Canterbury Law Review
The decision of the House of Lords in Foskett v McKeown will have a profound impact on the process of tracing. Foskett brings into sharp focus the relationship between the terms of a contract under which a substitution takes place and the rules of tracing into the substitute asset. The concept of causation, which is evidenced by an emphasis on the terms of the contract, has now been overcome. The new concept of attribution appears to be no more than a sophisticated form of the concept of attribution, and should also be rejected.
Foskett also has a bearing on the scope and effects of the rule in Re Hallett's Estate with regard to the process of tracing. The speeches of the majority in Foskett overturn the perceived rule in Hallett, bringing greater coherence to process of tracing.
The decision in Foskett can be justified on the basis of tracing into direct substitutes. However, it may also have an impact on the swollen assets theory. This alternative set of tracing rules has been doubted in some jurisdictions. However, in New Zealand, the theory may still be valid. The decision in Foskett may force a re-examination of the theory in the near future.
A company controlled by Mr Murphy and Mr Deasy had entered into agreements with 220 purchasers for the sale of plots of land in Portugal. The purchasers paid deposits amounting to £2,645,000 to Mr Deasy. This money was held on trust for the purchasers. Although Mr Murphy was not an original trustee with regard to the purchasers, he must have become a trustee for them at some point. All of the money was eventually dissipated.
Some of the money had been used by Mr Murphy to pay premiums for a policy of insurance on his own life. The policy was held on trust for Mr Murphy's children. Pursuant to this policy, a Death Benefit would be paid at the time of Mr Murphy's death. The Death Benefit would be either £1,000,000 or the aggregate value of units that had accrued under the policy. The policy contained a formula for working out how many units were in a "fund" of such units. From the payment of the second premium onwards, units were allocated to this fund.
The fund constituted the investment element of the policy. The other part of the policy was the insurance element. The insurance element was initially paid for by part of the annual premiums, however, from the third premium onwards, it was paid for through cancelling units from the fund. After the first two premiums had been paid, and if a subsequent premium was not paid then the policy would be converted into a "paid up" policy. Annual premiums would cease to become due, and the insurance part of the policy would be funded through cancelling units in the fund. This did not occur because five premiums were paid, and Mr Murphy died in the fifth year. Mr Murphy paid the first two premiums out of his own money. While the source of the third premium was disputed, it was agreed that the last two premiums had been paid with the purchasers' money.
The House of Lords decided by a narrow majority that the purchasers were entitled to a share of the proceeds equal to the proportion of the premiums paid with their money. Lord Millett gave the leading speech, followed by Lord Browne-Wilkinson and Lord Hoffmann. Lord Hope and Lord Steyn delivered dissenting speeches. Even within the majority, however, there was a great disparity of views.
Tracing is a process that allows value to be tracked through substitutions. The purpose of this part of the paper is to explore, from the perspective of Foskett, the way in which this occurs. In particular, the terms of the contract under which a substitution takes place may have some effect on the way in which value is tracked through it. This is relevant to the argument that tracing requires some kind of causative link to be proven. This issue cannot be allowed to have any effect on the tracing process. Furthermore, although separate from the issue of tracing, the effect of a substitution on proprietary rights shall be considered.
The ability to trace in Foskett was complicated by the fact that the insurance policy contained different divisions of value. The mixed fund of units might be considered to be the mixture of value in this case. However, it is more accurate to recognise that the true mixture of value was the insurance policy itself. This view recognises the contributions made to the insurance element of the policy.
Lord Millett, in contrast to the rest of the majority, was convinced that the terms of the policy were of paramount importance. His Lordship looked at the exact terms of the policy in order to determine the value of the respective contributions to it. The units were considered an integral part of the tracing chain. Lord Millett determined that the proportions in which each party had contributed to the value of the units was the correct way to determine the extent of their respective contributions to the policy. This argument overlooks the fact that the first two premiums contributed to the insurance element of the policy even though they did not purchase units.
This emphasis on the terms of the contract suggests that Lord Millett applied a sophisticated form of the causation theory of tracing. This theory holds that "tracing just helps to show a causal link". Care, however, must be taken to distinguish the sophisticated form of causation from the basic form. The basic form is equivalent to a "but for" test: the payment of premiums using the purchasers' money did not cause the payment of the Death Benefit because it would have been paid even if no subsequent premiums had been paid. By contrast, close reliance upon the terms of the contract may be equivalent to applying a more sophisticated concept of causation.
It has been argued that the majority in Foskett held that tracing was dependent on "attribution not causation." In other words, the majority decision was not based upon any concept of causation. However, this is not correct if the concept of attribution is simply the sophisticated form in disguise.
Lord Millett was concerned with the fact that some units contributed to the insurance part of the policy (which, in the events that happened, caused the Death Benefit to be paid). His Lordship further held that the Death Benefit was paid because all the premiums were paid. Surely this is equivalent to holding that payment of the premiums caused the Death Benefit to be paid?
It has been assumed that "the use of trust money to pay [the later premiums] had no effect on the death benefit payable." While it is true that sufficient units were allocated to the fund after the first two years premiums had been paid, it is certainly not true to say that those specific units were the only ones that were cancelled to pay for the insurance part of the policy.
It is necessary to realise that units were cancelled to pay the cost of insurance in later years. It is quite wrong to say that the Death Benefit would have been paid even if no further units were cancelled from the fund to pay for the insurance part of the policy. The insurance would run out at the end of the second year, and without cancelling any more units, and with no further injections of premiums, there could be no possible way to pay for insurance for the later years.
The only way that an argument for a basic form of causation can be supported is on the basis that the units cancelled to pay for the later insurance premiums were those paid for by the second premium. However, it is not possible to know which units were cancelled. There was no separate fund of second-premium-units that could be used without contamination from later-premium-units. Therefore some units that paid for the later insurance premiums may well have come from the payment of the later policy premiums. So the money of the purchasers probably did help to cause the payment of the Death Benefit, in a sophisticated sense of causation. It is literally true that in this case "we can see what has been paid out and why it was paid."
Lord Millett therefore applied a sophisticated form of causation in drawing a distinction between what actually happened and what could have happened if the later premiums had not been paid.
It must be realised that the sophisticated form of causation is not necessarily excluded by the majority decision in Foskett that it was irrelevant that the second premium had provided sufficient units. For example, assume the policy had a clause that clearly expressed that the first units into the fund were also the first units cancelled to pay the cost of insurance. In that situation, it would be correct to say that the later premiums had not caused the Death Benefit to be paid. In such circumstances, if his Lordship's reasoning were taken to its logical conclusion, Lord Millett might have disallowed a proportionate share of the proceeds.
However, if the rules of tracing were dependent on issues of causation then the following situation might arise: A steals $50,000 from B and then enters into a contract to exchange the money for an exercise machine, with a Toyota car thrown in for free. In this situation, the exercise machine is the substitute for the money but the car is not, despite the car being worth many times the value of the exercise machine. However, it does not seem appropriate to deny B's ability to trace into the car simply because, on the terms of the contract, it was not "bought" by the money. The point is that the rules of tracing cannot be affected by issues of causation. Lord Millett therefore placed too much emphasis on the exact terms of the insurance policy.
It is vital to ascertain exactly what was substituted for the premiums. Were the premiums substituted for the policy as a whole, or were they substituted for the units of account that were, under the terms of the policy, cancelled to pay the cost of life insurance? If the question is one of attribution, it is not entirely clear why there should be any concern with the intricate terms of the policy.
Lord Hope, in his dissenting speech, framed his argument in terms of attribution, by holding that the entire Death Benefit was attributable to the payment of the first premium. This statement is puzzling because the first premium did not contribute to the fund of units. The first premium did nothing more than pay for the cost of insurance for the first year. It did not confer any right to recover the Death Benefit in subsequent years. If no further premiums had been paid after the first, then there is no question that the Death Benefit would not have been paid on Mr Murphy's death.
Is it important to rely on the terms of the policy? As Lord Hope recognised, the policy stated that the Death Benefit was to be paid "in consideration for all the premiums." His Lordship believed that this was simply an answer to the argument that there was no causative link between payment of the later premiums and the payment of the Death Benefit. It was due to the vague nature of the concept of attribution that his Lordship was able to place greater significance on to the first premium than the later ones.
It is important to see through such obfuscation. It appears that both Lord Hope and Lord Millett placed too much reliance upon the terms of the policy. The better view may be to ignore the terms of the policy and to concentrate solely upon the inputs and outputs related to the policy. Looking at the contributions to the policy, it can be seen that the premiums were the only input. While Lord Hope thought that it was proper to regard only part of this input as attributable to the payment of the Death Benefit, this would defeat the purpose of tracing through mixtures of value. The mixture of value was the policy itself, not the fund of units under the terms of the policy.
Lord Hope employed the argument that everything depends upon what is "fair, just and reasonable" to deny the ability of the purchasers to assert that their value was now represented by a proportionate share of the proceeds. However, care must be taken to distinguish this argument from tracing. Tracing is a separate issue and should not be confused with the kind of claim that can be brought by the plaintiff.
It appears, therefore, that Lord Hope was not in favour of any role for causation or attribution in the process of tracing. While attribution could be relevant to the type of claim that the purchasers could finally assert, it would not be relevant to the process of tracing. This is consistent with the argument that "tracing is not coterminous with causality." The concepts of causation and tracing are quite separate and care should be taken not to confuse the two.
The result in Foskett upholds the approach that it is unnecessary to be ruled by the terms of the contract under which a substitution has taken place. Both Lord Browne-Wilkinson and Lord Hoffmann displayed welcome affection for the view that the terms of the policy were unimportant. Their Lordships held that it was important to take account of the relative contributions to the premiums only. The policy could not force the rules of tracing into its own mould. The purchasers were therefore entitled to share in the proceeds of the policy in the same proportion that they had provided toward the payment of the premiums. Causation and attribution are thus distanced from the law of tracing.
Contrary to the argument of Lord Hope, Lord Browne-Wilkinson pointed out that
[t]his case does not depend on whether it is fair, just and reasonable to give the purchasers an interest as a result of which the court in its discretion provides a remedy. It is a case of hard-nosed property rights.
Although this is not a statement about tracing, it has implications for tracing. If equitable property rights are discretionary, why is there any need to trace them? Lord Hope clearly thought that it was possible to trace from the payment of the premiums into the proceeds. However, the consequence of tracing was not, in his Lordship's view, an automatic creation of equitable property rights in the new asset. His Lordship held that even if the purchasers could trace into the proceeds, they would not be able to establish a right to either an equitable lien or a proportionate share unless equitable considerations favoured it. Against this view, Lord Millett held that the right of the purchasers in the policy dated from the time that their money was used to pay the premiums. This means that the equitable interest dated from the time of substitution, not from the time ofjudgment. There is difficulty with the idea that perfect proprietary rights (such as equitable ownership) spring into existence at the time of substitution. When a substitution occurs, there is some value in the new asset that can be ascribed, on the rules of tracing, to the old asset. It does not follow that the plaintiff will be able to make a successful claim in respect of the new asset. Even if the plaintiff could make a successful claim, she might wait until she could assess whether the new asset had appreciated or depreciated in value. In this way she could see which type of claim would be more advantageous. Therefore, how can the right that the plaintiff finally obtains have existed in the new asset from the time of substitution?
The answer could be that the right that the plaintiff gains at the time of substitution is a generic proprietary right. This right is not created by the process of tracing. It is created by the simple fact of substitution. The right cannot yet be legal or equitable because there is no way of knowing what type of claim the plaintiff is eventually going to be successful in proving. Indeed, the plaintiff might fail to prove any claim to the new asset. In this situation, the generic proprietary right would be extinguished. However, if an appeal were to be successful then the generic proprietary right would never have been extinguished.
At the time judgment is granted in favour of the plaintiff, the generic proprietary right would be converted into a perfect proprietary right. It is at this point that the right becomes legal or equitable. The generic proprietary right serves only as a placeholder. It preserves the priority of the proprietary right that is eventually granted. When the perfect proprietary right is finally granted, this is really a conversion of the generic proprietary right. Since the generic right existed from the time of substitution, the perfected right is considered to have existed from the time of substitution. The speech of Lord Millett in Foskett can be justified on the basis of a generic proprietary right. However, that is not the only way in which it can be justified.
Lord Millett held that where a beneficiary can claim a lien, this "is an exception to the primary rule...". This could mean that a lien arises only from the date of judgment, while equitable ownership arises at the time of substitution. However, this would make a lien much more susceptible to intervening third party property rights than under the generic property right analysis.
Where there was no mixture of value (a clean substitution), tracing was rights-neutral but where there was a mixture (a mixed substitution), it was not. This part of the paper demonstrates that tracing is rights-neutral in all situations. Foskett clears up a misinterpretation of earlier cases that suggests any other conclusion.
This issue was clouded by "the ambiguous dicta of Sir George Jessel [MR], in Re Hallett's Estate." Lord Millett overcame the perceived difficulty in Hallett by holding that "the time has come to state unequivocally that English Law has no such rule." However, it is not at all clear that the result in Foskett is contrary to the rule in Hallett. If this is correct, there was no need for his Lordship to disapprove so strongly of the reasoning in Hallett.
The reasoning in Hallett is not necessarily inconsistent with a pro-rata sharing of the product of the mixed substitution. Sir George Jessel MR was not considering the situation in which the product of the mixed substitution had increased in value. His Lordship was concerned that, because a plaintiff has not paid for the whole of the product of a mixed substitution, she should not be entitled to take the entire product of the mixed substitution in satisfaction of her claim. There has to be some kind of division of the product of the mixed substitution. In Hallett, this division was achieved by imposing a lien. It does not necessarily follow that a lien is the only way of making a division. This argument has been criticised as being a "strained" interpretation of the case.
The decision in Dexter Motors Ltd v Mitcalfe gives further insight into this issue. In that case, Fair J expressed the opinion that: 
a cestui que trust or principal is entitled to regard property which has been purchased with his own money as his property, but not so to regard property the purchase moneys of which were only partly provided by him.
This statement must be read in the context of the case. It means that the principal cannot be entitled to the whole of the property purchased where his value did not provide the whole of the purchase price. Some kind of division must be made. The statement of Fair J can be read as holding that there must be a division because the principal did not pay for the whole of the product of the mixed substitution.
The time at which a right comes into existence is a difficult issue. Does the right exist from the time of substitution onwards? One possibility is that the best the plaintiff got at the moment of substitution was a restitutionary power in rem? This is a power to take steps to perfect one's legal or equitable ownership of an asset. This concept is specifically designed to prevent several difficulties with the idea that a right in the new asset springs into existence immediately upon the act of substitution. As will be seen, it is not the same as the idea of a generic proprietary right that was earlier introduced.
First, if the plaintiff wishes to trace for the purpose of asserting a proprietary right in an asset, she must show that she has a perfect proprietary right at each stage of the tracing chain. The idea of a power in rem would solve this difficulty because it would eliminate the need to show a proprietary right at every link in the chain. The power in rem would therefore allow the plaintiff to assert that a certain asset was the traceable product of his value despite an inability to assert a perfect proprietary right in respect of one or more assets in the tracing chain. This assumes that a perfect proprietary right must be shown at every link. However, if all that needs to be shown is a generic proprietary right, this argument for a power in rem loses its force.
Second, the rights gained in certain assets at each link in the chain might be different rights from those gained in relation to other assets in the tracing chain. However, this difficulty is easily dismissed because we trace value, not rights. The rights exigible against an asset are merely a consequence of the value inherent in that asset. When a substitution occurs, a generic proprietary right is the only right that arises in favour of the plaintiff.
Third, there is the possibility of a "multiplication of ownerships". It has been argued that "there is an obvious objection to [a plaintiff] being owner of both the original and the substitute product." This is because, with a large number of substitutions, the plaintiff could quickly become owner of a multitude of assets. The existence of a restitutionary power in rem would prevent the multiplication of ownerships because once a plaintiff had asserted his ownership in one asset, he would then be prevented from claiming the substitute. However, it has been observed that there is no requirement that a plaintiff must lose legal or equitable property rights in order to acquire new ones. There should be no difficulty with allowing the plaintiff to gain a new property right because of the actions of a thief. A generic proprietary right is a more satisfactory analysis because such a right would not be extinguished simply because a successful claim had been made in relation to a substitute asset.
Another issue relates to the time of the creation of the new property right. Conceptually, the restitutionary power in rem analysis is forward looking. Any rights acquired upon the 'ratification' of the action of the thief exist from the time of ratification onward. The only possible alternative has been suggested to be a kind of retroactive ratification. Upon ratification the plaintiff is considered to have had the rights from the time of substitution onward. It is unclear whether or not equitable property rights could be retroactively validated. However, as the general rule in equity is that the first in time prevails, it appears unlikely that retroactive ratification could work in relation to equitable property rights. With equitable rights then, the plaintiff who had only a restitutionary power in rem would never have rights in the substitute immediately upon the act of substitution. Such rights could exist only from the time of "ratification" onward.
The retroactive ratification principle could play havoc with intervening third party rights. Even if a third party found out about the defect after acquiring the asset, he would still not know what his position was until the plaintiff had completed his action of ratification.
Of course, the idea that a plaintiff acquires a generic proprietary interest (neither legal nor equitable) in any substitute product immediately and without any need for ratification, may not make the position of a third party any more certain. The plaintiff who gained such a generic right might choose to later vindicate that right. Any rights successfully claimed would exist from the time of substitution onward, even if they were equitable rights. No earlier equity could intervene because the generic right is being converted into an equitable right, and the generic right arose at the time of the substitution, rather than being retrospectively created, for example; by ratification. The third party in this situation would not know what kind of claim was to be brought, so his position is no more certain than with the power in rem.
These issues were raised in Lipkin Gorman v Karpnale Ltd? A firm had to prove that it could trace its interest from a bank account into the money paid to a club. This was in order to show that the club had gained something at the expense of the firm. It was held that tracing property into its product 
involves a decision by the owner of the original property to assert his title to the product in place of his original property. This is sometimes referred to as ratification.
It was pointed out that the firm had a legal proprietary right in the original bank account. Since a bank account is nothing more than a contract of debt, this legal property right was held in a personal right. The firm was allowed to trace their interest from that bank account into the money that was exchanged for it. The value that the club received was the traceable product of the firm's original property.
The firm's decision to trace for the purposes of proving that the club had been enriched by their value may have diminished the value of the claim that they could have against the bank. If the bank had acted wrongfully when it debited the account, the firm would have to make an election. Either it could claim the bank had wrongfully debited the account, or it could claim "ownership" of the money the club had received. The power in rem would not allow a right to be claimed in respect of the full value of both assets. If the firm claimed "ownership" of the money, it could only claim the value of the account after the money was withdrawn.
Of course, the firm did not make a claim in respect of either the account or the money. It made a claim against the club, but this claim could only succeed if the firm could show that it "owned" the money that had been paid to the club. In an action for money had and received, enrichment has to be at the expense of the plaintiff. The power in rem must have been exercised in order to show that the firm had some right to the money withdrawn.
It has been argued that, "in Lipkin Gorman, the plaintiff's claim was... for the sum of money received by the defendant and which belonged at law to his firm." This cannot mean that the firm owned the money. It must mean that the firm had some inferior proprietary right in the money. This is some legal "ownership" right that is less than full ownership because it can exist side by side with legal ownership. It has been noted that the power in rem was never exercised in Lipkin Gorman? However, if it was not then the firm could never have shown that it "owned" the money that Mr Cass withdrew from the account. If a power in rem was truly applicable to this situation then the assertion of rights in respect of the money withdrawn must have precluded assertion of rights to the original asset (the bank account) to the extent of the amount of money that was withdrawn by Mr Cass.
The obvious objection to the application of a power in rem in this context is that to "perfect" its interest in the money the firm must give up its property right in the original asset (the bank account) to the extent that it was diminished by the money paid to the club. In this situation, the power in rem is doing more that preventing multiplication of ownerships, it is actually downgrading the type of ownership that can be asserted because the right asserted in the money was less than full legal ownership. On that basis, there is no justification for the adoption of a power in rem in this kind of situation.
A generic proprietary right analysis would be applicable to Lipkin Gorman. At the time the money was withdrawn, a generic proprietary right in the money would have arisen in favour of the firm. This generic right was later converted into a legal ownership right less than full legal ownership. There is nothing in this analysis that requires loss of rights in the full value of the bank account.
Lipkin Gorman did not involve equitable interests, as in Foskett. The position taken in Foskett militates against the existence of a restitutionary power in rem. In that case, Lord Browne-Wilkinson held that the purchasers' interest under the original express trust was a continuing one that did not cease even though the identity of the trustee had changed. In addition, Lord Millett held that the plaintiff's 
interest arises by reason of and immediately upon the payment of the premiums, and the extent of his share is ascertainable at once.
While neither Lord Browne-Wilkinson nor Lord Millett adopted the idea of a restitutionary power in rem, their Lordships supported their arguments on different grounds. Lord Millett was clearly contemplating the creation of a new right that springs into existence as soon as a substitution is made, without requiring any act of ratification. Lord Browne- Wilkinson, on the other hand, thought that the interest was a continuing one that always existed from a point in time prior to the substitution.
It may be the case that the different approaches taken by Lord Browne- Wilkinson and Lord Millett reflect different views as to the type of trust to which the purchasers were entitled. In Boscawen v Bajwa, Millett LJ (as he then was) suggested that where a plaintiff succeeds in tracing, and if he proves that the property is still in the hands of the defendant, a constructive trust arises in favour of the plaintiff. This constructive trust was held to arise by operation of law, and was not dependent on any discretion vested in the court. The correct classification of the type of trust that applies, whether it is express, constructive or resulting, is not an issue of tracing, and is outside the scope of this paper.
The swollen assets theory is one way of "moderating the rigours of the requirements of tracing." It would allow the plaintiff to assert proprietary rights in any asset of the defendant. Orthodox tracing rules, on the other hand, confine the plaintiff to asserting a right in a substitute asset only.
It has been argued that "adoption of the [swollen assets theory] would be, in effect, a decision that [proprietary] rights are indestructible." However, this argument is conceptually flawed. The exercise of tracing does not allow a pre-existing proprietary right to be asserted against a new asset. It allows a new proprietary right to be asserted against a new asset. This is consistent with the position taken in Foskett. In that case, Lord Millett held that a proprietary interest arises at the time of the substitution. This militates against the idea that tracing merely allows a pre-existing proprietary right to be asserted against a new asset. Lord Browne- wilkinson, however, was of the opinion that the proprietary rights in the substitute asset were the same rights that had originally existed under the purchasers' express trust. The argument of Lord Browne-Wilkinson does not explain the situation where the right that the plaintiff wishes to assert against the new asset is a different type of right than the right he held in the original asset. In Boscawen v Bajwa, Millett LJ held that the type of remedy is not prescribed by the process of tracing.
In the United States, the swollen assets theory arose in litigation resulting from the failure of several banks in the 1930s. Some decisions demonstrated limited support for the theory, in allowing the plaintiff to assert a lien over any cash items of the defendant. Orthodox rules of tracing would confine the plaintiff to assets that could be said to be the direct substitutes for his original asset. The swollen assets theory would extend the power of tracing if it allowed the plaintiff to treat any cash items of the defendant as the substitute for his original asset. one way of further extending the rules of tracing would be to allow the plaintiff to treat any assets (cash or non-cash) of the defendant as the substitute for the asset of the plaintiff.
The Privy Council examined a version of the swollen assets theory in Space Investments Ltd v Canadian Imperial Bank of Commerce Trust Co. (Bahamas) Ltd. In that case, a bank was a trustee of a sum of money. The express terms of the trust allowed the bank to deposit the trust fund with itself. The bank did not, therefore, act in breach of trust when it put the money in with its general assets. Lord Templeman, however, made an interesting observation regarding the situation in which a trustee wrongfully mingles the trust fund with his own assets. In those circumstances, "equity allows the beneficiaries... to trace the trust money to all the assets of the bank".
Lord Templeman was not applying the ordinary rules of tracing. The dicta of Lord Templeman clearly advocate a wider role for the rules of tracing. Where the wrongdoer mixed another's asset with items of his own, the swollen assets theory would provide a connection with any of the wrongdoer's assets.
The New Zealand Court of Appeal may have approved this principle in Liggett v Kensington. The claimants wished to assert equitable charges over one asset in particular (a stock of gold). Cooke P allowed them to do this, although it appears that they could have asserted exactly the same interests in any asset of the defendant company. However, Gault J would have restricted the claimants to the ordinary rules of tracing (which in this case also led to the stock of gold). McKay J was in the minority in holding that the claimants could not trace at all because there was no misappropriation of moneys in which the claimants had an interest.
The argument of Lord Templeman in Space Investments was mentioned by the Privy Council in Re Goldcorp Exchange Ltd (in rec). This case was an appeal from Liggett. A company had entered into contracts with customers. Pursuant to the contracts, the company became the absolute owner of the money paid to it. The customers did not retain any kind of interest in the money. Thus the proprietary rights of the customers had been extinguished prior to any mixture or substitution. The assets of the company were not swollen by the receipt of any value from the customers. Nevertheless, Lord Mustill made a curious obiter statement that the judgment in Space Investments would not have assisted the customers in their efforts to trace their value because the money was paid into an overdrawn bank account.
This observation goes against the grain of the swollen assets theory. The advantage of the swollen assets theory is that it allows the plaintiff to trace into a new asset even where the original asset has been dissipated. Ordinary tracing rules would either confine the plaintiff to tracing into what was acquired in exchange for incurring the debt, or would completely deny the ability to trace in this situation.
One further aspect of the Goldcorp case deserves mention. One group of plaintiffs were in a different position from the main body of customers. This special group was referred in the case as the "Walker & Hall claimants". Walker & Hall Commodities Ltd had kept a separate stock of gold for its customers. When Goldcorp Exchange Ltd acquired the business of Walker & Hall, it absorbed this separate stock of gold into its general stock. This constituted a wrongful dealing with the gold of the Walker & Hall claimants, and those claimants sought to trace their interest from the original stock of gold owned by them into other assets of Goldcorp Exchange Ltd.
An interesting issue was whether or not they were constrained by the "lowest intermediate balance rule." Where the swollen assets theory applies, the lowest intermediate balance rule should be determined by treating all the defendant's assets as one single fund. In that respect, if the swollen assets theory had been applied in Goldcorp, it would have allowed the Walker & Hall claimants to claim perhaps the entire value of their gold. However, Lord Mustill opined that it was 
not necessary or appropriate to consider the scope and ambit of the observations in the Space Investments case... or their application to trustees other than bank trustees because all members of this Board are agreed that that it would be inequitable to impose a lien in favour of the Walker & Hall claimants.
It appears from this that the Privy Council is favouring the swollen assets theory in relation to bank trustees at the very least. It is also clear that the Privy Council considered that equitable proprietary rights were in some way discretionary. This is contrary to the majority decision in Foskett where it was decided that equitable proprietary rights are not discretionary.
Furthermore, in Tinsley v Milligan, Lord Browne-Wilkinson held that property rights, whether they exist at law or in equity, are not discretionary and both depend on fixed principles.
The Privy Council made no decision about the swollen assets theory in Goldcorp. The speech of Lord Mustill in Goldcorp is favourable to the existence of the swollen assets theory. This is because the theory could have been applied in that case, and a lien would have been granted if equitable considerations had favoured this.
Another case that impacts on the swollen assets theory is Bishopsgate Investment Management Ltd v Homan, where the appellant attempted to claim an equitable charge over all the assets of the respondent to gain priority over the mass of creditors. In the English Court of Appeal, Dillon LJ purported to apply the comments of Lord Mustill in Goldcorp through holding that
Lord Templeman's observations in the Space Investments case... were not concerned... with the situation. where trust moneys have been paid into an overdrawn bank account, or an account which has become overdrawn.
This appears to be a misunderstanding of both Space Investments and Goldcorp. Lord Templeman in Space Investments was clearly in favour of the ability to trace even where the original value has been dissipated (such as when money is paid into an overdrawn bank account). Lord Mustill in Goldcorp merely suggested that it would be impossible to assert a proprietary right when no asset had been misappropriated.
The decision in Liggett was considered in Hongkong and Shanghai Banking Corporation Ltd v Fortex Group Ltd (in rec & in liq). In the High Court, Master Hansen purported to apply the decision of Cooke P in Liggett. It was held that Cooke P in Liggett based his decision on the plaintiff retaining a persisting equitable proprietary right that could give rise to a constructive trust. This may not be an accurate interpretation of Liggett because Cooke P had held that the rules of tracing could provide a link with any of the assets of the wrongdoer, rather than just the substitute. It could be that Master Hansen in the instant case would have allowed the plaintiff to have a constructive trust over any asset of the defendant. This case could therefore be analysed as approval for the swollen assets theory. However, a closer look reveals greater differences between the two cases. Liggett was a case in which the rules of tracing were considered. There was no argument in the instant case about the rules of tracing. Nevertheless, the money received by the company would not have been retained by it in specie. It would certainly have been deposited into a bank account. Tracing must have been implicitly allowed. It is not entirely clear which asset of the company was subject to the equitable interest of the plaintiff. This makes the case very unclear as to its precedent value.
This uncertainty was further propagated by the decision of the Court of Appeal in Fortex Group Ltd (in rec & in liq) v Macintosh. In this case it was held that the decision in Space investments could not assist the respondent because it was simply an application of the rule in Hallett. Again, this is a misreading of the rule in Hallett. In Hallett, the plaintiff was allowed to have a lien only over the substitute asset. In Space investments, it was suggested that tracing could provide a link with any of the assets of the wrongdoer, not just the substitute asset. Neither type of tracing would have been available in Fortex because the company did not receive any asset from the respondents. Although Tipping J appeared to approve the decision in Space investments, that case does not seem to have been considered in depth. It is therefore unlikely that this case will be considered a binding precedent in relation to the swollen assets theory.
In Re First investments Ltd (in liq), the swollen assets theory was examined. Hammond J held that, because the money sought to be traced had been paid into an overdrawn bank account, it was not possible to trace into any substitute asset. The next issue was to consider the impact of the swollen assets theory. Hammond J thought that there might be "some wider tracing principle" but that principle did not apply on the facts of this case. However, a charge over all the assets of the appellant would have been granted had there been unconscionable conduct by the appellant.
First investments Ltd therefore gives a degree of legitimacy to the swollen assets theory in New Zealand. The decision in Foskett that equitable property rights are not discretionary may go some distance toward elevating the importance of the swollen assets theory. This is because there may no longer be a requirement of unconscionable conduct before an equitable lien could be granted after the conclusion of the tracing process.
One way of looking at the result in Foskett is to say that it upholds, and even extends the reach of the swollen assets theory. This is because the plaintiffs in Foskett succeeded in obtaining equitable ownership of a proportionate share of the proceeds of the insurance policy, and equitable ownership is a superior right to an equitable lien. In Foskett, Lord Millett opined that it was unnecessary 
to consider whether there are any circumstances in which the beneficiary is confined to a lien in cases where the fund is more than sufficient to repay the contributions of all parties. It is sufficient to say that he is not so confined in a case like the present.
On the face of it, this statement does not necessarily refer to the swollen assets theory. This is because Lord Millett thought that the proceeds of the insurance policy were the indirect substitute for the purchasers' money. As seen above, it is possible that the ordinary rules of tracing could provide a remedy in Foskett. The advantage of the swollen assets theory is that it eliminates the need to show that the proceeds were the substitute for the original asset. The purchasers would need to show that the assets of the children were swollen through receipt of the value of the purchasers. Then, on the swollen assets theory, the purchasers could assert a proprietary right in any asset of the children. The obvious target would be the proceeds of the insurance policy.
The difficulty with this analysis is that the purchasers would be asserting a proportionate right to an asset that is worth far more than the amount of their value that had originally been used to pay the premiums. The only way to achieve this is to claim ownership of a proportionate part of the new asset. If the swollen assets theory could allow the purchasers to retrospectively claim that they had ownership of any asset of the children from the time at which their money was dissipated then they could claim any increase in the value of that asset. If, however, the swollen assets theory does not allow an interest to arise immediately at the time of substitution then it is less likely that the purchasers would be allowed to claim ownership of an asset that has increased in value. A generic proprietary right would be a satisfactory way to analyse the situation. A generic proprietary right in favour of the plaintiff would arise over all the defendant's assets at the time of mixing.
The swollen assets theory was not the basis of the decision in Foskett. To that extent, it is unlikely that the case extends the principles discussed above. Nevertheless, it is not yet clear that the swollen assets theory has disappeared completely. If the result in Foskett was to be criticised, it may be necessary not only to argue that tracing was impossible on ordinary principles, but also that the swollen assets theory could not have provided a way for the purchasers to trace their value.
The process of tracing, after Foskett, will be significantly changed. Although the case is not the final word on all aspects of the law of tracing, it gives valuable guidance on some important issues that harrow this area of the law.
With regard to the weight to be ascribed to issues of causation, speeches in Foskett uphold the idea that it is unnecessary to be ruled by the terms of a contract when determining substitutions for the purposes of the law of tracing. A closer look at Foskett reveals that the new concept of attribution may be no more than a sophisticated form of the concept of causation.
After Foskett, it is clear that the rule in Hallett has been exposed. No longer is it necessary to be bound by the myth regarding tracing through mixtures of value.
The swollen assets theory may assume greater importance as a result of the decision in Foskett. This would be a significant addition to the orthodox rules of tracing, and would cause a re-think of many of the concepts surrounding the division of assets in insolvency. Recent decisions in this jurisdiction have not been completely hostile to the theory and the way is now paved for an enlightened decision on this issue.
[*] Department of Commercial Law, The University of Auckland. I am extremely grateful to Professor Charles Rickett, and Alexandra Sims, for their assistance in the preparation of this paper. This paper is based on a dissertation submitted in partial fulfilment of the requirements of The University of Auckland for the degree of Master of Laws.
  UKHL 29;  1 AC 102.
 Re Hallett's Estate, Knatchbull v Hallett (1880) 13 Ch D 696.
 Foskett v McKeown  UKHL 29;  1 AC 102 at 129 (per Lord Millett).
 Ibid at 111 (per Lord Browne-Wilkinson); 116 (per Lord Hoffmann); 145 (per Lord Millett).
 P Birks, "The Necessity of a Unitary Law of Tracing" 239 at 242 in Making Commercial Law, Essays in Honour of Roy Goode ( Clarendon Press, New York, R Cranston ed, 1997).
 Foskett v McKeown  UKHL 29;  1 AC 102 at 142.
 Ibid at 143.
 L Smith , "Unjust enrichment, property, and the structure of trusts" (2000) 116 LQR 412 at 417.
 D Fox , "Vindicating property in substituted assets"  LMCLQ 1 at 3; Foskett v McKeown  UKHL 29;  1 AC 102 at 137 (per Lord Millett).
 Foskett v McKeown  UKHL 29;  1 AC 102 at 133.
 C Rotherham, "Trust property and unjust enrichment: Tracing into the proceeds of life insurance policies"  CLJ 440 at 441.
 J Stevens, "Equitable Tracing into the Proceeds of a Life-insurance Policy Partially Funded with Misappropriated Trust Money"  Conv 406 at 410.
 Foskett v McKeown  UKHL 29;  1 AC 102 at 124 (per Lord Hope).
 Ibid at 137 (per Lord Millett).
 Foskett v McKeown  UKHL 29;  1 AC 102 at 122.
 Ibid at 119 (per Lord Hope); 133 (per Lord Millett).
 Foskett v McKeown  UKHL 29;  1 AC 102 at 119.
 Ibid at 121.
 L Smith, "Unjust enrichment, property, and the structure of trusts" (2000) 116 LQR 412 at 417.
 A Burrows, "Proprietary restitution: Unmasking unjust enrichment" (2001) 117 LQR 412 at 417.
 Foskett v McKeown  UKHL 29;  1 AC 102 at 111 (per Lord Browne-Wilkinson); 116 (per Lord Hoffmann).
 Foskett v McKeown  UKHL 29;  1 AC 102 at 109.
 Ibid at 119.
 Ibid at 134.
 T Wu, "Foskett v McKeown, Hard-nosed property rights or unjust enrichment?"  MelbULawRw 8; (2001) 25 MULR 295 at 302.
 Foskett v McKeown  UKHL 29;  1 AC 102 at 132.
 D Fox, "Vindicating property in substituted assets"  LMCLQ 1 at 2.
 Op cit n 26 at 131.
 L Smith, The Law of Tracing (Clarendon Press, Oxford, 1997) 182 at n 65.
  NZLR 804.
 Dexter Motors Ltd v Mitcalfe  NZLR 804 at 826.
 P Birks, "Mixing and Tracing: Property and Restitution" (1992) 45(2) CLP 69 at 89.
 See Section III.
 P Birks, "Overview: Tracing, Claiming and Defences" 289 at 309 in Laundering and Tracing (Clarendon Press, Oxford, P Birks ed, 1995).
 P Birks, "Mixing and Tracing: Property and Restitution" (1992) 45(2) CLP 69 at 95.
 A Burrows, The Law of Restitution (Butterworths, London, 1993), p 65.
 R Grantham & C Rickett, Enrichment and Restitution in New Zealand (Hart, Oxford, 2000), p 445.
 Ibid at 447.
  2 AC 548.
 A Burrows, "Restitution of payments made under swap transactions" (1993) 143 NLJ 480 at 481.
 Lipkin Gorman v Karpnale Ltd  2 AC 548 at 573 (per Lord Goff).
 Ibid at 574.
 Ibid at 572.
 R Goff & G Jones, The Law of Restitution (Sweet & Maxwell, London, 5th ed, 1998), p 102.
 P Birks, "Overview: Tracing, Claiming and Defences" 289 at 311 in Laundering and Tracing (Clarendon Press, Oxford, P Birks ed,1995).
 Foskett v McKeown  UKHL 29;  1 AC 102 at 108.
 Ibid at 134.
  4 All ER 796.
 Boscawen v Bajwa  EWCA Civ 15;  4 All ER 769 at 777.
 P Birks, "Overview: Tracing, Claiming and Defences" 289 at 317 in Laundering and Tracing (Clarendon Press, Oxford, P Birks ed,1995).
 L Gullifer, "Recovery of misappropriated assets: orthodoxy re-established?"  LMCLQ 446.
 L Smith, The Law of Tracing (Clarendon Press, Oxford, 1997), p 313.
 Foskett v McKeown  UKHL 29;  1 AC 102 at 134.
 Ibid at 108.
 Boscawen v Bajwa  EWCA Civ 15;  4 All ER 769 at 777.
 D Oesterle, "Deficiencies of the Restitutionary Right to Trace Misappropriated Property in Equity and in UCC s9-306" (1983) 68 Cornell L Rev 172 at 189.
 K Taft, "A defense of a limited use of the swollen assets theory where money has wrongfully been mingled with other money" (1939) 39 Colum L Rev 172 at 182.
  UKPC 1;  1 WLR 1072.
 L Gullifer, "Recovery of misappropriated assets: orthodoxy re-established?"  LMCLQ 446 at 447.
 Space Investments Ltd v Canadian Imperial Bank of Commerce Trust Co. (Bahamas) Ltd  UKPC 1;  1 WLR 1072 at 1074.
  1 NZLR 257. This case was reversed on other grounds in Re Goldcorp Exchange Ltd (in rec)  1 AC 74.
 Liggett v Kensington  1 NZLR 257 at 275 (per Cooke P).
 Ibid at 280-281( per Gault J).
 Ibid at 303 ( per McKay J).
  1 AC 74.
 Re Goldcorp Exchange Ltd (in rec)  1 AC 74 at 105.
 L Smith, The Law of Tracing (Clarendon Press, Oxford, 1997), p 147.
 Ibid at 201 and 228.
 K Taft, "A defense of a limited use of the swollen assets theory where money has wrongfully been mingled with other money" (1939) 39 Colum L Rev 172 at 188.
 Re Goldcorp Exchange Ltd (in rec)  1 AC 74 at 110.
 Foskett v McKeown  UKHL 29;  1 AC 102 at 109 (per Lord Browne-Wilkinson); 134 (per Lord Millett) but cf 120 (per Lord Hope).
  UKHL 3;  1 AC 340.
 Tinsley v Milligan  UKHL 3;  1 AC 340 at 371.
  Ch 211.
 Bishopsgate Investment Management Ltd v Homan  Ch 211 at 218.
 (1995) 5 NZBLC 103,869.
 Hongkong and Shanghai Banking Corporation Ltd v Fortex Group Ltd (in rec & in liq) (1995) 5 NZBLC 103,869 at 103,816.
  3 NZLR 111.
 Fortex Group Ltd (in rec & in liq) v Macintosh  3 NZLR 111 at 118 (per Tipping J).
 (1999) 8 NZCLC 261,932.
 Re First investments Ltd (in liq) (1999) 8 NZCLC 261,932 at 261,935.
 Ibid at 261,936.
 Foskett v McKeown  UKHL 29;  1 AC 102 at 132.
 Ibid at 134.