What did McPhail v Doulton decide? To what problems has it already given rise?

To what
problems may it give rise in future?

The case of McPhail v Doulton is a landmark decision of the House of Lords in the field of English
Trusts law as it redefined the certainty test for discretionary trusts. The case concerned a deed
executed by Mr Betram Baden which settled a non-charitable trust for the benefit of the staff of
Matthew Hall and Co Ltd and their relatives and dependents. The clause at issue was clause 9, the
objects clause, which stated "The trustees shall apply the net income of the fund in making at their
absolute discretion grants to or for the benefit of any of the officers and employees or ex-officers or
ex-employees of the company or to any relatives or dependants of any such persons in such
amounts at such times and on such conditions (if any) as they think fit." It was challenged that this
was not sufficiently certain to fulfil the certainty of objects test for a valid trust. The decision which
followed altered the path of trusts law relating to discretionary trusts, but has also given rise to
many problems regarding the mechanism of the revised certainty test, issues of administrative
unworkability and whether the test conflicts too dramatically with orthodox trust theory.

As mentioned, the decision in Mcphail v Doulton changed the certainty of objects test for
discretionary trusts. Prior to the decision, the rule relating to discretionary trusts mirrored that of
fixed trusts in that every potential beneficiary had to be ascertainable meaning it would have to be
possible to make a complete list of all possible beneficiaries. However, Lord Wilberforce in the case
of McPhail v Doulton (whilst maintaining the complete list test for fixed trusts) stated that in the
case of discretionary trusts the test would be the same as that for powers outlined in the Re
Gulbenkian Settlement case. This test is the ‘in or out test’, meaning that you need only ask "can it
be said with certainty that any given individual is or is not a member of the class” in order for the
certainty test to be satisfied. In order to adopt this test Lord Wilberforce had to overcome two
arguments for the maintenance of the stricter test. First, that a trustee’s duty to distribute could
only be performed if he was able to consider every possible claimant and secondly that the court
could only execute the trusts, on failure of the trustees to do so, by equal division of the fund. Lord
Wilberforce met these concerns head on by explaining that it was only necessary to consider every
possible claimant if one was fully distributing the fund, i.e. essentially winding it up. In such cases
there would necessarily be a wider and more systematic survey of all potential beneficiaries when
deciding to make grants, yet still there was no requirement to draw up a complete list of names, as
the law did not require the exercise of a discretionary power. Further he felt that the court being
called upon to execute the trust if the trustee would not do so was a theoretical rather than a
practical difficulty as in the cases which had reached the courts; there were no examples of a trustee
refusing to act. Further, it was not the case, in his view, that distribution was impossible unless there
as an equal division, and he cited several older cases, prior to 1801, in which the court exercised
discretion in relation to the making of distributions. By overcoming these arguments, Lord
Wilberforce found the more relaxed powers test to be more appropriate. There was a strong case
for assimilating the tests for powers and discretionary trusts as it is often difficult to ascertain which
exists in a certain case and so a relaxation of the test was needed to prevent many trusts from
failing. On the other hand, it has been argued that this assimilation of the tests for powers and
discretionary trusts destroys what used to be one of the most important reasons for distinguishing
between trusts and powers, which could give rise to difficulties considering the line between trusts
and powers is already a thin one and so this decision could make distinguishing between the two
even more difficult in future.
However, the real problems stemming from the decision in McPhail v Doulton are concerned with
the mechanisms of the test itself. It is stated that the test is concerned with conceptual as opposed
to evidential certainty. Therefore as long as a class of beneficiaries is conceptually certain then it is
not an issue that it may be evidentially difficult to state whether a person is or is not a member of a
class. The problems or difficulties with this test were highlighted in the case of Re Baden’s Deed
Trusts (No.2) when it had to be decided whether the words “dependants” and “relatives” were too
conceptually uncertain to be given effect too. The problem with using a conceptual certainty test is
pointed out by Hanbury and Martin, who highlight that different minds may take different views on
the question of whether a particular description is conceptually certain or not and there are likely to
be many borderline cases. Moreover, the test leads to very fine, arbitrary distinctions, so as to give
the result that “friends” would not be conceptually certain whereas “old friends” was stated to be
sufficiently certain in Re Gibbard. It can be argued that by using conceptual certainty for the basis of
the test for discretionary trusts the law is being opened up to potential conflict as different judges
may have different views on what is or is not conceptually certain. It would seem ironic that a
certainty test is in fact itself subject to uncertainty in its application. However, arguably, it is possible
to find a solution to the problems created by a need for conceptual certainty as provisions could be
included which leave the definition of a term to a third party meaning he can give effect to the
words preventing the need for conceptual certainty from frustrating the trust. This was the position
in Re Coxen whereby it was held a testator could validly make the trustee’s opinion the criterion
provided he had sufficiently defined the state of affairs on which they were to form their opinion.
Hence, there are methods of circumventing some of the potential problems which could be caused
because of the use of conceptual certainty.

However, more fundamentally, it is suggested that there are problems with the very idea of an ‘in or
out test’. It seems to require demonstrating that a person either definitely is or definitely is not part
of the class. But in some circumstances it may be extremely difficult to do this, for example trying to
prove you are not somebody’s relative. Essentially this is the evidential difficulties argument and the
problems it creates have been demonstrated in the case law as judges have been unable to decide
on the best way to get around the problem. In Baden No 2 Megaw LJ outlined that the test was
satisfied if it is possible to say whether at least a substantial number of objects certainly fit within
the trust, even if the answer of ‘in or out’ for a number of other persons would be “it is not proven”
rather than that they are definitively not in the class. This indicates the problem of evidential
difficulties which require the test in Mcphail to be watered down slightly so that definite proof is
perhaps not needed. Yet this has not been universally accepted; Stamp LJ instead stuck to the
stricter test, stating you had to be able to say whether they were or were not within a class. Yet in
this case, Stamp was able to save the trust from failure by construing relatives as meaning next-of-
kin, or nearest blood relations and so he narrowed down the words enough to enable the evidential
difficulties to be overcome. Yet Stamp was in the minority here, with the majority holding that
relatives meant a descendant from any common ancestor. This would seem to be a far broader
definition and one which would again make it difficult to prove whether you were or were not a
descendant, giving rise to evidential difficulties. Hence the requirement of conceptual certainty
means that evidential problems can then arise which makes the ‘in or out test’ (literally read)
difficult to satisfy and could lead to beneficiaries not being included and the trust failing for
uncertainty. It also highlights the effect of the decision on orthodox trust law principles such as the
beneficiary principle, which states there must be clearly identified beneficiaries. By allowing a class
such as relatives to satisfy a certainty test, the class could be so vast that the beneficiary principle
has been reduced to near vanishing-point, which has led to criticisms of the Mcphail decision.
Another problem surrounding the Mcphail decision is administrative unworkability. This problem
arises from the judgment of Lord Wilberforce in the Mcphail case where he indicated that there may
be a basis for differentiating between the test for a discretionary trust and a mere power as there
will be cases where a description of beneficiaries, which although complying with the certainty test
might be so wide as to not form anything like a class. He gave an example of the “resident’s of
Greater London” which would be too wide for the trust to be practical. This issue was considered in
relation to powers in the case of Re Hay’s Settlement trusts where it was held that a mere power is
not invalid on the ground of the size of the class unlike a trust- due to the more stringent nature of
the duties of the discretionary trustee. However, although this may seem like a reasonable
proposition it has led to confusion as to whether it is numbers of potential beneficiaries at issue or
the capriciousness of the trustee which is at issue. Hence, in Re Manisty’s settlement it was stated
that the Greater London Trust would not be void on the basis of numbers but on the ground that the
terms of the power negatived any sensible intention on the settlor’s part and any sensible
consideration by the trustees leading them to simply divide the trust in a capricious manner. This
begs the questions whether capriciousness is the same test as administrative unworkability. It seems
it is not as capriciousness has been held to apply to both mere powers and discretionary trusts and
so we are left with a proposition that despite the certainty test being fulfilled there are cases where
a class is so wide that there is an additional duty on the trustees to make a survey of the range of
objects so as would enable them to carry out their fiduciary duty. In cases of wide ranging
discretionary trusts, the trustee must not simply proceed to distribute the funds amongst the
beneficiaries who happen to be at hand, but must first consider what person or persons are objects
of the power in order to enable him to use his trust power effectively. This seems to complicate the
certainty test and suggest it is no longer sufficient on its own in certain cases to decide if a trust is
valid- it must be assessed against administrative unworkability also. It could be argued that the
original test was far more straightforward and would not have called for policy rationales to help
decide if a trust is valid. Indeed, Hardcastle has been keen to point out that the logical and juridical
basis of the administrative unworkability test is uncertain, as “it is unclear from Lord Wilberforce's
formulation whether administrative unworkability is a source of definitional uncertainty or a ground
of invalidity in its own right”, suggesting the basis of the administrative unworkability test is
unsound.

It could be said that problems may arise in the future as a result of Mcphail v Doulton. Lord
Wilberforce indicated that a lot of the potential problems of the court having to be able to enforce
the trust were theoretical as no case had become before them where a trustee had refused to
exercise his duty. However, if a case did come, this could create problems, as how is the court to go
about enforcing the trust and on what basis, when the beneficiaries are not clearly identified? It may
be that the court would enforce the trust in the manner best calculated to give effect to the
testator’s intentions. As stated by Lord Wilberforce in Mettoy Pension Trustees Ltd v Evans the court
may appoint new trustees or authorise direct representatives of the classes of beneficiaries to
prepare a scheme of distribution or may even direct the trustees to distribute. There could also be
issue regarding the conceptual certainty test if a case were to come to the court in which a long
series of catagories did comply with the Mcphail test and yet there was a single added category
which did not comply. It seems from the authorities that this would be likely to cause the entire trust
to fail which could be seen as a harsh rule indicating a need to come up with a severance principle
whereby the conforming parts of the trust can still be upheld.
Overall, it is clear that the new test for discretionary powers has not been without its problems or
criticism. Conceptual certainty and evidential difficulties have been at the core of the problem, but
so too has the perception that the new test infringes upon key orthodox trust principles such as the
beneficiary principle. However, Grbich has suggested that these issues could be overcome by a
realisation that the court in Mcphail was actually dealing with a new type of trust, which he defines
as the “modern discretionary trust”. He feels that the minority and majority in the case of Mcphail
were talking at cross-purposes, with the majority talking of the modern discretionary trust and the
minority dealing with the more traditional direct gift with a power to divest. He states by seeing it as
a modern discretionary trust the assimilation of the powers and discretionary trust test is better
understood as there is no imposed obligation to hold property for individual objects in a modern
discretionary trust and no obligation to allocate during the selection period, there is merely a power
to allocate and so there is no need to have a list of every beneficiary. Perhaps by recognising this
new type of trust, the logical basis behind the Mcphail decision can be better understood and the
reason why these trusts do not need to conform to the orthodox beneficiary principle can also be
gathered.