt it was the subsequent behaviour of Mr Wyatt which allowed the Court to determine that he had no intention of creating a trust.

It is unfortunate that most commentary on sham trust seems to start with the frequently cited passage of Diplock LJ in Snook -v- London and West Riding Investments Limited [1967] 2QB786802.

In a case concerned with hire purchase he said that this popular and pejorative word (sham),

`If it has any meaning in law, it means acts done or documents executed by the parties to the `sham` which are intended by them to give to third parties or the Court the appearance of creating between the parties legal rights and obligations different from the actual legal rights and obligations (if any) which the parties intend to create. One thing I think, however, is clear in legal principle, morality and the authorities… that for acts or documents to be a `sham`, with whatever legal consequences follow from this, all the parties thereto must have a common intention that the acts or documents are not to create the legal rights and obligations which they give the appearance of creating. No unexpressed intention of a `shammer` affect the rights of a party whom he deceived.`


It is unfortunate that there has been so much focus on that passage as it does not lend itself much to trusts, nor in fact need to. The focus is understandable given that the passage was mentioned in the Wyatt judgment. Counsel seem to have referred to the passage in Rahman though it was not mentioned in the judgment.

Diplock LJ’s passage creates the obvious problem, in attempting to apply it to trusts, that he speaks of the parties having a common intention to create a sham. That cannot be correct. Granted in Rahman the Court focussed on the trustee’s contribution to the sham as much as Mr Rahman’s, and granted further that in practice it may be much more difficult to show a sham where there a bipartite trust deed. Nevertheless, I cannot see that ultimately the trustee’s intention is anything other than irrelevant. If a settlor intends to create a trust, and the other certainties and requirements of constitution are present then the intentions of the trustee will be irrelevant. In the absence of willing trustees Equity ultimately would provide that trustee. It does not allow a trust to fail for want of a willing trustee. Equally, if there is no thought of a sham in the mind of the trustee and the trustee fully intends to accept a trust, nevertheless, if the settlor does not intend to create a trust, then, questions of proof and to what extent a man is bound by an instrument he signs aside, one of the core certainties is missing and there cannot be a trust.

That a trust is a sham is not a dedicated or innovative invalidation of a trust. It is not the trust as such which is a sham. There is no trust to be a sham. It is the trust documentation which is the sham. The sham trust fails simply because one or more of the certainties and/or formalities of constitution are not present. The focus in Rahman was very much on the administration, or lack of it, of the trust. However, the administration did not somehow cause the Rahman trust to become a sham subsequent to the date of signature. The method of administration was found to be evidence of a sham such that the court held that there had never been an effective transfer of assets to a trustee as opposed to a nominee, ie. the trust was found in retrospect to be a sham but that sham finding meant it had never existed and had not ever been constituted. The point is easier to see inWyatt. Mr Wyatt was found from his subsequent behaviour never to have intended there to be a trust. His wife seems to have been content to sign whatever her husband might have put before her and similarly to have been lacking in intention to create a trust.

Therefore, what it comes down to I think is that sham trusts can only be caused by events at the inception of the trust albeit subsequent matters may be the more likely method of proof of the sham. In day to day terms, there can only be a sham caused by the take-on and acceptance procedures of the trustee, in culmination with the intention and understanding etc of the supposed settlor. The focus is on the setting up of the trust The question is whether the trust has been created at all. A trust, if it is going to be a sham, is going to be a sham from the moment it is taken on. Take the following slightly extreme but not uncommon scenario as an example.


A US citizen prefers to visit inconspicuously therefore cruises down from Miami to a Carribean tax haven. He gets off the cruise boat and visits three or four trust providers in town to obtain quotations. Price is a significant factor in who is going to look after his money. The companies he visits have long abandoned value (and to an extent therefore risk and responsibility) related fees because of competitive pressures. A cut rate annual fee is offered simply to secure the business. This may well be a negotiated fee below even the trustee’s advertised minimum annual fee. The whole relationship is off to the wrong start. The intending `settlor` has been given a fee more representative of running an account than a trust will its inherent risks. That is the value he will put on it and the attitude he will have of it.

2.1     Client

The prospective settlor asks how he knows that the trustee will do what he wants. He is assured that he is the settlor and therefore the client and that, of course, his instructions as the client will be followed. There is liberal use of the word client and, of course, it will subsequently be splashed all over the file, as will words like `instructions`.

There is a genuine belief in some quarters that somehow the settlor has some special standing and rights towards a trust, to the point even that there is a contract between the settlor and the trustee for the administration of the trust. That is not so. A settlor has whatever rights are reserved to him in the trust deed. The law will also require that if the trustee knows of the settlor’s purpose in creating the trust then discretions etc. have to be exercised bearing in mind that purpose. This may be formalised through a letter of wishes. But that is the full extent of the settlor’s rights.

Letters of wishes are regarded almost as a necessity with offshore trusts. Yet, in reality, many are an absolute nonsense and simply advertise the settlor’s intended dominion and the trustee’s immediate acknowledgement of that. Many are (preprinted) standards of the trustee with little scope for variation. They may cover matters which have no relevance, eg. cover real property when there is no real property in the trust. They are unlikely to give much insight into the creation of the trust and sometimes simply request to be consulted.

2.2    Shelf Trusts

A trust will not quite be a shelf trust in the sense of a shelf company, but will frequently be a trust company standard run off the computer, and will not have been seen, not have been considered and frequently will have had no professional advice taken on it. In this regard it is well worth looking at West v. Lazard [1993] JLR 165. The case touched many issues and the judgment is very lengthy. However, the following illustrates the point,

`Mr West cannot be bound by the terms of a `standard form trust` or `shelf trust` of which he had no knowledge. …….. It cannot be the fault of Lazard that Mr West told us that he never fully understood the terms of the trust; it is the fault of Lazard that he took down a shelf trust without attempting to give Mr West an explanation of the terms of it and (as trustee) to ensure that it conformed with his wishes.`

That was in the context of whether the trustee could take the benefit of a particular clause. Nevertheless, it illustrates the point that intending settlors will have little concept of what they are signing. It might be thought that if a settlor puts his signature to a formal settlement deed he is bound by it. There is not a huge amount of authority but that would seem not to be the case. While it will certainly seem that a bipartite deed would be less capable of being set aside than a unilateral declaration of trust, the judgment in Commissioners of Stamp Duties (Queensland) v. Jolliffe (1920) 28 CLR 178 suggests not.

The point is compounded when a trust is created not by settlement but by unilateral declaration of trust by the trustee being in possession of the initial trust fund, as is frequently the fashion. There is still, in substance, a settlement. The declaration of trust is still the settlor’s declaration of the terms of the trust on which the trust assets are to be held. Pending the declaration of trust, the trustee holds the trust assets on resulting trust for the settlor. Theoretically the trustee declares the trust at the direction of the settlor. However, it is even less likely that the settlor will know the terms of the trust that he has instructed the trustee to declare. There is no document that he has put his signature to. At best, he will have put his signature to a proposal form which itself may only be a brief menu choice, eg. ticks for discretionary, protector, special companies clauses etc. The settlor will conceivably never see the trust, or a draft of the trust, declared for him.

2.3     Understanding

The trust concept is notoriously difficult to explain to the layman. A settlor has tointend to create a trust as one of the three certainties. In many situations this will not be the case simply because the settlor has no idea of what a trust is. Historically, the cases and the text books have treated certainty of intention as a matter of whether the settlor was serious about creating a trust. However, in the offshore context it will frequently be the case that while the settlor intends to create something, he does not intend to create a trust. It will be accepted that he intended to do something, ie. the seriousness of the intention is there. However, the intention will not always be specifically to create a trust. The postulated US citizen above may simply be looking for someone to ‘look after’ his money offshore. This will be particularly so where foreign settlors are concerned, especially those from non-common law jurisdictions, to whom English may be a second or third language, who is usually being provided with an in-house standard document, which document may be filled with gobbledygook and legalese, and which document he might well not understand and, not infrequently, not even read.

2.4    Holding Company

A holding company may be required to hold the underlying trust investments, for whatever reason – sometimes trustee policy. Here again there is great scope for error. Nominee directors – of which strictly there is no such thing – will likely be provided. As part of the provision of the underlying company and the nominee directors (and shareholders) the trustee company may well ask the settlor to sign a management agreement under which the trustee agrees to manage the company and keep it in good order, good standing and generally compliant. This is inappropriate. The settlor is not the beneficial owner of the company as it is a trust owned company, but that is indicative of the reality of things; he settlor as client mentality Again, the trustee has declared its hand.

2.5    Credit Cards

As he is about to put his funds into an offshore trust the intending US settlor wants to know how he may access his money in the USA without there being a remittance or trail. Technical questions on the remittance point aside, the suggestion will sometimes be that the client should use a credit card. He will complete the application form at the same time as the trust and company proposal forms. This can cause various technical problems and again offer indications of the true nature of the overall arrangements. The following is an extreme example but illustrates the point. The so-called client has a blind trust established for him, with the Red Cross of some small and distant jurisdiction showing as the only beneficiary on the face of the deed. He is offered and readily takes a credit card so that he can spend his money in the USA. The trustee obviously cannot take out a credit card in the name of the trust as it is not an entity and the trustee will not want the liability. Therefore a company will be incorporated and held underneath the trust. The trust will lend funds to the company. The client will be issued with a card, probably in his personal name rather than the name of the company. The client will not be a director, employee or even a shareholder of the company. He then uses the card in the USA. The company receives the bill from the credit card company and pays it. The client has no intention of paying. It is his money in his trust which he is accessing through his credit card; the card is a conduit to his money and the trust is to pay. As intended, therefore, the underlying company will clear the credit card bill. The expenditure cannot have been for any corporate purpose. The client, therefore, owes the company money. Local regulation means that accounts have to be produced. The company accounts will show a loan due from the client; it cannot even show a loan from a beneficiary as the trust is blind. In time, a sizeable deficit loan account may accrue and in theory at least an insolvent company – which possibly ought to require the directors to wind it up -will be the immediate trust asset. The only way to clear and balance this will be for the trust to provide funds. The simplest way to clear the debt might be for the trust to appoint funds to the client, as a beneficiary, without any actual transfer. The trust will simply set off what it then owes the beneficiary against what the beneficiary owes the company which in turn owes the trust. The client is clearly a contemplated beneficiary from the very start. It will be a paper exercise, though one which will first require the client to be added as a beneficiary. The whole thing cries out sham from inception.


3       Statutory Cure?

Rahman is likely to be only the tip of an iceberg. Even without Rahman the offshore trust world is becoming increasingly litigious. Rahman offers a significant additional piece of ammunition to those seeking to attack trusts. It is probably no surprise, therefore, that one or two jurisdictions have begun statutorily to address Rahman. The Bahamas has introduced what arguably is the most revolutionary (or innovative according to one’s view) statutory response. This is section 3 of its Trustee Act 1998 set out in full below.

`3. (1) The retention, possession or acquisition by the settlor of any one or more of the matters referred to in subsection (2) shall not invalidate a trust or the trust instrument or cause a trust created inter vivos to be a testamentary trust or disposition or the trust instrument creating it to be a testamentary document.

(2) The matters referred to in subsection (1) are –

(a) any powers to revoke the trust or the trust instrument or any trusts or powers granted thereby, or to withdraw property from the trust;

(b) any powers of appointment or disposition over any of the trust property;

(c) any powers to amend the trust or the trust instrument;

(d) any powers to appoint, add or remove any trustees, protectors or beneficiaries;

(e) any powers to give directions to trustees in connection with the exercise of any of their powers or discretions;

(f) any provisions requiring the consent of the settlor to any act or abstention of trustees;

(g) any such other powers as are referred to in subsection (2)(a) to (h) of section 81;

(h) the appointment of the settlor as a protector of the trust;

(i) any beneficial interests of the settlor (including absolute beneficial interests) in the capital or income of the trust property or in both such capital and income; and

(j) any interests of the settlor in any companies or assets underlying the trust property and any control of the settlor over such companies or assets.

(3) Subject to any contrary intention expressed in the trust instrument and subject to its other terms, a power in a trust instrument to amend, alter or vary a trust shall include (without limitation) a power to add as beneficiaries any persons whatever (including the settlor and any private or charitable trusts or foundations) and to remove any beneficiaries.`

Section 3 is intended to do four things, in summary:

  • to counter sham trust arguments and to address the Rahman judgment and its ramifications
  • to salvage mini-trusts (see below)
  • to counter the argument that some trusts are in substance wills, and improperly executed, and therefore invalid (a related area but not the subject of this article)
  • to attract business and to make the Bahamas competitive

The mini-trusts referred to were not any particular type of trust. Typically they were a short form trust, usually revocable, which gave settlors substantial control over the trust and which were, like life insurance, sold rather bought, and substantially marketed to South and Central Americans and others.

Some of the powers in section 3(2) which section 3(1) allows the settlor to retain, possess or require are barely objectionable and would not, in the absence of section 3, ordinarily lead to the invalidity of the purported trust. This would include the powers of appointment and the appointment of the settlor as a protector of the trust in particular. At the other end of the scale some would, in the absence of section 3, be completely objectionable, eg. the power to givedirections to trustees in connection with the exercise of any of their powers or discretions.

Section 3 is said to counter sham arguments and to protect trusts from being found to be shams. It does not specifically say this. It is implied from, `shall not invalidate`. The permitted matters would not actually lead to a trust being a sham. Where the settlor is exercising control over a trustee because of the matters in section 3(2), that is not a sham; the trustee is behaving as per the trust deed. It may be likely that the trust fund might be treated as being beneficially owned by the settlor, but that is not a sham.

Section 3 is said to address Rahman. It is doubtful that it does as Rahman was essentially about trustee behaviour and settlor attitude. Rahman from the sham point was not particularly about the contents of the trust deed.

One of the reasons for section 3 is said to be to salvage or attempt to offer some sort of statutory protection to the thousands of mini trusts sold from the Bahamas in recent years. Many of those trusts are so settlor-friendly that they were thought to be in significant danger of being called shams.

The essence of section 3 is an attempt to facilitate trusts which have the protections of being trusts (with all the benefits that brings) but at the same time to allow the settlor various protections and controls and potentially to give settlors complete control over trustees. That is, to offer settlors have your cake and eat it.

3.1 Sanctioned Powers

Sub-section 3(2) of the Act contains a number of matters which are said not to invalidate a trust or to cause it to be treated as a will. This is not a definitive or exhaustive list.

3.1.1 Power to revoke the trust or to withdraw trust property. Without the protection of section 3, this power (unless explicable by some other good reason) would normally be dangerous both from the point of view of sham attacks and will attacks. The obvious explanation for revocability is to allow the settlor to change his mind and effectively to cancel the trust. Occasionally there might be tax reasons for revocability but few other reasons. If the settlor wants the facility to have the trust fund back it would be preferable for the mechanism to be a positive one and to have it through such as an appointment.

3.1.2 Powers of appointment or disposition over any of the trust property. This is not an unusual sort of power and, even without section 3, not usually too problematic.

3.1.3 Power to amend the trust or the trust instrument. This is essentially a dangerous power. On the face of section 3(2)(c) the settlor can retain the power to amend the trust unilaterally. In trusts where the power to amend is seen, it usually is a two party mechanism, ie. power to the trustees to change with settlor’s consent or perhaps for a protector to be involved. It would be very unusual for a trustee to have the unilateral power to amend. It is also unusual for a settlor to have unilateral power.

3.1.4 Powers to appoint or remove any trustees, protectors or beneficiaries. Again, the permitted power is unilateral. I do not know if it is of any great significance but ‘trustees’ is the first main category in the list, ie. it is not alphabetical. On the face of section 3(2)(d) the settlor can have a power unilaterally to remove the trustees of the settlement. The settlor can also have the power unilaterally to remove beneficiaries. Normally that power is needed in sophisticated offshore trusts, but perhaps not unilaterally by the settlor. This is a dangerous provision.

3.1.5 Power to give directions to trustees in connection with the exercise of any of their powers or discretions. On the face of it, that is a contradiction in terms. Directions hints of instruction. If it is the case that the settlor can direct the exercise of discretions then the discretions simply are not discretions. Possibly section 3(2)(e) is intended only to emphasise that a settlor can give his input but settlors have always had the ability to offer their wishes etc. Even prior to the Trustee Act, a power in the trust deed to allow the settlor to give his non-binding directions would not have invalidated the trust. Therefore, for various reasons, one has to presume that directions in section 3(2)(e) does mean wishes of the settlor which the trustees must take account of. Such power can be very dangerous in the context of sham type attacks.

3.1.6 Provisions requiring the consent of the settlor to any act or decision of the trustees. This is a familiar type of provision. Many trust deeds require that the trustees must have the settlor’s prior written consent to certain actions. Such prior written consent provisions are not particularly dangerous. The act or decision etc. has to be initiated by the trustees. The settlor just has the right to veto. Bear in mind that the trust deed in Rahman was full of settlor consent provisions. They did not contribute to the finding that the Rahman trust was a sham. The sham was down to behaviour and attitude.

3.1.7 The appointment of the settlor as a protector of the trust. Conceptually, there is absolutely no reason why the settlor should not be a protector, or at least the settlor as a protector is not inconsistent with the concept of a trust. There may be jurisdictional reasons against it. However, the settlor as a protector does offer scope to anyone attacking their trusts. Some of the stronger protector powers look positively dangerous if exercisable by the settlor. For example, the power to remove and appoint new trustees. If the settlor, as protector, has that power it gives an immense level of control over the trust with consequent risks of attack. If a third party protector has those powers, even if that third party is connected to the settlor, at least there is a layer between the settlor and the trustees.

3.1.8 Retention of an interest. Section 3 sanctions the retention by the settlor of an interest in the capital or income of the trust property. Section 3 apart, this could be positively dangerous. If a settlor retains an interest in particular in the capital of a fund there is scope for an argument that the trust fund should be treated as part of the beneficial estate of the settlor. For example, if the settlor has the power to require distributions of capital from the trust, there will be strong arguments. Depending on the wider terms of the trust provisions operating after the settlor’s death, there might also be a risk of will-type attacks.

3.1.9 Interest in underlying company. Section 3 also sanctions an interest of the settlor in any underlying company or trust assets as well as any control the settlor might have over such companies or assets. Such interests or control are not so much sham trust and invalid will points. They are more points of investment and management. That is, the extent to which trustees allow outside control or interest in underlying trust companies and assets, particularly that of the settlor, is more a matter trustees might answer for in litigation concerning investment performance or breach of trust, but not necessarily sham litigation. Possibly settlor involvement in the underlying company might be relevant in genuine sham litigation, ie. where the behaviour and attitudes of the trustee and settlor are pertinent.

3.2 Retention, Possession or Acquisition

Retention will usually comprehend actual powers retained by the settlor through the trust deed. Possession means possession as a matter of fact rather than formal authorisation through the trust deed. For example, a settlor’s interest in an underlying company of the trust is likely to come under this category. Acquisition is most likely to mean after-acquired, eg. the appointment of a settlor as protector might not always be the starting position in the trust document. The trust document might result in the settlor becoming protector under the appointment mechanism of the trust document.

3.3 Limiting Arguments

There has been much adverse comment on section 3. It is yet to be seen to what extent it will be successful. My view is that section 3 is unlikely to work or offer any great protection, for various reasons.

  • Section 3 is Bahamian legislation. It will only be applied by Bahamian Courts and Bahamian Courts can only be effective to the extent that Bahamian Courts have control of the relevant trust assets. Therefore, one can only be certain of its application in respect of the Bahamian assets of Bahamian trusts. Even if the immediate trust asset is a Bahamian IBC, one cannot be certain that underlying assets of a trust contained in the IBC will be sheltered by section 3. If the trust is attacked in another jurisdiction the corporate veil of the IBC might well be lifted. Section 3 is going to receive very little sympathy outside the Bahamas. It is an attempt to say that something which no longer resembles a trust still is a trust. That will not be well received.
  • There is some circularity in section 3. `shall not invalidate a trust` suggests that there has to be a trust in the first place. If the trust is so full of settlor-reserved powers that it no longer resembles a trust – the hypothesis which section 3 seeks to counter – then there is an argument that the protection of section 3(2) is not available because there is not a trust in the first place. Therefore, there is no protection. Such a construction would render section 3 purposeless and I can see that argument being advanced in litigation.
  • Even in the Bahamas one has to wonder what the judicial attitudes to section 3 might be. The section is radical and it almost tries to turn trust law on its head. It is well worth comparing the approach in the Orange Grove litigation. See the relevant passages in Professor Willoughby’s book, Misplaced Trust. The Court bent over backwards to give a limited construction to the `offending` statute.
  • Section 3 is fundamentally flawed. A trust is an institution of immense pedigree. It is the essence of a trust that there is separation of the trust property from the personal estate of the settlor. If the settlor’s creditors seek to have access to the trust assets the answer is that the assets are no longer part of his estate but belong to the trustees. Section 3 is perilously close to altering that. It attempts to allow settlors to have control of their trustees – either directly through the power to give directions to trustees in the exercise of the discretions or indirectly through the ability to remove and appoint trustees. This is compounded by the ability to get the trust property back, as permitted, and the ability to have an interest (even an absolute interest) in the capital of the trust. Ultimately, therefore, there is no real separation between a settlor and the trust assets. Section 3, therefore is very close to trying to stipulate that something which patently is not a trust on general principles nevertheless still is a trust.
  • Section 3 does not prevent sham attacks because of behaviour. The point needs to be repeated that Rahman was about trustee behaviour and acquiescence on the one part and settlor attitude, dominion and control on the other part. The contents of the trust deed had little to do with the sham finding. The trust deed had many powers reserved to Mr Rahman. Not all were of the type canvassed in section 3 but some were of the degree contemplated by section 3. Whilst certainly those powers as much as the behaviour etc. supported the Court’s finding of donner et retenir ne vaut, the alternative finding of a sham depended upon behaviour etc. as indicative of and probative of the sham finding. The reserved powers went to the funding of donner et retenir ne vaut not shown. There is absolutely nothing in section 3 which legislates against that and, indeed, it would be very difficult to do so.
  1. Conclusion

One can understand the reason why trusts of the Rahman type are found to be shams. Yet they are so unnecessary. The consequences of a trust being a sham will be addressed in the next edition of the journal, but are adverse to both the settlor and the trustee. It is very easy to run a trust to the satisfaction and expectation of the settlor as ‘principal’ yet for it to be a secure trust not a sham. It simply needs a little settlor education at the outset. Yet even the self interest of both parties seems not always to be a controlling factor. I suspect many more sham trusts will end up in the courts.

We perhaps ought to drop the phrase sham trust. It is an easy shorthand . If a sham no trust. If a trust no sham. But putting the two together suggests therer can be a type of trust called a sham trust. There cannot.

Halliwell Landau