Three Certainties

A key requirement of an express trust is that the settlor intended to create a trust. Equity does not allow a trust to fail for want of a trustee': if the intended trustee does not accept the trust property and act as a trustee, the trust will still be valid and the beneficiaries will still have a beneficial interest, although the settlor will be the trustee until the new trustees are appointment. A trust can exist even if the trustee or beneficiary does not know about the trust. 

The word trust does not have to exist in the declaration of trust as equity looks to intent, not form.(Meggary J in Re Kayford: the question is whether in substance significant intention has manifested)  In Richards v Delbridge 1874 a settlor who intends to declare himself a trustee need not use the words 'I declare myself trustee' but he must do something which is equivalent to it, and use expressions which have that meaning. For example, the cases of Paul v Constance and Rowe v Prance who's actions established a declaration of trust. 

Before the Executors Act 1830, the courts were more willing to establish a trust as the property that was unattached would remain in the executor's hands. Thus, cases, where the settlor used mere precatory words, were sufficient. However, the attitude of the courts changed in Lambe v Eames, where the CA found that imperative words were necessary. 

Whether a trust is intended is always a question of interpretation. 

- Re Adams and the Kensington Vestry 1884: the testor provided in his will that he left his estate to the absolute use of his wife in full confidence she will do what is right as to the disposal between his children. Cotton LJ held that he intended to leave the property to his wife absolutely. Previous case law had gone too far in accepting precatory language as sufficient evidence of intention. 

Must look at true effect of words: “we must not rely upon the mere use of any particular words, but, considering all the words which are used, we have to see what is their true effect, and what was the intention of the testator as expressed in his will

- Comiskey v Bowring-Hanbury 1905: held that the words in full confidence that at her death his wife would devise it to one or more of his nieces was sufficient to establish a trust. But, only because immediately following this direction was a statement saying that if the wife failed to devise the property it would be divided equally amongst them. (Compare Lord Davey and Lord Lindley's views of this case) 

The above cases concerned finding the intention to create a trust in the wording of a document. However, in cases where there is no document to construe, conduct will show a trust if, objectively assessed, the conduct was enough to show that a trust was intended.

Paul v Constance 1977: the deceased was separated from his wife and began a relationship with P until his death. The deceased deposited money into a bank account in his name which P could withdraw money. He said to P that the money is as much yours as mine. P claimed the money was held on trust for her. 

Scarman LJ in the Court of Appeal stated that there was a trust in favour of P. The conduct through repeated assertions that the money was both of theirs was enough to create an express trust. This is an objective test of whether the conduct is enough to show that a trust was intended. 'There must be a clear declaration of trust and that means there must be clear evidence from what is said or done of an intention to create a trust. However, he did note that it might be thought that this was a borderline case.  --> Virgo suggests this is probably the limit as to how far the courts will go to find a sufficiently certain intention

--> Gardener: this shows the importance of context - in family circumstances, where there is no legal advice, the courts may be more ready to find necessary intention. 

There are two elements to the certainty of subject matter: (i) it must be clear what property is held on trust and (ii) the beneficial interest must be clear. It is important to establish what property is unavailable to creditors upon insolvency. If the trust fails for lack of certainty of subject-matter then no property can be transferred, and the beneficiaries have no priority over creditors. 

There are two areas of difficulty in regards to subject-matter: 

(i) Vague or general descriptions of the trust property 

(ii) Trust as part of a bulk

Although the property is broadly defined to include both tangible and intangible property such as shares, and even milk quotas, it will still fail if the property is described in general. For example, 'the bulk' of the property is not sufficiently clear (Palmer v Simmonds) In Sprange v Barnard the 'remaining part of what was left' was not sufficiently clear, neither was 'such parts not sold' (Re Jones). Although, Re Last anything left was certain and so was reasonable income (Re Golay's Will Trust)

Even if the property is defined it must be identifiable (Westdeutsche 1996). If not, the trust will be void for uncertainty (Re Goldcorp). 

It is presumed that the leading case law is Hunter v Moss although this is considered a significant case, it has not been without academic and judicial criticism. It is important to consider the case law before Hunter v Moss to understand the reason why academics such as Penner has suggested that Dillon J has sent the law into 'turmoil.'

Re London Wine Co 1986: LWC owned a large bulk of wine in which it declared to hold on trust for future buyers. However, no steps were taken to separate the property from the mass. Therefore, the court held that there was no valid trust created as LWC had failed to segregate the property from the large bulk. Oliver J in his discussion compared the case to that of a farmer who declares himself trustee over two sheep from a large bulk, neither is identifiable. It is not perfect or a correct trust. As a result, the property must be identifiable and separate from a large bulk of property in order for the subject-matter to be certain. 

Re Goldcorp Exchange 1995: This concerns a gold dealership that went into insolvency. The purchasers sought ownership of the property which they had previously purchased. However, similarly, they did not take steps to separate the gold purchased from the large bulk of the property. The court held that this was not a valid trust. The judgement came from the Privy Council in which they stated that even if the company had used the current stock, there was no valid trust. There was no trust even though Goldcorp had advertised in a brochure that they would have legal title to the gold upon purchasing, the court found that there was no declaration of the sort.  --> These cases would be decided differently due to s.20A sale of goods act 1979 which was added in 1995 which provides the purchaser of the bulk of good becomes an owner in common of a share of the bulk. 

In Hunter v Moss, Moss owned 950 shares of a company and declared himself a trustee for 50 of those shares for Hunter, without specifying which of these 950 shares were held on trust. 

CA (Dillon LJ): trust was valid, despite shares not being segregated.

1. The shares were intangible assets of identical value 

2. Testimonial Trusts? 

3. Distinguished from other case law?

Despite Hunter v Moss being followed by Re Harvard Securities 1998 and Pearson v Lehman Brothers 2010, there have been many criticisms of the law in this area. 

If the settlor had 50 shares from two companies, then the trust would fail. However, as the 50 shares are from the same company they are considered completely identical by Dillion J. This is especially because it is intangible property - if tangible and there was 50 sheep, one of them could have an unknown disease. (Ignores the fact that shares have their own unique certification numbers.)

Sensible conclusion? But what if one of the shares are sold or if the value has increased? One suggestion is that tracing should be an appropriate model to determine this issue. However, if Martin's tracing is such a brilliant remedy why should it not work for a tangible property? 

Alastair Hudson agrees with this criticism as it contradicts the element of property law which states that there be specifically identifiable property to be subject to a property right. He also states that the distinction between tangible and intangible is spurious and that having different rules for such is unnecessary and will contribute to the uncertainty in the law. 

--> MacJordan Construction v Brookmount - decided after Hunter but surely the money would come under intangible property? However, despite this, the court stated that just because the property is intangible does not mean that the subject-matter is certain. In this case, the building contract provided that the client would retain 3% of the contract price as a trustee for the builder depending on the work being satisfactory. However, the client had never assumed an obligation to establish the retention fund from that account.

Dillion J argued that if the shares were left in a will, then the trust would be valid. However, this decision causes practical difficulties, primarily because it is not possible to identify any clear rationale as to how this works in practice. David Hayton states that the decision was hurried and that an inter vivos trust could not be the same as a testamentary trust. With the latter, the testor does everything necessary to ***** himself of all his legal and equitable title in favour of the executor which makes sense. With an inter vivos trust, the settlor does everything necessary in order to create the trust, but until he has segregated the assets which are to be held on trust then the subject matter is uncertain. This is a valid point as Moss should have separated his 50 shares from the remainder of his 900 shares and told the company registrar of this at the time of the agreement. 

Parkinston stated that London Wine involved a fluctuating mass, so the court could not identify the assets beneficially owned by each customer by reference to a proportion of the whole bulk - therefore, the beneficial interest in the wine stock, at any time, might not be the same as her contractual entitlement to the wine, whereas property in Hunter could be identified as proportion of the bulk of shares - Pearson v Lehmans Brothers 2010 approach. Hayton argues that this is flawed. He rejects a Lehman Bros tenants in common analysis because this involves substituting the employer’s intention to give the employee 50 shares for a different intention to create a tenancy in common. 

o   Distinguished Re London Wine: a principle that applies only to the tangible property as wine (and other chattels) is not completely identical — i.e. part of the wine could be damaged, so identification of specific shares is necessary. 

Re Harvard Securities [1998] Neuberger J: Clients had a beneficial interest under the trust, despite the shares not being segregated. As long as the total and proportions were clear, that was enough. Distinction between London Wine and Hunter approved. However, applied reluctantly? States that shares are identical but chattles are not. 

Hudson argued that the only way to reconcile the two cases is public policy – the companies were insolvent but the employer in Hunter v Moss was not. In Hunter, the stakes were different – the question was not is there enough property to go around, but it was rather should the defendant have to transfer the property he has got?

For instance, in cases such as Re Gold, it has been a reccuring public policy concern that if the court enforces the trust in question, the beneficiary interet will take priority and defeat the creditor’s interet. Not only does this unsettle the basic tent on insolvency law by conferring priority to an unsecured creditor, but it words of Lord Mustill, to enable the claimants priority it would give them an adventitious benefit devoid of foundation in logic and justice.’ Whilst Hudson’s public policy argument can explain the divergence in case law, it does not give satisfactory legal closure.

White v Shortall explicitly rejected the approach taken by Dillion LJ in Hunter v Moss.

The argument raised that, like H v M a distinct trust could take effect even though there was no segregation of those shares from the entire holding. The court upheld the validity of the trust, but explicitly disavowed the rationale. Instead, the inference which the court took from the facts was that a single trust took effect over the entire holding and the trustees had the power to elect which shares out of that entire shareholding were to be treated as being held by the claimant.

This avoided the problem of certainty of subject-matter as the trust took effect over the identified whole fund and there was no need to segregated. Dillon LJ sought to justify dispensing with the need for certainty of subject matter on the basis that most of the authorities had concerned chattels and on the basis that executors of a will trust do not know the property which they are to hold on trust and yet such trusts are valid. Neither argument holds water in that executors know that they hold on trust any property owned by the testator at the time of death, and that there is no reason in principle for distinguishing between tangible and intangible property except that it led conveniently to the answer Dillon LJ wanted to reach and only if one ignored the earlier decision of the Court of Appeal in MacJordan v Brookmount. Doctrinally, it is suggested that the decision in Hunter v Moss is wrong; the alternative analysis in White v Shortall is more satisfying. 

The certainty of objects relates to identifying the beneficiaries of a trust. Every trust (except charitable trusts) must satisfy this requirement. If the trust fails for lack of certainty of objects, then there is a resulting trust for the settlor. 

It is important that the objects of the trust be defined with sufficient certainty that the trustees, or default the court, know to whom the trust property can and cannot be appointed. 

In Re Gulbenkain's Settlement Trust, Lord Upjohn stated that the principle is that the donor must make his intention plain to the object of the trust. 

Under a fixed trust, the trustees have no discretion regarding how the trust property should be distributed. In order to satisfy the certainty of objects requirement, both conceptual certainty and evidential certainty need to be established. 

The relevant test is the one person's test established in Re Barlow's Will Trust 1979 in which the onus is placed on the object to show that they satisfy the condition.

Under the fixed trust the trustees have no discretion regarding how the trust is distributed, exactly how much has already been determined by the settlor. This means that the trustees must know who the property is for.  The appropriate test is the 'complete list' test set out in IRC v Broadway Cottages in which fixed trusts need to be conceptually and evidentially certain. It use to be thought that if the trustees failed to carry out a discretionary trust then it would be for the courts to choose between the possible beneficiaries and so the court needed a complete list to be drawn up. However, this is no longer the position in regards to discretionary trusts (McPhail v Doulton) Although, for class disposition, this is still the correct approach. 

Ascertainability is not necessary. No reason why all beneficiaries must be ascertained; if one of the beneficiaries cannot be found, then his share could simply be paid into the court.   Re Gulbenkinan: “if the class is sufficiently defined by the donor the fact that it may be difficult to ascertain the whereabouts or continued existence of some of its members at the relevant time matters not. The trustees can apply to the court for directions or pay a share into court.”Nor should a fixed trust be afflicted by administrative unworkability: if a complete list can be drawn up, then it should always be practicable for the trustees to give effect to the settlor’s intention.

Under a discretionary trust, the trustees enjoy a discretion as to which objects should receive the trust property, and how much each object should receive. Potential objects must be defined with sufficient certainty to enable the trustees to exercise their discretion appropriately. 

Originally the complete list test applied (IRC v Broadway Cottages 1955) However, the house of lords abandoned the complete list approach and adopted the given postulant test: only need a complete list where the trust property is to be divided equally. 

McPhail v Doulton: settlor set up a fund for the benefit of employees and relatives at the ‘absolute discretion’ of trustees. Questions: (i) whether this was a trust or power; (ii) the appropriate test for certainty of objects requirement.  Lord Wilberforce: “the rule recently fastened upon the courts by IRC v Broadway Cottages Trust ought to be discarded, and the test for the validity of trust powers ought to be similar to that accepted by this House in Re Gulbenkian’s Settlements for powers, namely that the trust is valid if it can be said with certainty that any given individual is or is not a member of the class.”

Wilberforce: ‘given postulant’ test includes conceptual/evidential certainty and administrative workability. Trust will fail where “the definition of beneficiaries is so hopelessly wide as not to form ‘anything like a class’.” Case was remitted the court of first instance to apply the test, made it to CA:

Issue on appeal was whether the groups ‘relatives and ‘dependents’ satisfied the new test — could it be said with certainty that any given individual was a member of the class. CA: held the terms satisfied the test, but were uneven in their application of the McPhail test. All agree that the language used must be conceptually certain, but differed as to the level of evidential certainty required.

o   Sachs LJ (liberal approach): the trust was valid because the court could always determine who was a dependant / relative. For him the McPhail test required only conceptual clarity: “The court is never defeated by evidential uncertainty... once the class of persons to be benefited is conceptually certain it then becomes a question of fact to be determined on evidence whether any postulant has on inquiry been proved to be within it: if it is not so proved, then he is not in it.”

o   Stamp LJ (restrictive approach): evidential as well as conceptual certainty is required: McPhail test means the court must be able to tell with certainty whether a given person is / is not an object of the trust. He reached his decision, in this case, by defining relatives as ‘next of kin’ (cf. Sach’s definition: ‘descended from a common ancestor’).

o   Megaw LJ (middle ground): would find the test satisfied if, “as regards at least a substantial number of objects, it can be said with certainty that they fall within the trust.”

Re Baden No.2 case, the court was required to consider the provision in a discretionary trust for “relatives”. The court was required to follow principles set out in McPhail v Doulton. Nonetheless, to have done so on the basis of a purely literal application of the test may have led to the invalidity of trust. Therefore their Lordships sought to add their own gloss to those principles. However, 3 judges sought to give their own versions of interpretations of “relatives.”‘Dependants’ were defined as those who are wholly or partly financially dependent on somebody else. ‘Relatives’ were defined by Sachs and Megaw LJJ as descendants from a common ancestor, whereas Stamp LJ defined them as the next of kin or nearest blood relations. Additionally, they each had different interpretations of the relevant test. 

In the case of Re Barlow's Will Trusts, Browne-Wilkinson J had to discuss the validity of a trust which was intended to benefit 'friends.'  Miss Barlow, the testatrix had a large collection of pictures which she specifically declared them to be held by her executor on trust to sell them, but her family and friends could buy them first. Browne-Wilkison J held that the trust was valid because the concepts of friends could be given a workable meaning. Although 'friend' could have a wide variety of meaning, the minimum requirements were that (a) the relationship had to be long-standing (b) be a social and not a business or professional relationship, and (c) although they may not have met for some time, when circumstances allowed, they would meet frequently. 

Not conceptually certain for discretionary trusts or fiduciary powers because there are many degrees of friendship ranging from an intimate relationship to a mere acquaintance. Rather, in this context, it was sufficient that one person undoubtedly fulfilled the condition and it did not matter whether or not anybody else did (Re Allen). Browne-Wilkinson J further stated that he did need to decide whether Re Gibbard's Will Trust was good law, in that it applied the Re Allen test to powers of appointment. But, it does show that the Re Allen test is the correct test, the word 'friends' is not too uncertain to be given effect. He also disregarded 'my old friends' by Meggary J to not be applied to such a case. 

If it is a discretionary trust: When applying the is or is not test, certainty is established if it could be said of any given candidate that she was or was not within the class of beneficaries. Lord Wilberforce in Mcphail v Doulton suggested that a discretionary trust for the benefit of the 'residents of Greater London; would be administratively unworkable and so void. 

This was applied in West Yorkshire Metropolitan County Council, the council wished to create a discretionary trust of £40,000 to be applied for a lift of purposes for any benefit of the inhabitants of the county of West Yorkshire. Taylor J held that this trust was invalid because it was administratively unworkable to distribute such small amounts to all people. It was not, however, capricious, but merely too difficult and constly for a court to enforce. In Re Hay Settlement Trusts Megarry recognised that a discretionary trust for anyone other than a specific few would be administratively unworkable. 

If it is a power of appointment: Fidcuary powers have been recognised as valid where the trustees are able to exercise the power in favour of the whole world. It follows that a fiduciary power cannot be struck down because of administratively unworkable simply because of the breadth of class (Re Hay's Settlement Trust) - although can be invalidated if it is capriciousness Templeman in Re Manistry

Re Sayer 1957 - if there is a fixed trust then this would lead to a failure of evidential certainty and will be invalid. 

If it is a discretionary trust then evidential certainty is not usually a problem (Re Baden's Deeds Trust) as the trustees only need to distribute to those beneficiaries of whom they have noticed, provided that they have advertised their intention to do so in the press for a specific time (Trustee Act 1925, S.27) Where one beneficiary is missing, trustees of a testamentary trust may ask for the court for a Benjamin Order (Re Benjamin 1902) in which it would allow them to distribute to other beneficiaries or otherwise must take out insurance - this is not applied in inter vivos trusts, only testamentary trusts. Or, if one beneficiary is missing it can pay the share of the missing B into the court, and allocate the shares equally between the 5 children (Re Gulbenkian's settlement)

In Boyce v Boyce, the testator left his two houses on trust for his daughters, under the condition that his daughter Maria should choose the one she wanted, and the remaining one would then go to his other daughter, Charlotte. Maria died before her father, and it was unknown which house she would have chosen. The Vice-Chancellor, held that the trust failed because it was uncertain which house Maria would have chosen, and which would go to Charlotte. 

This, where the beneficial shares are unclear and it is clear that the trustee was not meant to take property absolutely, the property is held on resulting trust for the settlor or held for the beneficiaries of the residue. 

In Re Tuck’s Settlement Trusts, Sir Adolph Tuck, a baronet who had run the art publisher Raphael Tuck & Sons, created a trust for future baronets who were married to a wife 'of Jewish blood' and who 'continues to worship according to the Jewish faith'. If in doubt, 'the decision of the Chief Rabbi in London of either the Portuguese or Anglo German Community… shall be conclusive'. It was contended that the concepts of being of Jewish faith and of Jewish blood were too uncertain for the trust to be valid. Lord Denning MR in obiter held the trust was valid, and the Chief Rabbi could resolve any uncertainty. The trust would have been valid even if the Chief Rabbi had not been identified. But Eveleigh LJ contrasted and said the trust was valid, but only because the Chief Rabbi’s opinion of who was Jewish was part of the definition of the class of beneficiaries. He felt he would not have been able to resolve an uncertain class. On the facts, it might be too capricious for the president to determine who is the fan of the club. 

The appropriate test is the 'is or is not test' established in Re Gulbenkian's Settlements 1970 in which the objects of the power included those with whom a particular person had been residing. It was held that it was not conceptually uncertain because the court could determine whether a particular person matched a description. This seems at odds with the strict application to conceptually certainty in Re Baden (No 2) in which you needed a clear definition to be valid. However, it might be because Re Baden involved a discretionary trust and was decided later by which time the courts had adopted a more sophisticated approach to the problem of certainty of objects. 

Assumes the same test for evidential certainty as discretionary trusts.