Auckland University Law Review
The Role of Estoppel as a Defence to Claims in Unjust Enrichment
Since recognition of the ‘change of position’ defence in Lipkin Gorman (a firm) v Karpnale Ltd the role of estoppel in claims of unjust enrichment has been questioned. Three recent English Court of Appeal decisions have cast further doubt on the continuing role of the estoppel defence, as the courts creatively strain to limit its application. The decision in Dextra Bank & Trust Co. Ltd v Bank of Jamaica, that widened the defence of change of position to include anticipatory reliance, is ostensibly yet another nail in the estoppel coffin.
The principal concern of the courts is that, successfully pleaded, estoppel will defeat the plaintiff’s claim in its entirety. This may in some circumstances allow the defendant to retain a ‘windfall’, resulting in injustice to the plaintiff. In contrast, the change of position defence determines the extent to which the defendant’s position has changed in reliance on the receipt; it is only the corresponding ‘disenrichment’ that is protected. This is considered to be a more “just” result.
The apparent direction of the judiciary is reflected in Jonathan Parker J’s statement in Phillip Collins Ltd v Davis: “[t]he law has now developed to the point where the defence of estoppel by representation is no longer apt in restitutionary claims where the [more] flexible defence of change of position is in principle available”. Should the estoppel defence indeed be consigned to the legal archives?
This article explores the role and function of estoppel in the modern law of unjust enrichment, considers the relationship between the defences of estoppel and change of position, and examines the decisions in Scottish Equitable plc v Derby and National Westminster Bank plc v Somer International (UK) Ltd in detail. The article concludes that the retention and proper application of the estoppel defence is essential to the integrity of the law in this complex area.
Unjust enrichment at the expense of another may be categorised as a causative event that gives rise to a legally recognised right to restitution. Similarly, a representation in estoppel by representation is an event arising from a manifestation of consent. The representation generates a primary obligation to honour the promise. The response of the courts to a breach of this obligation has been to hold the representor to her promise by denying her legal rights to restitution.
Accordingly, where estoppel is raised as a defence to a claim in unjust enrichment, two different events, both giving rise to legally recognised rights, are implicated. Because estoppel operates as an absolute defence, the plaintiff’s right to recover his mistaken payment is “trumped by the justice of being bound by seriously intended representations subsequently relied upon”. In effect, the defendant’s interest in the security of her receipt takes priority over the plaintiff’s interest in the restitution of his wealth.
Successfully pleaded, estoppel operates as a denial of unjust enrichment (that is, the defendant’s retention of the mistaken payment can no longer be regarded as unjust) rather than a denial of enrichment, which would simply require that the measure of subsequent disenrichment be assessed. Key observes that it is only if the representation is limited to the extent of the defendant’s “detrimental reliance” that it might be regarded as affecting the enrichment condition of the plaintiff’s action.
The question remains whether, in view of the rapid development of the change of position defence, the courts will continue to allow the estoppel defence to prevail over repayment of the defendant’s surviving enrichment. Birks has observed that to do so, “a judge would have to find a reason why the justice of allowing the plaintiff to recover the still un-eliminated enrichment should be trumped by the justice of holding the plaintiff to his representations”. The answer to this complex question is explored below; first, by considering the doctrinal underpinnings of estoppel by representation; secondly, by analysing the ingredients necessary to establish an estoppel; and finally, by considering the modern role of estoppel as a defence to a claim in unjust enrichment.
Estoppel by representation provides a complete defence to a claim in unjust enrichment where (i) the plaintiff makes a representation of fact that leads the defendant to believe that he was entitled to the payment, and (ii) the defendant acts to his detriment in reliance on the representation.
The historical development of the doctrine of estoppel by representation has been well documented by a number of academics and commentators. However, it is useful to outline briefly the history of estoppel as a basis for understanding the potential role of the doctrine in modern law.
1. The Origins of Estoppel by Representation at Common Law
Although the origins of the doctrine of estoppel by representation were equitable, it was developed by the common law in the early nineteenth century following the decision of Lord Mansfield in Montefiori v Montefiori. The broad principle behind the defence was described by Lord Denman CJ in Pickard v Sears:
[W]here one by his words or conduct wilfully causes another to believe the existence of a certain state of things, and induces him to act on that belief, so as to alter his own previous position, the former is concluded from averring against the latter a different state of things as existing at the same time.
The original aim of estoppel was essentially the same as that of the developing law of contract: to hold people to their promises or undertakings. However, the two doctrines conflicted in the mid nineteenth century, primarily due to the operation of the Statute of Frauds 1677 and the tenacious requirements of the doctrine of consideration. Following Jorden v Money estoppel, both at law and in equity, was no longer concerned with the enforcement of promises because of the courts’ insistence that it operate only on statements of fact. The function of common law estoppel was finally clarified in Low v Bouverie, where it was explained that:
Estoppel is only a rule of evidence; you cannot found an action upon estoppel. Estoppel is only important as being one step in the progress towards relief on the hypothesis that the defendant is estopped from denying the truth of something that he has said.
Accordingly, the opportunity to form a doctrine of binding reliance-based promises in the developing law of obligations was stifled: a promise was enforceable only as a contract or not at all. Estoppel by representation was confined to a role as a rule of evidence, a purpose that was affirmed by the English Court of Appeal in the leading case of Avon County Council v Howlett (“Avon CC”) and more recently in Somer International. The consequence of the rule is to preclude the representor from asserting facts contrary to his representation; an estoppel does not confer substantive rights on the representee.
However, although estoppel at common law does not technically confer substantive rights, in effect it acts to ensure that the expectations of the representee are met. As a result, its remedial consequence is essentially the same as that of contract. The substantive effect of estoppel relief has been recognised by the courts. In Somer International, Potter LJ commented that:
Nonetheless, because of the decisive impact which estoppel by representation may have upon the outcome of any individual case, whether as a step on the way to establishing a cause of action, or as defeating a prima facie valid claim based on facts which (absent the representation) would entitle the claimant to recover, such estoppel undoubtedly gives rise to substantive legal consequences.
Earlier, Lord Wright similarly observed in Canada and Dominion Sugar Co Ltd v Canadian National (West Indies) Steamships Ltd that “[e]stoppel is often described as a rule of evidence, as, indeed, it may be so described. But the whole concept is more correctly viewed as a substantive rule of law”.
Conceptually, therefore, estoppel by representation is, in effect, a binding promise or undertaking that arises by consent. In essence, the court is seeking to enforce an obligation. As Birks has observed:
Estoppels have all along been binding promises. But until recently, the rare word [estoppel] has also been useful for something else, namely to identify their peculiarity in being binding only for one purpose, for the purpose of being used as a shield in litigation .... What we have been doing in the last 50 years is ... recognising that promises as to past, present and future are all binding in that defensive role ... .
Accordingly, estoppel by representation can be viewed as a “detrimental reliance contract”. If this were to assume full legal status, the remedy for breach would undoubtedly be contractual, a remedy which is “flexible, adaptable and relatively certain”. Contractual damages are measured from the expectation base unless, exceptionally, a reliance-based award is considered more appropriate.
Paradoxically the evidential rule, which operates as a form of strict liability, serves to ensure that the defendant’s expectation is indeed borne out: the courts have traditionally had no discretion to assess the measure of relief (whether contractual or otherwise) in estoppel by representation. As will be discussed below, most of the academic and judicial criticism of estoppel in the context of unjust enrichment is focused on this consequence of the defence.
2. Equitable Estoppel
‘Equitable’ estoppel, which includes both proprietary and promissory estoppel, has developed in parallel to common law estoppel despite the attempt by nineteenth century judges to confine estoppel at both law and equity to a rule of evidence. Proprietary estoppel has a long history spanning centuries, culminating in the first major decision recognizing the doctrine in its modern manifestation, Re Basham. Promissory estoppel has undergone rapid development and expansion since Lord Denning’s judgment in Central London Property Trust Ltd v High Trees House Ltd, (“High Trees”) to the extent that it has been used effectively as a cause of action.
In contrast to estoppel by representation, it is acknowledged that both forms of equitable estoppel modify or suspend the substantive rights of the parties. However, the courts’ “minimum equity to do justice” approach to equitable estoppel means that the remedy is usually reliance-based rather than expectation-based.
All forms of estoppel have a common underlying purpose: to prevent the plaintiff from refusing to recognise (or seeking unjustly to deny or avoid) an assumption or belief that she has induced or encouraged in the defendant, where the defendant has subsequently acted or regulated his affairs in reliance on that assumption. This principle has prompted attempts to integrate estoppel by representation into a unified doctrine modelled on equitable estoppel: the evidentiary rule would be replaced with a more flexible regime based on notions of fairness, justice, and conscionability. However, with the exception of the Australian High Court, the judiciary has so far resisted any rationalisation of the various forms of estoppel. One of the principal attractions of a unified doctrine in the context of unjust enrichment claims, as will be examined below, is the notional availability of equitable remedies.
Had it not been for the doctrine of consideration, estoppel by representation might have developed as a branch of contract law in which promises that are relied on to the detriment of the promisee are treated in much the same way as contracts. Unless and until estoppel by representation becomes subsumed into the law of contract, it seems correct, in principle, to treat it as an ‘almost contract’ rather than as a species of equitable estoppel – it is essentially a manifestation of consent. This is certainly the direction borne out by the authorities. However, whether this analysis causes injustice to the plaintiff in cases of mistaken payment and, if so, what the law’s response should be, remains to be considered.
The following section analyses the strict requirements of estoppel by representation: the representation, reliance, and detriment. If the defendant to a claim in unjust enrichment is unable to establish an estoppel, the alternative is to fall back on the defence of change of position.
Properly understood and applied, the estoppel defence will only be available in a very narrow range of circumstances. The elements required to establish the defence are as follows:
1. The Representation
In accordance with the common law position and authorities, the representation must be a clear and unambiguous statement of fact. In Lim v Ward McCulloch Solicitors Nominees Ltd, Tipping J stated that “[t]here is a conceptual similarity between the clarity necessary before a representation can found an estoppel and what is required for contractual certainty”.
It is apparent that the payment of money by mistake without any form of collateral representation will not be sufficient to ground an estoppel. There is nothing to justify the defendant’s reliance on any inherent representation, as the defendant is in as good a position as the plaintiff to know whether the payment is due. Goff and Jones state: “A mistaken payment is not in itself a representation which gives rise to an estoppel. There must, therefore, normally be some further indication of the defendant’s supposed title other then the mere fact of payment.” It is also plain that the plaintiff must make the representation to the defendant: representations made by third parties without the knowledge or consent of the plaintiff will not be sufficient.
The common law has recognised several exceptions to the general principle that representations must be made expressly and unequivocally. These include cases where a duty of accuracy exists, and where the representation may be implied from special circumstances surrounding the transfer.
In advancing a unified doctrine of estoppel, Cooke maintains that representations need not necessarily have contractual certainty, because of the different nature of the two jurisdictions. However, if the focus of estoppel by representation is to hold the representor to her promise, the standard of the representation must without doubt be clear and unequivocal. Grantham and Rickett have commented that to found the defence of estoppel “it is necessary to point to some factor which renders the position of the parties unequal”; for example, where the responsibility for the accuracy of the payment lies with the plaintiff. However, merely establishing some degree of inequality may not be sufficient unless the court considers that another category of exception is warranted.
Following the emergence of the change of position defence, it is almost certain that the courts will discern a representation only in the most unequivocal circumstances. As Robert Walker LJ has predicted, “the court will no longer feel constrained to find that a representation has been made, in a borderline case, in order to avoid an unjust result”. Accordingly, representations that fall short of contractual certainty are likely to be disregarded. This is a welcome clarification of the law that will restrict the estoppel defence to deserving cases. It is therefore difficult to concur with the statement in Somer International that “whether or not an actual representation as to entitlement was made or can be spelt out is largely fortuitous and ex hypothesi the result of accident rather than deliberate conduct”. For a representation to be valid there must indeed be a “deliberate” consensual act by the representor.
The defendant’s reliance on the plaintiff’s representation is a critical enquiry in establishing the estoppel defence. Reliance has been described as “believing the representation to be true”. For example, in United Bank v Jiwani the defence of estoppel failed, as it was held that the defendant had not honestly believed that the mistaken payment was due to him. In addition to being honest, the belief engendered by the representation must also be reasonable in the circumstances.
Whether a defendant honestly believed the representation to be true, and whether such belief was reasonable in the circumstances, is both a subjective and an objective inquiry. Usually the court at first instance will rule on the evidence before it, and appeal courts are often reluctant to overturn the determination at first instance. The onus is on the defendant to establish his honest belief, a burden that, in cases of mistaken payment, will often be very difficult to discharge.
Cooke has commented that the successful use of estoppel in cases of mistaken payment is rare, as usually the payee is well aware that the payment is not his. In cases where the payment represents a ‘windfall’ it may be almost impossible for the defendant to establish that he honestly believed the truth of the representation (for example, that “the money is yours” or some equivalent statement), or that this belief was reasonable. Increased focus by the courts on the honesty requirement may well address the concern described by Slade LJ in Avon CC:
I recognise that in some circumstances the doctrine of estoppel could be said to give rise to injustice if it operated so as to defeat in its entirety an action which would otherwise lie for money had and received. This might be the case for example where the sums sought to be recovered were so large as to bear no relation to any detriment which the recipient could possibly have suffered.
In contrast to the requirement in estoppel that the defendant must hold an honest belief, knowledge of the mistake is not invariably fatal to the change of position defence. All that is required is that, notwithstanding his knowledge, the defendant has acted reasonably and with a clear conscience.
Estoppel is a reliance-based liability: the defendant must establish that she relied on the representation to her detriment. There are two components to detrimental reliance. First, the defendant must have taken some action in relation to the receipt of the money, usually described as a change of position. Secondly, the defendant must have suffered a detriment by taking that action. The measure of detriment for this purpose is clearly reliance loss, not expectation loss. Accordingly, a mere change of position that does not incur detriment will be insufficient to found the defence of estoppel.
The detrimental reliance inquiry is conceptually complicated by the relationships between change of position and reliance-based detriment; and between the reliance loss necessary to establish detriment, and the failure to meet the expectation of the representee. In cases of mistaken payment, the action taken in reliance on the representation is often not a disadvantage to the representee until the representor realises his mistake. However, once demand is made for return of the money, the representee’s change of position becomes detrimental, as he must incur the inconvenience of refunding the mistaken payment. It is the potential requirement to refund the money (as distinct from the failure of the expectation) that establishes the necessary reliance loss. Dixon J explained this proposition in Grundt v Great Boulder Proprietary Gold Mines Ltd, in which detriment was described as:
[T]hat which would flow from the change of position if the assumption were deserted that led to it. So long as the assumption is adhered to, the party who altered his situation on the faith of it cannot complain. His complaint is that when afterwards the other party makes a different state of affairs the basis of an assertion of right against him, if it is allowed, his own original change of position will operate as a detriment.
Detriment may also consist of an omission to act, in reliance on the representation; for example, failing to sue on time.
Conceptually, reliance-based detriment is an abstract commodity: it is not necessary, nor indeed desirable, to quantify the loss in monetary terms. Rather, “detrimental reliance is a factor that persuades and provokes a response, that leads the judge to hold that it would be wrong not to estop the representor from doing as he wishes”. Early decisions of the courts manifestly regarded any such quantification process as irrelevant and inappropriate.
Consequently, there is no minimum level of reliance loss necessary to found an estoppel, although the presumption is that any loss that is more than de minimis will qualify. Clearly what is de minimis in any particular case is a question for debate, and the de minimis principle may well be a means to avoid extreme results. The extent of the reliance loss relative to the amount of the mistaken payment is the critical test.
In estoppel, change of position refers to the action taken in reliance on the representation that is required to establish a causal connection between the representation and the defendant’s reliance. It is not a measure of the detriment or reliance loss suffered. For example, while expenditure on a Pacific cruise in reliance on the receipt is clearly a change of position, it is not a detriment per se. Rather, detriment is established when a restitutionary claim is made, and the defendant would be required to refund the mistaken payment. However, if the receipt had instead been invested in shares that increased in value (again, a change of position), no detriment will be suffered, as there has been no reliance loss.
The tendency of both courts and commentators to confuse the defendant’s change of position with the measure of reliance-based detriment, and their further failure to appreciate the role of the detrimental reliance inquiry in estoppel, has been an unfortunate development in this area of the law, particularly with the emergence of the (quite unrelated) change of position defence to claims in unjust enrichment. While both estoppel and the change of position defences have a significant and valid role in restitutionary claims, the two are conceptually quite different, as illustrated by the following points:
(i) The change of position defence is concerned with the quantum of the defendant’s disenrichment rather than the extent of reliance-based detriment. For example, repayment of a debt is not generally regarded as a change of position in the sense of disenrichment, yet clearly the inconvenience of refinancing would constitute a reliance loss for the purposes of founding an estoppel. Similarly, the loss of an opportunity to sue, which is recognised as constituting reliance-based detriment, will not necessarily be quantified in the change of position defence. Conversely, on the ‘wide’ view of the change of position defence, disenrichment is not necessarily limited to action taken in reliance on the receipt. For example, an innocent recipient of money that is later stolen will be protected by the change of position defence.
(ii) Perhaps more importantly, while the change of position defence operates to relieve the defendant of the obligation to repay the quantum of his disenrichment, the reliance-based detriment identified in estoppel is only an element in establishing the defence of estoppel. Once the full requirements of estoppel are met, the remedy is focused on making good the defendant’s expectations rather than compensating him for the detriment suffered, irrespective of whether such detriment is measured by disenrichment or reliance losses.
On the basis of the above analysis, it is impossible to see any merit in the “novel and ingenious” argument advanced in Scottish Equitable that the defence of change of position “pre-empts and disables the defence of estoppel by negativing detriment”. In estoppel, the defendant’s change of position is simply the action taken in reliance on the representation. It is not a measure of disenrichment, nor does it equate to the reliance-based detriment required to found an estoppel. Compensating for the defendant’s disenrichment will have no effect whatsoever on the determination of detriment in estoppel; a fortiori it cannot negate that detriment. Arbitrarily adjusting for any disenrichment will, at best, purport to destroy the causal link between the representation and the defendant’s reliance – which would be like shutting the stable door after the horse has bolted. It is difficult to conceive of a parallel situation where the operation of an alternative defence might break the chain of causation required to found an unrelated and potentially more just defence.
4. Fault of the Defendant
The defence of estoppel will fail if the payment was caused by the fault of the defendant; for example, where the defendant is a wrongdoer, or misleads the plaintiff by misrepresentation.
The unconscionability factor in all forms of estoppel is concerned with “whether or not it is wrong for the maker of the representation to go back on it”. However, this is not a test in itself: it is not a licence for the courts to dispense “palm tree justice”. Rather, the elements required to found an estoppel (as discussed above) must be examined with this underlying principle in mind.
In Lim Teng Huan v Ang Swee Chuan, Lord Browne-Wilkinson noted that the unconscionability arises from the withdrawal of the statement; that is, the inconsistency – there is no need for it to have been unconscionable to make the representation itself. Birks has stated that: “[t]here is no other kind of unconscionable behaviour involved other than that which consists in failing to honour one’s promises”. Resort to a subjective evaluation of “all of the circumstances of the case” is contrary to principle and “a counsel of despair. ... The danger is in crossing unawares from a contract unconscionability to a tort unconscionability”.
Traditionally, most actual references to unconscionability are found in cases of equitable estoppel, where the court may be required to make a value judgment to achieve a just and fair result. In such cases the representation may be less clear, or may involve, for example, a common assumption. However, as Cooke points out, the court nearly always bases its decision on a thorough examination of the criteria required to found an estoppel. While concern with unconscionability may be less apparent in cases of estoppel by representation (where the promise is more obviously contractual in nature), it nonetheless remains the underlying principle on which estoppel is founded.
It is essential to distinguish the unconscionability factor in establishing an estoppel from the remedial jurisdiction the court exercises once estoppel is found to exist on the facts. The unconscionability of the representor in departing from his representation does not, in itself, confer on the court a discretion as to the remedy that it awards. It simply establishes whether the defence of estoppel is made out. The relief granted by the court is a separate inquiry that must be governed by doctrine and principle, and is fully explored in the following section.
Recent developments in estoppel as a defence to unjust enrichment have suggested that a ‘remedial’ unconscionability test is warranted to “avoid the potential injustice which would arise from the defence of estoppel operating as automatically a complete bar to the plaintiff’s restitutionary claim”. This advance followed obiter comments in Avon CC that suggested it might be unconscionable, in certain circumstances, for the defendant to retain the whole of the mistaken payment. The response advocated is that estoppel should operate pro tanto, in much the same way as the change of position defence, and accordingly should focus on the extent of the defendant’s detriment. While prima facie this approach may seem to more satisfactorily achieve a balance of justice and fairness between the parties, it has a number of critical shortcomings. These shortcomings fall into two main categories:
(i) The representation ‘event’ is almost entirely overlooked in the analysis, in particular the consensual nature of the event and the consequential expectation engendered in the representee.
(ii) There is no basis, either in doctrine or in principle, for introducing an element of judicial value judgment and discretion into an area where the court has sufficient flexibility (albeit within principled constraints) to achieve a just result.
The following discussion and case analysis will support and expand on these observations.
Although it has been viewed as a rule of evidence, the substantive effect of estoppel is to hold the representor to his promise, and, in doing so, to meet the expectations of the representee who has relied on the representation. The plaintiff is, in essence, prevented from making out his cause of action. However, it is somewhat ironic that despite the attempts of nineteenth century judges to distinguish representations from contractual promises (which require the element of consideration) by the introduction of the evidentiary rule, the substantive outcome, at least in estoppel by representation under the common law, has been the equivalent of the award of expectation losses (the contractual measure). Accordingly, Cooke has observed that “[e]stoppel is about the meeting of expectations”.
Where estoppel by representation is successfully pleaded as a defence to unjust enrichment the usual measure of relief, by operation of the evidentiary rule, is the equivalent of expectation losses – until the perhaps aberrant decision in Somer International, the courts in England resisted any move to compensate reliance loss only. The extent of the detrimental reliance suffered by the defendant is not considered relevant; rather the relief focuses on making good the expectations that the defendant has legitimately formed. In cases of mistaken payment, this has meant that the defendant is entitled to retain the entire enrichment, irrespective of the extent of his detriment.
There has been widespread criticism of the ‘all or nothing’ relief provided by estoppel in cases of unjust enrichment, particularly since recognition of the change of position defence. As Grantham and Rickett have observed:
There is a growing recognition that while the immediate effect of the estoppel may be to prevent the party making the representation from alleging contrary facts, this of itself should not entail that the representor is held to all of the consequences of that statement.
Indeed, the development of a flexible pro tanto remedial regime in estoppel is considered by some to be essential to the survival of the defence. The justification for a pro tanto approach stems from two interrelated sources: both advocate reliance-based detriment as the appropriate remedial response to an estoppel.
1. The Limited Defence approach
The limited defence approach reflects the contention that awarding a complete defence to the representee nearly always leads to a disproportionate and unfair result. In the interests of achieving justice and fairness between the parties, the courts must have the flexibility and discretion to limit the defence to the extent that is deemed necessary, having regard to all of the circumstances of the case. In general, this will mean limiting the defendant’s counter-claim to reliance-based detriment. The relevant test is one of unconscionability: is it unconscionable for the defendant to retain the enrichment?
Doctrinal support for the limited defence approach is tenuous at best. The principle relied on is an equitable one, as summarised by Potter LJ in Somer International:
[I]t is clear that the doctrine of estoppel by representation stems from and is governed by considerations of justice and equity. That being so, it is difficult to see why equity should, as between the parties, be impotent in an appropriate case or category of case to require a person relying on the defence of estoppel by representation to rely upon it only to the extent of any detriment suffered.
In support of this view, advocates point to the equitable roots of estoppel by representation, and its description by Lord Cranworth LC in Jorden v Money as “a principle well known in the law, founded upon good faith and equity, a principle equally of law and equity”.
Key, a strong advocate of the limited defence approach, argues that “as a matter of principle, there is little to support the traditional view of estoppel by representation as affording a complete defence to restitution”. Provided that the defendant’s detriment is reversed, it is not unfair to make her disgorge any remaining enrichment, as she has not acted in detrimental reliance. He further comments that it is “inappropriate to regard the plaintiff’s innocent misrepresentation as affecting the unjust nature of the enrichment”.
2. The Unified Doctrine Approach
Proponents of this approach, which originated in Australia, maintain that the difference in analysis and result evident between estoppel by representation and equitable estoppel leads to inconsistency and potential injustice. Rather, a single unified doctrine of estoppel should be developed in which the court has the remedial flexibility that is apparent in equitable estoppel. The remedial focus of the unified doctrine is on “the minimum equity to do justice”, which has been interpreted as reliance-based relief.
Support for this approach is based on the view that estoppel by representation, in terms of its substantive effects, has “expanded beyond its evidentiary function”. Accordingly, in cases of unjust enrichment estoppel would provide a defence only to the extent of the defendant’s detrimental reliance. As Mason CJ has stated:
A central element of ... [the unified] doctrine is that there must be proportionality between the remedy and the detriment which it is its purpose to avoid. It would be wholly inequitable and unjust to insist upon a disproportionate making good of the relevant assumption.
While the unified doctrine approach has a strong foothold in Australia, the English Court of Appeal has not followed this lead. The current position, affirmed in Somer International, was aptly summarised by Millet LJ in First National Bank plc v Thompson:
[An attempt] to demonstrate that all estoppels other than estoppel by record are now subsumed in the single and all-embracing estoppel by representation and that they are governed by the same requirements has never won general acceptance. Historically unsound, it has been repudiated by academic writers and is unsupported by authority.
However, the courts’ future position remains relatively uncertain: in Scottish Equitable, Robert Walker LJ raised the possibility that the House of Lords might adopt a unified doctrine of estoppel to obtain the remedial flexibility evident in equitable estoppel.
Neither of these two approaches is convincing. The limited defence approach has several fundamental flaws. First, contrary to Key’s view, there is little ‘in principle’ to support the view that estoppel should not operate as a complete defence in most circumstances. If it is accepted that the strength of the representation is equivalent to that of a contractual promise, it is correspondingly appropriate in principle that the defendant’s expectation be fulfilled in its entirety. Equity is unlikely to intervene in this situation, especially to reduce arbitrarily the award to reliance-based relief.
Secondly, the focus in estoppel by representation is on the representation itself, and therefore the plaintiff’s conduct, not on the detriment suffered by the representee. Although fault (quite rightly) does not act as a bar to a claim in unjust enrichment, it may well be a relevant factor in the operation of the estoppel defence. For example, the representation may have been negligently made, or may be in breach of a duty; it is not at all clear that a representation in mistaken payment cases will necessarily be ‘innocent’, as Key asserts. If the court considers that it is unconscionable for the representor to resile from his promise, then the defendant is entitled to have her reasonable expectations met.
Thirdly, there is no justification, either in doctrine or in principle, for limiting the defendant’s remedy to an enrichment-related defence restricted to his reliance interest. Such an approach fails to consider the conduct of the representor in creating a legitimate expectation in the mind of the representee. Contrary to Key’s understanding, estoppel does not “operate by considering whether or not it is unconscionable or inequitable to require restitution”. Rather, estoppel is concerned with whether it is unconscionable for the representor to depart from his representation, which the defendant has relied on to his detriment. Detriment has no role other than as a necessary ingredient in establishing the defence: the remedial response of estoppel is plainly concerned with meeting the expectation generated as a result of the plaintiff’s representation.
The unified doctrine approach assumes that reliance-based relief will automatically flow in cases of estoppel by representation, due to the intervention of equity and its overriding objective of doing justice between the parties. However, this is neither a logical nor a necessary corollary. Where a representation is clear and unequivocal, as is almost certainly the case in estoppel by representation, it is quite likely that equity will continue to favour the satisfaction of the defendant’s expectations. Indeed, this has been the observed practice of the courts in Australia (where the unified approach doctrine has its roots), despite the High Court’s strong obiter comments suggesting otherwise. In Giumelli v Giumelli, the High Court of Australia affirmed that the expectation measure of relief was available, relying in part on the judgment of Deane J in The Commonwealth v Verwayen:
Prima facie, the operation of an estoppel by conduct is to preclude departure from the assumed state of affairs. It is only where relief framed on the basis of that assumed state of affairs would be inequitably harsh, that some lesser form of relief should be awarded.
Accordingly, in estoppel by representation justice is done by meeting expectations, not by remedying detriment. There is no reason for this to change, even if a unified doctrine is adopted, unless the requirement for an express promise or representation is relaxed. It is only if it becomes unnecessary to identify an express promise or representation – when it is merely necessary to ascertain whether it would be unconscionable for the plaintiff to deny an expectation that the defendant has acted on – that the court, in its equitable jurisdiction, may reasonably have a discretion to decide to what extent that expectation should be fulfilled.
In favouring expectation-based relief in estoppel by representation, there are two points worth observing:
(i) Although potentially exceptional, there may well be cases where the defendant’s reasonable expectation, properly assessed, does not amount to the full extent of the mistaken payment. This is an area that has not yet been explored by the courts, as it would require redefining the role of estoppel from a ‘strict liability’ evidentiary rule to a doctrine that enforces consensual obligations. Perhaps it is time that the courts recognised the substantive remedial role that estoppel legitimately plays by removing the evidentiary ‘bar’ and moving to a more transparent analysis of the obligations and rights of the parties.
(ii) There may be scope for limiting the overall effect of the representation to the time intervening between its being made and its subsequent ‘revocation’. This potential approach is conceptually analogous to that in High Trees, where the contractual rights of the plaintiff were suspended during the period of operation of the promise, but reinstated once the promise was retracted. Similarly, the defendant in an unjust enrichment claim could be limited to the expectation interests that he could establish were related to the period between the making of the representation and its eventual withdrawal. While perhaps theoretically appealing, this approach is likely to suffer from (not insurmountable) quantification difficulties: at what point does the defendant’s legitimate expectation terminate?
In rejecting the defence of estoppel by representation, the English Court of Appeal relied on what it interpreted as ‘exceptions’, expressed by the Court in Avon CC, to the traditional view that a defendant should be granted a complete defence. Clearly the Court considered that it was inequitable for Derby to retain the greater part of the mistaken payment of ₤172,400 when his ‘change of position’ was only ₤9,600.
Analysing the obiter comments of the Court in Avon CC in context, it is unclear whether the learned judges’ observations were intended to give rise to exceptions to the general rule, or, alternatively, to Harrison J’s subsequent interpretation:
[H]aving regard to the dicta of the Court of Appeal [in Avon CC] ... I do not consider that I am bound ... to hold, in the circumstances of this case, that estoppel must operate as a complete defence.
The Court in Avon CC strongly endorsed the evidentiary role of estoppel and its operation as a complete defence based on what it regarded as “established legal principles governing the doctrine of estoppel”. Further, the Court specifically considered the proposition that the defence might operate pro tanto and concluded that this was contrary to both principle and authority. Consequently, the elevation of their obiter remarks to the status of ‘exceptions’ seems to be a somewhat dubious interpretation, particularly as this interpretation is founded on a vague and undefined notion of unconscionability.
While there might prima facie be a possibility that “on the facts it would clearly be inequitable to allow a party to make a profit by pleading estoppel”, it is surprising that the Court in Scottish Equitable did not take the opportunity to examine fully the ingredients necessary to establish estoppel before resorting to the unconvincing ‘exception’ rationale. For example, did Derby hold an honest belief that he was entitled to the funds? Was his detrimental reliance (as distinct from his disenrichment) sufficient to found an estoppel? The facts of the case lend themselves to a detailed consideration of the de minimis threshold. It is unfortunate that the Court did not recognise this opportunity.
Of further concern is that the Court did not consider the conduct of the plaintiff in relation to the making of the representation. This is an essential aspect of the estoppel inquiry. As Fung and Ho have commented, “Harrison J should have considered the fact of the claimant’s repeated and negligent representations, its disregard of the contrary information provided by Mr Derby during the telephone conversation, and its failure to make a proper enquiry”. In addition to this, the time taken to discover the mistake (some sixteen months) is surely of relevance. It seems from the judgment that once the Court rightly decided that carelessness does not preclude recovery in cases of mistaken payment, the conduct of the plaintiff in the estoppel inquiry was overlooked. However, these are two discrete inquiries: the irrelevance of fault to the event of unjust enrichment does not preclude its consideration in establishing the estoppel defence.
Finally, the Court’s merging of Derby’s disenrichment with the detrimental reliance required to found an estoppel seems to have been influential in its conclusion. However, a proper inquiry is likely to have established that Derby indeed suffered substantial detrimental reliance once the mistake was discovered. For example, the inconvenience of refinancing his mortgage debt, together with the expectation of an improved lifestyle, may well have been relevant.
On balance, it is difficult to conclude that Scottish Equitable was rightly decided. Properly analysed, it seems that the requirements of estoppel were clearly established. Although the Court did not address the unconscionability of the plaintiff, it appears from the facts that it was unconscionable for Scottish Equitable plc to resile from their unequivocal representation. Far from obtaining an inequitable windfall, there seems to be no legitimate reason why, in these circumstances, the defendant was not entitled to have his reasonable expectation met.
VIII. The Decision in Somer International
In Somer International the English Court of Appeal reaffirmed the traditional classification of estoppel by representation as a rule of evidence. However, in order to avoid a result that it regarded as unjust, an “unconscionability test” was applied, which was described as a further “general” exception to the strict rule of evidence based on the decisions in Avon CC and Scottish Equitable. The test was expressed by Peter Gibson LJ as “whether it would be unconscionable and inequitable for the recipient of the moneys mistakenly paid to retain the moneys having regard to what the recipient did in reliance on the representation made to him”. The tenuous authority advanced in support of this proposition seems to rest on a hybrid of the limited defence and unified doctrine approaches, and is based on compensating the defendant for the disenrichment suffered.
If the unconscionability exception is adopted, it will “probably deprive the traditional rule of any significant role in the future”. However, this unprincipled development in the law of estoppel surely cannot survive. The almost exclusive focus by the Court on the defendant’s disenrichment is completely contrary to the established principles of estoppel by representation. Rather than considering the circumstances giving rise to the representation and the conduct of the representor, the Court chose instead to focus on the conduct of the representee, and the extent of the detriment suffered.
In analysing the elements necessary to found an estoppel on the facts given, there is little doubt that this case was wrongly decided. The defendant company proceeded to conduct its business on the assumption that the representation was correct: it suffered substantial detriment when the mistake was discovered, albeit that not all of the detriment could be quantified in monetary terms (such as the loss of a chance to sue). In view of the conduct of the plaintiff, there is no apparent reason why the defendant company should not have been entitled, both at law and in equity, to have its reasonable expectations honoured.
This case highlights the grave dangers of adopting a remedial ‘unconscionability’ test in estoppel cases. By departing from established principles, the court has watered down the significance of the representation to the point where it is virtually extinguished. The ‘justice’ of the plaintiff being entitled to recover his mistaken payment has trumped the legitimate expectations of the defendant who relied on the plaintiff’s promise to his detriment. As a result, the consensual strength of the representation has been arbitrarily reduced to one more akin to tort, with a remedy correspondingly focused on compensation for reliance loss. The Court has not given any adequate explanation for this shift in principle: the introduction of the notion of unconscionability in this context is nothing other than a front for judicial creativity.
Estoppel as a defence to claims in unjust enrichment is at a crossroads. This article has attempted to demonstrate that the defence of estoppel by representation has an important continuing role to play in the developing law of unjust enrichment. Recent attempts to restrict its application are misguided and may lead to potential injustice. The decisions in Scottish Equitable and Somer International aptly illustrate this point.
By careful and thorough analysis of the criteria required to found an estoppel, most of the perceived injustice to the plaintiff may be alleviated by restricting the application of the defence to deserving cases. Developing and controlling the estoppel framework – rather than resorting to the vague and woolly notion of unconscionability – more appropriately accommodates the potential ‘exceptions’ recognised in Avon CC. It is unlikely that, properly analysed, the defence will be available in true cases of windfall.
Although estoppel operates as a rule of evidence that prevents the plaintiff from making out his cause of action in unjust enrichment, the substantive effect of the defence is the meeting of the defendant’s expectations. By acknowledging that the plaintiff’s promise has the equivalent of contractual force, the courts may decide to remove the evidential bar, and accordingly have recourse to the range of ‘contractual’ remedies available. Such a move would stem criticism of the ‘all or nothing’ effect of estoppel, and permit recovery by the plaintiff in circumstances where the defendant’s reasonable expectation loss did not exhaust his enrichment. The alternative would be to subsume estoppel by representation into the law of consent. However, the doctrine of consideration remains a formidable hurdle to any development in this direction.
Restricting the remedial consequences of estoppel to a pro tanto award based on some measure of reliance loss is neither sound in principle, nor will it lead to justice and fairness between the parties. This approach ignores the critical factor of the representation, its concomitant of reliance, and the protection of the defendant’s legitimate expectations. Because detrimental reliance may comprise intangible factors (such as the loss of a chance), the extent of reliance-based detriment often cannot be adequately quantified in monetary terms. Furthermore, this approach will potentially lead to confusion with the change of position defence as the courts attempt to include or exclude factors that may not strictly constitute ‘disenrichment’. Accordingly, there is a danger that the development of the change of position defence may be muddied quite unnecessarily in order to accommodate the remedial consequences of estoppel.
Although estoppel is the ‘ancestor of the change of position defence, the two defences perform very different functions. Change of position seeks to rectify the overpayment by measuring the precise extent of the defendant’s surviving enrichment. In contrast, estoppel seeks to protect the legitimate interest of the defendant arising from his detrimental reliance on the representation. Put simply, change of position involves acting on the faith of the receipt, whereas estoppel is concerned with acting on the faith of the representation. In cases of mistaken payment the principles of estoppel require that the expectation of the defendant be protected, irrespective of the plaintiff’s detriment. Where estoppel cannot be justified on the facts, change of position offers an alternative defence based on the defendant’s disenrichment. The “tension” between the two defences identified by Goff and Jones ceases to exist on proper analysis and application of the criteria necessary to found an estoppel.
Finally, should the estoppel defence trump the plaintiff’s right to recover his mistaken payment? The right to recover for unjust enrichment is rapidly moving towards strict liability. Accordingly, the defences available to the defendant have assumed increasing importance as a means of mitigating what may, in some circumstances, be a potentially unjust prima facie right. The specific criteria required to found an estoppel explain why the plaintiff, rather than the defendant, should bear the loss that results from the defendant’s reliance. It is the element of consent in the plaintiff’s actions that makes it fair to estop him from pursuing an action for restitution: because of the plaintiff’s consent, the defendant’s enrichment can no longer be considered unjust. The failure of the courts to hold plaintiffs to their representations in cases involving mistaken payments (for example, where an express promise is made forbearing to sue) may potentially bring the law of unjust enrichment into disrepute.
* LLB(Hons), DipAcc (Victoria), CA. This article has been written from a United Kingdom perspective, as the law in New Zealand in this area is not well developed. However, the few cases available on estoppel by representation in New Zealand seem to indicate a preference for the position taken by the English courts, notwithstanding the acknowledgment of a unified element of Runconscionability’ that is common to all estoppels: see Gillies v Keogh  NZCA 168;  2 NZLR 327 (CA); Rattrays Wholesale Ltd v Meredyth-Young & A’Court Ltd  2 NZLR 363 (CA); Lim v Ward McCulloch Solicitors Nominees Ltd (1999) 8 NZCLC 261, 922 (CA).
1  2 AC 548 [“Lipkin Gorman”].
 Goff and Jones, The Law of Restitution (6th ed, 2002) 843; Key, “Excising Estoppel by Representation as a Defence to Restitution”  CLJ 525.
 Scottish Equitable plc v Derby  EWCA Civ 369;  3 All ER 818 [“Scottish Equitable”]; National Westminster Bank plc v Somer International (UK) Ltd  EWCA Civ 970;  1 All ER 198,  3 WLR 64 [“Somer International”]; Phillip Collins Ltd v Davis and another  3 All ER 808 [“Phillip Collins”].
 (26 November 2001) unreported, Privy Council, Appeal No. 26 of 2000 [“Dextra”].
 Prior to this case, it was generally accepted that estoppel would continue as a potential defence in situations where the defendant’s change of position preceded the receipt of the benefit: see Fung and Ho, “Change of Position and Estoppel”  117 LQR 14, 17; Jaffey, “Change of Position and Estoppel”  pt 1 Lloyds Mar & Com LQ 1; Somer International, supra note 3, .
 See Lipkin Gorman, supra note 1, 579 per Lord Goff.
 Goff and Jones, supra note 2, 838.
 Supra note 3.
 Although estoppel is potentially a defence to all types of restitutionary claim, it is mainly concerned with mistaken payments: see Virgo, The Principles of the Law of Restitution (Oxford, 1999) 694.
 Supra note 3.
 Birks, The Foundations of Unjust Enrichment: Six Centennial Lectures (2002) 47.
 See Birks, “Equity in the Modern Law: An Exercise in Taxonomy” (1996) 26 UWALJ 61 – 64 [“Equity in the Modern Law”].
 Birks, “Change of Position: The Nature of the Defence and its Relationship to Other Restitutionary Defences” in McInnes (ed), Restitution: Developments in Unjust Enrichment (1996) 68 [“Change of Position”].
 See Grantham and Rickett, Enrichment and Restitution in New Zealand (Hart, 2000) 318-319.
 This is the situation with the quite separate change of position defence.
 Key, supra note 2, 530. Note that the measure of detrimental reliance is problematic and does not necessarily correspond to disenrichment.
 Birks, “Change of Position”, supra note 14, 68.
 Skyring v Greenwood  EngR 621; (1825) 4 B & C 281 [“Skyring”]; Holt v Markham  1 KB 504 [“Holt v Markham”]; Avon County Council v Howlett  1 All ER 1073;  1 WLR 605 [“Avon CC”].
 See Cooke, The Modern Law of Estoppel (2000); Thompson, “From Representation to Expectation: Estoppel as a Cause of Action”  CLJ 257; MacDougall, “Consideration and Estoppel: Problem and Panacea”  15 Dalhousie LJ 261.
 See Cooke, supra note 20, 19 – 22.
  EngR 363; (1762) 1 Black W. 363; 96 ER 203.
  EngR 195; (1837) 112 ER 179; 6 Ad & E 469, 474.
 See Freeman v Cooke  EngR 687; (1828) 2 Exch 654, 664 per Parke B.
 See discussion in Cooke, supra note 20, 24-25.
 (1854) 5 HLC 185; [1843-60] All ER 350 [“Jorden v Money”].
  3 Ch 82 [“Low v Bouverie”].
 Ibid 105 per Bowen LJ.
 Jorden v Money, supra note 26.
 Supra note 19.
 Supra note 3,  and .
 Fung and Ho have commented that: “[j]udicial and academic statements that estoppel is a rule of evidence are predominantly made in order to emphasise that it does not create a cause of action”: see Fung and Ho, supra note 5, 18.
 Somer International, supra note 3, .
  AC 46, 56.
 Birks, “Equity in the Modern Law”, supra note 13, 63.
 MacDougall, supra note 20, 296.
  1 WLR 1498.
  KB 130 [“High Trees”]
 Waltons Stores (Interstate) Ltd. v Maher (1988) 164 CLR 387 [“Waltons Stores”].
 Crabb v Arun District Council  EWCA Civ 7;  3 All ER 865, 880;  Ch 179, 198.
 This is a contentious area; for example, MacDougall advocates a contractual approach to all forms of estoppel: see MacDougall, supra note 20.
 See Somer International, supra note 3, -.
 See Cooke, supra note 20, 150- 169. Note that the decisions of the Australian High Court in The Commonwealth v Verwayen (1990) 170 CLR 394 [“Verwayen”] and Waltons Stores, supra note 39, are ostensibly exceptions to the weight of authorities, at least in theory if not in practice (see discussion infra).
 Spencer, Bower and Turner, Estoppel by Representation (3rd ed, 1977) 4; Low v Bouverie, supra note 27, 106.
 (1999) 8 NZCLC 261, 922,  (CA) [“Lim”].
 Lipkin Gorman, supra note 1, 579; Phillip Collins, supra note 3; National Bank of New Zealand Ltd v Waitaki International Processing  2 NZLR 211 (CA) [“Waitaki”].
 Grantham and Rickett, supra note 15, 317; Birks, An Introduction to the Law of Restitution (Clarendon, Oxford, 1985) 403.
 Goff and Jones, supra note 2, 839.
 Lipkin Gorman, supra note 1.
 Skyring, supra note 19; cf R E Jones Ltd v Waring and Gillow Ltd  AC 670.
 Holt v Markham, supra note 19.
 Cooke, supra note 20, 75.
 Grantham and Rickett, supra note 15, 318.
 Scottish Equitable, supra note 3, .
 Somer International, supra note 3, .
 Cooke, supra note 20, 90.
  1 WLR 964.
 Lim, supra note 44, .
 See for example Scottish Equitable, supra note 3.
 Cooke, supra note 20, 90.
 Avon CC, supra note 19, 1089; 624-625. Emphasis added.
 Waitaki, supra note 46, 228. See also Birks, “Change of Position”, supra note 14, 59. In Dextra the Privy Council affirmed that “good faith on the part of the recipient” is sufficient to entitle the defendant to invoke the change of position defence: see Dextra, supra note 4, -.
 This stage of the enquiry is the causation test: see Cooke, supra note 20, 93.
 See Ibid 96.
 See Ibid 99 for a full discussion of this point.
 See Skyring, supra note 19, 289-90.
  HCA 58;  59 CLR 641, 674.
 Pacol Ltd. v Trade Lines Ltd. and R/I SIF IV (The ‘Henrik Sif’)  1 Lloyd’s Rep 456; Greenwood v Martins Bank Ltd.  AC 51.
 See Cooke, supra note 20, 98.
 Skyring, supra note 19; Holt v Markham, supra note 19.
 See Scottish Equitable, supra note 3.
 See Virgo, supra note 9, 701.
 Ibid 700.
 Scottish Equitable, supra note 3, -.
 See Somer International, supra note 3. Although this was discussed on the facts, it was left open as to whether loss of a chance would be relevant to the change of position defence, which was not advanced as a potential defence.
 See Burrows, The Law of Restitution (1993) 425-428; Scottish Equitable, supra note 3,  per Robert Walker LJ.
 Scottish Equitable, supra note 3, .
 This is the traditional view and is controversial in itself: see the section on Relief in Estoppel infra.
 Supra note 3, -.
 See generally Goff and Jones, supra note 2, 831; Virgo, supra note 9, 700-701.
 Cooke, supra note 20, 84.
 Ibid 86.
 Lim, supra note 44, 261, 927 per Tipping J.
  1 WLR 113, 117.
 Birks, “Equity and the Modern Law”, supra note 13, 64.
 See Deane J’s comments in Verwayen, supra note 42, 445. These words are also found in the New Zealand test in National Westminster Finance New Zealand Ltd v National Bank of New Zealand Ltd  1 NZLR 548, 550: “(6) In all the circumstances it would be unconscionable to allow the other party to resile or depart from the assumption”.
 Birks, “Equity and the Modern Law”, supra note 13, 64 n 148.
 Cooke, supra note 20, 85.
 But see Fung and Ho, supra note 5, 19, who hint at this possibility.
 See Scottish Equitable, supra note 3; Somer International, supra note 3.
 Virgo, supra note 9, 703; Somer International, supra note 3.
 Supra note 19, 1078 per Everleigh J; 1076 per Cumming-Bruce J.
 Somer International, supra note 3, is a good example of this shortcoming: see the discussion infra.
 Reaffirmed in Somer International, supra note 3, .
 Jackson, “Estoppel as a Sword” (1965) 81 LQR 84, 87-95.
 Avon CC, supra note 19.
 Cooke, supra note 20, 151.
 Avon CC, supra note 19; Holt v Markham, supra note 19.
 Grantham and Rickett, supra note 15, 320.
 Fung and Ho, supra note 5, 19; Key, supra note 2.
 Somer International, supra note 3, .
 Ibid .
 Supra note 26, 210.
 Key, supra note 2, 531.
 Ibid 532.
 See Waltons Stores, supra note 39; Verwayen, supra note 43.
 Following the comments of Scarman LJ in Crabb v Arun District Council  EWCA Civ 7;  3 All ER 865, 880;  Ch 179, 198.
 Verwayen, supra note 43: see generally Cooke, supra note 20, 158-165 for a detailed account.
 Ibid 412 per Mason CJ.
 Ibid 413.
 1 All ER 140, 144;  Ch 231, 236.
 Supra note 3, .
 Indeed, this has been the attitude of the courts to date with few exceptions: see Cooke, supra note 20, 150-169.
 This is borne out by the “preponderance of legal authority and judicial dicta at the highest level”: see Somer International, supra note 3, .
 See Scottish Equitable, supra note 3.
 As was arguably the case in Scottish Equitable: see Fung and Ho, supra note 5, 19.
 Key, supra note 2, 534.
 See Cooke, supra note 20, 163.
 Verwayen, supra note 43, 430 per Mason J; 429 per Brennan J.
  161 ALR 473.
 Supra note 43, 443 per Deane J.
 See generally Cooke, supra note 20, 158 – 169.
 Fung and Ho acknowledge that, if estoppel were to operate pro tanto, “the defendant’s expectation interest may also call for protection”: see Fung and Ho, supra note 5, 18.
 Scottish Equitable plc v Derby  3 All ER 793, 806. Harrison J’s position was affirmed by the CA: see Scottish Equitable, supra note 3.
 Avon CC, supra note 19, 1086-1087.
 Ibid 1088.
 Ibid 1075-1076 per Cumming-Bruce LJ.
 Fung and Ho, supra note 5, 19.
 Scottish Equitable, supra note 3, 824.
 Supra note 3, .
 Ibid -.
 Ibid .
 Ibid .
 Key, “Estoppel by Representation as a Defence to Restitution: The Exception Proves the Rule?”  CLJ 465, 468.
 The potential outcome of any litigation in this sense is not relevant (even if it were open for the judge to decide this as a matter of fact – a point that is contestable): it is the loss of the chance that causes the reliance loss.
 The evidential rule is sometimes described as an “historical accident”: see Fung and Ho, supra note 5, 18.
 Goff and Jones, supra note 2, 841-844.