Certainty of Objects for Discretionary Trusts

A) Certainty of Objects for Discretionary Trusts

Lord Denning in Re Vandervell’s Trusts (No.2)^[1] stated:

“It is clear law that a trust (other than a charitable trust) must be for ascertainable beneficiaries.”

The difficulty that arises in respect of a discretionary trust is that no-one who is an object can claim a defined interest or, in the words of Todd and Wilson^[2], “indeed any interest at all unless and until the trustee’s discretion is exercised in his favour.” However, the power of such trustees is not so absolute as it might seem since the courts will restrain them form acting contrary to the terms of the trust instrument and in the event of a capricious refusal to distribute at all will intervene in order to ensure distribution either by the appointment of new trustees or by exercising the power itself. An example of this may be found in Re Murphy’s Settlements^[3] in which an order was obtained against a settlor requiring the identification of the trustees so that a potential beneficiary could assert his rights.

By contrast, a fixed trust is one in which the interest and thus the share to be taken by the beneficiaries is specified in the trust instrument. In such a situation, it is necessary for the trustees to be able to ascertain exactly how many beneficiaries there are. It was originally supposed that the same principle applied to discretionary trusts on the basis that for the court to be able to intervene in the event that the trustees failed to perform their duties it would have to be able to ascertain the beneficiaries and distribute on the basis of the maxim that “equality is equity”. This argument was unravelled by the House of Lords in McPhail v Doulton^[4]. One Bertram Baden, the Chairman and Managing Director of a company had by Deed established a trust fund the relevant clause of which provided:

“The trustees shall apply [emphasis supplied] the net income of the fund in making at their absolute discretion grants to or for the benefit of any of the officers or employees or ex-officers or ex-employees of the company or to any relatives or dependants of such persons in such amounts at such times and on such conditions (if any) as they think fit.”

The trustees were not obliged to exhaust the income in any given year and there was provision for the accumulation of income that was not exhausted. In the light of the obvious width of the class of potential beneficiaries, the validity of the trust appeared to turn upon whether the trustees had been given a discretionary power or a duty of distribution. Baden died in 1960. Enter the grasping relatives. Some two years following his death, the executors challenged the validity of the trust deed. The trustees issued an originating summons to determine whether the deed was void for uncertainty. At first instance, Goff J held that the instrument created a mere power that was not void for uncertainty. In the Court of Appeal, the apparently mandatory nature of “shall apply” was disregarded and the decision at first instance (Russell LJ dissenting) was upheld. It was argued that where there were competing interpretations of a decision, the court should favour that which produced validity in accordance with the maxim ut res magis valeat quam pereat. The executors appealed to the House of Lords.

Lord Wilberforce is regarded^[5] as having “broken new ground”:

“It is striking how narrow and in a sense artificial is the distinction, in cases such as the present, between trusts or, as the particular type of trust is called, trust powers, and powers…A layman and, I suspect, also a logician, would find it hard to understand what difference there is.”

He continued:

“Differences there certainly are between trusts (trust powers) and powers, but as regards validity should they be so great as that in one case complete, or practically complete ascertainment is needed, but not in the other?”

While acknowledging that a trustee charged with the distribution of the whole of the fund must make a “wider and more systematic survey” of the objects, His Lordship warned that in the case of a trust power “the danger lies in overstating what the trustee is required to know or to enquire into before he can properly execute his trust”. However, the pre-existing orthodoxy was defended by Lord Hodson (dissenting) who relied upon the previous dicta of Lord Upjohn in Re Gulbenkian’s Settlements^[6]:

“The trustees have a duty to select the donees of the donor’s bounty from among the class designated by the donor; he has not entrusted them with any power to select the donees merely from among claimants who are within the class, for that is constituting a narrower class and the donor has given them no power to do this.”

Notwithstanding the above, the case was remitted to the High Court for reconsideration. At that stage, issues of conceptual certainty and evidential difficulty arose. At first instance^[7], Brightman J held that the controversial terms “dependants” and “relatives” were not too uncertain and upheld the deed. The Court of Appeal unanimously upheld the validity of the deed but differed as to the meaning of the certainty test. The reasoning that appears to emerge concerns the difference between the ability to arrive at conceptual certainty and the evidential ability to apply that concept to any given individual. If the “concept”, e.g. “someone under a moral obligation” cannot be ascertained with certainty then the gift will fail. However, the court should not be defeated by the evidential problems that might be encountered in determining whether someone is or is not a member of a stipulated class such as, for example “first cousins”.

This approach has, however, proved controversial. Todd and Wilson^[8] argue:

“We would suggest, therefore, that no good basis can be found for the conceptual/evidential distinction. Proponents of Sachs LJ’s view would respond that the test of administrative unworkability is sufficient to ensure that the trust is workable, but in that case is it reasonable to ask why any test at all is needed apart from administrative unworkability. In any case, evidential issues would now need to be dealt with by the administrative unworkability test, and t is not obvious that that is an improvement over dealing with them as part of the individual ascertainability test itself.”

These authors (p.134) then argue that in the case of fixed trusts there should be a more rigorous test:

“…it does not follow that merely because a discretionary trust is administratively workable with a given class of objects, a trust with the same class of objects will necessarily be so if the trustees’ discretion is removed. It is probably more satisfactory, however, to have a stricter certainty test for fixed trusts, albeit perhaps a test not as strict as the class ascertainability test.”

In conclusion it might be argued that McPhail v Doulton has produced more heat than light. The reasoning behind the test to be applied in respect of fixed trusts is easy to ascertain and defend. The tortuous reasoning that his developed in respect of discretionary trusts might be regarded as the unnecessary product of over-analysis and the anticipation of potential conceptual difficulties that are more real than supposed. The protracted, high-level litigation that was generated by Mr Baden’s Deed Trusts could be said to be more the product of the rapidly increasing size of the fund which fuelled the enthusiasm of the executors to pursue the claim than the product of genuine jurisprudential difficulty arising from the issues in question.

B) Property Holding by Unincorporated Associations

Problems frequently arise in respect of the distribution of surplus funds upon dissolution of an unincorporated association. Such funds will frequently have been accumulated by a variety of fund-raising activities such as, for example, the holding of raffles and small donations via collection boxes. The issue that thus arises is the basis upon which such funds are held where the rules of such an association make no provision for distribution of surplus upon dissolution. The courts have adopted a variety of approaches. For example, in Re Printers ad Transferrers Society^[9], weekly subscriptions had been taken for the benefit of a certain category of workers. Upon dissolution it was held that there was a resulting trust in favour f those who had subscribed and that the remaining assets should be distributed in proportion to the contributions made. A similar conclusion was reached in Re Hobourn Aero Components Air Raid Distress Fund^[10] save that the distribution on this occasion was in favour of contributors in proportion to contributions with an offset for benefits received. This decision might be criticised for the fact that a much more complex distribution occurred but it is still a finding in favour of a resulting trust approach. It is perhaps significant that in both the above cases, there was no assertion by the Crown of bona vacantia.

A contractual model was favoured in Cunnack v Edwards^[11]. A Friendly Society had been established to provide annuities for widows. A claim to the remaining assets was made by the Personal Representatives of the surviving members but it was held that there was no resulting trust on the basis that the contributors had effectively relinquished their claim to their contributions in return for a guarantee of benefits to their widows. Matthews^[12] argues that the necessary distinction is between donating to people rather than purposes. Where the former occurs, there is an absolute gift but the recipients are bound by the terms of an express or implied contract between them stemming from the nature of the society to apply such monies in a particular way.

The deficiency of this approach is highlighted in the leading case of Re Recher’s Will Trusts^[13] in which the deceased had given a share of her residuary estate to the London and Provincial Anti-Vivisection Society. That society had ceased to exist some 5 years before the death of the testator but Brightman J considered as a preliminary issue whether the gift would have been valid had the unincorporated association still existed. He concluded that a contract-holding interpretation was appropriate:

“…so it must follow that the life members, ordinary members and associate members of the London and Provincial Society could, at any moment of time, by unanimous agreement (or by majority vote if the rules of the society so prescribe) vary their multi-partite contract.”

The sting is in the tail:

“There would be no limit to the type of variation or termination to which all might agree…No-one would have locus standi to stop them doing so.”

Gardner^[14] however reassures us that the potentially shocking corollary of such a conclusion is unlikely to occur in practice:

“The relevant part of a club’s rules will provide, either expressly or [presumably] impliedly, that its members must devote to its purposes all property owned by them “on account of the club” or “in their capacity as members of the club” or something of that kind.”

Sadly, the waters have been muddied by the comparatively recent decision of Scott J in Davis v Richards & Wallington Industries Ltd^[15]. In that case a company had gone into liquidation and there was a substantial surplus in the remaining pension fund. At about the time of the winding up the fund’s trustees had purported to execute a new trust deed which allowed such a surplus to be devoted to the increase of pensions and the payment of any remaining money to the company. The first part of the case turned on the validity of the deed but His Lordship then proceeded to consider how the surplus should have been dealt with had the deed not been valid. He approached the issue by analogy to the position of an unincorporated association and considered the possibility of either a resulting trust or a Recher-style contract model. On the basis of this approach, the resulting trust should have been ruled out and the ability to identify a form of contract holding should have prevented a finding of bona vacantia. However, the conclusion reached was to the contrary: the money contributed by the employers was now held on resulting trust whereas the contributions of the employees and other funds which had become merged into the pension fund were regarded as bona vacantia. Gardner^[16] complains:

“The judgment itself is unfortunately not a model of clarity. It appears somewhat as though Scott J did not realise that he was departing from the usual treatment, so we do not have the benefit of his own account of the distinctions to be drawn.”

The conclusion of this Gardner article is attractive (pp.51-2):

“A perhaps more impressive candidate may be seen in an overtly result-oriented doctrine not far from the area presently under discussion: application cy-prs upon subsequent failure of charities.”

It is suggested that such “nearest purpose” application would allow the public and the economy to benefit from the return of funds that otherwise would remain tied up in such funds for very substantial periods of time.

Accordingly it is submitted that although it may be superficially attractive to favour the contract-holding model of property possession by unincorporated associations, this is ultimately an unwelcome development. The traditional view has it that the difficulties of analysis of the method of receiving funds by many such associations preclude a trust approach – where a donor has given a modest donation or participated in an event such as a raffle, it does indeed appear strained to find an intention to create a trust such as that which might formally be produced by a deed. Equally, it would be ludicrous to argue that such a gift should fail for uncertainty of intention. This dilemma gives rise to the seductive reasoning that the donor has parted with possession absolutely but that the members of the association remained constrained as to its use by the existence of the contract between them. The major deficiency in this approach is illustrated by the principles described above: if the only limitations upon the use of funds are those prescribed by the contract between the members, that contract can be varied without reference to the donor perhaps to the absurd extent of the members voting to share the money between themselves for their personal benefit. Such an outcome cannot be regarded as in any way consistent with the purpose of the gift. For this reason, it is submitted that the resulting trust analysis – while appearing initially over-elaborate – should remain the favoured model.

Bibliography

Gardner, S., A Detail in the Construction of Gifts to Unincorporated Associations [1998] Conv 8

Gardner, S., New Angles on Unincorporated Associations [1992] Conv 41

Martin, J., Hanbury and Martin Modern Equity, (17^th Ed., 2005)

Matthews, P., A Problem in the Construction of Gifts to Unincorporated Associations [1995] Conv 302

Moffat, G., Trusts Law: Text and Materials, (4^th Ed., 2005)

Smart, P., Holding Property for Non-Charitable Purposes: Mandates, Conditions and Estoppels [1987] Conv 415

Todd, P. & Wilson, S., Textbook on Trusts, (6^th Ed., 2003)

---

Footnotes

^[1] [1974] Ch 269 at 319

^[2] Todd, P. & Wilson, S., Textbook on Trusts, (6^th Ed., 2003), p.125

^[3] [1998] 3 All ER 1

^[4] [1970] 2 All ER 228

^[5] Moffat, G., Trusts Law: Text and Materials, (4^th Ed., 2005), p.215

^[6] [1970] AC 508

^[7] Re Baden’s Deed Trusts (No.2) [1972] Ch 604

^[8] Op. Cit., p.130

^[9] [1899] 2 Ch 184

^[10] [1946] Ch 86

^[11] [1896] 2 Ch 679

^[12] Matthews, P., A Problem in the Construction of Gifts to Unincorporated Associations [1995] Conv 302 at pp.303 et seq

^[13] [1972] Ch 526

^[14] Gardner, S., A Detail in the Construction of Gifts to Unincorporated Associations [1998] Conv 8 at p.9

^[15] [1990] 1 WLR 1511

^[16] Gardner, S., New Angles on Unincorporated Associations [1992] Conv 41