An express trust requires certainty of subject matter. This involves two limbs: the beneficial interests of the beneficiaries being certain, and the trust property being clearly identified and segregated. This second limb has traditionally been viewed by the courts as particularly important. Judges have frequently held that a failure to segregate trust property from a bulk renders a purported trust void. They have done so in order to protect the basic function of the trust. Namely, if trust property is to be successfully vested in the trustee, clarity is needed as to what the intended trust property actually is. A corollary of this is that clarity is further needed so as to know the scope of the trustee’s duties and their liability for breach.

Whilst this principle of segregation from the bulk was developed specifically with regard to cases involving tangible assets, it was assumed that it would apply likewise to those assets, such as shares, that are intangible in nature. Hunter v Moss, in belying this assumption, threw a spoke in the wheel of academic expectation, provoking in turn a fierce debate as to whether the case was correctly decided.

This essay argues that despite the criticisms of Hunter, Jill Martin is right to say that it’s solution is fair, sensible and workable.

This essay begins by surveying the cases that gave rise to the presumption that trust property need be segregated from a bulk. Following this, it turns to Hunter, specific attention being devoted to rebuking Hayton’s criticisms of the decision. 

The orthodox view: Re London Wine

The principle that trust property should be clearly identified and segregated if in bulk originates in the case of Re London Wine Co. Here, a company owning a stock of wine stored in a storehouse, entered into sale contracts with customers for its delivery. Before the latter was performed, however, the company went into receivership and it became apparent that the stock of wine remained as a single bulk, the bottles not having been allocated to the various sale contracts. Oliver J held that since particular wine bottles had not been segregated from the bulk, there was no certainty of subject matter and as such there was no trust.

This decision was endorsed by the Privy Council in Re Goldcorp Exchange, which similarly involved a company going into receivership, having failed to segregate the majority of units of gold bullion it had sold, but not delivered, to customers.

It is in the context of these cases that Hunter should be examined.

Hunter v Moss

The defendant was the registered owner of 950 shares in a company. He executed a declaration that he held 50 of them on trust for the claimant, who was an employee. The defendant subsequently sold the 950 shares and kept all the proceeds. The issue for the Court was whether the claimant had beneficial rights in the 50 shares, as these had not been segregated from the other shares.

If the orthodox approach in relation to tangible assets fell to be applied, judgment would have been entered against the claimant. However, this was not the case. At first instance, Colin Rimer Q.C. citing the case of Rollestone v National Bank of Commerce, a case from the Supreme Court of Missouri, upheld the trust. He distinguished Re London Wine Co, and drew an analogy to Re Clifford in holding there was nothing extraordinary in a testator leaving a legacy in an unspecified number of share as bottles of wine, as a chattel, could be different from one another: one bottle might be corked, another not. In contrast, shares were identical, with it therefore not being required to specify which individual shares were to be trust property.

In the Court of Appeal, Dillon LJ endorsed this reasoning and stated that unlike the wines in Re London Wine Co, shares were essentially identical and indistinguishable, and therefore, any 50 shares in the company were capable of forming the subject matter of the trust. Dillon LJ then concluded “as a person can give by a will a specified number of shares in a certain company, so equally, in my judgment, he can declare himself a trustee of 50 of his ordinary shares and that is effective to give a beneficial proprietary interest to the beneficiary under the trust”.


Dillon LJ’s decision has been criticised by a number of scholars- Penner and Selden, for example, finding in its willingness to depart from rigid precedent a “roguish thing” where again the unit of measurement is a “chancellor’s foot.” That said, this essay concentrates its analysis on the arguments of Hunter’s most cited critic: David Hayton.

Hayton criticises Hunter on two grounds. First, echoing the arguments of scholars Hudson and Parkinson, he states that there is no sensible basis for distinguishing between tangible and intangible goods. Second, Hayton argues that Dillon LJ’s will analogy implies some remarkably unorthodox views about the nature and function of trusts, Hayton’s suggestion being that it overlooks the fundamental difference between testamentary and inter vivos trusts.

Hayton first criticism can be critiqued on three grounds. Firstly on the ground that Hayton, whilst acknowledging that much of his argument on the indistinguishable nature of tangible and intangible goods is borrowed from a 1987 article of Roy Goode, fails to engage with the fact that Goode had since rebuked his own reasoning on Hunter in a series of articles.

Secondly, on the ground that Hayton’s first criticism also falls short in suggesting that Re Wait – a case in which a trust failed where wheat had not been segregated- is of general application. However, it is submitted that Hayton’s argument is incoherent as Re Wait looks curiously out of line with modern commercial practice in relation to bulk shipments of commodities as trusts of goods are governed by the Sale of Goods Act, and trusts of intangibles are not. Hunter’s recognition of this only adds strength to the conclusion that its judgement is a sensible one.

Thirdly, on the ground that Hunter can be distinguished from Re London Wine and Re Goldcorp, thus showing there is a sound basis for making a distinction between trust of intangibles and trust of goods. In Re Goldcorp the gold bullions were traded ceaselessly, thus the trust property was not incessant and it could not be contended that any specific bullion was held on trust for a particular client (Wilson). Similarly, in Re London Wine Co, it could not be argued that a specific wine bottle was being held for a particular client. This argument can be further magnified by the fact that it would be inequitable to treat a gold bullion or wine bottle in a similar manner, as some might be faulty or of inferior quality. In Hunter however, the shares formed part of a completely identical ‘bulk’, and were constant. Although it has been suggested that shares are different because they have an earmark (Ockleton), the numbering is used for the purpose of accounting for the shares rather than being any reference to the quality of property (Goode). Therefore, they can be treated in the same way as one another, and there is no need for segregation. Thus, the solution in Hunter is fair, sensible and workable.

However, it is conceded that there are two parts to Hayton’s criticisms that hold weight.

First, Hunter does create one uncertainty. Namely, it is unclear in the scenario where a settlor makes an oral declaration of trust of unspecified shares, and part of these are sold in the intermittent period, who is to pay taxes and claim the proceeds. In order to combat this criticism, Dillon LJ suggests that the ordinary tracing rules should apply and, this proposition has support from Martin as she argues that there no difference between one pound and another pound and one share and another share of the same class. That said, the tracing argument is unconvincing, presupposing the existence of what in fact requires to be established. The ability to trace presupposes an equitable property right (under a trust). But Hunter would only have this right if a trust arose in his favour, and here we return to square one in that for a valid trust to arise in his favour, the subject-matter would need to be certain.

Secondly, with regards to Hayton’s second criticism, it is accepted that Dillon LJ’s will analogy is problematic. With a testamentary trust, the testator divests himself of all legal and beneficial ownership in the shares – it then becomes the responsibility of the executor (who has full ownership of the trust property), in compliance with his duties to the beneficiaries, to administer the estate correctly. Conversely, where a settlor intends to declare an inter vivos trust of property of which she is currently the full beneficial owner, her attempt to constitute a trust of a part of this property is complete only when she has taken the additional step of identifying the specific property to be held on trust. Until that point, the settlor has simply made an “imperfect gift” (Hayton and Mitchell). This argument has support from Worthington who uses Re Rose as an authority to explain that when a donor intends to make a gift of specifically identified legal property, equity will never assist with the gift of part of an identified bulk unless the donor has physically segregated the portion to be given away.

Hayton might, therefore, be correct in these last two criticisms. But to argue that Hunter becomes unworkable in light of them is to overlook how the stumbling blocks in Dillon LJ’s judgement have been smoothed out by the courts in post-Hunter case law.

Hunter was applied in the cases of Re Harvard Securities and Pearson v Lehman Brothers. In the former case, it was argued that a failure to segregate the trust property might lead a court to find insufficient intention to create a trust in the first place. In order to deal with this criticism, the court Pearson v Lehman Brothers adopted an alternative approach suggested by Goode. Goode stated that a trust should be treated as having been declared in relation to a proportionate share of the trust fund i.e. 50 shares of 950 = 1/19th, and according to Briggs J, this upholds the settlor’s true intention. This solution is indeed workable in relation to shares rather than wine bottles or gold bullions as a declaration of 40 of 80 wine bottles or gold bullions cannot be interpreted as a declaration of trust of 50 per cent of the settlor’s right to the whole 80 gold bullions or wine bottles, as each bottle or bullion is “a specific thing, and has a specific existence independent of any broader right” (Hayton and Mitchell). This therefore shows that the solution in Hunter is fair, sensible and workable. In fact, this workable approach was followed in White v Shorthall – an Australian case similar to Hunter involving a trust for shares being upheld despite not being segregated from their entire shareholding, on the basis that the declaration of trust can be explained as an equitable tenancy in common since it seems highly arbitrary to find no trust where the settlor declares himself trustee of 50 units and a valid trust where he says 5%. Moreover, given the practical difficulties involved in physically segregating intangible property, it is submitted that the solution in Hunter is fair, sensible and workable.


In light of the aforementioned arguments, the judgment in Hunter has admittedly caused bewilderment leading me to believe that is it in fact the correct decision but due to Dillon LJ’s will analogy as well as his tracing argument, has been decided in the wrong way. Notwithstanding the academic criticism, the solution that shares do not need to be segregated is fair, sensible and workable, through the application of the alternative solutions proposed in the cases of Pearson v Lehman Brothers and White v Shortfall, and as such, it should continue to be followed in subsequent cases.