by Thomas Krebs1
The English law of unjust enrichment has always insisted that a claimant asking for restitution must show that there is a good reason why restitution should be available to him. It has differed from its more ancient Continental counterparts in that the ground for restitution was not an abstraction like “lack of juridical reason for the enrichment”, but a pragmatic, positive requirement or “unjust factor”, as Professor Birks has called it.
One question which lies at the heart of the recent swaps litigation is whether there can be restitution in English law for the mere reason that money was paid pursuant to a contract which has subsequently turned out to be void. If this question is answered in the affirmative, it will mean that the English law of unjust enrichment has come close to adopting the civilian condictio indebiti – the claim to restitution on the basis that money was paid which was not due. In this chapter, it is argued that that would be the beginning of the end of the system of “unjust factors”. The English law of unjust enrichment would move considerably closer to civilian legal systems.
Before the question can be faced, it is necessary to look very briefly at the two different approaches. The leading text-book in Comparative Law, Zweigert & Kötz’s Introduction to Comparative Law, identifies “the entrenched position of the institution of unjustified enrichment” as one of the distinctive features of German law.2 For this reason amongst others, German law will serve as a model of the Civilian approach.
In the latest edition of their book The Law of Restitution, Goff and Jones write:
“If money has been paid under a contract which is or becomes ineffective, the recipient is evidently enriched. It is a distinct question whether that enrichment is an unjust enrichment.”3
This “distinct question” is answered by English law by reference to a list of what Professor Birks has called “unjust factors” – factors which render an enrichment unjust in the eyes of the law. It is characteristic of these unjust factors that they give us an immediately intelligible reason why restitution should follow. Professor Birks has identified three families of unjust factors:4 the “I did not mean to give” family of claims, “unconscientious receipt” and “policy-motivated restitution”. Where the claimant was mistaken in making the payment, as in Barclays Bank v Simms,5 where a bank honoured a cheque which had been countermanded, or in Chase Manhattan Bank v Israel-British Bank,6 where a payment had, by mistake, been made twice, it is readily apparent that the defendant’s enrichment is unjust and should be reversed. Similarly, where the claimant is forced to part with his money because of the defendant’s illegitimate pressure, the latter’s enrichment will be considered “unjust” by the law – this time, “duress” is the unjust factor.
In cases of “failure of consideration” payment is made on a certain condition, normally that the defendant will counter-perform. Where that counter-performance fails, as it did in the famous Fibrosa case,7 it is clear that restitution should be available to the disappointed claimant.8
It is controversial whether there exists an additional family of claims called “unconscientious receipt” or “free acceptance”. The reason for restitution here is argued to be the fact that the defendant exploited the claimant’s weakness in some way,9 or freely accepted a benefit which he knew was not proffered gratuitously.10
Classifications and taxonomies often contain a residual category “others”, and this taxonomy is no exception. In some cases, restitution is available even though the case does not fit either of these two broad categories. Invariably, this is because the law pursues a policy favouring restitution in this kind of case. It will be necessary to discuss “policy-motivated restitution” in more detail later. For now, the well-known case of Woolwich v IRC may serve as an example.11 The Woolwich had been assessed for taxes which, in its view, the Revenue had no power to raise. It nevertheless paid under protest. It then successfully challenged the tax, showing that the assessment had been ultra vires and void. The Revenue repaid the overpaid tax, but refused to pay interest on it. The Woolwich therefore had to show that, as soon as it paid over the money, it had a restitutionary claim against the Revenue, which would naturally attract interest. The difficulty was that the Woolwich had not made a mistake – it had known the assessment was ultra vires from the start, and had said so. Nor could it be seriously contended that the payment had been made under duress: any pressure exerted was the pressure of due legal process, and as such it was not “illegitimate”. The Woolwich nevertheless succeeded. The principle “no taxation without Parliament”, the policy in favour of the legality of government and the rule of law strongly favoured restitution.12
Within this taxonomy there is no room for “restitution of moneys paid under void transactions”. Nullity as such does not lead to restitution, although this does not mean that restitution will not be available in circumstances in which a payment is made under a void transaction. In the majority of cases some unjust factor will be available: thus, if the defendant has not yet performed, the claimant might be able to rely on failure of consideration. Even where this is not the case, policy-motivated restitution may well be available. The Woolwich case is an example in point: the Revenue’s claim was void, but it was not that voidness which led to restitution, but the policy favouring constitutional legality. Often, and this is an important point to grasp at the outset, the policy favouring nullity will also favour restitution. That is not to say, however, that this will always be the case in English law.
It is striking that German law refers to “unjustified” rather than “unjust” enrichment. This is more than a mere semantic difference. Although one should not fall into the error of supposing that in German law it is the defendant who has to show that he is entitled to keep the enrichment, the starting point in German law is that all enrichments are prima facie unjust, unless they can be justified. The fact that it is the claimant who has to show that the defendant’s enrichment cannot be justified does not change this fundamental characteristic of German law, which is inherent in the formulation of the general clause of unjust enrichment in § 812 I of the German Civil Code, the Bürgerliches Gesetzbuch (BGB):
|Wer durch die Leistung eines anderen oder in sonstiger Weise auf dessen Kosten etwas ohne rechtlichen Grund erlangt, ist ihm zur Herausgabe verpflichtet. (…)||He who obtains something through somebody else’s performance or in another way at his expense without a legal cause, is obliged to make restitution to the other. (…)|
By § 812 I 1 an enrichment is considered unjust if it lacks a ‘legal cause’. In most cases, this will mean that a benefit has been transferred under a supposed legal transaction. That transaction turns out to have been void, either because it was void from the beginning, or because it has since been avoided ab initio. In such cases, § 812 I 1 provides that the recipient has to return the benefit to the transferor.
The problem with § 812 I 1 is, however, that it is considerably wider than this. As Markesinis, Lorenz and Dannemann point out, the main dilemma with general clauses ‘is that with them one tends to get more than one has bargained for: their wording will often cover more than it should. The main attention within legal systems based on general clauses will, therefore, be geared towards excluding certain categories from the application of the general rule.’13 The section has thus been narrowed down considerably, first by jurists, then by judges who adopted the proposed academic solutions. German law today differentiates between ‘enrichments by performance’ and ‘enrichments in another way’. Both have precisely defined requirements, which it has taken German jurists almost half a century to work out, and even today there is considerable controversy whether the distinction is appropriate, and where precisely it lies.
In this chapter, we will concentrate on “enrichments by performance”. While it is argued in Germany that in these cases restitution will follow because the purpose of the transfer, namely the discharge of an obligation, has failed,14 it is clear that in practice restitution will normally follow where a benefit has been transferred pursuant to an obligation which has later turned out to be void.
After this brief overview of the different approaches to unjust enrichment in England and Germany, it is now time to consider the swaps litigation. It is argued in this chapter that the effect of that litigation has been to move English law appreciably closer to German law.
Many readers will be, in Lord Goff’s words, “seasoned warriors in the continuing battle of the swaps”.15 They may forgive me for the brief explanation of the background which follows. Interest swaps are traded on the financial markets. Two parties lend each other a notional sum on terms that one pays interest at a fixed, the other at a floating rate. Both are speculating that interest rates will develop in their favour. Local authorities widely engaged in interest rate swaps from the early eighties onwards. In Hazell v Hammersmith and Fulham London Borough Council16 the House of Lords held that such transactions were ultra vires local authorities and that any swaps that local authorities had entered into were therefore wholly void. The consequence was a wave of litigation in which losing parties to interest rate swaps sought restitution of their losses. In this chapter, we are mainly concerned with the first and the last of these cases, namely with Westdeutsche Landesbank v Islington LBC17 and Kleinwort Benson v Lincoln LBC.18
The cases which reached the Commercial Court can be broadly divided into those which concerned “interrupted swaps” and those which concerned “closed swaps”. In interrupted swaps the risks had not been allowed to run their full course. One party was ahead when the blow of Hazell fell, but there was no telling whether the roles of winner and loser might not have been reversed as the swap continued, sometimes over several more years. Closed swaps had been fully performed, winners and losers conclusively determined. The risks had run their course, the parties had got what they had bargained for.
Litigants asking for restitution of moneys paid pursuant to interest rate swaps involving local authorities were faced with a number of problems. In most cases payments had been made both ways. The losing party had got at least something from the winner. This made it more difficult to argue that there had been a total failure of consideration, otherwise the obvious ground on which restitution could have been based. This problem was compounded in cases involving closed swaps. In these cases there had been no failure of consideration at all. The parties had got exactly what they had bargained for. It was furthermore difficult to base the claim for restitution on the ground of mistake. If there had been a mistake at all, and this is a question of some controversy, it had been one of law, and only the House of Lords could depart from the long-standing rule that there could be no restitution in English law based on mistake of law only. It was thus tempting to argue that restitution should be awarded for the simple reason that the interest rate swaps under which the money had been paid had been declared void by the House of Lords. This approach avoided all of the above difficulties at a stroke.
Westdeutsche Landesbank v Islington LBC was the first swaps case to reach the High Court. Two cases, Westdeutsche itself and Kleinwort Benson v Sandwell were heard together by Hobhouse J. In Westdeutsche, the relevant swaps had all been interrupted. One swap in Sandwell, however, had been fully completed. Irrespective of whether the swaps had been completed or not, Hobhouse J ordered restitution because there had been “no consideration” for the payments, the swap agreement having been void from the start. In other words, he based restitution on the nullity of the underlying transaction alone, without looking for substantive reasons for restitution, thus putting Professor Birks’s taxonomy into doubt and moving English law considerably closer to the Leistungskondiktion.
Professor Birks has since shown19 that restitution in interrupted swaps can be explained on the ground of failure of consideration: the risks have not run their course, the parties did not get what they bargained for. The requirement that a failure of consideration must be total can be avoided by ordering restitution of the net-balance of payments, as Hobhouse J indeed did in Westdeutsche itself. When Islington appealed to the Court of Appeal, this analysis was confirmed, although in a somewhat half-hearted fashion, and while arguably endorsing the “no consideration” approach at the same time. As Sandwell did not appeal, the Court of Appeal did not get the opportunity to decide the crucial question whether restitution was available in completed swaps until much later, in Guinness Mahon & Co v Kensington and Chelsea London BC.20
In a completed swap, there can be no question of restitution based on failure of consideration. While the mistake of law bar was still in place, the only conceivable ground for restitution was one of policy: the ultra vires doctrine pursues a policy in favour of a responsible administration of public finances. This policy is best realised by making sure that any ultra vires transactions entered into by local authorities are reversed, at least where the local authorities are the losers. A legal system committed to the rule of law, however, cannot countenance the possibility of compelling the non-suspecting banks to repay their winnings to the local authorities while not putting them in a position to likewise recoup their own losses.21 The policy underlying the ultra vires doctrine in this type of case could thus have justified ordering restitution in all cases. This argument, however, was never seriously advanced.22 Once Hobhouse J had ordered restitution on the ground of “no consideration”, it was no longer necessary to discuss the policy underlying the nullity in these particular cases. “No consideration” was far more convenient.
In Guinness Mahon & Co v Kensington and Chelsea London BC the Court of Appeal confirmed that “absence of consideration” was available as a ground for restitution in cases of completed swaps. Thus, there is now unequivocal Court of Appeal authority that nullity as such can trigger restitution. It is nevertheless open to doubt whether this line of authority will in future need to be relied on. This is because of the recent case of Kleinwort Benson v Lincoln CC, to which the discussion will now turn.
“Absence of consideration” was unlikely to assist the claimants in the latest and probably last instalment of the swaps saga. Some of the payments had been made more than six years before issue of the writ. The only hope was recovery on the ground of mistake, because s. 32(1)(c) of the Limitation Act 1980 meant that time would only have begun to run when the mistake could have been discovered, in other words when Hazell was decided.
As mentioned above, the obstacle in the way of such a claim was the long-standing rule that there could be no restitution for the sole reason that a transfer had been made under a mistake of law. A mistaken belief in the legal capacity of the local authority was, after all, clearly a mistake of law rather than fact. Only the House of Lords could abrogate that rule, and by a three to two majority it did.23 Their Lordships’ disagreement did not concern the question whether the rule should be abolished or not: they were unanimous in their condemnation of it. However, Lords Browne-Wilkinson and Lloyd felt that the bank could not be described as having paid the money under a mistake at all: at the time the payment was made the law was indeed what they thought it was, and most lawyers would have advised them that they were liable to pay. It was only later, when the Divisional Court declared swap transactions ultra vires local authorities, that the law was changed. While both Law Lords accepted that the decision of an English court was to some extent retrospective, they both felt that “retrospection cannot falsify history”.24 The “paradigm case” which they relied on is one in which the House of Lords overrules a line of Court of Appeal authority. In that situation, according to Lord Lloyd,25 prior to the House of Lords ruling, the law was as the Court of Appeal had said previously. The House of Lords is unable to change this. “The House of Lords can say that the Court of Appeal took a wrong turning. It can say what the law should have been. But it cannot say that the law actually applied by the Court of Appeal was other than what it was.”26 The majority, however, took the view that the declaratory theory of judicial decision meant that the law applicable at the time of the payment indeed had been what the House of Lords now declared it to be. It followed from this that at the time of the payment the payer was labouring under a mistake of law.
The question what constitutes a mistake in this context will exercise the minds of unjust enrichment and contract lawyers for some time to come. It will also be of great interest to legal philosophers. This is not the place to discuss the question in any detail. Whatever view one takes on this issue, however, the result of the decision of the majority is that where the law is changed by judicial decision this may well have restitutionary consequences. The emphasis is taken away from the state of mind of the payer and is transferred to the validity or otherwise of the obligation he sought to discharge by making the payment. Restitution follows, not so much because the payer was mistaken, but because the obligation he sought to discharge has now been declared void.
This was always going to be one difficulty with abolishing the mistake of law bar. Professor Birks had foreseen it. He had argued that, where both parties had got precisely what they had bargained for, in other words where they could no longer be prejudiced by their mistake of law, restitution should not be available – the mistake, if any, had been “spent”.27 Professor Birks thus focused on mistake, not on the nullity as such. Had his argument been accepted, it would have been necessary, in cases of “spent” mistakes, to ask why the underlying transaction had been void, and whether the policy favouring nullity also favoured restitution. Restitution in such cases would thus have been policy-based, not mistake-based. The House of Lords unfortunately rejected Professor Birks’s analysis without discussing it in any detail, and apparently without realising that there was much more to it than the question at what point in time the cause of action in money had and received arose.
In consequence, whenever a payment is now made under a void contract, a defendant will face an uphill struggle if he is to convince the court that the claimant was not labouring under a “mistaken” belief in the validity of the contract. The result will be that in the overwhelming majority of cases nullity by itself will be sufficient to trigger restitution. While the dismantling of the mistake of law bar has no doubt taken the wind out of the sails of the “no consideration” doctrine, the end result is very much the same, with nullity by itself leading to restitution of any benefits received pursuant to a void transaction.
Lord Goff, true to form, referred to comparative law in order to justify his reasoning. German law took pride of place in his analysis:
“It is of some interest that, in German law, recovery is not dependent on proof of mistake (whether of fact or law) by the claimant. Para 812(1) of the BGB (BürgerlichesGesetzbuch) confers a right to recover benefit obtained without legal justification (ohne rechtlichen Grund). […] Para 814 of the BGB however provides that a person cannot reclaim a benefit conferred by him if he knew that he was not bound to confer it; but it seems that the burden rests on the recipient to prove the existence of such knowledge (a striking contrast with the common law, which requires the plaintiff to prove mistake). […] For present purposes, however, the importance of this comparative material is to reveal that, in civil law systems, a blanket exclusion of recovery of money paid under a mistake of law is not regarded as necessary. In particular, the experience of these systems assists to dispel the fears expressed in the early English cases that a right of recovery on the ground of mistake of law may lead to a flood of litigation, while at the same time it shows that in some cases a right of recovery, which has in the past been denied in civil law countries on the basis of a narrower ground of principle or policy.”
There is much in this brief extract to stimulate the thought processes of a comparative lawyer. In particular: if it is true that the decision in Kleinwort Benson means that a claimant no longer needs to prove a mistake properly so called, but that the voidness of the transaction itself triggers restitution, the words of Lord Goff assume a new significance. Is it truly possible to conclude from the experience of German law that restitution for lack of legal ground will not lead to a flood of litigation in a common law system? The remainder of this chapter will examine this question.
In German as in English law, there are a number of reasons why a supposedly valid obligation might later turn out to be void. As a general rule, the use of the word “nichtig” (“void”) triggers the application of § 812 ff, the law of unjust enrichment. This might be thought to mean that the voidness of the obligation itself is the reason for restitution, or at least that it is the mistaken belief in the validity of the obligation which could be described as the “unjust factor”. In a seminal essay, which to this day can be described as the basis of the modern German law of unjust enrichment, Professor von Caemmerer pointed out that this is not so. He wrote: “The actual reasons for restitution lie outside the law of unjust enrichment. The law of restitution would otherwise be overburdened”.28 Whether an obligation is valid, void or voidable is determined by the general law. Where the general law determines that an obligation is void, or can be rendered void by one party’s unilateral act, restitution is the consequence. The general law is drafted with that consequence in mind. Generally, when the law uses the word “nichtig“, it is aware that this might have restitutionary consequences. Where restitution is not intended, the word “nichtig” can be avoided or “disarmed”, by making it clear that though an obligation is wholly void, any payments made under it are to be irrecoverable.
One way in which a transaction can end up void is because one or the other of the parties decides to exercise a right of rescission. According to § 142 I, if a transaction is rescinded, it becomes void ab initio, or, as a German lawyer would say, void “ex tunc“. A party may have a right to rescind on the grounds of mistake (§ 119 I), duress (§ 123 I) or fraudulent misrepresentation (§ 123 I). There are significant differences between English and German law in terms of what mistakes count, what kind of threat is unlawful and amounts to duress, and when rescission is possible for misrepresentation. These differences do not concern us here. Where both systems are largely similar is in the mechanism by which these vitiating factors are translated into reasons for restitution. If a party was coerced into contracting by duress, and paid out money under that contract, he will be able to extricate himself from the contract by rescinding it. In English law, rescission has the automatic consequence of restitution, although it is a matter of debate in what way this is to be conceptualised. In German law, rescission renders the contract void ab initio, which in turn triggers § 812 I. Restitution follows. The duress which caused the contract to be voidable is also the underlying reason for restitution.
Illegality is a particularly good example of the point I am trying to make. Where a party has contracted to do an act which is illegal, the law cannot be seen to enforce that contract. If it did, this would give rise to an inherent contradiction within the legal system. It is therefore imperative that obligations to do an illegal act will not be enforced by the law. To that extent at least such obligations must be invalid.
If an illegal contract has already been performed, on the other hand, the law is no longer faced with the danger of an inherent contradiction: it is not faced with the difficulty that giving the contract validity will imply using the legal process for illegal purposes. The question which must now be faced is whether the illegality should lead to restitution. This question is a complex one and need not be fully discussed for present purposes. It will be the subject of later chapters of this book.29 For now it suffices to point out that German law is sufficiently flexible to examine the nature and purpose of the infringed law in deciding whether or not restitution should follow. § 134 reads: “A transaction which contravenes a legal prohibition is void, unless the contrary appears from the legal provision in question”.
In order to decide whether or not a transaction should be held void, the court will thus be able to look at the relevant statute and ask whether its purpose would be furthered by awarding restitution. For example, German law makes it an offence for a shop to trade after opening hours as laid down by federal statute. If this rule is breached, it would make little sense to hold the contract of sale void, requiring the parties to make restitution to each other. The purpose of the statute, namely to prevent the exploitation of shop assistants, would not be furthered by such a rule. On the contrary, the unwinding of the contract would require them to work even longer hours.30 On the other hand, German law also makes it an offence for an unlicensed craftsman to contract for work. His customer also commits an offence by engaging him. These rules attempt, with very limited success, to prevent the evasion of VAT and income tax. That purpose is best served by rendering such contracts absolutely void. The worker cannot be sure that he will be paid, and a restitutionary claim will not be available to him.31 The aim is to dissuade him from entering into such illicit bargains.
The voidness of an illegal contract, particularly if it has already been executed, is thus not a foregone conclusion. It depends on a myriad of policy considerations. The judge retains a great degree of flexibility. Again, the word “nichtig” could only be used as long as the judge retained that flexibility. Otherwise restitution might have to be awarded in circumstances in which it might not be appropriate.
It should further be pointed out that the law of unjust enrichment itself contains a special provision concerned with illegality and immorality. Broadly, this focuses on the legality and morality of the receipt of the benefit. If the receipt itself is contrary to law or morals, § 817 provides an independent restitutionary “unjust factor”. Restitution is wholly barred, however, where both parties were guilty of infringing law or morals. This provision will be examined in greater detail in a later chapter. For now it makes it clear that, at least in the area of illegality, to say that restitution will follow when a performance has been made under a void transaction is not very informative.
We can summarise: the German legislator is well aware of the restitutionary consequences of rendering a transaction or obligation void. The code attempts not to use the word “nichtig” where such consequences are not desired.
Sometimes, however, performances are made under transactions which are void, but where restitution is for one reason or another not desired by the legal order. This is an effect of the very wide negative formulation of § 812 I, focusing on the lack of legal ground. In order to identify such cases, German law has had to construct a typology of factors barring restitution. We might call them “grounds for retention” or indeed, slightly cynically, “just factors”.
Formality requirements may be imposed by the law in the interest of the parties themselves or in the public interest. The parties themselves may benefit because they are prevented from taking ill thought-out, hasty steps which might have serious consequences, and because a formal transaction is easier to prove subsequently. The frequent requirement of notary attestation also ensures the provision of appropriate legal advice, so that the parties are prevented from entering blindly into a transaction the consequences of which they do not fully understand. Formalities also benefit the state: transactions are recorded in order to keep public registers up to date and to levy taxes and fees. Generally, § 125, 1 provides that where legal formality requirements are not complied with, the transaction is absolutely void.32 This general rule, however, does not apply across the board. In particular, in certain cases the provision imposing the formality requirement makes sure that § 125, 1 will have no restitutionary consequences.
§ 313 requires that land contracts must be attested by a notary. If they are not, they should be void according to § 125, 1. The second sentence of § 313, however, provides that where the land contract has been fully performed and the new owner registered at the land registry (Grundbuchamt), the lack of form is “cured” and the contract becomes binding. Full performance is therefore seen as a factor militating against restitution. Registration itself is a formal act, and the purpose of § 313 is thus largely achieved by registering the new owner’s title.33 Unwinding the transaction at this stage would mean compromising the integrity of the register, which is a further reason to bar restitution.
Similarly, § 518 I provides that gift promises require attestation by a notary. If they are not attested, they are thus void according to § 125. However, § 518 II provides that again lack of form is cured by full performance, which must be a relief to all children on their birthdays. Again, once the gift promise has been fully performed, the gift handed over, there are obvious strong reasons militating against restitution which make § 518 II necessary.
A final example is provided by § 766. This lays down the requirement that a guarantee or surety agreement must be in writing. Otherwise § 125 will apply. The purpose of this requirement is that, since the surety does not have to pay any money at the time of agreeing to guarantee the main debt, he should be warned against the potential adverse consequences of that step.34 The second sentence of § 766 again provides that once the surety pays up and satisfies the creditor, the lack of formality is cured. There is no longer any need to warn the surety at that stage. Indeed, even if the surety mistakenly believes to be bound by the guarantee, this will not help him.35 The payment is, in other words, a ground of retention which trumps the prima facie voidness of the guarantee. The fact that the obligation was void at or just before the time of payment does not trigger restitution.
§ 762 I provides that a bet does not give rise to an obligation. This is commonly justified not by moral objections but by the inherent dangers of gambling contracts.36 The provision goes on, however, to say that any performances under a bet cannot be recovered on the basis that no obligation existed. The second sentence of § 762 I thus effectively disapplies § 812 I.
§ 656, which applies to marriage brokerage, is in almost identical terms. Again, it provides that a promise to pay in consideration of marriage brokerage services does not give rise to an obligation, and again it goes on to say that if such a promise is nevertheless honoured the fact that no obligation existed cannot be relied on to demand restitution of any payments made.
These so-called “natural obligations”37 are good examples of cases in which German law considers that even the strongest form of invalidity, the total absence of an obligation, should not lead to restitution if performances are made. It would go too far to discuss the policies underlying these rules as long as it is pointed out that the legislator was quite clearly aware that both forms of “natural obligations” would be caught by the very wide § 812 I unless it was made clear that the latter provision should not apply. Although they are not obligations properly so-called, “natural” obligations thus form legal grounds; in other words, they entitle the recipient to retain the benefit conferred – they are reasons for retention.
Time-barred claims, often also placed in the category of “natural obligations”, are not void, but unenforceable. Again, the legislator makes careful use of terminology in § 222 I: “After the limitation period is expired, the debtor is entitled to refuse to perform.” Given that the word “nichtig” is not employed, it is arguable that § 812 I would not have applied in any event. § 222 II nevertheless puts it beyond doubt that where a time-barred claim is satisfied, the fact that the limitation period had expired does not entitle the payer to restitution, even where he was unaware that the claim was time-barred and thus unenforceable. The time-barred claim, though not enforceable, nevertheless constitutes a legal ground, a reason for retention.
If a widely formulated ground for restitution based on mistake of law has a saving grace, it is that it is clearly restricted to cases in which the claimant made a conscious decision to transfer the relevant benefit to the defendant. It cannot be said to extend to cases in which the benefit accrued to the defendant in any other way. However, we must not forget that the Court of Appeal in Guinness Mahon enthusiastically endorsed the “absence of consideration” approach, and although Kleinwort Benson may have stolen the thunder of that idea for the time being, it is still lurking at the back of the minds of legal advisers looking for a way to frame the otherwise hopeless claims of their clients.
Professor Schlechtriem, in his textbook on the law of obligations, writes: “The more abstract the requirements for enrichment claims, the more difficult becomes the necessary limiting exercise; this is especially true if one assumes one basic general requirement for all enrichment claims.”38 He writes in the context of three-party situations, a particular problem area in German enrichment law,39 but his comment is of more general validity. To base restitution on the absence or otherwise of a “legal ground” invites confusion. In particular, if the abstraction is rendered, as has been done so far in this chapter, as “basis in the law of obligations”, ie as an obligation by which the transferor is compelled to render a specific performance, one will run into problems when the enrichment is based not on a performance, but has come about in “another way”, as § 812 I puts it. Cases which English lawyers would now categorise as falling into the group of “restitution for wrongs”, or cases in which the claimant was wholly unaware that the defendant was enriching himself at his expense, called cases of “ignorance” by Professor Birks, simply do not involve a supposed obligation. Some other criteria must be found in order to decide when restitution should be available. This is not the place to go into this discussion. It should merely be pointed out that § 812 I as it was interpreted in the early decades of this century, just after the enactment of theBGB, represented an open invitation to dispense palm-tree justice, and was in fact interpreted as such. It is only thanks to the self-denying ordinance of theReichsgericht and the brilliance of two academics that this did not happen. Professors Wilburg and von Caemmerer, in 1934 and 1957 respectively, developed a typology based on the way in which the enrichment was obtained which was accepted by the Bundesgerichtshof as late as the 1960s. It was only then that German enrichment law was on a secure intellectual and typological footing.
The development of the Wilburg/von Caemmerer typology, presented as an example for English law by Professor Zimmermann,40 took several decades. It is not obvious to me that English law would simply adopt this typology wholesale, were it to accept a similarly broad general requirement for restitution. It is to be hoped that the potential time bomb of restitution for “absence of consideration” has been defused by Kleinwort Benson v Lincoln. However, the “absence of consideration” idea is still out there, it has never been expressly disapproved by the House of Lords and is thus, in Professor Birks’s words, left “lying casually around”,41 to be relied on by a litigant who is not even able to rely on the new and very broad unjust factor “mistake of law”, for the simple reason that he never made a conscious transfer in the first place.
I do not propose to argue that English law, contrary to Professor Zimmermann’s allegation, is “distinguished by its elegance”. I do argue, however, that English law serves its purpose: to identify those cases in which restitution should be available by requiring the claimant to prove that his claim falls within one of the established categories. In other words, the claimant has to show that there is a reason, recognised by the law, why he should have restitution.
Before the swaps litigation, the fact that a benefit was transferred under a void or non-existent obligation did not by itself trigger restitution. Lord Goff has recognised the superficial attractiveness of the proposition that where money is paid under a void contract it ought to be recoverable. One objection to this “simple call of justice” he saw in the structure of the English law of restitution:
“That law might have developed so as to recognise a condictio indebiti – an action for the recovery of money on the ground that it was not due. But it did not do so. Instead, as we have seen, there developed common law actions for the recovery of money paid under a mistake of fact, and under certain forms of compulsion.”42
It has been argued above that Westdeutsche in effect introduced the condictio indebiti into English law. It is arguable that Kleinwort Benson v Lincoln achieved very much the same result by dressing the Leistungskondiktion in English sheep’s clothing: by interpreting mistake of law to encompass cases where the “mistake” arose by subsequent overruling of existing law alone, in effect placing the emphasis on the invalidity of the obligation rather than the mistake of the claimant.
The great beauty of German law, as we have seen, is that the law of unjust enrichment works in tandem with the general law. The word “nichtig” works, if the simile is not too far-fetched, like a catalyst, referring the claimant to the law of unjust enrichment wherever it occurs. We have seen how this result can be achieved within the framework of a coherent codification, namely by the careful use of the word “nichtig“, by providing a number of “reasons for retention” which I have cynically called “just factors”, and last not least by developing a typology which has successfully limited the potentially very wide general clause in § 812 I, dispelling fears of palm-tree justice.
Outside the law of rescission, there is no such working in tandem in English law. In most cases, even following Kleinwort Benson, there will be a certain duplication of inquiries. It is clear that where a contract is valid, the law of restitution can have no role to play.43 The contract must therefore be shown to be void before the claimant can begin to establish a restitutionary claim. The “unjust factor” he relies on will often be the same as the vitiating factor he relied on to show that the contract was void. Thus, if the contract is void for mistake, it is likely that the mistake will also be relied upon as an unjust factor. Yet does this really matter? Elegance is not the be all and end all of the law. It is unlikely that where a contract is void for mistake much argument would be directed at the question of whether that mistake should also trigger restitution: it would appear that that would be the obvious conclusion.
Where contracts turn out to be invalid, failure of consideration will often be relied upon as an unjust factor. The contract being void, it is no longer possible to compel the defendant to fulfil his part of the bargain. The restitutionary claim enables the claimant at least to get back his own performance.
Some contracts are rendered void or voidable for policy reasons which favour restitution. In such cases the reason for restitution is that restitution favours the policy in question. One example is section 127 of the Insolvency Act 1986. That section, re-enacting section 522 of the Companies Act 1985, provides that in a winding up by the court any disposition of the company’s property made after the commencement of the winding up is void unless the court orders otherwise. It is not possible nor necessary to analyse the effects of the section in detail here. One effect of the section might be that an alienee never obtains the property in the benefit transferred. This may be so, for instance, where the property in goods is supposed to pass at the time of contracting. The contract and the transfer of property are void from the beginning and the company (whether through liquidator, administrator or administrative receiver) will be able to recover the goods. Whether the company’s claim to the goods is characterised as restitutionary is a difficult question, given that the property in the goods remains the company’s throughout. Where money is paid by the company after the commencement of the winding up, however, the company’s right under section 127 is clearly restitutionary. This is so because it is one of the characteristics of money that it is fully negotiable.44 Once the payment passes into currency the company’s title is lost.
Where a transaction transfers good title, notwithstanding section 127, the transferee will be subject to a claim to make restitution of the benefit obtained to the company, unless the court orders otherwise.45 It is difficult to explain this using mainstream unjust factors. “Mistake” may sometimes work, because the company may not have been aware of the presentation of the winding-up petition at the time of the payment, but this will only be so in some cases. Likewise, “failure of consideration” will only explain restitution in rare instances. No, the reason for restitution must be the policy of the bankruptcy laws to ensure the pari passu distribution of the company’s assets amongst creditors.46 Section 127 has this in common with other rules designed to protect the interests of creditors, in particular section 238 (transactions at an undervalue), section 239 (preferences) and section 423 (transactions defrauding creditors).47 Sometimes, restitution will favour that policy. In such cases, the court will refuse to ratify the disposition. In others, the disposition might have been beneficial to the company, and thus to the company’s creditors. In such a case, the court may ratify the disposition under s 127. The court’s power is similar to the German court’s discretion under §134 BGB to uphold an illegal transaction, refusing to declare it void where restitution would not favour the policy pursued by the infringed rule of law.48
It is tempting, however, to conclude that the word “void” will always have similar consequences, that it will always imply not only that an obligation cannot be relied on to found a claim to performance or compensation, but that it will also always have restitutionary consequences. In other words, one might be tempted to generalise that the policy underlying the decision to render a transaction void will also favour restitution in all cases. Hobhouse J drew that conclusion in Westdeutsche: the ultra vires doctrine rendered the swaps transactions void. Hobhouse J relied on nullity as such in granting the banks restitution. This made it unnecessary to analyse the ultra vires doctrine itself. Hobhouse J explicitly held that the ultra vires doctrine could not possibly assist the banks.49 He thus never asked himself whether the policy underlying it, namely the aim of protecting the integrity of public finances, made restitution necessary. Therefore, although it is possible to defend the outcome of Westdeutsche on that basis,50 the problem is that the inquiry was never undertaken.
In English law, it is not always possible to conclude that where an obligation is void, the policy which underlies the nullity will also favour restitution. For example, under section 2 of the Law of Property (Miscellaneous Provisions) Act 1989, a contract for the sale or other disposition of an interest in land can only be made in writing, all the terms must be contained in one document, or, where contracts are exchanged, in each, and the relevant documents must be signed by or on behalf of each party to the contract. It must follow that where these formalities are not complied with, the agreement in question must be void and of no effect. Where one party refuses to perform such an agreement, the courts will not assist the other by ordering specific performance or by awarding him damages. The question in the present context is, however, whether such an agreement, once performed, will subsequently have to be reversed, or will be reversible at the instance of one party or the other. If restitution must necessarily follow where performances have been exchanged pursuant to a void contract, that should clearly be the result.
Tootal Clothing Ltd v Guinea Properties Ltd51 concerned a lease under which the tenant was to carry out certain shop-fitting works to the demised premises. The landlord, Guinea Properties, was to contribute to the costs of these works. Its obligation to do so, however, was contained in a separate document. Having taken possession of the property, and having completed the works, the tenant sued for recovery of the landlord’s contribution. The landlord argued that the agreement was void and unenforceable because under section 2 of the 1989 Act all the relevant terms should have been in one document. This argument succeeded at first instance before Douglas Brown J, dealing with the question as a preliminary issue. He held “without any enthusiasm at all” that section 2 applied to the contract and that therefore all the terms of the contract had to be incorporated in one document. In the Court of Appeal, Scott LJ, with whom Boreham J and Parker LJ agreed, held that Guinea’s defence missed the point of section 2 of the 1989 Act. If the argument was sound, the result would be that executed agreements would have to be unravelled. It was therefore necessary to ask what policy was pursued by section 2. This question had been addressed previously by Hoffmann J (as he then was) in Spiro v Glencrown Properties Ltd.52 He had held that section 2 “was intended to prevent disputes over whether the parties had entered into a binding agreement or over what terms they had agreed.” As such, argued Scott LJ, section 2 was of relevance only to executory contracts. He continued:
“If parties choose to complete an oral land contract or a land contract that does not in some respect or other comply with section 2, they are at liberty to do so. Once they have done so, it becomes irrelevant that the contract they have completed may not have been in accordance with section 2.”53
While Tootal Clothing did not concern a restitutionary claim, the conclusion that section 2 had no relevance to executed contracts was necessary to avoid the conclusion which Douglas Brown J had felt compelled to draw at first instance.
For present purposes, Tootal Clothing can be interpreted in two ways: on the one hand it can be argued that the case clearly demonstrates that nullity as such will not lead to restitution where both parties have obtained what they bargained for, in other words, where there has been no failure of consideration or other substantive reason for restitution. On the other hand, it is arguable that lack of formality renders a contract unenforceable rather than void, or that the lack of formality is “cured” by subsequent performance. This would mirror the position in German law.
The traditional inquiry in English law would be turned right around. Instead of asking for reasons why an enrichment should be reversed, the defendant, enriched pursuant to a void contract, would have to show reasons justifying retention. English law, however, is not contained in a coherent code. Nor is language always used with the same degree of precision in English statutes as in the German civil code. When the word “void” is used in English law, it is far from clear that the policy underlying nullity can be made to extend to restitution of benefits conferred. The simple truth is that English law lacks the system of “just factors” which the fathers of the BGB were able to draft into their code. In the wake of Kleinwort Benson, this is likely to give rise to a number of problems.
One obvious difference between English law and German law is that English law still insists on the presence of some consideration before a promise will be enforced as a contract. In the well-known case of Eastwood v Kenyon54 a guardian decided to improve his young ward’s matrimonial prospects by renovating a house which belonged to her. He raised the necessary funds on a promissory note. The scheme bore fruit, and soon the ward was engaged to be married. The bridegroom, feeling morally obliged to do so, promised to pay the promissory note. Once the couple were married, however, he changed his mind. When sued upon his promise it was held that, contrary to Lord Mansfield’s view,55 a moral obligation was not sufficient consideration to support a promise. Any consideration furnished by the guardian was past consideration, and as such no consideration at all. In order to make my point, I have to change the facts slightly: what if the defendant had already honoured his promise, in the mistaken belief that he was bound to do so? English law has never had to address this question, because the mistake of law bar has prevented it from arising. Now, following Westdeutsche and Kleinwort Benson, the husband, in a restitutionary claim against the guardian, would be able to rely on two compelling arguments in favour of recovery: first, he could say that, in paying, he had made a mistake of law. Unknown to him, his promise did not bind him. He should therefore recover. Secondly, he could argue, the contract which he thought compelled him to pay never came into being because the three requirements of contractual liability, offer, acceptance and consideration, were never present. There should thus be restitution based on “absence of consideration”.
Restitution in such a case, however, is not desirable. The general policy of the law should be that people should honour their promises, the doctrine of consideration being merely a means to make sure that promises will not be enforced which were not given seriously, with intention to give rise to legal relations. English law will thus have to deal with this case in some way, following the recent developments in the law of unjust enrichment. Dr Sonia Meier suggests that the only way to deal with this situation is a wholesale adoption of the condictio indebiti. She argues that the underlying promise, though not binding as a contract, would still be sufficient to give rise to a legal ground. In other words, payment in accordance with an unenforceable promise will have to be considered a “just factor”, a reason for retention. As Dr Meier rightly recognises, this would constitute the final acceptance by English law that the German solution is superior.
The modified Eastwood v Kenyon raises another question, addressed in German law by § 814: the defendant was under a moral obligation to reimburse the claimant. The latter, although he had spent his money “voluntarily”, had spent it on improving his ward’s property. English law would probably deny him a restitutionary claim against the ward based on some kind of “moral compulsion” or necessity: Nicholson v Chapman,56Falcke v Scottish Imperial Insurance.57 It is nevertheless easy to see why the ward’s later husband would feel obliged to his wife’s former guardian and reimburse him for his expenditure. Should he be able to reclaim that reimbursement on the basis that he paid for “no consideration” or because he had believed58 himself legally obliged to pay? Again, English law might have to make an exception to the general rule, might have to recognise yet another “just factor”.
We can see that “restitution for nullity”, whether it be referred to as restitution for “absence of consideration” or restitution for mistake of law, will give rise to problems in English law, problems which go further than the obvious difficulty that all gifts might become recoverable. English law does not yet have the mechanisms which have enabled German law to function correctly. It will most likely have to develop them now.
Comparative law can be a powerful tool of legal analysis. Superficial comparative law, however, can be dangerous. The comparative lawyer must never lose sight of “the bigger picture”. Just because a legal rule operates well in one legal system does not necessarily mean that it will operate equally well in another. This is particularly true of the law of unjust enrichment. We may admire the German solution, we may be struck by its elegance, impressed by its logical coherence. But we must not be blinded by our own enthusiasm. The two legal systems, the two legal cultures are fundamentally different. These differences reflect differences in legal history and in cultural and intellectual development. I cannot put it any better than Tony Weir in his brilliant translation of Zweigert & Kötz’s Introduction to Comparative Law:
“If we may generalize, the European is given to making plans, to regulating things in advance, and therefore, in terms of law, to drawing up rules and systematizing them. He approaches life with fixed ideas, and operates deductively. The Englishman improvises, never making a decision until he has to: ‘we’ll cross that bridge when we come to it’. As Maitland said, he is an empiricist. Only experience counts for him; theorizing has little appeal; and so he is not given to abstract rules of law. Convinced, perhaps from living by the sea, that life will controvert the best-laid plans, the Englishman is more at home with case-law proceeding cautiously step-by-step than with legislation which purports to lay down rules for the solution of all future cases.”59
It fits in well with these broad general differences that German law should lay down a general clause of unjust enrichment, designed to operate within a coherent code, made possible by the rigorous and uniform use of terms like the word “nichtig“, while English law, less confident about the coherence of the legal system as a whole, should put its trust in less abstract, less well-defined “unjust factors”. The German law of unjust enrichment operates deductively, its starting point is a wide general principle, which is limited and narrowed down both by the codification itself and by subsequent academic analysis and interpretation. English law, on the other hand, has chosen to extend the incidence of restitution for unjust enrichment little by little, case by case, always, until now, on the basis that there must be some good reason why restitution should be available. In Kleinwort Benson v Lincoln and in Westdeutsche Landesbank v Islington the English courts have for the first time chosen a different approach. The new rule in English law is that whenever a payment is made under a void transaction, restitution will be available. The rule is very wide, and very indiscriminate. It no longer asks whether there is a good reason for restitution. The rule does fit well into a continental legal system. Our own system is likely to have problems with it. The rule is so wide that it will need to be narrowed down. Litigants will be encouraged by it to come to court with cases which previously would have been struck out as disclosing no reasonable cause of action. Now such cases are likely to proceed to trial until a new taxonomy of unjust enrichment is found. German law may be instructive in shaping that taxonomy.60 It is nevertheless a shame that the English law of restitution, having just settled down, having just established its place on the English legal map, should now be thrown into renewed turmoil.
1 I am grateful to Professor McKendrick for his comments on an earlier draft.
2 Zweigert & Kötz, An Introduction to Comparative Law, (trans. Weir), 3rd edition, Oxford 1998, 71
3 Goff & Jones, The Law of Restitution (5th ed, 1998), 499
4 See Birks & Chambers, Restitution Research Resource, 2nd Edition, § 113
5 Barclays Bank Ltd v WJ Simms Son and Cooke (Southern) Ltd  1 QB 677
6 Chase Manhattan Bank NA Ltd v Israel-British Bank (London) Ltd  Ch 105
7 Fibrosa Spolka Akcyjna v Fairbairn Lawson Combe Barbour Ltd  AC 32
8 A complete account of Professor Birks’s taxonomy should not fail to mention “ignorance”, where a benefit is taken from the claimant without his knowledge, and “helplessness”, which is the more extreme form of duress in which the claimant is incapable of preventing the defendant’s enrichment, eg by being tied to a chair: see Birks, Introduction, 140 ff
9 E.g. Louth v Diprose (1992) 175 CLR 621 (HC of Australia). It is arguable that some duress cases should equally be categorised as belonging into this category.
10 See Birks, “In Defence of Free Acceptance” in Burrows (ed.) Essays on the Law of Restitution (1991), 109
11 Woolwich Equitable Building Society v IRC  AC 70
12 Professor Birks had strongly argued that claimants in the Woolwich’s position should succeed on this ground: see Birks, “Restitution from the Executive: a Tercentenary Footnote to the Bill of Rights” in Finn, Essays on Restitution (1990) 161
13 Markesinis, Lorenz, Dannemann The Law of Contracts and Restitution (1997) 713
14 For many: Reuter, D and Martinek, M, Ungerechtfertigte Bereicherung (1983), 110
15 Kleinwort Benson v Lincoln London Borough Council  2 AC 349, 367
16 Hazell v Hammersmith and Fulham London Borough Council  2 AC 1
17 Westdeutsche Landesbank Girozentrale v Islington London Borough Council  4 All ER 890, QBD and CA, 924 (references will be to the All England Law Reports, as both first instance and Court of Appeal decisions are conveniently reported together there); varied  AC 669, HL
18 Kleinwort Benson v Lincoln London Borough Council  2 AC 349
19 Birks, “No Consideration: Restitution after Void Contracts”, (1993) 23 UWALR 195
20 Guinness Mahon & Co Ltd v Kensington and Chelsea Royal London BC  QB 215
21 Not surprisingly, the banks had been the overall winners of the swaps entered into with local authorities – Hazell turned them into the overall losers!
22 It is referred to in the bank’s skeleton argument in Westdeutsche and some of the other lead cases, but since in those cases the banks were asking for restitution, the argument was perceived to be a weak one.
23 The way in which the case reached the House of Lords was unusual. In the Commercial Court, Langley J made a consent order dismissing the claim but granting a certificate for an appeal directly to the House of Lords. In this way, their Lordships did not have the benefit of reasoned judgments below.
24 At 517 (Lord Browne-Wilkinson), 548 (Lord Lloyd)
25 At 548
27 Birks, “No Consideration: Restitution after Void Contracts” (1993) UWALR 195, 230, n 137
28 Ernst v Caemmerer, ‘Bereicherung und unerlaubte Handlung’ in Festschrift für Rabel, vol 1 (Tübingen 1954) 333, 343
29 Cross reference to illegality chapters
30 Cf. Medicus, Allgemeiner Teil des BGB, 7th ed., No. 648
31 But see BGHZ 111, 308, 312 ff.
32 This can lead to hard cases, which have indeed led to bad law: the courts have tried to temper the inflexibility of the rule in cases in which its application would lead to “simply unacceptable results”: OGHZ 1, 217; BGHZ 85, 315.
33 Cf Larenz, Lehrbuch des Schuldrechts, vol I, 14th ed., 73
34 BGH NJW 93, 1127, 1262
35 Cf Larenz/Canaris, Lehrbuch des Schuldrechts, vol II, 2, 13th ed., 5
36 Jauernig/Vollkommer, § 762, No. 1
37 For criticism of the term see Larenz, Schuldrecht I (n 31 above), 21
38 P Schlechtriem, Schuldrecht Besonderer Teil, 4th ed, 310
40 Zimmermann, “Unjustified Enrichment: The Modern Civilian Approach” (1995) 15 OJLS 403
41 Birks, ‘Restitution after Void Contracts’, 23 UWALR 195, 231
42 Woolwich Building Society v IRC (No 2)  AC 70, 172
43 This has been pointed out by Sonja Meier in her book on the English law of unjust enrichment “Irrtum und Zweckverfehlung” (Tübingen 1999), 99. She argues that English law here employs legal ground reasoning without realising it.
44 Miller v Race (1758) 1 Burr 452
45 See Merton and another v Hammond Suddards and another  2 BCLC 470. In that case, somewhat contrary to the argument taken in this paper, His Honour Judge Kolbert sitting as a High Court Judge said, in interpreting section 127: “I take the view that there is much to be said for a plain interpretation of a provision in plain language, especially as ‘void’ is a word of unmistakable meaning. Once an argument enters even marginally upon the proposition that ‘void’ bears any meaning other than ‘of no effect’, or the court is urged that ‘void’ really means ‘voidable’, that is, that a disposition takes effect until challenged and cancelled or avoided, the court should be very wary of accepting that argument.”
46 Although it may sometimes have the opposite effect, namely when assets have come into the hands of the liquidator and are claimed by an administrative receiver on behalf of a secured creditor: see Merton v Hammond Suddards, above.
47 See Birks, Introduction to the Law of Restitution, 308
49 Westdeutsche, at 915
51 Tootal Clothing Ltd v Guinea Properties Ltd 64 P & CR 452
52 Spiro v Glencrown Properties Ltd  Ch 537, 541
53 64 P & CR 452, 455
54 (1840) 11 A & E 438
55 Expressed in Lee v Muggeridge (1813) 5 Taunt 36, at 46, when he was still Sir James Mansfield CJ
56 Nicholson v Chapman (1793) 2 H Bl 254
57 Falcke v Scottish Imperial Insurance (1886) 34 Ch D 234
58 In German law, that belief might well have been right, because German law recognises claims based on negotiorum gestio: cf §§ 677 ff
59 Zweigert & Kötz, An Introduction to Comparative Law, (trans. Weir), 3rd edition, Oxford 1998, 70
60 Professor Zimmermann and Dr Meier go so far as to suggest that the only way forward for English law is now to resort to a legal ground analysis: see Zimmermann & Meier, “Judicial Development of the Law, Error Iuris, and the law of Unjustified Enrichment – a View from Germany”, (1999) 115 LQR 556
© 2000 T. Krebs. 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.