(ii) The Essence of Tracing
Where the original property cannot be followed (because, for example, it has been dissipated) it is necessary for the claimant to show that the value of the property in which he or she originally had a proprietary interest can be identified in property that has been received by the defendant. In other words, the tracing rules enable the claimant to identify substitute property in the defendant’s hands which the claimant had not previously owned but which can be considered to represent the claimant’s original property. Only once the claimant has done this can he or she claim the property in the hands of the defendant or the value of that property. So, for example, if the claimant paid ?10,000 to a friend because of a fundamental mistake and the friend then used that money to buy a car, which she then sold for ?15,000 and used the proceeds of sale to buy shares which she gave to her daughter, the claimant may wish to claim the shares from the daughter. To establish such a claim, the claimant will need to show both that he retained a proprietary interest in the original ?10,000 which he paid by mistake and that this proprietary interest can be traced into the car, the proceeds of the car and ultimately into the shares, so that the value in the original ?10,000 now subsists in the shares. Whether the claimant can establish this depends on the application of the tracing rules360 which are evidential rules and presumptions that enable the claimant to prove that value in the original property is now represented in substitute property. The essence of tracing was identified by Lord Millett in Foskett v McKeown:361
Tracing is thus neither a claim nor a remedy. It is merely the process by which the claimant demonstrates what has happened to his property, identifies its proceeds and the persons who have handled or received them, and justifies his claim that the proceeds can be regarded as representing his property. Tracing is also distinct from claiming. It identifies the traceable proceeds of the claimant’s property. It enables the claimant to substitute the traceable proceeds for the original asset as the subject matter of his claim. But it does not affect or establish his claim.
Further, in Shalson v Russo362 Rimer J described tracing as ‘the process by which a claimant seeks to show that an interest he had in an asset has become represented by an interest in a different asset’.
In some cases it may be relatively easy for the claimant to show that value in one asset is now represented in another asset, such as where shares have been misappropriated from a trust and are sold for cash, and the cash is then used to buy a car. In such a case, value is cleanly transferred from one asset to another via the cash. But there will be other cases that are much more factually complicated, such as where many different amounts are credited to a bank account, including the claimant’s money, and many different amounts are paid from this bank account. In this situation it will not be obvious whether the value of the claimant’s money that has been credited to the account remains in the account, so the tracing rules are needed to identify where the money contributed by the claimant can be considered to be located.
The operation of the tracing rules is illustrated particularly well by the facts of Foskett v McKeown. In that case, the beneficiaries sought to recover a proportionate share of the payment of a death benefit from the children of the trustee, where the fourth and fifth annual premium had been paid from the beneficiaries’ trust. The initial proprietary base was established by virtue of the money being held on an express trust for the beneficiaries. It was then necessary to establish by reference to the equitable tracing rules that the money from the trust fund could be traced from that fund, through various bank accounts, into the premiums which were paid to the insurance company and from there into the payment of the death benefit following the suicide of the trustee. The first part of the tracing exercise was straightforward, since it could be shown that the claimants’ money had been used to pay two premiums. The difficulty in the case concerned tracing from the premiums into the payment of the death benefit, via the insurance policy, which turned on the appropriate analysis of the function of the premiums in the light of the unusual nature of the insurance policy. This was a unit-linked life policy under which the premiums were used to pay for the cost of life cover through the allocation of units that were exhausted over time. Each premium bought a number of units and each unit kept the policy going for a bit longer. The policy would lapse only once all of the units had been used up. Paying a premium was rather like topping up a parking meter: each time a payment is made, the car can be parked for an additional period of time. Since the first three premiums, which were paid from the trustee’s own money, had purchased a substantial number of units, it followed that even if the fourth and fifth premiums had not been paid, the policy would not have lapsed at the time of the trustee’s death, because the units purchased from the first three premiums were still operating. So what was the effect of the fourth and fifth premiums?
For Lord Steyn, who dissented, the fact that the policy would not have lapsed had the fourth and fifth premiums not been paid meant that there was no link between the payment of those premiums and the receipt of the death benefit, so tracing was not possible. This involved a simple causative approach to the tracing exercise: the fourth and fifth premiums had not contributed to the death benefit being paid. The majority disagreed and adopted a different approach to the tracing exercise. Although they acknowledged that, in the events that happened, the premiums paid from the trust fund were not required to prevent the insurance policy from lapsing, they also recognized that this need not have been the case. If, for example, the trustee had lived longer, the premiums would have contributed to the maintenance of the policy. Consequently, Lord Browne-Wilkinson recognized that367 ‘the beneficial ownership of the policy, and therefore the policy moneys, cannot depend on how events turn out. The rights of the parties in the policy, one way or another, were fixed when the relevant premiums were paid when the future was unknown’. It followed that it was possible to trace the two premiums into the insurance policy, which was property in its own right because it consisted of a bundle of rights to which the policy holder was entitled in return for payment of the premiums, and from that to the death benefit, which represented the traceable proceeds of the policy and indirectly of the premiums.
In reaching the conclusion that it was possible to trace into the death benefit, the majority identified two fundamental principles of tracing.
(1) Attribution Rather Than Causation
If the tracing rules were to depend on establishing a causal link between the receipt of the original asset and obtaining a substitute asset, in the sense that, but for the receipt of the original asset the substitute would not have been obtained, tracing into the death benefit could not have been possible in Foskett v McKeown because the fourth and fifth premiums did not cause the death benefit to be obtained. But, since the majority recognized that tracing was possible on the facts, it follows that tracing need not depend on identifying a causal link between the original and substitute asset. Rather, tracing depends on attribution. In Foskett v McKeown it was sufficient that the death benefit could be attributed to the fourth and fifth premiums, and this could be shown because the death benefit was to be paid, according to the terms of the insurance policy, in consideration for all of the premiums paid, which therefore included the fourth and fifth pre- miums. This shift away from causation to attribution is important to our understanding of tracing, especially when it is coupled with a second key conclusion about the nature of tracing.
(2) Tracing Value Rather Than Identifying Property
All of the judges in Foskett v McKeown recognized that tracing was not concerned with the identification of chains of property, but instead focused on the identification of value within property. It is this value that is the essence of the claimant’s proprietary right and it is this value that is traced. This was expressly recognized by Lord Millett:
We speak of tracing one asset into another, but this too is inaccurate. The original asset still
exists in the hands of the new owner, or it may have become untraceable. The claimant
claims the new asset because it was acquired in whole or in part with the original asset.
What he traces, therefore, is not the physical asset itself but the value inherent in it.
The recognition of these two principles in Foskett v McKeown means that the operation of the tracing rules should be easier. Tracing does not depend on causation in any meaningful sense. Rather, we are concerned only with logical progression: with the identification of value in various locations without regard to the effect of that value on particular property, in that it need not be shown that the property was acquired because of that value. In Foskett v McKeown the claimants’ value could be traced from the trust fund, through bank accounts, into two premiums, then into the policy itself, and finally into the proceeds of that policy. In a telling phrase, Lord Millett talked of establishing ‘transactional links’;371 this is now the essential feature of tracing.
Although the decision of the majority in Foskett v McKeown is very significant to the modern understanding of the function of the law of tracing, the conclusion that it was possible to trace into the death benefit was dubious for two reasons. First, the assertion that the court is not concerned with how events turned out, but rather with proprietary rights at the time at which the premiums were paid, is inconsistent with the key conclusion that tracing is simply a matter of evidence. Consequently, the court should have regard to all of the evidence, and so, if the premiums paid from the trust fund might have contributed to the payment of the death benefit but did not actually do so, this should have defeated the tracing exercise. Secondly, even though the majority relied on the fact that the terms of the insurance policy stated that the ?1 million was paid in consideration of all of the premiums, it is not clear why this contractual term was sufficient to influence proprietary rights to the money. In particular, the fourth and fifth premiums were used to buy units that formed part of a mixed fund of units, since the first three premiums had also purchased units that had not all been used up. Until the trustee’s suicide, these units were gradually used to prevent the policy from lapsing. But which units would have been used first? The logical answer is that the first in time would have been used first, which would mean that the units attributable to the fourth and fifth premiums remained outstanding, and this conclusion could not be changed by the inclusion of a contractual term that the premiums were paid in consideration of all premiums, unless that term stated explicitly that the most recently paid premiums were to be treated as used first. It follows that, for both of these reasons, the preferable view is that it should have been possible to trace the trust funds into the fourth and fifth premiums and into the purchase ofunits, but it should not have been possible to trace into the payment following the trustee’s death.
-  359 Smith, The Law of Tracing, 3. 360 See p 613, below. 361  1 AC 102, 127.
-  362  EWHC 1637 (Ch),  Ch 281, .
-   1 AC 102. See p 568, above.
-  For tracing into and through bank accounts in Equity, see p 618, below.
-  Foskett v McKeown  1 AC 102, 113.
-  Ibid, 111 (Lord Browne-Wilkinson) and 138 (Lord Millett). 367 Ibid, 111.
-  Ibid, 137 (Lord Millett). Compare Lord Hope who, dissenting, expressly stated that the death benefit wasnot attributable to the payment of the premiums: ibid, 122.
-  Ibid, 116 (Lord Hoffmann), 119 (Lord Hope), and 133 (Lord Millett).
-  Ibid, 128 . 371 Ibid.