The Role of Illegality in the English Law of Unjust Enrichment.

by William Swadling

(2000) Oxford U Comparative L Forum 5 at | How to cite this article

I. Introduction

When English unjust enrichment lawyers talk of “illegality”, they generally do so in the context of the unravelling of partly-performed illegal contracts. Two parties enter into an illegal contract. What generally happens is that one pays for a service to be performed but fails to receive the agreed exchange. Can that party recover the value he has transferred to the non-performing recipient? The law in this area is complex and difficult to state with any accuracy.
There are a number of reasons why the judges do not find it easy to come to an agreed answer to this problem. The consequence of a finding of illegality in English law is that the contract is null and void. No action may be brought for compensation for non-performance, nor will an order for specific performance be available. With such a harsh attitude taken to contractual performance, the pressure falls entirely on the law of unjust enrichment to sort out the mess. And the difficulties are only exacerbated by the fact that English law adopts an extremely wide view as to what amounts to an illegal contract.1 Not only does it include contracts to commit crimes, as, for example, a contract to kill or to injure another person, but also contracts of which performance, though not illegal in any criminal sense, will not be enforced for various reasons of public policy. Examples of such contracts are marriage brokage contracts, contracts to commit civil wrongs, contracts to indemnify another against liability for unlawful acts, contracts in restraint of marriage, contracts promoting sexual immorality, contracts of insurance where there is no insurable risk, contracts purporting to oust the jurisdiction of the courts, trading with the enemy, and contracts restricting personal liberty. The merits of the plaintiff who pays a hit-man to murder a business rival are clearly different from one who pays a premium on an insurance policy in which he has no insurable interest.
A further difficulty in this area is caused by the fact that in many cases the role the contractual illegality is playing is ambiguous. Though in some cases it clearly provides a defence to what would be an otherwise valid claim for restitution of unjust enrichment, in others the illegality, at least at first sight, appears to provide a ground of claim in itself. In other words, the illegality gives rise to a liability to make restitution of an unjust enrichment which, absent the illegality, would not be present. Unfortunately, this distinction, between illegality as a defence to a claim for restitution of unjust enrichment, and illegality as a cause of action in unjust enrichment, is not always cleanly drawn, both in the case-law2 and in the academic literature. The cases on illegality as defence and illegality as cause of action tend simply to be run together, making an already difficult subject almost impossible to comprehend. This essay, adopting the scheme first suggested by Professor Birks,3 proceeds on the basis that the role of illegality in the context of unjust enrichment claims cannot be understood unless a separation is first made between cause of action and defence. The test which will be adopted here to decide on which side of the line any particular case falls is as follows. If we put aside the illegality, will the plaintiff still have a good cause of action? If the answer to that question is yes, then the illegality is operating as a defence. But where, conversely, the removal of the illegality would cause the claim to fail, the illegality goes to the existence of a cause of action, and is not operating as a defence.
But there is yet another difficulty of a similar nature. Little attempt is made by the majority of writers, both judicial and academic, to distinguish between claims based on the unjust enrichment of the defendant and those in which plaintiffs are seeking to enforce rights not generated by the unjust enrichment of the defendant, more precisely, property rights not generated by the unjust enrichment of the defendant.4 Claims for the enforcement of rights generated by the defendant’s unjust enrichment and claims for the enforcement of property rights not generated by unjust enrichment are clearly two different things,5 for they are rights generated by different causative events, and to which different defences may or may not apply. As we shall see, it was precisely this failure to distinguish between the two species of rights in Bigos v Bousted6 which led the court into error.
The aims of this essay are relatively modest. What it does not purport to do is to provide an answer to the difficult question, “When will illegality bar an otherwise valid claim for restitution of unjust enrichment?”.7 What it instead seeks to do is much more basic. First, it will demonstrate the need to keep distinct the operation of the defence of illegality where the claim is one to the enforcement of property rights not generated by unjust enrichment from one where the claim is to the restitution of unjust enrichment. Secondly, it will seek to establish whether in the law of restitution of unjust enrichment, illegality really does have the bivalent role ascribed to it above, viz, that it operates as both a defence and a cause of action (or “unjust factor”). Though well established as a defence to claims for restitution of unjust enrichment, the authority for saying that illegality also operates as a cause of action is weak. The so-called “repentance” cases apart, the decisions in which the illegality looks to be operating as an unjust factor can all be explained on alternative grounds. And what is more, even the repentance cases themselves provide no authority for claims in unjust enrichment, for they are in the main concerned with claims in respect of property rights not generated by unjust enrichment, and, moreover, are cases in which the illegality of the transaction is operating, albeit unsuccessfully, as a defence, and not a cause of action. The conclusion which is reached is that in English law illegality operates only as a defence to claims for restitution of unjust enrichment and never as a cause of action.

II. Illegality as a Defence

The defence of illegality is not unique to claims for restitution of unjust enrichments. We have already seen that it will bar claims in respect of the non-performance of contractual promises. It can also bar claims for compensatory damages for torts,8 and, more controversially, for the enforcement of property rights not generated by unjust enrichment. And since cases in the latter category are often run together with claims for restitution of unjust enrichment, it is to those which we must first turn.

1) Illegality as a defence to claims for the enforcement of property rights not generated by unjust enrichment

The operation of the defence of illegality to claims for the enforcement of property rights not generated by the unjust enrichment of the defendant is complex. In order to make any sense of the law in this area we need to examine separately the position at law from that obtaining in equity.

a) Common law property rights

So far as the common law is concerned, illegality does not operate as a defence to claims for the enforcement of property rights not generated by the defendant’s unjust enrichment. The reason seems to be that, unlike claims for the enforcement of contractual rights, the owner does not seek to extract his rights from any illegal act, but rather from rights which were in existence before the illegal act. In asserting such rights, therefore, the holder has no need to rely on any unlawful conduct on his part. An example is Bowmakers v Barnet Instruments.9 Machine tools were delivered pursuant to an unlawful hire-purchase agreement. Not all the instalments under the agreement were paid by the purchaser and the seller brought conversion when his demand for the return of the tools was not met. The purchaser defended the claim by pointing to the illegality of the hire-purchase agreement. The Court of Appeal held that this was no defence. Du Parcq LJ, delivering the judgment of the court, said:

“… a man’s right to possess his own chattels will as a general rule be enforced against one who, without any claim of right, is detaining them, or has converted them to his own use, even though it may appear either from the pleadings, or in the course of the trial, that the chattels in question came into the defendant’s possession by reason of an illegal contract between himself and the plaintiff, provided that the plaintiff does not seek, and is not forced, either to found his claim on the illegal contract or to plead its illegality in order to support his claim”.10

The same result was reached in Singh v Ali11 The plaintiff haulier wanted to acquire a lorry but knew that, because he did not satisfy the certain government conditions, he would not be granted a haulier’s permit by the relevant authorities. He therefore entered into an agreement with the defendant haulier, who did satisfy those conditions, under which the defendant would buy the lorry, register it in his own name, and then sell it on to the plaintiff, all the while concealing the second sale from the authorities by keeping the registration unchanged. The defendant did acquire such a lorry, registered it in his own name, and then sold and delivered it to the plaintiff. The parties later fell out, and the defendant seized the lorry from the plaintiff, who thereupon sued him in detinue. The defendant set up the illegality of the second sale as a defence. The trial judge, Smith J, held that the defence succeeded, that there was a “moral estoppel” generated by the illegal design which prevented the plaintiff from recovering. The Judicial Committee of the Privy Council disagreed. Although the transaction between the plaintiff and the defendant was illegal, property in the lorry passed through the act of delivery. The plaintiff could assert his title to the lorry against all the world, not because he had any merit of his own, but because there was no-one who could assert a better one. As Lord Denning explained:

“The court does not confiscate the property because of the illegality – it has no power to do so … The parties to the fraud are, of course, liable to be punished for the part they played in the illegal transaction, but nevertheless the property passes to the transferee.”12

That must be right. As Lord Denning makes clear, it is not for the civil court to punish the plaintiff for his illegal conduct. And in any case, the “punishment” which would be meted out might bear no relation to the seriousness of the conduct involved. As Singh v Ali demonstrates, a fairly minor criminal infringement, which might attract a fine of only a few hundred pounds from a criminal court, could well be visited with a confiscation of property worth many thousands of pounds were the court to allow illegality to operate as a defence to the enforcement of property rights not generated by unjust enrichment.

b) Equitable property rights

The position here is more complex. Until recently, the opposite position to that which obtained at common law prevailed. Applying the maxim “he who comes to equity must do so with clean hands”, the courts refused relief to a plaintiff who was party to an illegal design where what was claimed was an equitable, as opposed to a legal, property right. An example is provided by Chettiar v Chettiar,13 a decision of the Privy Council on appeal from the Court of Appeal of Malaya of almost the same vintage as Singh v Ali, with Lord Denning once again delivering the opinion of the Board. There, a father owned 139 acres of land. In order to evade administrative regulations as to its use, he transferred 40 acres into his son’s name, on the express understanding that the son was to hold it for him on trust. The father later contracted to sell the land and asked his son for a power of attorney to do so. The son refused and the father sought a declaration that the son held the land for him on trust, founding his claim not on the express trust14 but on the resulting trust which is presumed in the case of gratuitous transfers of property rights.15 The trial judge, who had also decided Singh v Ali at first instance but who, as we have seen, had been reversed on appeal, held that the plaintiff’s possible turpitude was “no reason for denying him the orders which he seeks”. The Judicial Committee of the Privy Council, however, disagreed:

“In Singh v Ali the plaintiff founded his claim on his right of property in the lorry and his possession of it. He did not have to found his cause of action on an immoral or illegal act. He was held entitled to recover. In the present case the father has of necessity to put forward, and indeed, assert, his own fraudulent purpose … He is met therefore by the principle stated long ago by Lord Mansfield “No court will lend its aid to a man who founds his cause of action upon an immoral or an illegal act,” see Holman v Johnson (1775) 1 Cowp 341, 343. Their Lordships are of opinion that the courts should not lend their aid to the father to obtain a re-transfer from the son.”16

Being entirely at odds with the attitude taken by the common law, and with its tendency to impose a disproportionate punishment on the plaintiff, it is not surprising that the equitable position is subject to a number of exceptions. The first is that relief will be given where that purpose for which the illegal trust was created fails to take effect. So, in Symes v Hughes17 the plaintiff, who was in financial difficulties, conveyed leasehold property to a widow with whom he was “on intimate terms” to hold for him on trust, the intention being to keep the leasehold property out of the hands of his creditors should he become bankrupt. Not surprisingly, the conveyance made no mention of the trust, instead falsely declaring that the widow had given valuable consideration for it. Although he did become bankrupt three years later, the widow had in the meantime conveyed the lease to her son-in-law. After her death, the plaintiff sought a reconveyance from the son-in-law, alleging that the latter took the lease with notice of the former’s equitable interest. Before the cause came on for hearing, the plaintiff was adjudicated bankrupt, and at the first meeting with his creditors, he persuaded them to revest his property in him so that he could, at his own expense, prosecute the claim against the son-in-law. Lord Romilly MR held that the plaintiff’s claim was not defeated by his illegality. He said that where the purpose for which the assignment was given was not carried into execution, and nothing was done under it, the mere intention to effect an illegal object when the assignment was executed did not deprive the assignor of his right to recover the property from the assignee.18
A further inroad into the confiscatory attitude of equity was made very recently by a majority of the House of Lords in Tinsley v Milligan,19 where the clean hands rule was held to be confined to the situation in which the plaintiff had to somehow plead his own illegality as part of the evidential basis of his case. If he could plead his case without mentioning the illegality, then even though that illegality might later be disclosed during the trial of the action, the plaintiff would be given equitable relief.
In Tinsley v Milligan the plaintiff and the defendant bought a house with joint funds. In order to enable both parties to make fraudulent claims to welfare benefits, the property was conveyed into the name of the plaintiff alone, although the agreement between the parties was that it would be held by her on trust for the two of them. Fraudulent claims were in fact made by both parties, but having now made her peace with the authorities, and having fallen out with the plaintiff, the defendant sought a declaration that the plaintiff held the property on trust for them both. As in Chettiar v Chettiar, she relied not upon the express trust, but on the presumed trust which arose through the device of a purchase money resulting trust.20
A majority of the House of Lords allowed the plaintiff’s claim. Lord Browne-Wilkinson said that the defendant won her case simply because it was not necessary on the pleadings for her to rely in any way on the illegality. He said that the common law principle enunciated in Bowmakers Ltd v Barnet Industries Ltd21 should apply with equal force to claims in equity:

“… English law has one single law of property made up of legal and equitable interests. Although for historical reasons legal estates and equitable estates have differing incidents, the person owning either type of estate has a right of property, a right in rem not merely a right in personam. If the law is that a party is entitled to enforce a property right acquired under an illegal transaction, in my judgment the same rule ought to apply to any property right so acquired, whether such right is legal or equitable.”22

His lordship said that the mere fact of her contribution to the purchase price of the house was enough to give the plaintiff an interest under a resulting trust and her motive for leaving her name off the title could not affect that result.
But the differences between common law and equity have not been completely eradicated, as the subsequent case of Tribe v Tribe23 demonstrates. There, recovery was allowed, but only under the principle enunciated in Symes v Hughes.24 Had the illegal purpose there been achieved, the plaintiff’s interest would have been forfeit, something which, as Singh v Ali25 demonstrates, could not have happened at common law. In Tribe v Tribe a father owned 459 out of 500 shares in a family-run clothing company. The father also held a lease of two shops, which the company occupied as licensee. The shops were in a state of disrepair and, worried about his liability under the repairing covenants and that he might, in order to meet those liabilities, be forced to dispose of his shares in the family company, the father transferred the shares to his son to hold for him on trust, the purpose of the trust being to keep the shares out of the hands of the father’s creditors. But the crisis passed, the matter being settled with the landlords by a surrender of the lease in the one case and a purchase of the reversion in the other. At no stage did the fact that the shares were in the hands of the son affect the outcome of the settlements. The father later sought the return of the shares from his son, again, not on the basis of the express trust,26 but by arguing that the gratuitous nature of the conveyance gave rise to a resulting trust in his favour. Not surprisingly, the son in turn argued that a presumption of advancement (gift) operated in his favour,27 which presumption could only be rebutted by the father leading evidence of his illegal purpose, which Tinsley v Milligan28 prevented him from doing.
The Court of Appeal found for the father. Tinsley v Milligan was a liberalising decision, and was certainly not to be seen as abolishing the Symes v Hughes exception, with the result that because the father had not in fact defrauded his creditors he would be allowed to lead evidence of the illegal purpose in order to rebut successfully the presumption of advancement in favour of his son. Any other result, said the court, would have been nonsensical. As Nourse LJ remarked, “If Miss Milligan was able to recover against Miss Tinsley even though she had succeeded in defrauding the Department of Social Security over a period of years, it would indeed be a cause for concern if a plaintiff who had not defrauded his creditors in any way was prevented from recovering simply because the defendant was his son.”29
It will be apparent that there are a number of difficulties with the operation of the illegality defence where a claim is made to the enforcement of equitable property rights not generated by the unjust enrichment of the defendant. If, as Lord Denning pointed out in Singh v Ali,30 a common law court has no power to confiscate such property rights because of illegality, the question which then arises is why the position in equity should be any different? For as Chettiar v Chettiar illustrates, the potential for confiscation still exists, the reasoning in that case being left untouched by Tinsley v Milligan. Second, the operation of the illegality defence in equity is arbitrary. Recovery depends upon who is forced to play the illegality card. As Millett LJ pointed out in Tribe v Tribe, the rule that the plaintiff wins so long as he does not have to rely on his own illegality:

“… is procedural in nature and depends on the adventitious location of the burden of proof in any given case. Had Mr Tribe transferred the shares to a stranger or distant relative whom he trusted, albeit for the same dishonest purpose, it cannot be doubted that he would have succeeded in his claim.”31

As it was, Mr Tribe only succeeded because he was able to bring himself within the Symes v Hughes exception. Had his illegal design succeeded, his interest would have been forfeited to his son.

2) Illegality as a defence to claims for restitution of unjust enrichment

The illegality of a contract will sometimes operate not only to prevent enforcement of that contract, but also to disqualify the plaintiff’s right to restitution of benefits transferred pursuant to it in what would otherwise be a valid claim based on more familiar unjust factors such as mistake or failure of consideration. The operation of the defence is governed by two overlapping maxims, ex turpi causa non oritur actio (no disgraceful matter can ground an action) and in pari delicto potior est conditio defendentis (where the guilt is shared the position of the defendant is the stronger). As we have noted, the exact limits of the defence are the subject of much conjecture and we will not be entering into that debate. All we need do is note that illegality undoubtably does sometimes have the effect barring what would otherwise be valid claims for the restitution of unjust enrichments.
An example of a claim in respect of a mistaken payment barred by illegality is provided by Parkinson v College of Ambulance,32 where the secretary of a charity fraudulently misrepresented to the plaintiff that either he or the charity was in a position to undertake that the plaintiff could be got a knighthood in return for a large donation to the charity. The plaintiff made the donation but did not receive the promised knighthood. He sued to recover the money which he had paid on the ground that the payment had been induced by fraud (a species of mistake). Lush J refused the claim. Although it was true that the plaintiff had been defrauded, he knew that the contract into which he was entering was illegal and one which he ought not to have made. He could not therefore claim that he was not in pari delicto with the defendant. He had only himself to blame for his loss. Applying the test set out above,33 the illegality is here clearly operating as a defence since, if we put the illegality aside, the plaintiff would undoubtably have had a good cause of action based on his mistaken transfer.
The same is true of a case involving a failure of consideration, Berg v Sadler & Moore.34 The plaintiff tobacconist had been put on a stop-list by the Tobacco Trade Association. He tried to get supplies by putting up another person to buy them with his money, an action which amounted to a criminal offence. After the money had been paid over but before the goods had been delivered, the defendant wholesalers realised what was happening. They declined either to deliver the goods or to return the money. The Court of Appeal refused the tobacconist restitution of the moneys he had paid. Since, absent the illegality, the plaintiff would have undoubtably had a claim for restitution of unjust enrichment based on a total failure of consideration, the illegal nature of the design is undoubtably operating as a defence here too.
But the mere fact that the contract is illegal will not always operate as a bar. An exception is implicit in the wording of one of the maxims used in this area. If the rule is in pari delicto potior est conditio defendentis, then, in cases where the plaintiff does not share the defendant’s guilt, recovery ought to be allowed. So, for example, in Hughes v Liverpool Victoria Legal Friendly Society,35 the plaintiff was induced by a fraudulent misrepresentation to pay premiums on policies on lives in which she had no insurable interest. The contract of insurance, since it did not relate to an insurable interest, was an illegal one.36 The plaintiff sought recovery of the premiums which she had paid but was met with the defence of in pari delicto. She succeeded nonetheless. The Court of Appeal held that as the victim of a fraud, she was innocent and so entitled to say that she was not in pari delicto with the defendant who by a false representation had induced her to believe that the transaction was an innocent one and one which was enforceable in law. Once again, therefore, illegality is operating as a defence, though this time it is found to have no application to the facts of the case in hand.

III. Illegality as Cause of Action

Nowhere in any other branch of English private law does illegality per se provide a plaintiff with a cause of action that he would not otherwise have. Although illegality operates in contract and tort to bar what would otherwise be valid claims, it never creates a cause of action in itself. And nor does illegality ever give rise to property rights, at least not property rights not created as a response to unjust enrichments. But, it is said, this is not the case in claims for the restitution of unjust enrichment. Goff and Jones, for example, while acknowledging that it in general it operates as defence, assert that “there are situations where a plaintiff is able to rely on illegality itself as the ground to support his restitutionary claim”.37 This is also the view of Burrows, who devotes separate chapters to “Illegality as a ground of restitution” and “Defences”, one of which is illegality.38 The question which will now be addressed, and the point which is the main focus of this paper, is whether this is right. Does illegality, or more accurately, “transfer of benefits pursuant to an illegal contract”, ever constitute a ground of restitutionary claim?
We should immediately notice that the ground as formulated does not allege any vitiation or qualification of the plaintiff’s consent to the transfer. If it comes anywhere, therefore, it has to come under some head of policy-motivated restitution. But before addressing the question whether it does fall under that heading, we must clear away two potential distractions. The first is that there are some cases which are claimed as examples of restitution for illegality but which can be explained on more orthodox grounds, as cases in which the unjust factor is the vitiation of the plaintiff’s consent to the transfer. The second is that an illegal contract is a void contract, and there is an argument which says that void contract is a ground of claim.

1) Illegality within vitiated consent

There are some cases which are claimed as examples of illegality operating as a cause of action but which, on closer examination, are better explained as examples of restitution for vitiated consent. Some are clearer than others.
A clear case is Smith v Bromley,39 where the ground of claim was duress. The plaintiff’s brother was bankrupt. The defendant, his chief creditor, had taken out a commission against him, but afterwards, finding no dividend likely to be paid, refused to sign his certificate unless he was paid £40 and given a credit note for a further £20. The plaintiff paid the money and the defendant signed the certificate. When the plaintiff sought to recover the money, the defendant argued that the consideration for the payment was illegal and that the plaintiff was party to that illegality. Lord Mansfield nevertheless allowed recovery. The case was, he said, analogous to Astley v Reynolds40 (a case of duress of goods), and, the plaintiff not being in pari delicto, the illegality presented no bar to her claim.41 The fact pattern of this case is, therefore, exactly the same as that in Hughes v Liverpool Victoria Legal Friendly Society.42 Illegality is being raised as a defence, though a defence which fails on the particular facts of the case because the parties are not in pari delicto. In no way is the illegality of the transaction in this case operating as a cause of action.
Next, there is a group of cases in which the ground of claim can also be explained as non-voluntary transfer, though this is admittedly more controversial. We will call these cases the “protected class” cases.43 In Kiriri Cotton Ltd v Dewani,44 a landlord charged a premium for the grant to a tenant of a seven year lease. By virtue of section 3(2) of the Uganda Rent Restriction Ordinance, a landlord committed an offence in demanding or receiving such a premium. But the ordinance was poorly drafted, and both parties honestly and reasonably believed that the provision did not apply to their transaction. The tenant went into possession, but, upon discovering that the restriction did in fact apply to his lease, sought the return of his premium. In his defence, the landlord argued that the tenant’s claim was barred because he was guilty of aiding and abetting a criminal offence. He also denied the existence of any ground of claim, the mistake on the part of the tenant being a mistake of law.45 The Privy Council nevertheless allowed recovery. Lord Denning, who gave the opinion of the Board, said that the issue was whether the plaintiff was in pari delicto with the defendant. He said that he was not, because “The duty of observing the law is firmly placed by the ordinance on the shoulders of the landlord for the protection of the tenant; and if the law is broken, the landlord must take the primary responsibility.” From that it followed that the plaintiff won. “Seeing, then, that the parties are not in pari delicto, the tenant is entitled to recover the premium by the common law.”
It is not immediately obvious why a finding that the parties were not in pari delicto should lead to the conclusion that the tenant could recover his premium. As we saw with the case of Smith v Bromley,46 a finding that a plaintiff is not in pari delicto gives us a reason why illegality should not bar an otherwise valid claim; but it gives us no positive reason why a claim which is bad from the start should be allowed. Indeed, Kiriri Cotton Ltd v Dewani is a prime example of the phenomenon referred to above,47 of a court failing to distinguish between illegality as a defence and illegality as a cause of action. But if we look closely, there is in fact a ground of restitution contained in the decision which is independent of the illegality. A better explanation of the case is that, like Smith v Bromley, this too was a case on a non-voluntary transfer. Housing was in short supply in Uganda, with the result that landlords had the upper hand, and tenants were vulnerable to exploitation. And while we normally leave the market to mediate in such situations, statute had recognised the transactional inequality by forbidding this particular type of bargain. The law having decreed that the tenant was not capable of bargaining as an equal, restitution followed because of his lack of free will to the bargain. An alternative explanation, again within non-voluntary transfer, is mistake. Though the mistake was one of law, the fact that the statute was passed for the protection of the tenant against the landlord meant that the usual restrictions on recovery for mistake of law were removed.
But, since there was no evidence of a mistake, mistake of law cannot explain the case of Hermann v Charlesworth.48 Nor can failure of consideration, even if we remove the requirement that the failure be “total”.49 In this case a woman who had entered into a marriage brokage contract with the defendant was able to recover the fee she had paid, despite the fact that the defendant had partly performed the agreement. Indeed, the Court of Appeal said that even if there had been full performance, the plaintiff would still have been allowed to recover.50 This can again be explained as a case of transactional inequality, the rule that such contracts are illegal being for the protection of single women as a vulnerable class.

2) Void contracts

An illegal contract is a void contract.51 There is an argument which says that “void contract” is in itself a ground of restitution. That proposition is demonstrably wrong, but this is not the place to chase that hare. The only point to make is that illegal contracts and void contracts are not co-terminus. A contract may be void for many reasons other than illegality, for example, for incapacity52 or for want of formality.53 So, if restitution follows because of a contract’s invalidity, it must be the invalidity and not the illegality which is the trigger.

3) Illegality per se

Having now cleared away those potential distractions, we now reach the heart of our enquiry. The first point to note is that it is clear that “payment under an illegal contract” is not per se a ground of restitution. This can be illustrated by reference to two cases: Green v Portsmouth Stadium Ltd and Shaw v Shaw.
In Green v Portsmouth Stadium Ltd54 the plaintiff bookmaker had paid £2 for entry to a greyhound track. This was almost four times the amount the owner of the track was allowed by statute to charge, and the overcharging in fact amounted to a criminal offence on his part. The Court of Appeal held that there was no ground on which the plaintiff could recover his overpayment. Denning LJ said:

“… there is no allegation that the plaintiff was under any mistake of fact, nor is there any allegation that he was under a mistake of law; nor that he was oppressed or imposed upon in any way. We must assume on this pleading that the bookmaker knew perfectly well that the only lawful charge was 11s. 3d.; nevertheless he voluntarily chose to pay £2 to the stadium, and he now seeks to recover it back. He does not claim, and cannot claim, for money paid on a consideration that has wholly failed, for he has had the consideration. He has gone on the track and conducted his bookmaking operation there. The only ground on which he claims the money is that there was a breach of the statute in charging him too much.”55

No orthodox ground of unjust enrichment having been found, the next question was whether the statute made bookmakers a protected class. The Court of Appeal held that it did not. The statute was passed for the regulation of race-courses via the criminal law; it did not create a “bookmakers’ charter”. That being the case, the court held that the breach of the statute did not, standing alone, give rise to a claim for repayment.
In Shaw v Shaw,56 the plaintiff sought to recover £4,000 paid for a flat in Majorca. The contract of sale was illegal, as the Treasury consent required under the Exchange Control Act 1947 had not been obtained. The defendant’s application to have the claim struck out as disclosing no cause of action was granted by the Court of Appeal. Lord Denning MR said:

“It has long been settled that no person can found a cause of action upon his own illegal act. … If the plaintiff is to overcome this bar, he must [put] forward some reason why he should not be defeated by his own illegality. To take a simple illustration: supposing the flat in Majorca had not been conveyed to him and that it had not been handed over to him in return for the £4,000, then I can well see that he could make out a claim. He could say that the money had been paid over on a consideration which had wholly failed, but he does not attempt to do that. On this pleading, it may well be that he has got the flat and yet still wants his money back. He bases himself on nothing but the illegal payment. To my mind, it is clearly bad.”57

The starting point, therefore, is that “payment under an illegal contract” is not in and of itself a ground of restitution. Indeed, given that it entails no defect or qualification of consent, nor any inherent policy reason why restitution should issue, any other conclusion would be absurd.
But though “payment under an illegal contract” is not per se a ground of restitution, there are said to be two classes of case in which, were the illegality to be removed, the claim would fail. Thus, the argument goes, the illegality is operating as an ingredient of a cause of action. The two classes of case are the “protected class” cases and the “repentance” cases.
We have already dealt with the “protected class” cases, and have seen that they properly belong under the heading of vitiated consent, for in them there is a deemed transactional inequality. The “repentance” cases are more difficult to explain. In these there is said to be a policy-motivated head of claim in unjust enrichment concerned with encouraging the renunciation of unlawful contracts. But why should the availability of an unjust enrichment claim be considered as encouraging the renunciation of illegal designs? The reason given is that if money, goods or services have changed hands pursuant to a still executory illegal contract and restitution were not available, then the transferor will have no incentive to repent. If he cannot get restitution he will be more likely to go ahead with the unlawful design, for he will have nothing to lose. Whereas if he knows that if he repents he can recover his outlay, he will be more likely to do so. Though plausible, this explanation suffers from the obvious difficulty that it somewhat unrealistically assumes a fairly sophisticated knowledge of the law of restitution on the part of participants in illegal designs.58
But it is not the rationale of the principle which is in question, but its existence. That there is such a principle is said to stem from a dictum of Mellish LJ in Taylor v Bowers, where he said that:

“If money is paid, or goods delivered for an illegal purpose, the person who has so paid the money or delivered the goods may recover them back before the illegal purpose is carried out.”59

This “repentance principle” must, however, be read in context.
In Taylor v Bowers itself, the plaintiff, fearing his own insolvency, handed over the possession of his goods to a friend to keep them out of the hands of his creditors. The delivery was dressed up as a sale, with fictitious bills of exchange given by the friend in return. But there was no intention to transfer property in the goods to the friend, with the result that became only a bailee. Two meetings of the plaintiff’s creditors were held, but no compromise was effected. The friend later sold the goods to the defendant, who knew of the fraudulent arrangement. The plaintiff brought detinue against the defendant, and, in his defence, the defendant pleaded the illegal nature of the original bailment. The Court of Appeal held that the defence of illegality failed. The fraudulent purpose not having been carried out, the plaintiff was not relying on the illegal transaction, but was entitled to repudiate it, and recover his goods from the friend, and the defendant had no better title than the friend, as he knew how the latter had become possessed of the goods.
The point to note about this case is that all three judges in the Court of Appeal treated the illegality of the bailment as a possible defence, but held it inapplicable on the facts, the reason for the inapplicability being that the illegal purpose had not been carried out. In other words, the case is in the same mould as Smith v Bromley and Hughes v Liverpool Victoria Legal Friendly Society. In no sense was the illegality itself treated as the cause of action. The cause of action was instead one based on the plaintiff’s continuing property rights in the goods. Ironically, this comes out most clearly in a passage in Mellish LJ’s judgment, where he says:

“I think the only question open upon this rule is, assuming that the plaintiff had never really intended to part with his goods to Alcock or to Bowers, whether he was precluded from recovering the goods from Bowers on the ground that he could not do so without proving the illegal transaction to which he was a party.”60

There is, however, a dictum similar to that of Mellish LJ in the earlier case of Hastelow v Jackson.61 There, Littledale J said:

“If two parties enter into an illegal contract, and money is paid upon it by one to the other, that may be recovered back before the execution of the contract, but not afterwards.”62

But again, this statement has to be read in context. A and B had deposited equal sums with a stakeholder to abide the event of a boxing match between them. The match was played and B was adjudged the winner. A did not accept the result and told the stakeholder not to pay over his stake to B. The stakeholder nevertheless paid it to B, and A sued the stakeholder for money had and received. He won. But the court made it clear that the question was simply one of want of authority. Before paying the money over to B, A’s consent to the payment had been withdrawn. As Bayley J said:

“… if a stakeholder pays over money without authority from the party, and in opposition to his desire, he does so at his own peril.”63

The case is not, therefore, one in which the plaintiff’s repentance of an illegal design is founding a cause of action in unjust enrichment.
But the biggest problem for those who argue that Taylor v Bowers is authority for the proposition that repentance from an illegal design is a cause of action in unjust enrichment is that it is not actually a case involving restitution of unjust enrichment. The court was clear that property in the goods did not pass to the friend. The claim was not therefore made on the basis of rights arising from unjust enrichment but from rights arising in some other way, presumably, in this case, the consensual purchase of the goods by the plaintiff at some time prior to the bailment. Thus, even if the principle which it lays down is correct, Taylor v Bowers provides no authority for repentance as a ground of claim in unjust enrichment.
But a case which did involve restitution for unjust enrichment, and a case in which the claim actually failed, was Kearley v Thomson.64 The defendants were solicitors to the petitioning creditor in certain bankruptcy proceedings, and had incurred costs which were to be paid out of the estate. The plaintiff, a friend of the bankrupt, offered to pay the defendants a sum of money for these costs, which had not been paid owing to want of assets, on their undertaking not to appear at the public examination of the bankrupt, and not to oppose his order of discharge. The defendants, with the consent of their client, agreed to this, and received the money. They did not appear at the public examination of the bankrupt, but, before any application for his discharge had been made, the plaintiff brought an action to recover back the money from them.
The Court of Appeal65 held that the money was not recoverable. Fry LJ, who gave the only reasoned judgment, doubted the validity of the dictum of Mellish LJ in Taylor v Bowers, but anyway distinguished it on ground that in the instant case the illegal purpose had been partly carried out. But Fry LJ thought the better rule was that there should be no recovery, whether the contract was executed or executory. In other words, and keeping an eye on the distinction between illegality as a cause of action and illegality as a defence, he said that even in the case of a total failure of consideration, where the illegal purpose had in no part been carried out, the illegality of the contract should operate as a defence to an otherwise valid claim.
So far, then, the repentance principle, if seen as a principle dictating restitution of unjust enrichment, would seem to have little, if any, support. It forms no part of the decision in Taylor v Bowers, which was in any case not a claim for restitution of unjust enrichment, and its correctness was doubted by Fry LJ in Kearley v Thomson. Unfortunately, however, it was applied in Bigos v Bousted.66 Bousted had deposited a share certificate as security for a loan of money to be made in contravention of exchange controls. The loan was never made and Bousted sought the return of his certificate. He argued that though the contract was an illegal one, as it was still executory, he was allowed a locus poenitentiae, and was, therefore, entitled to claim the return of the certificate. He failed. Pritchard J held that Bousted could not succeed because he could not bring himself within the repentance principle. The parties were in pari delicto at the time of the making the agreement and Bousted was not entitled to seek the aid of the court to recover the certificate. Bousted had not withdrawn from the agreement because of repentance but rather because the illegal contract had been frustrated by the lender’s refusal to perform.
Despite the doubts cast upon the validity of Mellish LJ’s dictum, it was said by Pritchard J to be proved by the decision of the Court of Appeal in Hermann v Charlesworth.67 But, as we have seen,68 Hermann v Charlesworth was not a case explicable on any principle of repentance, for the court would have allowed recovery even had Miss Hermann contracted a valid marriage, and that the better explanation of the case is that it was a claim in respect of money transferred pursuant to a vitiated consent. Bigos v Bousted is a clearly different type of case. Being a claim to a common law property right not generated by unjust enrichment, the correct principle to be applied was that contained in Singh v Ali, which would have led to recovery by the plaintiff, however unrepentant he may have been.

IV. Conclusion

While there is no doubt that in English law illegality can operate as a defence to a claim for restitution of unjust enrichment, the argument which has been made is that there is no support in the case-law for the proposition that illegality can found a claim in unjust enrichment. Most of the cases normally cited in favour of illegality as a cause of action can be explained as cases of transfers under an impaired consent, including the “protected class” cases. That only leaves the repentance cases, which turn out on closer examination not to be concerned with claims for restitution of unjust enrichment at all but with actions for the enforcement of the plaintiff’s property rights generated by events other than the defendant’s unjust enrichment. As such, they can tell us nothing about the role of illegality within the law of unjust enrichment. And when those cases are extracted from the law of unjust enrichment there remains no authority for saying that illegality ever operates as a cause of action in the English law of unjust enrichment.


1 For a general account of illegality in the context of contractual undertakings, see Sir Guenter Treitel, The Law of Contract (10th ed, London, 1999), pp 392-452.

2 The worst offender in this regard is probably the decision of the Privy Council in Kiriri Cotton Ltd v Dewani [1960] AC 192.

3 Peter Birks, An Introduction to the Law of Restitution (Oxford, rev ed, 1989), pp 299-303, 424-432.

4 This is not to deny the possibility, at least in English law, of property rights being generated by the defendant’s unjust enrichment, as witness the decision of Goulding J in Chase Manhattan Bank SA v Israel-British Bank (London) Ltd [1981] Ch 105 (recipient of a mistaken payment said to hold it on constructive trust for the payee). On the question whether it is correct for the courts to create property rights as a response to unjust enrichment, see W J Swadling, “Property and Unjust Enrichment” in J W Harris (ed),Property Problems: From Genes to Pension Funds (London, 1997).

5 Though this is something which the word “restitution” tends to disguise: see P Birks, “Misnomer” in Cornish et al (edd), Restitution: Past, Present and Future (Oxford, 1998).

6 [1951] 1 All ER 92.

7 Those seeking an answer to this question should consult Peter Birks, “Recovering Value Transferred Under an Illegal Contract” [2000] Theoretical Inquiries in Law 155.

8 In National Coal Board v England [1954] AC 403, 429, Lord Asquith said:

“If two burglars, A and B, agree to open a safe by means of explosives, and A so negligently handles the explosive charge as to injure B, B might find some difficulty in maintaining an action for negligence against A.”

For a general account of the defence of illegality in tort, see W V H Rogers, Winfield & Jolowicz on Tort (15th ed, London, 1998), pp 866-871.

9 [1945] KB 65.

10 Ibid, at p 71.

11 [1960] AC 167.

12 Ibid, at p 177.

13 [1962] AC 294.

14 It is not clear why a claim was not made on the basis of the express trust. One probable reason is that there was no written evidence of the declaration, a statutory requirement in English law by virtue of Law of Property Act 1925, s 53(1)(b), and its predecessor, Statute of Frauds 1677, s 7. If the 1677 statute had not been in force in Malaya at the time the case was decided, it would have only been because it was replaced by a more modern piece of legislation to the same effect.

15 “If a man seised of land make a feoffment thereof and it appeareth not to what use the feoffment was made, nor it is not upon any bargain or other recompence, then it shall be taken to the use of the feoffor, except the contrary can be proved”: Christopher St Germain,Dialogues between a Doctor of Divinity and a Student of the Common Law (1532), ii, c 21.

16 [1962] AC 294, 303.

17 (1870) LR 9 Eq 475.

18 (1870) LR 9 Eq 475, 479.

19 [1994] 1 AC 340. The minority, Lords Goff of Chieveley and Keith of Kinkel, adhered to the old position that once the court was apprised of the illegality it should automatically refuse relief.

20 “… the trust of a legal estate, whether freehold, copyhold, or leasehold; whether taken in the names of the purchasers and other jointly, or in the names of others without that of the purchaser; whether in the one name or several; whether jointly or successive – results to the man who advances the purchase-money”: per Eyre CB inDyer v Dyer (1788) 2 Cox Eq 92, 93.

21 [1945] KB 65. See discussion at text to n 9 above.

22 [1994] 1 AC 340, 371.

23 [1996] Ch 107.

24 (1870) LR 9 Eq 475.

25 [1960] AC 167.

26 Although there are no formality requirements attaching to a declaration of trust of personalty, this is probably explained by the fact that the writ was issued before the decision of the House of Lords in Tinsley v Milligan [1994] 1 AC 340 had been handed down.

27 ” … the circumstance of one or more of the nominees being a child or children of the purchaser, is to operate by rebutting the resulting trust …”:per Eyre CB in Dyer v Dyer (1788) 2 Cox 92, 93.

28 [1994] 1 AC 340.

29 [1996] Ch 107, 122.

30 [1960] AC 167, 177.

31 [1996] Ch 107, 134.

32 [1925] 2 KB 1.

33 Text to n 3.

34 [1937] 2 KB 158.

35 [1916] 2 KB 483.

36 Life Assurance Act 1774, s 1.

37 Gareth Jones (ed), Goff and Jones: The Law of Restitution (5th ed, London, 1998), p 607 (hereafter Goff and Jones).

38 A S Burrows, The Law of Restitution (London, 1993), chaps 11 and 15.

39 (1760) 2 Doug 696n. Goff and Jones (at p 613) would appear to claim this as an example of illegality as a cause of action, though it must be admitted that their treatment of it is ambiguous.

40 (1731) 2 Str 915.

41 The case has parallels in the Roman law, where the condictio ob turpem causam, ob iniustam causam would have been available to such a litigant.

42 [1916] 2 KB 483.

43 Although the plaintiff in Smith v Bromley was in a protected class, this fact went to the inapplicability of the defence (because she was not in pari delicto) rather than to the cause of action itself, which in this case was duress.

44 [1960] AC 192.

45 It was not until 1998 that English law allowed recovery for mistake of law: Kleinwort Benson Ltd v Lincoln City Council [1998] 3 WLR 1095.

46 (1760) 2 Doug 696n.

47 Text to n 2.

48 [1905] 2 KB 123.

49 As Lord Goff of Chieveley recently did in Goss v Chilcott [1996] AC 788, at least in cases in which there are no computational difficulties are involved in valuing the benefits received by the plaintiff.

50 [1905] 2 KB 123, 133-134 (Sir Richard Henn Collins MR), 136 (Mathew LJ), 138 (Cozens-Hardy LJ).

51 Mogul Steamship Co v McGregor, Gow & Co [1892] AC 25, 39, per Lord Halsbury.

52 As in Hazell v Hammersmith and Fulham LBC [1992] 2 AC 1, where an interest-rate swap agreement between a local authority and a merchant bank was held to be beyond the powers of the local authority.

53 As in the case of contracts for the sale of interests in land which are not in signed writing: Law of Property (Miscellaneous Provisions) Act 1989, s 2.

54 [1953] 2 QB 190.

55 Ibid, at p 195.

56 [1965] 1 WLR 537.

57 Ibid, at p 539.

58 Cf the comment of Millett LJ in Tribe v Tribe [1996] Ch 107, 133-134, that:

“It is, of course, artificial to think that anyone would be dissuaded by the primary rule [that no court will lend its aid to a man who founds his cause of action on an illegal act] from entering into a proposed fraud, if only because such a person would be unlikely to be a studious reader of the law reports or to seek advice from a lawyer whom he has taken fully into his confidence.”

59 (1876) 1 QBD 291, 300.

60 Ibid, at p 298-299.

61 (1828) 8 B & C 221. Although there are earlier statements to similar effect, all before this case are obiter: R Merkin, “Restitution by Withdrawal from Executory Illegal Contracts” (1981) 97 LQR 420, 423.

62 Ibid, at p 226.

63 Ibid, at p 225.

64 (1890) 24 QBD 742.

65 Lord Coleridge CJ, Lord Esher MR, and Fry LJJ.

66 [1951] 1 All ER 91.

67 [1905] 2 KB 123.

68 Text to n 48.

© 2000 W. Swadling. This HTML edition © 2000 University of Oxford.
This research has been supported by the network ‘Common Principles of European Private Law’, within the Training and Mobility of Researchers project (TMR) funded by the European Commission (Contract No ERB FMRX-CT97-0118), administered from Münster by Professor Dr Reinhard Schulze. The partner universities are: The University of Oxford (UK), Universidad de Barcelona (Spain), Humboldt-Universität zu Berlin (Germany), Université Jean Moulin Lyon III (France), Westfälische Wilhelms-Universität Münster (Germany), Katholieke Universiteit Nijmegen (Holland) and Università di Torino (Italy).
The contents of this page may be downloaded and printed out in single copies for individual use only. Making multiple copies without permission is prohibited.